Citation
Alternating vs. direct current

Material Information

Title:
Alternating vs. direct current a transient study of the U.S. Coast Guard's 270 foot medium endurance cutter's electrical distribution system
Uncontrolled:
Transient study of the U.S. Coast Guard's 270 foot medium endurance cutter's electrical distribution system
Creator:
Hutton, Keoni Alexander ( author )
Language:
English
Physical Description:
1 electronic file (114 pages) : ;

Subjects

Subjects / Keywords:
Electric circuits -- Alternating currents ( lcsh )
Electric power distribution ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
While the United States Navy has conducted extensive research into the use of shipboard DC zonal electrical distribution systems (ZED), no project has analyzed the benefits for installation on a Coast Guard cutter which has a significantly different load profile than Navy warships. Simulink models of the existing 270 medium endurance cutter (WMEC) AC radial electrical distribution system and a proposed DC ZED system were created and tested with three transients. The result demonstrated a significant reduction in settling time and an increased robustness caused by the insulation provided by the introduction of power electronic converters. Beyond the transients, a DC ZED provides better standardized installation for any ship reducing construction costs and timelines, and simplifying training and support. Additionally, the DCZED increases a ship innate survivability by reducing longitudinal cables that penetrate watertight bulkheads increasing a boundaries effectiveness. The Coast Guard would be best served by pushing for a zonal distribution system on all future cutter acquisitions.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
by Keoni Alexander Hutton.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
953736572 ( OCLC )
ocn953736572
Classification:
LD1193.E53 2016m H97 ( lcc )

Downloads

This item has the following downloads:


Full Text
ALTERNATING VS. DIRECT CURRENT: A TRANSIENT STUDY OF THE U.S. COAST
GUARDS 270 FOOT MEDIUM ENDURANCE CUTTERS ELECTRICAL DISTRIBUTION
SYSTEM
by
KEONI ALEXANDER HUTTON
B.S., United States Coast Guard Academy, 2010
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Science
Electrical Engineering
2016


This thesis for the Master of Science degree by
Keoni Alexander Hutton
has been approved for the
Electrical Engineering Program
by
Jae-Do Park, Chair
Fernando Mancilla-David
Miloje Radenkovic


The views expressed herein are those of the author and are not to be construed as official or
reflecting the views of the Commandant or of the U. S. Coast Guard.


Hutton, Keoni Alexander (M.S., Electrical Engineering)
Alternating vs. Direct Current: A Transient Study of the U.S. Coast Guards 270 Foot Medium
Endurance Cutters Electrical Distribution System
Thesis directed by Assistant Professor Jae-Do Park.
ABSTRACT
While the United States Navy has conducted extensive research into the use of
shipboard DC zonal electrical distribution systems (ZED), no project has analyzed the benefits
for installation on a Coast Guard cutter which has a significantly different load profile than Navy
warships. Simulink models of the existing 270 medium endurance cutter (WMEC) AC radial
electrical distribution system and a proposed DC ZED system were created and tested with
three transients. The result demonstrated a significant reduction in settling time and an
increased robustness caused by the insulation provided by the introduction of power electronic
converters. Beyond the transients, a DC ZED provides better standardized installation for any
ship reducing construction costs and timelines, and simplifying training and support.
Additionally, the DCZED increases a ships innate survivability by reducing longitudinal cables
that penetrate watertight bulkheads increasing a boundaries effectiveness. The Coast Guard
would be best served by pushing for a zonal distribution system on all future cutter acquisitions.
The form and content of this abstract are approved. I recommend its publication
Approved: Jae-Do Park
IV


ACKNOWLEDGEM ENTS
The author would like to acknowledge the office of Naval Engineering CG-45, the staff at
the Office of Mission Support Workforce Management DCMS-81, and the staff at the Denver,
Colorado Coast Guard Recruiting Office for their assistance and support: LCDR Roger
Robitaille, LT Lucas Marino, BMC Stanley Rittner, MK1 Krista Beck, AET1 Charles McKenzie,
MST2 Whip Blacklaw, and Ms. Mary Fuata.
The author would also like to express gratitude to the staff and faculty of the University
of Colorado Denver College of Engineering and Applied Science Electrical Engineering
Department especially the committee members Dr. Jae Do Park, Dr. Fernando Mancilla-David,
and Dr. Miloje Radenkovic as well as the University of Colorado Denver Alumni Association.
Finally the author would like to express gratitude to his family and friends, especially his
wife Ashley, without whose support, this research could never have been completed.
v


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION............................................................ 1
Background.................................................................1
Scope of Research..........................................................4
Objectives.................................................................7
Organization of Thesis.....................................................7
II. AC DISTRIBUTION ARCHITECTURE AND MODELING................................8
270 WMEC Design Requirements..............................................8
ACDS Simulink Model.......................................................12
Simulink Simulation Using the ACDS Model..................................20
III. DC DISTRIBUTION ARCHITECTURE AND MODELING...............................23
DCDS Design...............................................................23
Modeling of DCDS..........................................................24
Simulink Simulation Using the DCDS Model..................................32
IV. SIMULATION RESULTS..................................................... 33
Trial 1: Loss of a Generator While Underway...............................33
Trial 2: Starting a Large Load While Underway.............................37
Trial 3: Battle Damage While Underway.....................................40
Trial 4: Loss of a Generator During General Quarters......................43
VI


Trial 5: Starting a Large Load During General Quarters.....................46
Trial 6: Battle Damage During General Quarters.............................50
Trial 7: Loss of a Generator While at Anchor...............................53
Trial 8: Starting a Large Load While at Anchor..............................56
Trial 9: Battle Damage While at Anchor......................................59
Simulation Results Analysis.................................................63
V. DISTRIBUTION SYSTEM COMPARISON..............................................65
Weight Benefits.............................................................65
Standardization Benefits....................................................67
Survivability Benefits......................................................69
VI. CONCLUSIONS AND RECOMMENDATIONS.............................................72
REFERENCES........................................................................74
APPENDIX
A. 270 WMEC Cable Calculations............................................... 78
B. Transmission Line Calculations..............................................79
C. DC ZED Load Distribution Calculations.......................................80
D. DC ZED Transmission Line Calculations.......................................87
E. MATLABCode..................................................................88
vii


LIST OF TABLES
TABLE
1. Statutory missions of the U.S. Coast Guard....................................... 1
2. Electric power system characteristics at the user interface.......................2
3. U.S. Coast Guard 270 WMEC principle characteristics..............................6
4. Model consolidated load specifications...........................................17
5. Loads eliminated from model consideration........................................18
6. Total power (kW) during each plant configuration.................................21
7. Proposed zonal load breakdown....................................................25
8. Trial 1 ACDS and DCDS comparison summary....................................36
9. Trial 2 ACDS and DCDS comparison summary....................................39
10. Trial 3 ACDS and DCDS comparison summary....................................42
11. Trial 4 ACDS and DCDS comparison summary....................................46
12. Trial 5 ACDS and DCDS comparison summary....................................49
13. Trial 6 ACDS and DCDS comparison summary....................................52
14. Trial 7 ACDS and DCDS comparison summary....................................56
15. Trial 8 ACDS and DCDS comparison summary....................................59
16. Trial 9 ACDS and DCDS comparison summary....................................62
17. DDG-51 weight comparison between radial and zonal architectures................66
viii


LIST OF FIGURES
FIGURE
1. Comparison of zonal and radial distribution architecture............................4
2. 270 WMEC ACDS simplified line diagram..............................................6
3. Ashore and shipboard receptacle potential...........................................9
4. Realistic shipboard ungrounded system..............................................10
5. Simple galvanic cell in a lead-acid battery........................................11
6. Three phase equivalent circuit model of synchronous generator......................14
7. Type DC1A-DC commutator exciter block diagram......................................15
8. Simulink implementation of 270 WMEC synchronous generator.........................16
9. Simulink implementation of load....................................................18
10. Transmission line equivalent circuit..............................................18
11. Simulink implementation of transmission line......................................20
12. 270 WMEC proposed DC Zonal Electrical Distribution System........................24
13. VSC-based rectifier for DC shipboard power system application.....................26
14. Voltage source converter controller block diagram.................................29
15. Simulink implementation of DCDS loads (Underway Steaming/Anchor configuration)...31
16. Simulink implementation of DCDS loads (General Quarters configuration)............32
17. Voltage and current for generator 1S and generator 2S.............................34
18. Generator 1S voltage before and after the transient introduction for ACDS.........34
IX


19. Generator 2S voltage before and after the transient introduction for DCDS......35
20. Trial 1 Voltage measured at selected loads for ACDS...........35
21. Trial 1 Voltage measured at selected loads for DCDS...........36
22. Voltage and current measured at selected loads during Trial 2..................37
23. Trial 2 Generator 1S and 2S voltage before and after transient for ACDS...........37
24. Trial 2 Generator 1S and 2S voltage before and after transient for DCDS...........38
25. Trial 2 voltage measured at selected loads for ACDS...........38
26. Trial 2 voltage measured at selected loads for DCDS...........39
27. Voltage and current measured at selected loads during Trial 3..................40
28. Trial 3 Generator 1S and 2S voltage before and after transient for ACDS.......40
29. Trial 3 Generator 1S and 2S voltage before and after transient for DCDS.......41
30. Trial 3 voltage measured at selected loads for ACDS...........41
31. Trial 3 voltage measured at selected loads for DCDS...........42
32. Voltage and current for generator 1S, generator 2S, and generator 1E during Trial 4... 43
33. Trial 4 Generator 1S and 2S and 1E voltage before and after transient for ACDS.44
34. Trial 4 Generator 1S and 2S and 1E voltage before and after transient for DCDS.44
35. Trial 4 voltage measured at selected loads for ACDS...........45
36. Trial 4 voltage measured at selected loads for DCDS...........45
37. Voltage and current measured at selected loads during Trial 5..................47
38. Trial 5 generator 1S and 2S and 1E voltage before and after transient for ACDS.47
x


39. Trial 5 generator 1S and 2S and 1E voltage before and after transient for DCDS.48
40. Trial 5 voltage measured at selected loads for ACDS.........................48
41. Trial 5 voltage measured at selected loads for DCDS.........................49
42. Voltage and current measured at selected loads during Trial 5..................50
43. Trial 6 generator 1S, 2S, and 1E voltage before and after transient for ACDS.50
44. Trial 6 generator 1S, 2S, and 1E voltage before and after transient for DCDS.51
45. Trial 6 voltage measured at selected loads for ACDS.........................51
46. Trial 6 voltage measured at selected loads for DCDS.........................52
47. Voltage and current for generator 1S, and generator 1E during Trial 7..........53
48. Voltage at generator 1E for ACDS during startup..............................54
49. Voltage at generator 1E for DCDS during startup..............................54
50. Trial 7 voltage measured at selected loads for ACDS.........................55
51. Trial 7 voltage measured at selected loads for ACDS.........................55
52. Voltage and current measured at selected loads during Trial 8..................56
53. Generator 1S voltage before and after transient for ACDS.....................57
54. Generator 1S voltage before and after transient for DCDS.....................57
55. Trial 8 voltage measured at selected loads for ACDS.........................58
56. Trial 8 voltage measured at selected loads for ACDS.........................58
57. Voltage and current measured at selected loads during Trial 9..................60
58. Generator 1S voltage before and after transient for ACDS.......................60


59. Generator 1S voltage before and after transient for DCDS.................61
60. Trial 9 voltage measured at selected loads for ACDS......................61
61. Trial 9 voltage measured at selected loads for DCDS......................62
62. Advantage of zonal construction..........................................57
63. Damage to USS Cole.......................................................58


CHAPTER I
INTRODUCTION
Background
The United States Coast Guard was founded as the Revenue Marine on 4
August 1790 with the explicit purpose of enforcing tariff and trade laws to finance the
creation of a new nation [1], From its humble origins with a fleet of ten cutters, the Coast
Guards missions and responsibilities grew as the organization merged with the U.S. Life
Saving Service, the Lighthouse Service, the Steamboat Inspection Service, and the
Bureau of Marine Inspection and Navigation [1], Today the Coast Guard operates a fleet
of 90 cutters and patrol craft in addition to over 1800 boats [2] in support of the eleven
statutory missions displayed in Table 1:
Table 1: Statutory missions of the U.S. Coast Guard [3],
Ports, Waterways and Coastal Security Drug Interdiction
Aids to Navigation Search and Rescue
Living Marine Resources Marine Safety
Defense Readiness Migrant Interdiction
Marine Environmental Protection Ice Operations
Other Law Enforcement
As the Coast Guards statutory responsibilities have expanded, so have the
requirements for its cutter fleet. Todays cutters feature sophisticated command, control,
communications, computers, intelligence, surveillance and reconnaissance (C4ISR)
systems and weapons systems requiring seamless interoperability with other
Department of Homeland Security (DHS) and Department of Defense (DOD) assets [4],
These C4ISR systems when coupled with the ever expanding use of technology for
personal and professional matters are driving up the electrical load requirements for
Coast Guard cutters. The U.S. Naval Sea Services Command describes the importance
of electric power in Chapter 320 of the Naval Ships Technical Manual:
1


Electric power is essential to a modern naval ships fighting and functional
effectiveness. Electric power trains [and] elevate[s] gun turrets and missile
launchers; operate[s] the rudders hydraulic system; run[s] auxiliaries; provide[s]
light; and power[s] interior communication, weapons control, radio, radar, sonar,
and missile systems. A ship without electric power is useless as a fighting or
supporting unit and is almost totally defenseless against enemy attack [6],
The electric distribution system on board a ship must be able to provide six basic
functions:
1. Control: The functional element of the power system that coordinates the
other functional elements. [7][8]
2. Power Generation: Converts fuel into electrical power. In the Coast Guard,
this is primarily accomplished with a generator utilizing either a diesel engine
or gas turbine as the prime mover. [7][8]
3. Power Distribution: The ability of the system to move the power generated to
the functional loads that will utilize the power. The distribution system
consists of cables, switches, and fault protection equipment. [7][8]
4. Power Conversion: The ability of the system to convert the power generated
into an acceptable form for each load. Three types of shipboard power are
defined by the NSTM 320: [6][7][8]
Table 2: Electric power system characteristics at the user interface [6],
Power Type Type 1 Type II Type III
Nominal Frequency (Hz) 60 Hz 400 Hz 400 Hz
Frequency Tolerance 3% 5% 1/2%
Freq Recovery Time 2 seconds 2 seconds .25 seconds
Nominal Voltage (V) 440 V or 115 V 440 V or 115 V 440 V or 115 V
Voltage Tolerance 5% 5% 2%
V Recovery Time 2 seconds 2 seconds .25 seconds
5. Energy Storage: The ability of a system to store energy that is generated to
be used during a loss of primary generation capability. [7][8],
6. Utilization: The ultimate purpose of the distribution system, the delivery of the
generated power to the user. Typical loads consist of lighting,
2


communications and navigation gear, weapons systems, motors for pumps,
galley equipment and personal use by the crew. [7][8]
These basic functions apply to all ships from the tugboat in the harbor to the cruise ship
sailing around the world.
The cutters of the U.S. Coast Guard, however, are expected to continue
performing in extremely hazardous conditions up to and including fighting a war.
Therefore, the cutters must be designed to continue to operate even after taking heavy
damage [7], To this end, the loads on a cutter are characterized based on their
importance to the cutters ability to perform its mission. These loads are broken down
into vital, semi-vital and non-vital categories [7], Vital loads will be connected to two
sources of power through an automatic bus transfer (ABT) switch. The ABT will
automatically switch the power source feeding the load when the primary power source
is lost. If a vital load is not considered particularly vital, it will still be fed with two power
sources, but the bus transfer switch will be operated manually [7], Semi-vital and non-
vital loads are fed by one power source and are only characterized differently for
simplicity in the event load shedding becomes necessary [7], The ultimate goal being to
seamlessly provide alternate paths to these vital loads to ensure the ships survivability
[9].
Two major architecture categories have emerged to meet these survivability
requirements: conventional, or radial, architecture and a zonal architecture. Today,
every U.S. Coast Guard cutter employs a radial ACDS architecture where a generator
provides 450 VAC 60 Hz Type I power directly to a ships service switchboard which
then distributes the power to a load center which distributes the power to the associated
equipment. The ships service switchboards can be fed by its own generator, or through
the connection of bus ties, by any other generator [9][10], The Type I power provided by
the generators are also used to run a motor-generator set that produces 400 Hz Type II
3


or Type III power. Conversely, the zonal architecture consists of two power distribution
busses that run the length of the ship. The two busses should be separated port and
starboard and vertically as far as possible to provide maximum survivability [10], Each
generator is connected directly to the main bus and feeds its power to the bus. The ship
is then broken down into zones utilizing existing water tight boundaries. Each zone
would contain two load centers to feed the associated equipment in that zone only [10],
Figure 1 provides a comparison of a radial and zonal distribution architecture for a
typical ship.
Figure 1: Comparison of zonal and radial distribution architecture [10],
Scope of Research
Over the last 25 years, the U.S. Navy has conducted multiple studies into the
benefits associated with a Direct Current Zonal Electrical Distribution System (DCZEDS)
for the next generation of war fighting ships and detailed numerous benefits that will be
4


expounded upon in Chapter 5 of this Thesis. However, no U.S. Coast Guard project has
examined the implications of this technology specifically for Coast Guard use.
In the late 90s, the Coast Guard embarked on a $21.1 billion acquisition program
designed to recapitalize aging cutters and prepare the surface fleet for continued
operations in the twenty first century [2][4], As of January 2015, the Coast Guard
received delivery of four National Security Cutters, with four more in development and
ten Fast Response Cutters with 24 more in development [4], The requirements for these
vessels are well defined and locked in to their respective acquisition contracts.
Additionally, the Coast Guard is planning for the acquisition of up to 25 Offshore Patrol
Cutters (OPC). A preliminary design contract was awarded in February 2014 to three
different vendors to design the OPC. The Coast Guards Acquisition Directorate will
then evaluate the proposed designs and award a final contract to one vendor to
construct the vessels [11], The requirements for the OPC are still generally defined and
are being interpreted differently by each of the preliminary design contractors; thus, the
OPC program could benefit most from an analysis of potential electrical distribution
architectures.
The OPC is designed to replace the Coast Guards twenty nine 210 and 270
Medium Endurance Cutters (WMEC) and accomplish the same mission set. This Thesis
will examine the existing 270 WMEC electrical distribution system, its missions, and
design requirements and use it as a case study for a hypothetical DCZED allowing for a
direct comparison between the system architectures based on the actual characteristics
of a well-defined, actively serving system.
The Coast Guards 270 WMEC fleet consists of 13 cutters built between 1979
and 1989. The 270s principle characteristics are shown in Table 3.
5


Table 3: U.S. Coast Guard 270 WMEC principle characteristics [5],
Specifications
Builder WMEC 901-904. Tacoma Boatbuilding Co.
WMEC 905-913. R E Derecktor of Rhode Island Inc. Middletown. R.l.
Length 270 feet overall
255 feet at Waterline
Beam 38 Feet
Draft 14 feet
Displacement 1.780 Tons
Propulsion Twin V-18 ALCO diesel engines each delivering 3.600 shaft horsepower.
twin reduction gears, twin shafts and rudders,
twin 9 foot diameter controllable pitch propellers
Capacity 89.000 gallons which allows an effective range of 10.250 nautical miles
83.000 gallons Diesel Oil
20.668 gallons JP-5 Fuel
8.248 gallons Fresh Water
Maximum Range 9900-10.250 nautical miles
Maximum Speed 19.5-20 knots
Primary Missions Law Enforcement Defense Operations. Search & Rescue
Typical Crew 100 Personnel (14 Officers. 86 Enlisted)
Armament 1 MK 75mm rapid fire gun
2 50 cal machine guns
2 MK36 Super RBOC chaff launchers
Aircraft: One HH-65Aor
one LAMPS I helicopter or
one HH-60J helicopter
Boats One 23' Over the Horizon MKllI
One 20" Rigid Hull Inflatable
Systems MK-92 Gunfire Control System
Dual surface search radars
Identification Friend or Foe OFF) transponder and interrogator
Tactical Air Navigation (TACAN)
AN/SLQ-32 Electronic Surveillance Receiver
Infrared and low light camera and video recording equipment
Satellite voice and data communications.
MF. HF. VHP. and UHF voice and data communications.
surface to surface, surface to airvoice and data links
Computer driven system using ship's systems such as RADAR. SATNAV. LORAN. OMEGA, and dead reckoning inputs to plot and track ship's position at all times and plot and track the positions of up to sixty- four surface and airtaraet
The 270 WMEC electrical plant consists of two Ship Service Diesel Generators and an
Emergency Diesel Generator connected to an AC radial distribution system (ACDS) that
feeds 51 load centers throughout the ship [12],
Figure 2: 270 WMEC ACDS simplified line diagram.
6


Objectives
The objectives of this Thesis are as follows:
1. Examine the existing electrical distribution system of the 270 Famous Class
Medium Endurance Cutter and outline the current load requirements.
2. Utilize Matlab-Simulink modeling software to create a virtual model of the
Famous Class Cutter's ACDS. Simulate common casualties in a variety of plant
configurations to validate the model against data drawn from the cutters history
and the requirements outlined by Coast Guard policy.
3. Design a DC ZED to meet the requirements outlined by the current load
configuration and future requirements outlined in the NSC and OPC acquisition
documents. Simulate the same casualties in the same plant configurations as the
ACDS simulations.
4. Compare and evaluate the DC ZED model against the model for the existing AC
configuration using the following criteria:
a. Transient Response
b. Weight
c. Standardization
d. Survivability
Organization of Thesis
Chapter 2 will outline the existing AC distribution architecture of the 270 WMEC
and the design requirements that were met during its construction. It will then discuss
building the model of the ACDS and the development of the simulation trials that both
models were tested under. Chapter 3 will outline the design requirements for the DC
ZED and the Simulink model construction. Finally, it will discuss the simulation trials and
problems encountered. Chapter 4 will examine the simulation results for the transient
study of the ACDS and DCDS trials. Chapter 5 will examine and compare the weight,
survivability and standardization potential of the DCZED system. Chapter 6 outlines the
analysis conclusions and recommendations for the OPC acquisition and future work.
7


CHAPTER II
AC DISTRIBUTION ARCHITECTURE AND MODELING
270 WMEC Design Requirements
As mentioned previously, Coast Guard vessels are designed to continue to
operate despite faults, failures and damage. In order to accomplish this feat, Coast
Guard vessel design is governed by numerous publications produced by the Department
of Homeland Security and Department of Defense. The electric plant of a military ship
must meet the requirements set forth in the Naval Ships Technical Manual Chapter 300,
Electric Plant-General, Chapter 310 Electric Power Generators and Conversion and
Chapter 320 Electric Power Distribution Systems. Those requirements will be discussed
in depth in this chapter.
Floating Ground
Coast Guard electrical distribution systems are ungrounded systems in that there
are no neutral or phase conductors intentionally connected to ground. On board a ship,
the ground is defined as the potential measured at the ships hull. An ungrounded
system will provide limited current during a single phase to ground fault thus allowing
equipment to continue to operate throughout the fault condition [13], This is most easily
seen at any 115 V receptacle where both prongs have a potential of approximately 60
Vrms relative to ground as seen in Figure 3.
HOME SHIP
UFIJTRAI * HOT
Figure 3: Ashore and shipboard receptacle potential [13],
8


A shipboard ungrounded system will contain stray resistance and capacitance
relative to ground caused by electrical equipment and cables in the system. Inherent
system resistance is the parallel sum of the inherent resistance caused by insulation
around generator, cable and load components. Similarly, the inherent system
capacitance is the parallel sum of the inherent capacitance of generators, cables and
loads relative to ground [13], While these are not physical components, the inherent
system resistance and capacitances can create a return path between the ground and
the conductor if a sailor comes into contact with a live conductor as in Figure 4.
Beyond the survivability benefits of an ungrounded system, avoiding ground
currents helps to prevent and minimize electrolytic corrosion of the ships hull [13][14],
Corrosion in its most basic form is the destructive alteration of metal by reaction with its
environment [14], In order for any type of corrosion to occur, four components must be
present to create a galvanic cell: an anode, a cathode, a metallic path that allows
electron flow, and an electrolytic path that allows ionic flow [14], Figure 5 illustrates the
galvanic cell as a basic lead-acid battery:
9


Figure 5: Simple galvanic cell in a lead-acid battery.
On board the WMEC, the electrolyte is created by the sea water upon which the
ship sails. The anode and cathode can be the ships hull and any underwater fitting
(propellers, rudders, transducer etc.). The only missing component is the electron flow
path. When any wire of the ships electrical system is grounded to the hull, the galvanic
cell is complete and electrolytic corrosion can occur. The resistance of this short-circuit
path is often high enough to restrict the current to a level insufficient to activate circuit
breakers...so the stray current condition can persist for some time [14], In this case,
the ground must be found and isolated as soon as possible.
Cable Insulation
In order to reduce the shock and grounding risk associated with the ungrounded
system, all cables must be properly insulated based on compatibility with other insulators
and the operating environment of the cable or equipment [13], Cables approved for use
on board Coast Guard cutters are listed in Military Standard 242J [31], The cables
10


utilized on board the 270 WMEC along with their characteristics are outlined in
Appendix A.
Power Generation
Two types of power are generated on board the Coast Guards 270 WMEC:
Type I and Type II. Type I power is produced by a 450-volt 60-hertz three phase
machine with either a Y-connected or delta-connected stator [25], The generator is
excited by a direct current revolving field type rotor with a salient pole. The DC current is
created through a rotating rectifier connected to the exciter armature [26], The prime
mover is directly connected to the generator rotor. Three generator parameters must be
controlled: frequency, voltage and load division [25]:
Frequency: Frequency depends on the speed of the prime mover and is
controlled by a mechanical hydraulic governor that is set by the generator watch
stander through an adjustment potentiometer.
Voltage: Voltage is controlled through a voltage regulator. The voltage regulator
will maintain the generators terminal voltage and ensure it does not exceed a
preset value during operation or load changes.
Load Division: The division of the electrical load between the generators
operating in parallel is controlled by the prime movers governor.
These characteristics are constantly monitored by watch standers through voltage,
current, power, frequency and power factor meters located near the ships service
switchboards and the prime movers.
Generator Protection
Overcurrent protection is provided through a circuit breaker connecting the
generator to its service switchboard. Significant damage can occur when generators are
11


operated in parallel if one generator takes the load and forces the other into motoring
mode. To prevent this occurrence, the generator circuit breaker is equipped with
reverse power protection which trips the generator breaker if it senses a reversal of the
power flow [25],
Load Characteristics
The electric load of the 270-WMEC consists of a wide variety of equipment that is
used (or not) during a wide variety of situations. The largest loads on the distribution
system consist of purely resistive loads such as heaters, and highly inductive loads such
as the motor-generator sets that produce 400 Hz Type II power [20], The power factor
of the system is tightly controlled by the generator watch stander to maintain a .8 lagging
power factor. Thus, the overall system tends to act as an inductive load.
ACDS Simulink Model
In this thesis, the SimPower Systems toolbox of the MATLAB-Simulink software
suite has been utilized to construct the model and run the simulations.
Synchronous Generator Electrical Model [15]
For ease of analysis, each generator on the 270 WMEC is assumed to be a 2-
pole, perfectly round rotor, synchronous machine driven by a diesel engine. The
electrical frequency produced by a synchronous machine is related to the mechanical
rotational speed of the rotor as seen in (1):
f = (1)
7e 120 U7
Where nm denotes the mechanical speed of the rotor, and P is the number of poles in
the machine. Thus, to produce the standard 60 Hz frequency, the generator must spin at
a constant 3600 RPM. The mechanical speed is controlled by a governor that is
12


manually controlled by the generator watch stander. The internal phase voltage
produced by the generator is given by (2):
Ea = V2tiNc where Nc represents the number of turns of wire in the stator coil, 0 is the flux in the
machine and f is the frequency of rotation.
The terminal phase voltage of the generator V

voltage Ea due to the following factors [15]:
1. The armature reaction in the air-gap magnetic field
2. The self-inductance of the armature coils
3. The resistance of the armature coils
The armature reaction induces a second voltage in the stator coils 90 behind and
directly proportional to the current Ia- Thus, the armature reaction can be modeled as an
inductor in series with the voltage Ea. The self-inductance and resistance of the armature
coils are similarly proportional to the armature current and can be modeled by an
inductor and resistor in series with the armature reaction. The 3-phase equivalent circuit
of the synchronous generator is thus given in Figure 6:
Figure 6: Three Phase equivalent circuit model of synchronous generator.
13


The voltage of each phase can then be calculated using (3):
V

(3)
The output active and reactive power measured at the terminals of the generator
can be expressed as in (4) and (5) respectively:
The output power of the generator is directly related to the input power provided to the
generator by the diesel engine prime mover, but mechanical and electrical losses must
be accounted for. The input mechanical power is calculated by (6):
From this power, the machine suffers friction and windage losses, core losses and other
stray losses that reduces the mechanical power delivered to the generator. After the
mechanical power is converted, the generator windings will produce additional copper
losses that further reduce the output power calculated as in (4) [15],
Generator Excitation [19]
The excitation system for the 270 WMEC generators is not well known due to
the age and lack of publicly available documentation. The DC1A excitation model is
utilized to represent a field-controlled dc commutator exciter with a continuously acting
voltage regulator thanks to its wide implementation throughout the Electrical Engineering
industry and simplicity of use, and is modeled by the following block diagram:
Pout = V:3VtIlcos6z = 3 V^IAcosOz
Qout 'j3VTILsin9z 31^ylj^sindz
(5)
(4)
(6)
V
Figure 7: Type DC1A-DC commutator exciter block diagram [19],
14


The terminal voltage transducer and load compensator output (Vc) is subtracted from a
reference voltage (Vref) while the stabilizing feedback (Vf) is subtracted and the system
stabilizing signal (Vs) is added to create an error signal. The major time constant (Ta)
and gain (Ka) incorporate limits representing saturation or power supply limitations. The
regulator output, (Vr) controls the exciter and allows for a self-excited shunt field when Ke
= 0, which accounts for the shunt field rheostat setting, and is automatically calculated
by MATLAB. Vx accounts for saturation in the exciter.
The complete generator model implemented in Simulink can be seen in Figure 8.
Figure 8: Simulink implementation of 270 WMEC synchronous generator.
Load Model
The 270 WMEC electrical distribution system feeds 51 load centers directly from
the ships service switchboards [20], In this thesis, the load centers were characterized
as either purely resistive or as RL loads:
1. Purely resistive loads. These loads consist primarily of lighting and heating
circuits that contain little or no inductive responses.
2. RL loads. This category contains the majority of the loads on board the 270
WMEC. It contains rotating equipment like the pumps for hydraulic,
lubrication, and damage control systems.
The Coast Guards Engineering Logistics Center commissioned a load analysis of the
270 WMEC in 2007 that analyzed the load draw during in port and underway conditions
15


[20], These measurements formed the basis of the analysis. In this work however, the
data set is incomplete in that it only contains the measured active power for each load
center. Reactive power on board USCG ships is tightly controlled by the Generator
Watch Stander to maintain power factor = .8 Lagging. As such, all reactive powers in
the model is assumed using the following relations:
P = V3 IVcosO (11)
Q = ^JSIVsinO (12)
PF = cosd (13)
Solving (11) for V3/F and then plugging the result into (12) yields:
Q = ^sin(acos(PF)) (14)
Modeling all 51 load centers created an unpalatable Simulink model that took
weeks to run to completion. As such, the load centers were combined into a single RL
load based on the switchboard feeding it, and the geographic location within the ship.
The nine resulting loads are summarized in Table 4.
Table 4: Model consolidated load specifications [20],
Generator Location (VIax Load (KW) Reactive Load (KVAR)
xs Forward 202. X X5X.575
xs Amidships X30.S 98.X
xs Aft 3X2.7 234.525
2S Forward 66.7 50.025
2S Amidships X57.8 XX8.35
2S Aft X70 X27.5
XE Forward XXO.6 82.95
XE Amidships 83.9 62.925
XE Aft 5X.2 38.4
Total Load X285.8 964.35
Installed Capacity 1425 X068.75
At no time would the ship actually be running at maximum capacity. The numbers in
Table 4 represent the maximum possible load that could be reached. Additionally, five
loads were eliminated from the study due to their infrequent use. The loads that were
eliminated are shown in Table 5.
16


Table 5: Loads eliminated from model consideration.
Load Reason for Elimination
Capstan Only used during mooring evolutions
Anchor Windlass Only used during mooring evolutions
'Anchor Windlass 2 Only used during mooring evolutions
jlMRS Heater No 3 Only used during winter months, purely resistive
OTHB Davit Control Station Only used for boat launch and recovery.
Every load was fitted with a circuit breaker that would allow it to be turned on or off
depending on the parameters of the simulation. The load model created for the Simulink
model can be seen in Figure 9. Additional resistors were placed in parallel with the RL
loads to eliminate errors associated with having an inductor and a synchronous machine
connected in series (through the ground).
Figure 9: Simulink implementation of load.
Line Model
The transmission line model used in this thesis comes from Bergen and Vittals
Power System Analysis [21], The model uses three elements to capture the dynamics of
a transmission line. A resistor captures the heat caused by current circulation. An
inductor models the magnetic field. The shunt capacitive effect caused by the potential
relative to the ground beneath the transmission line is modeled as a capacitor. All
distributed parameters were drawn from the data sheets for cables as identified in
Appendix A. Figure 10 shows the per phase equivalent circuit of a transmission line.
17


Examining a section of the entire transmission line of length dx, and applying Kirchhoffs
Voltage Law and Current Law yields the following equations:
(15)
(16)
Solving (15) and (16) yields:
dV
dx~ Z
dl T7
i;=yv
V = V2 cosh(yx) + Zcl2 sinh(yx)
/ = I2 cosh(yx) + sinh(yx)
(17)
(18)
Where Zc represents the characteristic impedance of the line and y is the propagation
constant:
(19)
ii N I (20)
As the ship is only 270 long, this is the longest any transmission line could be. As such,
all shunt admittances were neglected [21], Additionally, the resistive elements of the
transmission line would only increase the active power used and is captured within the
load model. Transmission line lengths were calculated using the following assumptions:
18


1. The ships service diesel generator terminals were located at frame 103 on deck
three.
2. The emergency diesel generator terminals are located at frame 197 on deck
three.
3. The fore/aft length of a transmission line is the difference between the frames
between the power source and the load center.
4. The athwartships length of a transmission line is 9.5 feet each if the source/load
is port or starboard of the centerline or 19 feet if the load/source is on the
centerline.
5. There are 8 feet vertically between decks.
6. The distance of each load in a geographic region of a ship were averaged
together to determine the distance.
7. All cables feeding a composite load are assumed to be TSGU-30, to provide the
greatest dynamic effects.
Figure 11 shows the final transmission line model utilized for the Simulink model:
Three-Phase
Parallel RLC Branchl
Figure 11: Simulink implementation of transmission line.
Detailed calculations used to determine the transmission line values can be found in
Appendix B.
19


Simulink Simulation Using the ACDS Model
The ACDS model was tested against three common transients in three plant
configurations.
Plant Configurations
Plant configurations are governed by the Coast Guards Naval Engineering
Manual and unit level standing orders. The Engineering Department on board the
USCGC MOHAWK (WMEC-913) provided valuable insight to determine the
requirements for the plant configurations to test:
1. Underway Steaming-This is the most common plant configuration when the
ship is underway. Both ships service diesel generators are running in parallel
and both bus ties are closed so the two generators are sharing the load. The
Emergency Diesel Generator is placed in standby and will start automatically if it
senses a loss of power at its switchboard [23][24],
2. General Quarters-This plant configuration is set during high risk situations
such as operating near shore or during battle conditions. All three generators are
online and both bus ties are open so each generator only supplies the loads on
its own switchboard. This ensures maximum survivability in case of damage
[2 3] [24],
3. Anchor-This plant configuration is set during low risk situations such as
general steaming at night when the load is low, or while at anchor. Only one
generator, either the 1S or2S, is online and supplying all the electrical demand.
The offline ships service generator is typically cold iron, meaning it would take
several minutes before it could supply power to a load. The Emergency Diesel
20


Generator is placed in standby and will start automatically if it senses a loss of
power at its switchboard [23][24],
Table 6 summarizes the load during each of these plant configurations in kW:
Table 6: Total power (kW) during each plant configuration [20][24],
Underway Steaming General Quarters Anchor
Generator IS 148.805 372.31 57.29
Generator 2S 136.27 202.26 92.28
Generator IE 42.3 166.6 14.77
Total Power 327.375 741.17 164.34
Transient Sources
The transients chosen to test against are either common for the system, or
notably severe.
1. Loss of Generator: This transient is tested by allowing the system to reach a
steady state, and then opening a generator circuit breaker, disconnecting it from the
distribution network. The loss of a generator could occur for reasons as complex as,
mechanical failure or as simple as a mistake by a watch stander. During this type of
event, if another generator is online, it would take the load. If another generator is not
already online, the emergency generator would start up to restore power.
2. Starting a Load: The electrical load on the 270 WMEC changes constantly
throughout the day as machinery is brought online and taken off line in the normal
course of duty. This transient is tested by allowing the system to reach a steady state
and then closing the circuit breaker to the aft 2S switchboard loads causing a sudden
increase of 30-40 kW.
3. Battle Damage. Potentially the most disastrous transient the electric
distribution system could face. This transient is tested by allowing the system to reach a
21


steady state and then opening the circuit breakers for all the forward loads causing a
sudden drop in the load between 30-300 kW (depending on configuration). This analysis
seeks only to examine the transients caused by this sudden shift in the load, and as
such, the operation of automatic bus transfers and/or system reconfiguration to restore
power go beyond the scope of this analysis and are not considered.
22


CHAPTER III
DC DISTRIBUTION ARCHITECTURE AND MODELING
DCDS Design
The zonal electric distribution system in this study was constructed within the
following constraints that were used in a similar U.S. Navy study [10]:
1. Electrical zones are delineated based on existing watertight boundaries on the
270 WMEC. Where zones between watertight bulkheads had a load smaller
than 50 kW, they were combined with adjacent zones to create a larger load.
2. The maximum electrical load within a zone should not exceed the trip coil
setting of the largest circuit breaker currently available within government stock.
3. The electrical load of a zone should permit the use of standardized distribution
equipment.
In order to create a direct comparison between the distribution architectures, the
following assumptions were used:
1. Generation equipment (prime movers, generators, controls, and switchboards)
remain in their original locations on the vessel.
2. Distribution feeders after the zonal load center remain as originally
constructed.
3. Shore power and casualty power systems are not affected.
4. Vital loads are fed from alternate buses within the same zone but bus transfer
equipment is the same as the ACDS model.
Utilizing these assumptions and rules as outlined produces a distribution system
architecture as seen in figure 12:
23


Figure 12: 270 WMEC proposed DC Zonal Electrical Distribution System
The total load (expressed in kW) found in each zone is outlined in Table 7. Full
calculations can be found in Appendix C.
Table 7: Proposed zonal load breakdown
Zone Load
1 71 .3
2 306.3
3 94.8
4 255.7
5 309.7
6 56.4
7 94.4
Modeling of DCDS
Generator Model
The ultimate purpose of this research is to provide a direct comparison between
the existing AC distribution system currently employed on the 270 WMEC and the
hypothetical DC distribution system that is designed under the assumptions and
constraints listed in Chapter 3.1. To this end, the ACDS model outlined in Chapter 2.2
and 2.3 is kept intact as much as possible for use in the DCDS model. The Generator
24


model is taken directly from the ACDS Simulink model and only the addition of power
electronic converters between each generator and the DC bus changes the nature of the
distribution system.
Voltage Source Converter Model [27][30]
Multiple strategies exist to convert the AC voltage/current produced by the ships
three generators into a DC voltage/current for transmission throughout the ship on the
DC bus. Diode and thyristor rectifiers have been popular but can inject harmonic or
reactive currents into the bus. To avoid this phenomenon, a pulse-width-modulated
(PWM) voltage source converter (VSC) is used to convert the AC voltage/current from
the generators to DC for transmission on the bus and will be implemented at each
generator as seen in Figure 13.
The VSC is controlled using the vector control technique which transfers the
three phase AC quantities onto a synchronously rotating reference frame (abc to dq) and
controls it as if it were a DC quantity. By aligning the d-axis of the reference frame with
the voltage vector, the resulting current components will directly control the active and
reactive power flow through the converter allowing for decoupled control of the power
components [28], Additionally, the DC components integrate easily with a proportional
integrator (PI) controller which ensures a fast rise time, a bounded-input bounded-output
stability, and no error once the transient has passed and the resulting quantity has
reached steady state.
The VSC model for this project is shown in Figure 13.
25


iioad
The VSC is connected to a balanced three phase system. The voltage and currents of
the system are expressed as
va + vb + vc = 0
i-a + ib + i-c = 0
(21)
(22)
Where va, Vb, and vc are the voltages produced in each phase at the source and, ia, k,
and ic are the phase currents at the input of the VSC.
The VSC can then be modeled by the following set of differential equations:
. dvdl
dt
--Lia+l[Va-Vdc(Sa-S-^^)] (23)
-~Llb+lL[vb-vdc(sb- (24)
--Lic + lL[vc-vdc(sa-^^)] (25)
^ai-a T + Scic iioad (26)
26


Where R is the series resistance in each phase, L is the inductance of the VSC input
filter, C is the capacitance of the DC-Link capacitor, iioad is the current fed to the load,
and Vdc is the voltage at the DC-link. Sa, Sb, and Sc are the switching functions of the
IGBTs on each of the phase legs. Note that the legs for each phase are independent of
the others, though the two switches in the phase operate opposite one another. That is,
when Si is on, S4 is off. Similarly, when S3 is on S6 is off and when S5 is on, S2 is off.
Applying the Park transform to the VSC model yields the equations in the dq
domain:
did ,
(x) Isj
dt %
R . Sd ,1
lld ^ vdc T ^ vd
(27)
dig , R , Sq 1
= (jOlrf-----------In-------- Vdc +~Vn
dt a L q L ac L q
(28)
dVdc 3Sd _____ 3Sg ____ 1 .
dt ~ 2C d 2C $ C load
(29)
Where id and iq are components of the input currents, Vd and vq are the components of
the input voltage, oj is the angular frequency of the rotating reference frame and Sd and
Sq are the relative components of the switching functions. The power transferred
through the VSC can be found by
p = \(ydid + vqiq) (30)
Q =l(.vdiq-vqid) (31)
As mentioned earlier, when the d-axis of the reference frame is aligned with the Vd
vector, the vq vector will be equal to 0 and the power equations simplify to
P = \(vdid) (32)
Q = \(vdiq) (33)
27


Thus, the active power is proportional to the direct component of the input current, and
the reactive power is proportional to the quadrature component of the input current
allowing for decoupled control of the active and reactive power.
Voltage Source Converter Controller [27]
The controller for the VSC consists of three PI controllers operating in the DQ
domain. One for the DC link voltage, one for the direct current component and the last
for the quadrature current component. Solving equations 24 and 25 for the voltage
components yields
vcL Rid + (x)Liq + urd (34)
ct Iq 0 = Rin + L odLirf + urn Q dt a rq (35)
Vq=0 since the d-axis of the reference frame is aligned with the vd vector, functions are replaced by The switching
^dvdc ~ urd (36)
Rqvdc ~ urq (37)
The voltage equations each contain a cross coupling term that depends on the opposite current from that which is being controlled. The controller takes this into account by adding in the cross coupled terms after the PI controller. These controller equations result in the control block diagram seen in Figure 14:
28


Figure 14: Voltage source converter controller block diagram [27]
The current PI controllers have a transfer function of the form [29]
Tpi(S) = Kp+^ (38)
Where Kp represents the proportional gain of the controller and K, represents the integral
gain. The transfer function of the VSC is of the form
Tvsc(S)=^-r (39)
This results in a controlled system with the transfer function
nS) = QCp+^)1^ (40)
The controllers purpose is to meet the following design goals [29]:
1. The systems output has zero steady-state error
2. A bounded input results in a bounded output (BIBO stability)
29


3. The system has a large bandwidth
To meet these goals, the following rules must be enforced [29]:
1. T(0) = 00
2. Gain Margin must be larger than 6 dB.
3. Phase Margin must be larger than 45.
4. The Phase Inversion Frequency must be larger than the Gain Crossover
Frequency.
These four rules result in the following gains for the direct and quadrature current
controllers:
v = -=^~
P 100TS
2nR
Kj -------
1 100TS
Similarly, the gains for the DC-Link voltage PI controller are as follows
-- _ 2Cvdc^OJn
PV ~ 3vd
Kir,
CVdc^n2
3vd
(41)
(42)
(43)
(44)
Where can be manipulated to control the maximum overshoot and settling time. Vd is the peak
of the supply voltage and C is the DC-Link capacitance.
Load Model
The DCDS model proved to be extremely volatile in the MATLAB Simulink
environment. In order to eliminate numerous MATLAB Simulink errors, the DCDS load
30


model was simplified to a larger degree than the ACDS model. The forward, amidships
and aft loads fed by each generator in the ACDS model were combined into a single RL
load with their transmission line parameters. The resistive load is calculated using
equations 45-47.
R
Lforward
P IS forward^^zS forward"^ P IE forward
(45)
R
amidships ~
____________________Vj____________________
PlS amidships +P2S amidships +P1E amidships
(46)
Raft
_________v2__________
P1S aft+P2S aft+p IE aft
(47)
The inductive part of the load is calculated using the transmission line
parameters as in Appendix D. The DCDS load is shown in Figure 15.
Figure 15: Simulink implementation of DCDS loads (Underway Steaming and Anchor
configuration).
To capture the isolated nature of the system in General Quarters configuration, a
separate model was created in which each load is connected to its generator only. The
Simulink implementation of the GQ configuration is shown in Figure 16.
31


Figure 16: Simulink implementation of DCDS loads (General Quarters configuration).
Simulink Simulation Using the DCDS Model
To provide a direct comparison between the ACDS and the DCDS models, the
DCDS model is tested using the same plant configurations and transients as outlined in
Chapter 2.3 which results in nine trails.
Plant Configurations
1. Underway Steaming
2. General Quarters
3. Anchor
Transient Sources
1. Loss of Generator
2. Starting a Large Load
3. Battle Damage
32


CHAPTER IV
SIMULATION RESULTS
Each Simulink model was tested with three transients under three plant
configurations resulting in nine total trials. In this chapter the results of each simulation
trial are presented side by side in order to more clearly analyze the differences between
the two simulations. The graphs presented are narrowed around the time the transient
was introduced and the time scale was limited to the extent of the transient recovery.
The configurations are evaluated based on the design requirements outlined in NSTM
320 [6],
1. Nominal voltage is 440 V.
2. Average of the three phase line to line voltages must not exceed 5% of
nominal.
3. Maximum voltage transient is 16% of nominal.
4. Voltage transient recovery takes less than 2 seconds.
5. Voltage spike does not exceed 2500 V.
Trial 1: Loss of a Generator While Underway
Trial 1 tests the effect of a generator circuit breaker being tripped open while the
system is configured for general underway steaming with two generators online and
splitting the load evenly. Figure 17 shows the transient inducing condition of the
generator tripping offline. The ACDS transient was introduced six seconds into the
simulation while the DCDS transient was introduced one second into the simulation.
33


Gen 1S Current
500 r
(0
c 0

0
Gen 1S Voltage
cn
cfl
o
>
-500
0
500 r
4 6
Gen 2S Voltage
Time (s)
Figure 17: Voltage (left) and current (right) for generator 1S (top) and generator 2S
(bottom).
600
CO 400
c
a)
^ 200
Gen 1S Voltage

Vabc

Gn 1S Voltage
| Vabc |
(O 44^
440
435
6 62 64 66 68 7 7 2 7 4 7 r; 7 8
cn
2
"6
>
600
CO 400
CM
c
u 200
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Gen 2S Voltage

Vabc
6 62 64 66 68 7 72
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 18: Generator 1S voltage before and after the transient introduction for ACDS.
34


>
CD
cn
3
O
>
600 r
0) 400 -
c
a>
13 200
0 -
0.9
600
0) 400
CM
c
a>
O
200
Gen 1S Voltage
Vabc
_________i_____i______i____________i_____i_____i______i_____i
0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1.1
Gen 2S Voltage
Vabc
0.9 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1.1
Time (s)
Figure 19: Generator 2S voltage before and after the transient introduction for DCDS.
The ACDS generator suffered a 2.1% voltage drop when the 1S generator was
tripped offline, then recovered to nominal voltage after 1.8 seconds. The DCDS
generators suffered a 2804 V spike when the generator was disconnected, but did not
experience any change in the voltage.
Forward Loads
^ 450 i-----1----1----1----1----1----1----1-----r
>
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Amidships Loads
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Aft Loads
Time (s)
Figure 20: Trial 1 Voltage measured at selected loads for ACDS.
35


1000
Forward Loads
v^mmMa/WWW^
____i____i___i___i____i___i____i_____
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
_ 1000
>
a> 800
05
ra
? 600
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
1000
800
600
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Time (s)
Figure 21: Trial 1 Voltage measured at selected loads for DCDS.
The ACDS loads suffered a 2.1% voltage drop when the generator tripped offline
then took 1.7 seconds to recover. The DCDS load voltage oscillated for .64 seconds
before returning to nominal. The oscillation peaked 20.9% below nominal value.
Table 8: Trial 1 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440 V 448 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 2.1 20.9
Max Recovery Time 2s 1.8 0.64
Max V Spike 2500 V 0 4000
Aft Loads

Amidships Loads

The loss of a generator causes a significant transient to be introduced to the
system. The introduction of power electronic converters allowed the loads to recover
three times faster than when the generator supplies the loads directly and insulated the
generator from the transient. However, the DCDS system transient was ten times
greater than the ACDS system transient and oscillated. Additionally, the DCDS
generator suffered from a voltage spike nearly twice as high as allowed by NSTM 320.
36


Trial 2: Starting a Large Load While Underway
Trial 2 tests the effect of a large load being started while the system is configured
for general underway steaming with two generators online and splitting the load evenly.
Figure 20 shows the transient inducing condition. The ACDS transient was induced six
seconds into the simulation while the DCDS transient was induced one second into the
simulation.
500 t
9 0
§
-500 1
Forward Loads
----1 50 i--
<
-50 1
10
Amidships Loads
50 i---
<
10
10
10
Time (s)
Figure 22: Voltage (left) and current (right) measured at selected loads during Trial 2.
Figure 23: Trial 2 Generator 1S and 2S voltage before and after transient for ACDS.
37


The ACDS generator voltage drops .16% from nominal before recovering after
1.9 seconds. The DCDS generator voltage remained steady throughout the transient
seen on the loads.
>
<1J
cn
03
o
>
600
^ 400
c
O
200 -
Gen 1S Voltage
.i___________i._.........1___________j________
0.92 0.94 0.96 0.98 1 1.02 1.04
Gen 2S Voltage
600 r
CO 400
CM
c
cu
u 200 -
Vabc
_ I______L______L -
1.06 1.08 1.1
Vabc
Q ___I____I_____I___I_____I____1_________1____I_____I
0.9 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1.1
Time (s)
Figure 24: Trial 2 Generator 1S and 2S voltage before and after transient for DCDS.
Forward Loads
448.5 -
448
447.5
447
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Amidships Loads
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Aft Loads
450 --1----1----1-----1----1-----1---
448 -
446 -
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 25: Trial 2 voltage measured at selected loads for ACDS.
38


1000
Forward Loads
O)
(0
0.9
1000
800
600
0.9
1000
800
600
0.9
1.1 1.2 1.3
Amidships Loads
1.2 1.3
Time (s)
Figure 26: Trial 2 voltage measured at selected loads for DCDS
The existing ACDS load voltage dropped .1% when the aft load breaker was closed
before recovering after 1.9 seconds. The DCDS load voltage oscillated to a peak of
31.7% from nominal when the new load was started, and came back to nominal after .47
seconds.
Table 9: Trial 2 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440 V 448 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 0.16 31.7
Max Recovery Time 2 s 1.9 0.472
Max V Spike 2500 V 0 0
When the load was started, the ACDS system remained fairly steady suffering a
tiny fraction of a voltage change. The DCDS conversely, saw an oscillating voltage that
lasted .47 seconds before returning to nominal. Despite the significant oscillation, the
recovery time of the DCDS was four times faster.
39


Trial 3: Battle Damage While Underway
Trial 3 is meant to simulate the system experiencing damage by losing the large
forward load while the system is configured for general underway steaming with two
generators online and splitting the load evenly. Figure 23 shows the load change to
introduce the transient. The ACDS transient was induced six seconds into the
simulation while the DCDS transient was induced one second into the simulation.
500
0
-500
500
0
-500
500
0
-500
Forward Loads
_ 50 i---
<
Amidships Loads
------1 ^ 100 I-
s. I
| 0
. O I
Aft Loads
- ?20r
Time (s)
Figure 27: Voltage (left) and current (right) measured at selected loads during Trial 3.
Figure 28: Trial 3 Generator 1S and 2S voltage before and after transient for ACDS.
40


Gen 1S Voltage
500 1
CO
c 0
CD
CO I
Vabc
-500
cu
oi
"O
>
0.9
500 -
0.95
1 1.05
Gen 2S Voltage
1.1
1.15
Vabc
CO
CM
C 0
CD
O
-500
0.9
0.95
1.1
1.15
1 1.05
Time (s)
Figure 29: Trial 3 Generator 1S and 2S voltage before and after transient for DCDS.
The ACDS generators see a voltage increase of 1.1% when the forward load is
lost and recovers back to nominal voltage after 1.87 seconds. The DCDS generators
voltage remained steady throughout the period of time that the loads experienced the
transient.
500
Forward Loads
455
£
O) 450 -
3
O
> 445
6.5 7 7.5
Amidships Loads
8.5
455
i----------1----------1----------1---------1----------1----------1---------1----------1----------r~
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Aft Loads
450 -
445
I----------1----------1----------1---------1----------1----------1---------1----------1----------1
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 30: Trial 3 voltage measured at selected loads for ACDS.
41


Forward Loads
Time (s)
Figure 31: Trial 3 voltage measured at selected loads for DCDS.
The ACDS loads voltage increased 1.1% when the forward load circuit breaker
opened and recovered after 2 seconds. The DCDS load voltage spiked at 10.8% above
nominal before recovering after .12 seconds.
Table 10: Trial 3 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440 V 448 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 1.1 10.8
Max Recovery Time 2 s 1.87 0.12
Max V Spike 2500 V 0 1938
The sudden loss of a load places a large strain on a generator. The kinetic
energy stored in the rotor causes the speed of the rotor to increase when the load is
removed. The introduction of power electronics allowed for a larger voltage transient,
but also recovered significantly faster. Additionally, the DCDS generator experienced a
large voltage spike that was not seen on the ACDS simulation.
42


Voltage (V)
Trial 4: Loss of a Generator During General Quarters
Trial 4 is designed to simulate the loss of a generator during a General Quarters
situation where all three generators are online, but the bus tie circuit breakers are open
and each generator is only supplying the loads attached to its own switchboard. The
configuration change to induce the transient is shown in Figure 26. The ACDS transient
was induced six seconds into the simulation while the DCDS transient was induced one
second into the simulation.
500 r
CO
Gen 0
-500
500
CO CM
c 0 0 0
-500
1000
LD
T
c o 0 0
1000
0
Gen 1S Voltage
Vabc
5 10
Gen 2S Voltage
Vabc
5 10
Gen 1E Voltage
Vabc
5 10
0 5 10
500
<
I 0
o
-500
Gen 2S Current
0 5 10
Gen 1E Current
0 5 10
Time (s)
Figure 32: Voltage (left) and current (right) for generator 1S, generator 2S, and
generator 1E during Trial 4.
43


450 -
CO
c 440
0
O
430
5.5
450 -
> CO
cn C 440
ca a>
O > V 430
5.5
LU
c
o
0
460
450
440
430
5.5
Gen 1S Voltage
Vabc
6 6.5 7 7.5 8
Gen 2S Voltage
Vabc
6 6.5 7 7.5 8
Gen 1E Voltage
Vabc
6 6.5 7 7.5 8
Time (s)
Figure 33: Trial 4 Generator 1S (top) and 2S (middle) and 1E (bottom) voltage before
and after transient for ACDS.
Gen 1S Voltage
500
CO Vabc
c 0 0 0
-500 .. ,
0.9 0.95 1 1.05 1.1 1
Gen 2S Voltage
500 ..
CO CM Vabc
C 0 O 0
-500 , ._L_
0.9 0.95 1 1.05 1.1 1
Gen 1E Voltage
500 -
HI Vabc
c 0 O 0
-500 , _L_
0.9 0.95 1 1.05 1.1 1.15
Time (s)
Figure 34: Trial 4 Generator 1S (top) and 2S (middle) and 1E (bottom) voltage before
and after transient for DCDS.
44


The generators did not suffer from any transient, as they were isolated from the
lost generator by the open bus tie circuit breakers.
5.9 6 6.1 6.2 6.3 6.4 6.5
Amidships Loads
5.9 6 6.1 6.2 6.3 6.4 6.5
Time (s)
Figure 35: Trial 4 voltage measured at selected loads for ACDS.
1000
g> 500
ca
o
>
1000
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Amidships Loads
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time (s)
Figure 36: Trial 4 voltage measured at selected loads for DCDS.
45


On both the ACDS and DCDS simulations, the unaffected loads did not
experience the transient as they were isolated from the transient by the open bus tie
circuit breakers. The loads attached to the 1S generator lost all power and would need
to be restored manually by closing one of the bus tie circuit breakers.
Table 11: Trial 4 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440V 445 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 0 0
Max Recovery Time 2 s 0 0
Max V Spike 2500 V 20 1977
The General Guarters configuration is designed to isolate the system from
potentially disastrous transients during the most volatile situations the ship may
encounter. The simulation demonstrates this concept works as designed and protected
the unaffected generators and loads from the transient caused by the total loss of a
generator. The aft loads did experience a voltage spike on the ACDS simulation,
however, it is caused by the shunt resistances built into the bus tie circuit breakers
Simulink model and would not appear in an actual system.
Trial 5: Starting a Large Load During General Quarters
Trial 5 simulates the transient caused by a large load starting when the
distribution system is set up in the General Quarters configuration. The load change to
introduce the transient is shown in Figure 29. The ACDS transient was induced six
seconds into the simulation while the DCDS transient was induced one second into the
simulation.
46


Voltage (V)
Gen 1E Gen 2S Gen 1S
1000
>
cd
05 0
CO
Forward Loads
---1 ^ 500 i
<
o
>
-1000
0
_ 500 r
>
CD
05 0
CO
+
o
> -500 L
0
5 10 0 5 10
Amidships Loads
____________i ^ 200 ----------------------
<
3
o
-200
10
10
500
CD
05 0
CO
o
>
-500
Aft Loads
-i ^ 50
<
§5 0
3
o
-50
10
0 5 10 0 5
Time (s)
Figure 37: Voltage (left) and current (right) measured at selected loads during Trial 5.
Gen 1S Voltage
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 38: Trial 5 Generator 1S, 2S, and 1E voltage before and after transient for ACDS.
47


Gen 1S Voltage
to
500 -
S 0
o
Vabc
-500 c-
CO
C\l
500 |
0.9 0.95 1 1.05
Gen 2S Voltage
1.1
CD
O) c
ca
_ CD
>
Vabc
-500 L_
0.9
500 |-
- [
s 0
o I
-500 b_
0.95
1 1.05
Gen 1E Voltage
1.1
1.15
Vabc
0.9
0.95
1 1.05
Time (s)
1.1
Figure 39: Trial 5 Generator 1S, 2S, and 1E voltage before and after transient for DCDS.
In the both configurations, when the load is started, the transient is isolated to
Generator 2S as both bus tie breakers are open. The ACDS generator 2S voltage
dropped .3% when the load was started. It then took 1.76 seconds for the voltage to
return to nominal. The DCDS generator voltage remained steady throughout the
transient on the aft loads.
Forward Loads
442
440
438
442
>
0
o> 440
J5
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Amidships Loads
438
446
444
442
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Aft Loads
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 40: Trial 5 voltage measured at selected loads for ACDS (right set).
48


1000
Forward Loads
800
600
0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2
_ 1000
>
800
O)
3
o 600
0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2
1000
800
600
0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2
Time (s)
Figure 41: Trial 5 voltage measured at selected loads for DCDS (right set).
As on the generator side of the systems, the loads are isolated and there is no
transient except on the load that is started. The AC load transient maxes out .31% low
before climbing to nominal voltage in 1.76 seconds. The DC system voltage drops 13%
before recovering after .15 seconds.
Table 12: Trial 5 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440V 445 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 0.3 13
Max Recovery Time 25 1.76 0.15
Max V Spike 2500 V 0 0
Aft Loads
Amidships Loads
T------1------1------1------1-----1------T
T I I I I I T
The General Quarters configuration isolates the unaffected sections of the
system leaving only a transient on the load that is started. The DCDS system allowed
the load to reach nominal voltage eleven times faster than the ACDS system despite
suffering a larger overshoot.
49


Trial 6: Battle Damage During General Quarters
Trial 6 is meant to simulate the system experiencing damage by losing the entire
forward load while all three generators are online and supplying the loads only attached
to their own switchboards. The load change to introduce the transient is seen in Figure
32. The ACDS transient was induced six seconds into the simulation while the DCDS
transient was induced one second into the simulation.
500 r
0
03 0
3
> -500
Forward Loads
^ 500 |---
<
-500 1
10
10
1000 r
0
O) 0
Amidships Loads
-----1 500 i---
<
| 0
1000 1
500 r
0
03 0
1 -500 L
10 0
Aft Loads
i ^ so r
<
| o
10
-500 1
-50 1
10
10
Time (s)
Figure 42: Voltage (left) and current (right) measured at selected loads during Trial 5.
Gen 1S Voltage
co
c
CD
CD
500
Gen S Voltage
Vabc
I ^
6 62 64 66 68 7 72 74 76
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Gen 2S Voltage
co
500
S <=
CO o CD
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Gen 1E Voltage
LD
c
a)
CD
500

0
----------Vabc
Gen g Voltage_________
6 62 64 66 68 7 72 74 76 78
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 43: Trial 6 Generator 1S, 2S, and 1E voltage before and after transient for ACDS.
50


Gen 1S Voltage
500 I
co Vabc
c 0 D o I -500 |
0.9 0.95 1 1.05 1.1 1.15
500 1 Gen 2S Voltage
CO CM 1 Vabc
cn Cfl p C 0 CD 0 1 -500 |
> 0.9 0.95 1 1.05 1.1 1.15
500 I- Gen 1E Voltage
LU Vabc
c 0 CD o 1- 1 -500 |
0.9 0.95 1 1.05 1.1 1.15
Time (s)
Figure 44: Trial 6 Generator 1S, 2S, and 1E voltage before and after transient for DCDS.
The ACDS system generator model shows a transient in all three generators
despite the isolation provided by the General Quarters configuration. This is likely
caused by the shunt resistances around the bus tie circuit breakers. Generator 1S
voltage maxes out at 3.5% above nominal. Generator 2S maxes out at 1.9% above
nominal. Generator 1E maxes out at 2.8% above nominal. All three generators take 1.7
seconds to recover from the transient. On the DCDS generators, the isolation is intact
and no transient is seen in any of the generator voltage waveforms.
Forward Loads
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Aft Loads
-i-i-mil Intj]-----1----1----1----1----1----1----
450
440
430
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 45: Trial 6 Voltage measured at selected loads for ACDS.
51


Forward Loads
0.8 0.9 1 1.1 1.2 1.3 1.4 1.5
Aft Loads
1000 --------1-------1--------1-------1--------1-------1--------
800 ;_________. ________________________________;
600 -
______I______I_____I______I_____I______I_____
0.8 0.9 1 1.1 1.2 1.3 1.4 1.5
Time (s)
Figure 46: Trial 6 Voltage measured at selected loads for DCDS.
The ACDS amidships load voltage showed an unexpected transient which peaks
at 3.5% of nominal before settling in 1.78 seconds. The aft loads peak at 1.9% above
nominal and settled after 1.78 seconds. The DCDS loads were completely isolated from
the transient due to the open bus tie circuit breakers, but the affected DC Bus saw an
18.5% overshoot before settling in .4 seconds.
Table 13: Trial 6 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440 V 445 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 3.5 18.5
Max Recovery Time 2 s 1.78 0.4
Max V Spike 2500 V 0 0
The ACDS system suffered a large voltage overshoot and despite the isolation
provided by the open bus tie circuit breakers, the transient was seen in every generator
and load waveform. This is most likely caused by the shunt resistances in the Simulink
circuit breaker model. The DCDS model was more realistic in that the sections were
isolated and only the affected section of the DC Bus saw a transient. The overshoot on
52


the DC Bus was three times larger than the transient on the ACDS, but recovered over
four times faster.
Trial 7: Loss of a Generator While at Anchor
Trial 7 simulates the situation when a generator is lost while the ship is
configured to be at anchor with one generator online and supplying a limited load. When
the generator is lost, the ship loses all power and the emergency generator comes
online to take the full load. The configuration change to introduce the transient is seen in
Figure 35. The ACDS transient was induced six seconds into the simulation while the
DCDS transient was induced at the beginning of the simulation. Both emergency
generators come online .5 seconds later
4000
>
CD
Ol
CO
o
>
2000
0
-2000
-4000 L
0
Gen 1S Voltage
Vabc
400
< 200
I 0
i_
o
-200
-400
Gen 1S Current
10
10
>
CD
)
co
"o
>
4000
2000
0
-2000
-4000
Gen 1E Voltage
Gen 1E Current
400
Vabc ~ 200 labc L
<
Hi
c _
4J 0
o
-200
-400
0 5 10 0 5 10
Time (s)
Figure 47: Voltage (left) and current (right) for generator 1S and generator 1E during
Trial 7.
53


460 r
Gen 1E Voltage
455 -
450
>
cn 445
CO
o
>
440
435
430 -------------
5.5 6 6.5
Vabc
7.5 8
Time (s)
Figure 48: Voltage at generator 1E for ACDS during startup.
>
O)
ra
o
>
480
470
460
450
440
430
420
410
400
Gen 1E Voltage
.J._________________________I.
Vabc
,j_______________i.
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Time (s)
Figure 49: Voltage at generator 1E for DCDS during startup.
When the ACDS generator 1E breaker closes, the system experiences a 2239
spike before experiencing a 3.3% voltage drop, finally stabilizing after 1.5 seconds.
When the DCDS 1E generator comes online, the voltage starts 1.3% below nominal
which stabilizes after .3 seconds.


460
440 -
420
460
0}
O) 440 -
to
o
>
420
460
440 h
420

Forward Loads
i------1-------r
~\----------------r
t--------r
i----------------r
6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Amidships Loads
n--------r

6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Aft Loads
I----------------T
6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 50: Trial 7 voltage measured at selected loads for ACDS.
1000
Forward Loads
1000
a) 800
O)
TO
o 600
1000
1 1 1 1 1 1 1
r~ -
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.
Amidships Loads

/, ' -
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.
Aft Loads

-
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.
Time (s)
Figure 51: Trial 7 voltage (left) and current (right) measured at selected loads for DCDS.
55


The loads on both the ACDS and DCDS startup from no voltage. When the
breaker closes, the ACDS loads experienced a 2284 V spike before dropping to 3.4%
below nominal before recovering after 2 seconds. The DCDS loads experienced a 6.1%
voltage overshoot which recovered after .25 seconds.
Table 14: Trial 7 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440V 450 440.5
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 3.3 1.3
Max Recovery Time 2 s 1.5 0.3
Max V Spike 2500 V 2239 0
Trial 7 is unique in that it is the only trial in which the system is completely de-
energized and the transient is caused by the startup of a generator. The DCDS system
provided full voltage to the loads 5 times faster than the ACDS and did not experience a
voltage spike when the 1E generator started up.
Trial 8: Starting a Large Load While at Anchor
Trial 8 simulates the situation when a large load is started during anchor
configuration and only one generator is online supplying all the loads. Figure 37 shows
the transient introduced. The ACDS transient was induced six seconds into the
simulation while the DCDS transient was induced one second into the simulation.
s
5
s
O
>
3
§
Forward Loads
---- 20
&
^ 0
3
____, o _
£
3
o
1 i 1 i
0 5 10
oads
i i i i
0 5 5 10
[ I J
Time (s)
Figure 52: Voltage (left) and current (right) measured at selected loads during Trial 8.
56


600
Gen 1S Voltage
- Vabc
500 -
400
iu
cn 300
ra
o
>
200
100
6 65 7 '5
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 53: Generator 1S voltage before and after transient for ACDS.
600 r
400
Gen 1S Voltage
Vabc
200
>
aj
cn 0
ffl
o
>
-200
-400
-600 ----------1----------------------1-----------1----------1
0.9 0.95 1 1.05 1.1 1.15
Time (s)
Figure 54: Generator 1S voltage before and after transient for DCDS.
The ACDS generator suffers a .2% voltage drop when the load is started and
then recovers in 2 seconds. The DCDS generator did not show any voltage drop during
the time the loads showed oscillating voltage.
57


Forward Loads
448 --1-----1----1-----1----1-----1----1-----1----1-----r
446
444 H
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Aft Loads
445 i-----1------1----1-----1----1-----1-----1----1-----1----
440 mimimmmm
6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
Time (s)
Figure 55: Trial 8 Voltage measured at selected loads for ACDS.
Forward Loads
Figure 56: Trial 8 Voltage measured at selected loads for DCDS.
58


On the load side of the ACDS system, the existing loads suffered a .2% voltage
drop when the load started and recovered after 1.5 seconds. Similarly, the load that
came online started .2% below nominal then reached nominal voltage after 1.5 seconds
The DCDS loads saw a voltage oscillation when the load came online which lasted .64
seconds and maxed out 23.6% larger than nominal.
Table 15: Trial 8 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440V 446 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 0.2 23.6
Max Recovery Time 25 2 0.64
Max V Spike 2500 V 0 0
The voltage transient on the ACDS system was minimal at only .2%, but took
nearly 2 seconds for the transient to recover. The DCDS system boasted a significantly
faster response time, approximately 4 times faster, but suffered from an oscillating
voltage on the DC Bus.
Trial 9: Battle Damage While at Anchor
Trial 9 is meant to simulate battle damage by suddenly opening the circuit
breaker to every forward load and removing a large load from the system. The load
change to introduce the transient is shown in Figure 40. The ACDS transient was
induced six seconds into the simulation while the DCDS transient was induced one
second into the simulation.
59


500
0
O) o
CO

-500
500
0)
O) 0
ca

-500
500
(D
ra 0
CO
o
>
-500
Forward Loads
_ 20 |--
<
5 0
5*>L
10 0
Amidships Loads
^ 10
<
5 o
d-ioL
10 0
Aft Loads
' 20
<
a?
3
o
-20
10
10
J
10
0 5 10 0 5
Time (s)
Figure 57: Voltage (left) and current (right) measured at selected loads during Trial 9.
Figure 58: Generator 1S voltage before and after transient for ACDS.
60


600
Gen 1S Voltage
Vabc
400
200 -
>
CD
CD
ra
o
>
0
-200 -
-400
0.9 0.95 1 1.05 1.1
Time (s)
Figure 59: Generator 1S voltage before and after transient for DCDS.
The ACDS generator voltage peaked at 2.3% above nominal before settling after
2 seconds. The DCDS generator voltage maintained its voltage throughout the period
that the loads displayed the transient.
Forward Loads
450 llllllllllllllllllllllllllllll-'-------------1-------------1---------------
400 ----------------1-------------1-------------1---------------
5.5 6 6.5 7 7.5 8
Amidships Loads
> 450
0
o>
co
1
400
5.5
6.5 7 7.5
Aft Loads
450
400
5.5
6.5
7.5
Time (s)
Figure 60: Trial 9 voltage measured at selected loads for ACDS.
61


1000
Forward Loads
800
600
0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2 1.25 1.3
_ 1000
>
oj 800
O)
TO
o 600
0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2 1.25 1.3
1000
800
600
0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2 1.25 1.3
Time (s)
Figure 61: Trial 9 voltage measured at selected loads for DCDS.
The ACDS loads peaked at 2.8% above nominal after the forward loads drop
offline before recovering after 1.7 seconds. The DCDS load voltage briefly peaked
12.4% above nominal before finally recovering .25 seconds after the load was lost.
Aft Loads

-
Amidships Loads
T-----1----1-----1----1-----1----1----1-----T
J_____________I___________I____________I____________I____________I____________I____________I____________L
Table 16: Trial 9 ACDS and DCDS comparison summary
Criteria ACDS DCDS
Nominal Voltage 440V 446 750
L-L Voltage Average 22V 0 0
Max V Overshoot 16% 2.33 12.4
Max Recovery Time 2s 2 0.25
Max V Spike 2500 V 0 0
When the load was lost, the remaining ACDS loads saw a 2.3% voltage increase
then recovered after 2 seconds. The DCDS generator did not experience any transient,
but the other loads saw the voltage overshoot but within requirements.
62


Simulation Results Analysis
Consistently across all the trials, the DCDS simulation provided significantly
faster recovery times than the ACDS simulation. Without the power electronic
converters, the generators governor is the only control structure in the system. The
governor is a mechanical control system, which has a response time limited by the
mechanical components. Thus, irrelevant of the type of transient introduced, or the
configuration of the system, the ACDS generators recovery was limited to between 1.5
and two seconds. Without changing the machinery installed in the system, there is no
way to increase the recovery speed of the generator.
The DCDS power converters, on the other hand, respond to transients
significantly faster thanks to the high switching frequency of the transistors. This allows
the PI controllers to identify changes and respond at a rate beyond the capability of the
generator governor. The DCDS simulation recovered an average of nearly five times
faster than the ACDS. Recovery time was the greatest advantage identified by the
simulation trials, in other areas however, the DCDS simulation fell short.
When the transients were introduced, the ACDS system suffered a voltage
change of 1.5% on average. The DCDS system, on the other hand, saw a voltage
change ten times larger, 14.7 V on average. Much of the volatility of the DC bus voltage
could be minimized by introducing an energy storage mechanism, such as a battery
bank or flywheel, on the DC bus, to provide additional constant power during the
transient.
The voltage on the DCDS simulations exhibited excessive noise in the voltage
waveforms compared to the ACDS simulation waveforms. That noise is created by the
switching action on the IGBT and could be eliminated with the installation of additional
filters on the power converter.
63


Unsurprisingly, in both the ACDS and DCDS simulations, when more generators
were online and sharing the load, the transients tended to be smaller thanks to the
kinetic energy stored in the generator rotor. While rotor inertia allows the generator to
maintain speed and voltage better than the power electronic converters, the generators
governor responds slowly to changes in the load. Conversely, while the power
electronics allow for significantly faster recovery, the lack of inertia allows a much larger
overshoot which could potentially be controlled with additional energy storage on the
bus.
64


CHAPTER V
DISTRIBUTION SYSTEM COMPARISON
Thus far, this thesis has focused on a transient analysis of the existing AC
distribution system and the proposed DC distribution system for the 270 WMEC.
However there are additional advantages to a zonal distribution system that are worth
discussion.
Weight Benefits
The current radial architecture of the 270 WMECs ACDS contains a multitude of
cables to distribute the power from the ship service switchboards to the load centers
throughout the ship. As each load center is fed directly from the ships service
switchboards located in either the engine room or emergency generator room, there are
many redundant cables running from fore to aft (longitudinally) that are eliminated with a
distribution bus. A permanent distribution bus consists of only two cables running
longitudinally thus eliminating all the extra cables that are ordinarily needed to distribute
the electrical power.
To analyze the weight difference, the following assumptions were made:
1. A DC Bus would operate at 1000 kV with 2 conductors and would require a
DSGU-9 cable [31].
2. The location of the generators and load centers remains the same as the
ACDS configuration.
3. All longitudinal cables feeding individual load centers are removed and the DC
bus cables are added.
4. All athwartships and vertical cables are left in place to feed from the
distribution bus to the load canter.
Following these assumptions, over 4500 lbs worth of cables could be removed from the
ship by using a distribution bus. Detailed calculations can be found in Appendix D.
65


In 1993, Naval Engineers Journal published a study conducted by Chester Petry
and Jay Rumburg who compared the radial AC electrical distribution systems on board
the Navys DDG-51 class destroyers to a zonal AC distribution system. In addition to
cable weight reductions, that study found additional weight savings by removing the
following equipment [10]:
Power distribution switchboard sections
Power distribution load centers
The removed equipment was replaced with smaller, more efficient equipment [10]:
Main distribution bus feeder switchboards
Two power distribution load centers in each new electrical zone
Load center to main distribution bus connection boxes
The overall weight change (in long tons) is summarized in Table 17:
Table 17: DDG-51 weight comparison between radial and zonal architectures [10]
System Description Removal (LT) Electrical System Install (LT) Net Change (LT)
Foundations 3.3 4.3 +1.0
Power Cables 116.7 79.8 -36.9
Switchgear 20.8 20 -.8
Totals 140.8 104.1 -36.7
A more complete weight comparison, taking into account switchgear and foundations,
was not completed as part of this thesis due to a lack of available information for the
270 WMEC. However, it is worth noting the significant weight reductions realized on the
DDG-51 class. These weight reductions translate to significant cost savings earned
through more efficient propulsion and a reduction in fuel usage.
66


Standardization Benefits
The benefits of standardization have already been realized during the Coast
Guards response to the devastation in New Orleans following Hurricane Katrina. CAPT
Bruce Jones, the commanding officer of Air Station New Orleans explained the fact that
you can take a rescue swimmer from Savannah and stick him on a helicopter from
Houston with a pilot from Detroit and a flight mech [sic] from San Francisco, and these
guys have never met before and they can go out and fly for six hours and rescue 80
people and come back without a scratch on the helicopter shows the importance and
benefit of standard equipment and training [34],
The zonal architecture is extremely flexible and can be expanded or contracted
to fit any size of vessel. The Office of Naval Research has continued the work to develop
a standard Power Electronic Building Block (PEBB) that can change any electrical power
input into any desired (Type I, II or III) output [33], These PEBBs can not only replace
the motor-generators currently employed on Coast Guard cutters, but in coordination
with a DC distribution system, could be employed across multiple ship classes. As the
PEBBs can change any type of power into any other, generation equipment could then
be standardized and only scaled for larger demand on larger cutters. This allows for
shipbuilders to stock a large number of the same parts which could then be employed
across multiple construction projects and the Coast Guard to reduce its logistics footprint
by stocking parts for a smaller variety of equipment.
Standardizing the equipment employed across all Coast Guard cutter classes
would also simplify training and qualification requirements for personnel. This opens a
wide range of options for personnel assignments, especially for short-term or emergent
personnel needs at a unit where a member assigned to any other ship would be
knowledgeable enough to operate the equipment.
67


Utilizing a zonal electrical distribution system would provide numerous
advantages during the acquisition phase of the ships lifecycle as well. At present, with a
radial architecture, electrical equipment can not be installed during construction until
after the hull is completely assembled. Only then can cables be run throughout the ship
from switchboards to loads [10], The ZED concept keeps all distribution cabling and
equipment isolated to a single zone while the zones are connected to each other only
through the distribution bus. This allows for shipbuilders to use a modular construction
technique where the entire zone could be completely outfitted and tested before zone
assembly as in Figure 43 [32],
ZONAL ARCHITECTURE
CABLES CROSSING ZONES
TWO BUSES
NO LONGITUDINAL FEEDERS
Figure 62: Advantage of zonal construction [32]
Outfitting an individual zone provides greater accessibility for workers running cables,
testing equipment, and troubleshooting problems before the zone is joined to the rest of
the ship [32], Additionally, since they do not rely on each other, multiple zones could be
built concurrently, and then joined together all of which reduce the overall cost and
period of performance to build the ship. Similarly, as the ship ages and receives
equipment upgrades, a radial architecture would simplify the upgrade as power cables
would only need to be run within a watertight zone rather than from the generator [10],
68


From reducing construction costs and timelines, to simplifying training and
support, the zonal electrical distribution architecture could revolutionize cutter acquisition
and sustainment logistics and maximize the Coast Guards strained budget in these
areas.
Survivability Benefits
On October 12, 2000, while refueling in Aden, Yemen, the USS Cole was
attacked by two al-Qaeda terrorists carrying a bomb in a small boat. The detonation
next to the USS Coles hull created a hole over 40 feet wide [37],
Figure 63: Damage to USS Cole [37]
Despite the significant damage and flooding the ship suffered, the USS Cole did not
sink thanks to the damage control strategy employed by the ships crew which exploited
the ships innate survivability [37],
The U.S. Navy defines a ships survivability as [a] measure of both the capability
of the ship, mission critical systems, and the crew to perform assigned warfare missions,
and of the protection provided to the crew to prevent serious injury or death [35], A
69


ships survivability has nothing to do with combat, necessarily, but also includes
accidents such as fire, flooding, groundings, etc. Survivability is broken down into three
disciplines:
Susceptibility: A measure of the capability of the ship, mission critical systems,
and crew to avoid and or defeat an attack and is a function of operational tactics,
signature reduction, countermeasures, and self-defense system effectiveness.
Vulnerability: A measure of the capability of the ship, mission critical systems,
and crew to withstand the initial damage effects from conventional, Chemical
Biological or Radiological (CBR), or asymmetric threat weapons or accidents,
and to continue to perform assigned primary warfare missions, and protect the
crew from serious injury or death.
Recoverability: A measure of the capability of the ship and crew, after initial
damage effects, whatever the cause, to take emergency action to contain and
control damage, prevent loss of a damaged ship, minimize personnel casualties,
and restore and sustain primary mission capabilities [35],
A zonal electrical distribution system drastically increases survivability by
decreasing vulnerability and increasing recoverability. The first step to combating any
casualty, and what saved the USS Cole, is to contain it by setting boundaries. A
boundary can be set at any watertight (for flooding emergencies) or fume tight (for fire or
CBR emergencies) bulkhead [36], By reducing the number of longitudinal cables that
repeatedly penetrate watertight bulkheads to two (the distribution bus) those boundaries
are more effective for stopping flooding and fire from spreading, allowing the ship to
continue its primary mission despite the damage that has been suffered. In the case of
the USS Cole, the flooding boundaries ensured that the damaged section of the ship
was the only section that suffered flooding, not nearly enough to overcome her innate
buoyancy and cause her to sink.
Additionally, a zonal electrical distribution system increases recoverability by
minimizing the damage that would need to be repaired to continue functioning. With two
busses running the length of the ship, separated port and starboard and separated
vertically as much as practical, the chance of both busses being damaged
70


simultaneously is minimized. If one bus is damaged, there is a second bus within the
watertight compartment that can provide power to necessary equipment without
breaching watertight boundaries with casualty power cables. This allows the ship to
maintain its high state of readiness without compromise. Repair to the damaged bus is
also simplified as casualty power can be rigged around the damaged section, restoring
power to the rest of the bus quickly and efficiently.
The simplicity of a zonal distribution system greatly increases a ships
survivability compared with a radial distribution system by reducing penetrations through
watertight boundaries and drastically simplifying casualty power routes. The attack on
the USS Cole vividly demonstrated that ensuring the integrity of boundaries and
preventing the spread of damage is the surest way to keep the ship in fighting shape.
71


CHAPTER VI
CONCLUSIONS AND RECOMMENDATIONS
Throughout the work completed in this thesis, the purported benefits of a direct
current zonal electrical distribution system have been tested and analyzed. When
compared with an existing alternating current radial distribution system the benefits are
striking.
The introduction of power electronic converters to rectify the current and voltage
from the generators not only provide insulation to the generators, but increases the
speed at which the system can recover. Additionally, switching from a radial-type
distribution system to a zonal system drastically reduces the ships weight by eliminating
redundant cabling, foundations, and transformers, which ultimately leads to significant
fuel savings over the life of the ship.
Unfortunately, the work on this thesis is incomplete. While the wide range of
computer models developed as part of this thesis create a solid basis for future deeper
mathematical analysis of each trial, the DCDS model was drastically simplified to
eliminate errors encountered in the MATLAB-Simulink software and meet necessary
timelines. An inverter model needs to be developed and incorporated on the load side of
the DCDS model and the trials tested again. The creation of a droop controller for the
generator models would ensure the load is more evenly divided between the online
generators. Additionally, incorporating power converters to produce Type II and Type III
power, which could ultimately replace motor-generator sets currently employed on
cutters, is a necessary addition to the simulation.
The benefits demonstrated by the work on this thesis suggest that taxpayer funds
would be saved in the acquisition phase of the ships lifecycle by reducing the amount of
work and materials needed to install the system. Additional expenditure would be
72


avoided during the sustainment phase with a reduced weight, increasing engine fuel
efficiency.
Despite these benefits, there are significant challenges that still need to be
addressed on a DC based system. A fault on the DC bus allows for a very large fault
current in a short amount of time. If not addressed quickly, the large fault current can
cause significant damage to sensitive electronics. Additionally, there were significantly
larger overshoots on the DCDS transients, though tuning of the power electronics
controllers could reduce the observed overshoot. Finally, DC grids are inherently
unstable due to the dynamics of the power electronic interfaces and the low frequency
oscillation when multiple converters are connected together. Though these challenges
remain, the purported benefits of a DC ZED show that these challenges are worth
solving.
In an era of constrained budgets and increasingly complex operating
environments, the Coast Guard must find sustainable ways to reduce operating costs. A
DC ZED system reduces costs in every phase of the ships lifecycle, simplifies personnel
and training requirements, and increases the ships innate survivability. However
significant challenges remain. Fault conditions and stability issues need to be further
studied and mitigation strategies developed before a DCZED can be incorporated into
Coast Guard use. Beyond the work considered for the Offshore Patrol Cutter, the Coast
Guard, and the American taxpayer would be best served by considering a zonal
distribution system for all future cutter acquisitions.
73


REFERENCES
[1] "U.S. Coast Guard History." U.S. Coast Guard History: Brief. U.S. Coast Guard, 16 Oct.
2014. Web. 01 July 2015.
[2] United States. Cong. Congressional Research Service. Coast Guard Cutter Procurement
Background and Issues for Congress. By Ronald O'Rourke. Cong. Rept. R42567.
Washington, DC: Congressional Research Service, Library of Congress, 2015. U.S. Coast
Guard Cutters & Craft Index. U.S. Coast Guard, 9 Apr. 2015. Web. 1 July 2015.
[3] "Missions." USCG: About Us. U.S. Coast Guard, 20 Mar. 2014. Web. 01 July 2015.
[4] "Recapitalization." U.S. Coast Guard Acquisition Programs. U.S. Coast Guard, Nov. 2014.
Web. 1 July 2015.
[5] "Famous Cutter Class 270-Foot Medium Endurance Cutter (WMEC)." Global Security.
GlobalSecurity.org, 2015. Web. 01 July 2015.
[6] United States. Department of Defense. Naval Sea Systems Command. NSTM 320: Electric
Power Distribution Systems. Naval Sea Systems Command, 1 Mar. 2005. Web. 14 Nov.
2014.
[7] Amy, John V. "Considerations in the Design of Naval Electric Power Systems." IEEE
Power Engineering Society Summer Meeting, (2002): IEEEXplore. Web. 14 Jan. 2015.
[8] Doerry, Norbert H., and James C. Davis. "Integrated Power System for Marine
Applications." Naval Engineers Journal 106.3 (1994): 77-90. Web. 15 Jan. 2015.
[9] Ciezki, John G., and Robert W. Ashton. "Selection and Stability Issues Associated with a
Navy Shipboard DC Zonal Electric Distribution System." IEEE Transactions on Power
Delivery IEEE Trans. Power Delivery 15.2 (2000): 665-69. IEEEXplore. Web. 21 Oct.
2014.
[10] Petry, Chester R., and Jay W. Rumburg. "Zonal Electrical Distribution Systems: An
Affordable Architecture for the Future." Naval Engineers Journal 105.3 (1993): 45-
51. NEJOnline. Web. 14 Jan. 2015.
[11] "Offshore Patrol Cutter." OFFSHORE PATROL CUTTER (n.d.): n. pag. Offshore Patrol
Cutter. U.S. Coast Guard Acquisition Directorate, May 2015. Web. 24 Aug. 2015.
[12] U.S. Coast Guard Headquarters, 270 FT WMEC (B) Electrical One-Line Diagram,
Naval Engineering Division Dwg No 905-WMEC-320-014-F.
74


[13] United States. Department of Defense. Naval Sea Systems Command. NSTM 300: Electric
Plant-General. Naval Sea Systems Command, 1 May 2012. Web. 14 Jan. 2015.
[14] Collier, Everett. The Boatowner's Guide to Corrosion: A Complete Reference for
Boatowners and Marine Professionals. Camden, Me.: International Marine/McGraw-
Hill, 2006. Print.
[15] Chapman, Stephen J. "Chapter 5 Synchronous Machines." Electric Machinery and Power
System Fundamentals. Boston: McGraw-Hill, 2002. 192-264. Print.
[16] Mohan, Ned. "10.6 Synchronous Generators." Electric Machines and Drives: A First
Course. Hoboken, NJ: Wiley, 2012. 187-90. Print.
[17] Wood, Allen J., Bruce F. Wollenberg, and Gerald B. Sheble. "10 Control of
Generation." Power Generation, Operation, and Control. Third ed. Hoboken, NJ: John
Wiley & Sons, 2014. 468-500. Print.
[18] MATLAB Simulink. Program documentation. Model and Dynamics of the Synchronous
Machine. Vers. R2015a. Mathworks, 2015. Web. 2 Sept. 2015.
[19] "IEEE Std 421.5 2005 IEEE Recommended Practice for Excitation System Models for
Power System Stability Studies," IEEE Std 421.5-2005 (Revision of IEEE Std 421.5-
1992), pp. 0___1-85, 2006.
[20] U.S. Coast Guard Headquarters, 270 FT WMEC Load & Power Analysis, Office
of Naval Engineering Dwg No 901-WMEC-311-001-G.
[21] Bergen, Arthur R., and Vijay Vittal. Power Systems Analysis. Second ed. Upper Saddle
River, NJ: Prentice Hall, 2000. Print.
[22] United States of America. Department of Homeland Security. United States Coast
Guard. Naval Engineering Manual. Washington DC: Office of Naval Engineering, 2011.
Print. COMDTINST M9000.6F.
[23] Sowers, Leigh G. Personal interview. 20 Mar. 2015.
[24] United States of America. Department of Homeland Security. U.S. Coast Guard. Standing
Orders for the Engineering Department-USCGC MOHAWK. By A. B. Morrison. N.p.:
n.p., 2014. Print. MOHAWKINST 9000.1G.
75


[25] United States. Department of Defense. Naval Sea Systems Command. NSTM 310: Electric
Power Generators and Conversion. Naval Sea Systems Command, 1 Mar 2005. Web. 14
Sep. 2015.
[26] United States. U.S. Coast Guard. Office of Naval Engineering. Technical Manual
Generator Section-SSG. N.p.: n.p., 1974. Print. TP 2820B.
[27] Teleke, Sercan, Subhashish Bhattacharya, and Mesut Baran. "Enhanced Control of Voltage
Source Converters for DC Shipboard Power Systems." Naval Engineers Journal 122.1
(2010): 81-91. Wiley Online Library. Web. 8 Dec. 2015.
[28] Sul, Seung-Ki. Control of Electric Machine Drive System. Hoboken, NJ: Wiley-IEEE,
2011. Print.
[29] Voland, Gerard. Control Systems Modeling and Analysis. Englewood Cliffs, NJ: Prentice-
Hall, 1986. Print.
[30] Teleke, Sercan, Subhashish Bhattacharya, and Mesut E. Baran. "A Novel PWM Voltage
Source Converter for a DC Zonal Shipboard Power System." Proc. of 33rd Annual
Conference of the IEEE Industrial Electronics Society, Taipei, Taiwan. N.p.: n.p., n.d. N.
pag. IEEE Xplore. Web. 12 Oct. 2015.
[31] United States of America. Department of Defense. Department of the Navy. Military
Standard: Electronic Equipment Parts Selected Standards Wire and Cable. 242J ed. N.p.:
n.p., 1986. Everyspec. Everyspec LLC. Web. 31 Mar. 2015.
[32] Cecere, Michael L., Ill, Jack Abbott, Michael L. Bosworth, and Tracy Joseph Valsi.
"Commonality-Based Naval Ship Design, Production, and Support." Journal of Ship
Production 11.1 (1995): 1-14.SNAME. Society ofNaval Architects and Marine
Engineers, Feb. 1995. Web. 20 Jan. 2016.
[33] Piff, Joseph C. "Power Electronics Building Blocks (PEBB) Program: PEBB Bringing a
Whole New Perspective to Power Control and Distribution." Program Manager 1 Mar.
2003: 46-54. DAU Publications and Catalogs. Web. 20 Jan. 2016.
[34] United States of America. U.S. Coast Guard. Historian's Office. The U.S. Coast Guard &
Hurricane Katrina. By Scott Price. U.S. Coast Guard, Feb. 2006. Web. 20 Jan. 2016.
[35] United States of America. Department of the Navy. Office of the Chief ofNaval
Operations. Department of the Navy Issuances. Department of Defense, 13 Sept. 2012.
Web. 25 Jan. 2016.
76


[36] United States of America. U.S. Coast Guard. Office of the Commandant. Machinery Space
Firefighting Doctrine for Class Bravo Fires. Washington DC: U.S. Coast Guard, 2009.
COMDTINST M9555.1B.CG-612 Directives and Publications Division. U.S. Coast
Guard, 23 Nov. 2009. Web. 25 Jan. 2016.
[37] Evans, Mark L. "Cole (DDG-67)." Cole (DDG-67). Naval History and Heritage
Command, 30 June 2015. Web. 25 Jan. 2016.
77


Cable Specifications JMIL-STD-242J &MIL-DTL-24643C)
Cable Q/1000 ft of cable R/Ym Max V (RMS) Conductor Conductor Diameter (in) GMD R' GMR for L GMRforC LjH/m) C (F/m] L (H/km) C jF/Ttm) wt/c o n d u ct d r/1000ft CableWt/ft
D5GU-14 0.8407 0.27582021 5000 2 0.67 0.67 0.260898654 0.418093408 0473761543 9.43146E-08 1.60518E-10 9.43146E-05 L60518E-07 43 0.036
DSGU-4 2.68 0.879265092 3000 2 0.427 0.427 0.166274217 0.266456545 0.301934596 9.43146E-08 1.60518E-10 9.43146E-05 1.60S18E-07 106 0.106
DSGU-50 0 205 0067257218 5000 2 0.911 0.911 0.354744289 0 568482231 0.644174278 9.43146E-08 1.60518E-10 9.43146E-05 1.60518E-07 158.1217 0.3162434
DSGU-9 1.06 0.347769029 3000 2 0.544 0.544 0.211834131 0.339466887 0.384666089 9.43146E-08 1.60518E-10 9.43146E-05 160518E-07 28 0.056
LSDS6U-4 2.58 0.846456693 1000 2 0 0 0 0 #DIV/0! ttDIV/0! ffDrV/0! 9DIV/0! 106 0.106
LST3GU-100 0.1 0032808399 1000 3 1.266 1.266 0.492981635 0 790009335 0895197185 9.43146E-08 1.60518E-10 9.43146E-05 1.6051BE-07 1618 1.618
L5TSGU-200 0.05 0.016404199 1000 3 1.669 1.669 0.649910228 1.0414394 1.130161218 9.43146E-08 1.60518E-10 9.43146E-05 1.6051BE-07 3086 3.036
LSTSGU-23 0.506 0.166666667 1000 3 0.812 0.812 0.316193592 0.50670425i 0.574170706 9.43146E-08 1.60518E-10 9.43146E-05 1.60518E-07 443 0.443
LST5GU-50 0.201 0.065944882 1000 3 0.969 0.969 0.377329545 0.604675392 0.685186471 9.43146E-08 1 60518E-10 9.43146E-05 1.6051BE-07 886 0.836
TSGU-100 0.102 0.033464567 5000 3 1.266 1.266 0.492981635 0 790009335 0895197185 9.43146E-08 1.60518E-10 9.43146E-05 160518E-07 320.7967 0.9623901
TSGU-125 0.102 0.033464567 5000 3 1.515 1.515 0.589942478 0.945390318 1.071266773 9.43146E-08 1.60518E-10 9.43146E-05 1 6051BE-07 320.7967 0.9623901
T5GU-150 0.0642 0.021062992 5000 3 1.515 1.515 0.589942478 0.945390318 1.071266773 9.43146E-08 1 60518E-10 9.43146E-05 1 60518E-07 490 1.47
TSGU-200 0.0509 0.016699475 5000 3 1.669 1.669 0.649910228 1.0414394 1.180161218 9.43146E-08 1.60518E-10 9.43146E-05 1 60518E-07 630 1.39
TSGU-23 0.5284 0.1733S958 5000 3 0.812 0.812 0.316193592 0.50670425< 0.574170706 9.43146E-08 1 60518E-10 9.43146E-05 1.60518E-07 70.2772 0.2103316
TSGU-30 0.5284 0.17335958 5000 3 0.449 0.449 0.174841038 0.280184935 0.317490945 9.43146E-08 1.60518E-10 9.43146E-05 1.6051SE-07 70.2772 0.2108
TSGU-350 0.035 0.01148294 5000 3 0.449 0.449 0.174841038 0.280184985 0.317490945 9.43146E-08 1 60518E-10I 9.43146E-05 1 60518E-07 1300 3.9
TSGU-4 2.68 0.B79265092 3000 3 0.449 0.449 0.174841038 0.280184985 0.317490945 9.43146E-08 160518E-10 9.43146E-05 1.60518E-07 3.4607 0.0253821
T3GU-40 0.5284 0.17335958 5000 3 0.969 0.969 0.377329545 0.604675392 0.685186471 9.43146E-08 1.60518E-10 9.43146E-05 1.6051BE-07 130 0.39
TSGU-50 0.205 0.067257218 5000 3 0.969 0.969 0.377329545 0.604675392 0.685186471 9.43146E-08 1 60518E-10 9.43146E-05 1.60518E-07 158.1217 0.4743651
TSGU-60 0.205 0.067257218 5000 3 1.134 1.134 0.441580706 0.707638694 0.80185909 9.43146E-OB 1.60518E-10 9.43146E-05 1.6051BE-07 190 0.57
T5GU-75 0.129 0042322835 5000 3 1.134 1.134 0.441530706 0.707638694 0.80185909 9.43146E-08 1.60518E-10 9.43146E-05 1.60518E-07 239.4592 0.7183776
TSGU-9 1.06 0.347769029 3000 3 0.575 0.575 0.225905561 0.358311507 0.406586599 9.43146E-08 160518E-10 9.43146E-05 1.60518E-07 28 0.034
D5GU-23 0.5284 0.17335958 5000 2 0.781 0.781 0.304122162 0.48755963 0.552250596 9.43146E-0B 1.60518E-10 9.43146E-05 1.60518E-07 295.3333333 0.59066667
Cables are outdated and no data is available. Data assumed next larger or smaller cable for worse case resistance and weight
00
270 WMEC CABLE CHARACTERISTICS


CD
TRANSMISSION LINE CALCULATIONS


08
SO 8 0 IS S60S
so so IS UJ3-dl-LZ-l9-lJ GSOS
01 01 IS (l)S-dl-(3-l9-ll ssos
OS 0'S IS U)V-dl(3-l9-l) ssos
so so OS U)l-dl-(!-09-l) scot
t'O oso OS d-dl-(S-09-l) uoc
0S 0'S OS N-dl-l3-09-lJ clog
6*0 so os lN-dl-lS-09-U GOOt
OS OS OS WH-di-(;-09-0 eeos
VO OSO OS S-dl-(309-l) 0L0C
so so os d-dl-lS-09-U sooc
so 30 OS 3-dl-(3-09H) zooc
so 30 OS a-dl-l3-09-lJ sooc
VO oso OS a-di-ls-09-U 6003
so szo OS V-dl-(3-09d) sooc
VI VI ss UJn-n-te-es-i) tois
0 l O'l ss s-n-lz-ss-i) l-SOS
01 O'l GS -ni-LE-es-iJ osos
01 O'l GS H-ni-ls-es-iJ GSOS
01 O'l GS 3-HI-C3-63-1) ssos
O'l O'l GS a-n-te-es-i) scos
VQ OS oso IQ -d3-(£-iS-l) SSOS
VQ OS oso IQ 3-dS-(£-iS-l) iSOS
VQ 0S oso IQ a-d3-(£-iSd) ssos
VQ OS oso IQ V-d3-(£-i9H) ssos
30 so 030 IQ (l)d-dl (l iSd) SOIS
VO oso IQ -dl-(l-iS-l) scos
VO oso IQ -dl-(l-iS-l) LCOS
to oso IQ a-di-(i-is-i) ocos
VO oso IQ W-dl-(l-iSH) GSOS
OS OS SS s-dt-O-so-z) l-OOG
O'OL 001 SS d-dt-(l-S9-S) cooc
O'Gt OGt ss (s)3-dt-(l-29-S) OOOG
3Sl 3SI ss -dt-0-£9-s) coot
01 O'l ss 3-dt-(l-39-2) soot
et C't ss a-dt-0-£9-j) loot
01 O'L ss V-dt-(l-S9-Z) ooot
von imcra von 31VU
dH UdlVd aaawnN iinoaiD
Hi leionsuoz
zz zz it (l-6t-£)-dHlit-£) ££0t
sz sz it (S-8t-£)-dl-a-it£) OlOt
9EI. 3ZI it (3-it-£)-dl(l-it-£) t£Ot
za za sz (8)0-dt-(J-8£-l) 8308
to so sz (s)a-dt-(j-8£-i) 8808
80 OO'l LZ (l)0-dt-(S-i£-l) 3018
8 i 8 i LZ tl)3-dt-(£-i£-l) 1308
to OS'O LZ (i)a-dt-(j-i£-i) £808
a Ot'O LZ (s)y-dt-(s-i£-i) £808
to OS'O LZ [tJv-dHs-is-U 1808
Gtl O'SlfO'OS 3Z (l)3-dt-S£ 1003
Gtl ooz 3Z (i)a-dt-si £00£
avon avon
NN03 31VU dH UdlVU snrerrg aaawriN imaaia
SNOiivnnonvo NOiinaiuisia avcn aaz oa
O XI0N3ddV


8006 (02-62-11-1P-A 62 2.0 2.0
6007 f02-62-1l-1P-B 62 2.0 2.0
6010 f02-62-11-1P-Cf11 62 0.5 0.5
5062 f2-62-1l-1L-E 62 1.0 1.0
4034 (2-62-11-1L-L 62 0.5 0.5
5063 (2-62-11-1L-M 62 1.0 1.0
6100 (2-62-1J-1L-R 62 0.1 0.1
5042 (2-62-1J-1L-S 62 1.0 1.0
4030 (02-66-21-1L-AA 66 0.5 0.5
3001 (02-66-2)-1L-F 66 1.0 1.0
6033 (02-66-2)-1L-M(1) 66 0.2 0.2
4033 r02-66-2l-1L-M 66 0.5 0.5
5033 f02-66-2l-1L-X 66 1.0 1.0
4023 (02-66-21-1L-Z 66 0.5 0.5
4017 f02-67-21-1EP-f01-102-2l 67 0.3 0.3
4011 f02-67-2l-1EP-f02-65-21 67 8.0 8.0
4012 f02-67-2l-1EP-f02-67-41 67 4.8 4.8
4013 f02-67-2l-1EP-f02-73-11 67 2.3 2.3
4014 f02-67-21-1EP-f02-85-11 67 10.5 10.5
4015 f02-67-2l-1EP-f02-36-21 67 1.4 1.4
4016 (02-67-21-1EP-D 67 3.0 3.0
6013 f1-70-1l-1EP-Af2l 70 0.2 0.2
6011 f1-70-1l-1EP-Bf21 70 0.3 0.3
6012 f1-70-1l-1EP-Cf2l 70 0.4 0.4
6014 f1-70-1l-1EP-Df2l 70 0.2 0.2
6016 f1-70-1l-1EP-Ef2l 70 0.2 0.2
6015 f1-70-1l-1EP-Ff2l 70 0.1 0.1
6017 f1-70-1l-1EP-Gf2l 70 0.2 0.2
6013 f1-70-1l-1EP-Hf2l 70 0.4 0.4
6020 f1-70-1l-1EP-Jf2l 70 0.2 0.2
6018 ri-70-1l-1EP-Kf2l 70 0.3 0.3
6022 f1-70-1l-1EP-Lf2l 70 0.6 0.6
6021 f1-70-1l-1EP-Mf2l 70 0.5 0.5
6023 f1-70-1l-1EP-Wf2l 70 0.3 0.3
6024 f1-70-1l-1EP-Pf2l 70 0.3 0.8
6027 f1-70-1l-1EP-Rf2l 70 0.4 0.4
6025 f1-70-1l-1EP-Sf2l 70 0.3 0.3
6030 f1-70-1l-1EP-Tf2l 70 0.4 0.4
6023 f1-70-1l-1EP-Uf2l 70 0.1 0.1
6028 f1-70-1l-1EP-Vf2l 70 0.4 0.4
6115 f1-70-1l-1EP-Wf1l 70 0.50 0.5
I 4037 f01-71-1l-1EL-E 71 0.6 0.6
4038 f01-71-1l-1EL-H 71 0.6 0.6
6117 f01-71-1l-1EL-Kf1l 71 0.2 0.2
4027 (02-71-21-1EL-G 71 1.2 1.2
4008 (02-71-21-1EL-H 71 1.3 1.3
5051 f02-71-2l-1EL-Jf1l 71 1.0 1.0
5052 r02-71-2l-1EL-Lf1l 71 1.0 1.0


6072 f01-72-11-4EP-Af61 72 3.00 2.2
6050 r01-72-1l-4EP-Ar7l 72 22.0 22.0
6026 101-72-1HEP-A181 72 11.6 11.6
6074 f01-72-1l-4EP-Bf10l 72 0.50 0.4
6057 f01-72-1l-4EP-Bf3l 72 33.3 33.3
6073 roi-72-n-4EP-Brei 72 0.75 0.6
6075 101-72-11-4EP-C131 72 0.75 0.6
6112 101-72-1HEP-D121 72 0.20 0.1
6113 f01-72-11-4EP-Ef21 72 0.20 0.1
6114 f01-72-11-4EP-Ff21 72 0.50 0.4
3014 M-74-11-1EL-E 74 1.0 1.0
4005 H-74-11-1EL-K 74 0.2 0.2
5037 (01-80-1J-1L-D 80 1.0 1.0
2004 f01-80-11-1L-FT1l 80 0.8 0.8
5038 (01-80-1J-1L-R 80 1.0 1.0
4020 r02-81-2l-1P-f02-67-6l 81 2.0 2.0
4021 f02-81-21-1P-f02-79-21 81 6.3 6.3
I 4022 f02-81-21-1P-f02-83-21 81 25.1 25.1
| 4018 102-81-21-1P-B 81 1.3 1.3
4018 102-81-21-1P-C 81 3.0 3.0
I Zone 2 Total 306.3
CIRCUIT NUMBER Frame RATED HR RATED CONN
LOAD LOAD

3000 (1-32-4J-4P-A 82 3.0 2.6
3002 M-32-41-4P-B 82 10.0 10.0
3012 11-82-41-4P-C 82 13.0 13.0
3015 M-82-41-4P-D 82 18.0 13.0
6004 12-82-11-4P-Cm 82 3.00 2.2
6005 f2-82-3l-4P-Df1l 82 3.00 2.2
6058 f01-83-2l-1P-Af2l 83 1.0 1.0
6060 f01-83-2l-1P-Bf2l 83 1.0 1.0
6061 f01-83-2l-1P-Cf21 83 1.0 1.0
6062 f01-83-21-1P-Df21 83 0.5 0.5
6063 r01-83-2l-1P-Ef2l 83 0.5 0.5
6064 101-83-21-1P-F121 83 0.5 0.5
6106 101-83-21-1P-G 83 2.0 2.0
4028 1E-4EP-f02-83-1l 33 8.5 8.5
6002 f2-102-2l-4EP-Df1l 102 7.50 5.6
6000 (2-102-2J-4EP-F 102 10.0 7.5
6080 T2-102-21-4EP-JfH 102 5.0 3.7
6001 (3-102-2J-4EP-A 102 10.0 7.5
6003 r3-io2-2i-4EP-cm 102 7.50 5.6
5045 (2-103-2J-1L-F 103 1.0 1.0
4006 12-103-41-1EL-A 103 0.7 0.7
4004 12-103-41-1EL-J 103 0.2 0.2
Zone 3 Total 84.8


CIRCUIT NUMBER Frame RATED HP RATED LOAD CONN LOAD

6078 1E-4EP-Fm 106 10.0 7.5
6079 1E-4EP-MH1 106 10.00 7.5
6048 n-116-21-1EP-Bf6l 116 3.2 3.2
6055 M-116-21-1EP-Df6l 116 17.5 17.5
6097 11-116-21-1EP-M161 116 0.33 0.3
6047 f1-116-2l-4EP-Af1l 116 5.00 3.7
6091 f1-116-2l-4EP-Bf21 116 5.0 3.7
6085 f1-116-2l-4EP-Cm 116 2.0 1.5
6087 f1-116-2l-4EP-Df2l 116 2.00 1.5
6077 f1-116-2l-4EP-Ef1l 116 5.0 3.7
6076 f1-116-91-4EP-Ff11 116 5.0 3.7
6054 n-116-2l-4EP-Qf21 116 3.00 2.2
6056 f1-116-2l-4EP-Hf1l 116 8.0 8.0
6099 f1-116-2l-4EP-Jf5l 116 7.5 5.6
6058 M-116-21-4EP-Jf61 116 12.6 12.6
6096 n-116-2l-4EP-Mr4l 116 9.5 9.5
1004 f3-117-1l-4P-Am 117 9.0 9.0
1008 f3-117-1l-4P-Bf1l 117 12.0 12.0
1009 f3-117-1l-4P-Cm 117 3.0 2.2
1030 13-117-11-4P-D 117 9.0 9.0
1095 f3-117-11-4P-Ff11 117 10.0 7.5
1016 f3-117-1l-4P-Gf1l 117 2.00 0.4
1031 13-117-11-4P-J 117 1.5 1.5
1039 13-117-11-4P-K 117 9.0 9.0
1005 f3-117-91-4P-Af1l 117 9.0 9.0
1009 f3-117-2l-4P-Bf1l 117 12.0 12.0
1097 f3-117-9l-4P-Ef1l 117 10.0 7.5
1003 f3-117-91-4P-Ff11 117 3.0 2.2
1033 f3-117-21-4P-Jf21 117 10.0 7.5
1019 13-117-31-4EP-BM1 117 5.0 3.7
1024 13-117-31-4EP-D111 117 10.0 7.5
1013 13-117-41-4EP-A111 117 5.0 3.7
1026 f3-117-41-4EP-Bf11 117 10.0 7.5
2005 2S-4P-EM1 117 52.8 52.8
Zone 4 Total 255.7
CIRCUIT NUMBER RATED HP RATED LOAD CONN LOAD

3001 (01-118-4)-4P-A 118 10.0 10.0
8000 (01-118-4)-4P-C 118 1.0 0.3
6008 r01-118-4l-4P-D 118 13.3 13.3
6008 f01-118-4l-4P-E 118 13.3 13.3
6052 n-118-1l-4P-Ar2l 118 37.1 37.1
6034 n-ii8-ii-4P-Br3i 118 2.00 1.5
6030 n-ii8-ii-4P-cr3i 118 3.00 2.2
6033 n-ii8-ii-4P-cr5i 118 1.2 1.2
7006 1S-4P-E 113 3.4 3.4
6044 f1-120-11-1EF-Af11 120 1.4 1.4
6045 ri-i20-ii-iEP-em 120 1.4 1.4
6046 f1-120-1l-1EF-Cm 120 1.4 1.4
6103 fl-120-ll-IEP-Dfll 120 1.3 1.3
83


5040 11-125-1-1L-0 125 1.0 1.0
0030 (1-125-1)-1L-E(1) 125 0.2 0.1
5066 11-125-11-1L-F 125 1.5 1.5
5023 11-142-21-4P-A111 142 12.0 15.0
5024 f1-142-2l-4P-Af 21 142 2.00 1.7
5004 (1-144-1J-1P-A 144 0.75 0.6
5005 (1-144-1J-1P-B 144 0.5 0.4
5010 f1-144-1l-1P-C 144 0.25 0.2
5013 M-144-11-1P-E 144 0.25 0.2
5017 M-144-11-1P-F 144 0.25 0.2
5013 M-144-11-1P-G 144 0.25 0.2
5012 f1-144-11-1P-Jf11 144 1.0 1.0
5022 f1-144-11-1P-Mf11 144 0.75 0.6
1017 12-145-21-4P-Am 145 2.00 0.4
1020 12-145-21-4P-Bm 145 6.0 6.0
1021 12-145-21-4P-Cm 145 6.0 6.0
1022 12-145-21-4P-Dm 145 0.75 0.6
1023 12-145-21-4P-EM1 145 0.75 0.6
1015 (2-145-2)-4P-F(1) 145 6.6 6.6
1014 (2-145-2)-4P-G(1) 145 6.6 6.6
6101 T2-145-4-1EP-Af1l 145 1.00 0.8
6036 12-145-4-1EP-J121 145 0.10 0.1
4035 12-145-41-1EL-C 145 0.5 0.5
1035 12-145-41-1EL-K 145 1.3 1.3
1011 12-145-41-IEP-Dm 145 2.0 2.0
1006 f2-145-41-1EP-Ff11 145 0.50 0.4
1007 f2-145-41-1EP-Hf 11 145 0.50 0.4
1010 12-145-41-1EP-M111 145 2.0 2.0
4026 12-145-41-1EP-W 145 0.5 0.5
1013 12-145-41-1EP-P121 145 0.0 0.03
1013 12-145-41-1EP-P131 145 0.0 0.03
1023 12-145-41-1EP-R 145 2.0 2.0
8004 H-150-1-1EL-R 150 0.8 0.8
8002 (1-150-1J-1EL-V 150 0.3 0.3
3003 (1-150-1J-1EL-V 150 1.0 1.0
5075 M-150-11-1EL-X 150 0.2 0.2
6043 1E-1EP-F171 153 1.1 1.1
1028 12-164-21-1EP-D116:21 164 3.2 3.2
5016 11-165-21-1P-A 165 1.3 1.3
5011 11-165-21-1P-B 165 3.6 3.6
5003 11-165-21-1P-C 165 3.0 3.0
5008 M-165-21-1P-E 165 0.33 0.3
5007 H-165-21-1P-F 165 0.25 0.2
5034 (1-165-2J-1P-J 165 0.75 0.6
5013 M-165-41-1P-K 165 5.0 5.0
5000 (1-165-4J-4P-A 165 24.0 24.0
5001 (1-165-4J-4P-B 165 16.2 16.2
5002 (1-165-4J-4P-C 165 12.0 12.0
5003 (1-165-4J-4P-D 165 22.0 22.0
5021 11-165-41-4P-E 165 10.0 10.0
5006 (1-165-4J-4P-G 165 24.0 24.0
5035 (1-165-4)-4P-H(1) 165 20.4 20.4
Zone 5 Total 303.7
84


CIRCUIT NUMBER Frame RATED HP RATED LOAD CONN LOAD

5067 f4-16S-2l-1P-Af1l 163 1.0 1.0
5066 (4-163-2l-1P-Bf2l 166 1.0 1.0
5014 (4-166-2)-1P-D 163 0.50 0.4
5066 r4-166-2l-1P-Ff1l 163 0.33 0.75 1.0
5015 (4-169-2)-1P-G 163 0.33 0.3
5070 f4-169-2l-1P-Jf 11 163 0.33 0.75 1.0
6036 (2-175-1)-1EP-A(2) 175 0.4 0.4
6031 (2-175-1)-1EP-B(2) 175 0.4 0.4
6033 (2-175-1)-1EP-C(2) 175 0.2 0.2
6035 f2-175-1l-1EP-Df2l 175 0.2 0.2
6034 (2-175-1)-1EP-E(2) 175 0.3 0.3
6040 r2-175-1l-1EP-Ff2l 175 0.3 0.3
6033 f2-175-1l-1EP-Qf2l 175 0.2 0.2
6036 f2-175-1l-1EP-Hf2l 175 0.6 0.6
6038 f2-175-1l-1EP-Jf2l 175 0.7 0.7
6037 r2-175-1l-1EP-Kf2l 175 0.5 0.5
6042 f2-175-1l-1EP-Lf2l 175 0.5 0.5
6041 f2-175-1l-1EP-Mf2l 175 0.4 0.4
6043 f2-175-1l-1EP-Nf2l 175 0.8 0.3
5041 (1-186-1)-1L-B 186 1.0 1.0
5057 M-186-11-1L-E 186 1.5 1.5
5056 f1-1S6-11-1L-F 186 0.4 0.4
6107 f1-186-1l-1L-Hf1l 186 0.2 0.2
6103 M-186-11-1L-K 186 0.3 0.3
6066 f1-186-1l-1L-Mf2l 186 1.0 1.0
6070 n-186-1l-1L-IVir2l 186 1.0 1.0
6067 f1-186-1l-1L-Pf2l 186 1.0 1.0
6071 f1-186-1l-1L-Rf2l 186 1.0 1.0
4031 M-188-41-1P-A 188 1.5 1.5
4032 M-188-41-1P-C 183 0.5 0.5
3004 f4-161-11-4P-A 131 10.0 10.0
5071 f4-181-1l-4P-Ef1l 131 5.00 3.7
5072 f4-161-1l-4P-Ff1l 131 5.00 3.7
5076 f4-133-2l-4EP-Cf1l 133 5.00 3.7
5043 f2-166-1l-1L-F 136 1.0 1.0
5065 f2-186-11-1L-J 136 1.0 1.0
5044 f2-196-11-1L-K 136 1.0 1.0
5064 (2-136-11-1L-L 136 1.0 1.0
6063 f2-166-1l-1L-Sf1l 136 0.2 0.1
6064 f2-136-1l-1L-Sf1l 136 0.20 0.1
5073 (2-206-01-1EL-E 206 1.2 1.2
1034 (2-206-01-1EL-H 206 1.3 1.3
4007 (2-206-01-1EL-M 206 0.2 0.2
4025 (2-206-01-1EL-NM1 206 0.6 0.6
10002 (2-206-0)-1EL-T 206 4.8 4.8
10003 (2-206-0J-1EL-U 206 2.3 2.3
Zone 6 Total 56.4


CIRCUIT NUMBER Frame RATED HP RATED LOAD CONN LOAD

£003 12-226-1HP-D 226 25.0 13.6
5020 f2-22S-11-4P-Ffn 228 3.00 2.2
2000 1S-+P-A(1) 228 20.0M5.0 14.9
1000 1SHE-4P-H(3) 228 30.0 22.4
1001 2SHE-4P-K(3) 228 30.0 22.4

6109 12-231-11-4EP-A111 231 12.5 12.5
6110 12-231-1l-4EP-Bm 231 1.0 0.3
6111 12-231-11-4EP-CH1 231 0.8 0.6
Zone 7 Total 94.4
'
Total Load Safety Margin
Gen 1S 364.0 0.76632211
Gen 2S 364.003 0.76632211
Gen 1E 460.5 0.96953684




499.4 1.05136842





Zone Load
1 71.3
2 306.3
3 94.8
4 255.7
5 309.7
6 56.4
7 94.4
.
86


Switchboard Load Starting Frame Ending Frame Transverse Distance Decks between Vertical Distance Total Length () Length (km) Cable Designator Cable weight/ft R' L c
IS Forward 20.9 3 24 44.9 0.013686 T5GU-30 0.5284 9.43146E-05 1.605E-07
IS Amidships 20.9 3 24 44.9 0.013686 TSGU-30 0.5284 9.43146E-05 1.605E-07
IS Aft 20.9 3 24 44.9 0.013686 TSGU-30 0.5284 9.43146E-05 1.605E-07
2S Forward 18.4 3 24 42.4 0.012924 TSGU-30 0.5284 9.43146E-05 1.605E-07
2S Amidships 18.4 3 24 42.4 0.012924 TSGU-30 0.5284 9.43146E-05 1.605E-07
2S Aft 18.4 3 24 42.4 0.012924 TSGU-30 0.5284 9.43146E-05 1.605E-07
IE Forward 18.26 2 16 34.26 0.010442 TSGU-30 0.5284 9.43146E-05 1.605E-07
IE Amidships 0.26 2 16 16.26 0.004956 TSGU-30 0.5284 9.43146E-05 1.605E-07
IE Aft 18.26 2 16 34.26 0.010442 TSGU-30 0.5284 9.43146E-05 1.605E-07
Busi 30 200 30 3 24 224 0.068275 DSGU-9 0.056 1.06
Bus 2 30 200 30 3 24 224 0.068275 DSGU-9 0.056 1.06
00
DC ZED TRANSMISSION LINE CALCULATIONS


APPENDIX E
MATLAB CODE
0,0,00000000000000000000000000000000000000000000000
ooooooooooooooooooooooooooooooooooooooooooooooooo
%Keoni Hutton
%270' WMEC AC Distribution System Model
%Simulations Code
%MSEE Thesis
%Started: 17 Aug 15
%Last Edited: 17 Aug 15
Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-
ooooooooooooooooooooooooooooooooooooooooooooooooo
clear all
clc
% Run Simulation
%Run Trial 1: Tripping Generator CB While Underwa
%Set Distribution System Configuration Parameters
ACDS 270 Parameters Underway Steaming
DCDS 270 TL Parameters
Teleke Converter Parameters
%Set Trial Parameters
WMEC_Model_Testl_Open_Gen_CB
Time Start=tic;
tt=sim('TELEKE Converter RLoad');
Time Elapsed = toe(Time Start);
Time Hours = Time Elapsed/(3600)
save('DCDS Trial 1 Variables');
%Run Trial 2: Start Large Load while underway
clear all
Teleke Converter Parameters
%Set Distribution System Configuration Parameters
ACDS 270 Parameters Underway Steaming
DCDS 270 TL Parameters
%Set Trial Parameters
WMEC Model Test2 Start Large Load
Time Start=tic;
tt=sim('TELEKE Converter RLoad');
Time Elapsed = toe(Time Start);
Time Hours = Time Elapsed/(3600)
save('DCDS Trial 2 Variables');
%Run Trial 3: Battle Damage while Underway
clear all
Teleke Converter Parameters
%Set Distribution System Configuration Parameters
ACDS 270 Parameters Underway Steaming
DCDS 270 TL Parameters
%Set Trial Parameters
WMEC Model Test3 Battle Damage
Time Start=tic;


Full Text

PAGE 1

ALTERNATING VS DIRECT CURRENT : A TRANSIENT STUDY OF THE U.S. COAST SYSTEM by KEONI ALEXANDER HUTTON B.S., United States Coast Guard Academy, 2010 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Electrical Engineering 2016

PAGE 2

ii This thesis for the Master of Science degree by Keoni Alexander Hutton has been approved for the Electrical Engineering Program by Jae Do Park, Chair Fernando Mancilla David Miloje Radenkovic April 21 2016

PAGE 3

iii The views expressed herein are those of the author and are not to be construed as official or reflecting the views of the Commandant or of the U. S. Coast Guard.

PAGE 4

iv Hutton, Keoni Alexander (M.S., Electrical Engineering) Alternating v s Thesis directed by As sistant Professor Jae Do Park. ABSTRACT While t he United States Navy has conducted extensive research into t he use of shipboard DC zonal electrical distribution systems (ZED), no project has analyzed the benefits for installation on a Coast Guard cutter which has a significantly different loa d profile than Navy warships. Simulink model ium endurance cutter (WMEC) AC radial electrical distribution system and a proposed DC ZED system w ere created and tested with three transients. The result demonstrated a significant reduction in settling time and an increased robustness caused by the ins ulation provided by the introduction of power electronic converters. Beyond the transients, a DC ZED provides better standardized installation for any ship reducing construction costs and timeline s and simplifying training and support. Additionally, the DCZED by reducing longitudinal cables that penetrate watertight bulkheads increasing a boundaries effectiveness The Coast Guard would be best served by pushing for a zonal distribution system on all future cutter acquisitions. The form and content of this abstract are approved. I recommend its publication Approved: Jae Do Park

PAGE 5

v ACKNOWLEDGEMENTS The author would like to acknowledge the office of Naval Engineering CG 45, the staff at the Office of Mission Suppo rt Workforce Management DCMS 81 and the staff at the Denver, Colorado Coast Guard Recruiting Office for their assistance and support : LCDR Roger Robitaille, LT Lucas Marino, BMC Stanley Rittner, MK1 Krista Beck, AET1 Charles McKenzie, MST2 Whip Blacklaw and Ms. Mary Fuata. The author would also like to express gratitude to the staff and faculty of the University of Colorado Denver College of Engineering and Applied Science Electrical Engineering Department especially the committee members Dr. Jae Do P ark, Dr. Fernando Mancilla D avid, and Dr. Miloje Radenkovic as well as the University of Colorado Denver Alumni Association. Finally the author would like to express gratitude to his family and friends, especially his wife Ashley, without whose support, t his research could never have been completed.

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. .. .. Objectives .. II. AC DISTRIBUTION ARCHITECTURE AND MODELING .. WMEC Design Requireme .. III. DC DISTRIBUTION ARCHITECTURE AND MODELING 23 Mo Simul IV. SIMULATION RESULTS Trial

PAGE 7

vii Trial 6: Battle Damage .. Trial 7: Loss of a Trial 9: Battle Damage Simulation V. DISTRIBUTION SYSTEM COMPARISON 65 65 .. 67 69 VI. CONCL USIONS AND RECOMMENDATIONS 72 74 APPENDIX A. 78 B. C. DC ZED Load Distrib D. DC ZED Tr E.

PAGE 8

viii LIST OF TABLES TABLE 1. Statutory missions of th 2. Electric power system characteristics a 3. 4. Model consolid 5. Loads eliminated f rom model 6. 7. Propose 8. Trial 1 ACDS an 6 9. Trial 2 ACDS and .. 10. Trial 3 ACDS and DCD 42 11. 12. Trial 5 ACDS and DCDS 13. 14. 56 15. Trial 8 ACDS and D 16. Trial 9 ACDS and DCDS comparison summary 17 DDG 51 weight comparison between radi 66

PAGE 9

ix LIST OF FIGURES FIGURE 1. 2. 3. 4. 10 5. Simple galvanic cell in a lead 6. Three phase equivalent circuit model 7. Type DC1A DC commutato 8. 9. 10. Tra nsmission li 11. Simulink implementat 12. 13. VSC based rectifier for DC shipbo ard power system application... 14. Voltage source converter controller block diagram.. 15. Simulink implementation of DCDS loads (Underway Stea 31 16. Simulink implementation of DCDS loads (General Qu 17. Voltage and current for generator 1S and generator 2S .. 1 8 Generator 1S voltage before and after the transient introduction for ACDS ...

PAGE 10

x 1 9 Generator 2S voltage before and after the transient intr oduction for DCDS 20. Trial 1 Voltage meas 35 21. 22 7 2 3 Trial 2 Generator 1S and 2S voltage before and a 24. Trial 2 Generator 1S and 2S voltage before and after transient for DCDS 2 5 Trial 2 voltage measured at 8 2 6 Trial 2 voltage measured at 9 2 7 Voltage and current measured at selec 40 2 8 Trial 3 Generator 1S and 2S voltage before and after transient for ACDS .. 40 29. Trial 3 Genera 30 Trial 3 voltage measured at selected loads for ACDS 31. 32 Voltage and current for generator 1S, generator 2S, an 33 Trial 4 Generator 1S and 2S and 1E voltage before and after transient for ACDS 4 34. 35 Trial 4 voltage measured at selected loads for ACDS .45 36 Trial 4 voltage measured at selected loads for D CDS ... .45 37 3 8. Trial 5 generator 1S and 2S and 1E voltage bef ore and after transient for ACDS

PAGE 11

xi 3 9. Trial 5 generator 1S and 2S and 1E voltage bef ore and after transient for DCDS 40 Trial 5 voltage meas 41 Trial 5 voltage meas 4 2 Voltage and current measured at selected 43 Trial 6 generator 1S 2S and 1E voltage before and after transient for ACDS 44 Trial 6 generator 1S 2S and 1E voltage before and after transient for D CDS 45 Trial 6 voltage measured at selected lo 46 Trial 6 voltage measured at selected lo 47 Voltage and current for generator 1S, and g ... 53 48. Voltage at generator 1E for ACDS .. 54 49. Voltage at generator 1E for DCDS 54 50 Trial 7 voltage measured at selected loads for ACDS 55 51 Trial 7 voltage measured at selected loads for ACDS 55 52 Voltage and current measured at sel 53 Generator 1S voltage before and after transient for ACDS 54 Generator 1S voltage before and after transient for D CDS 55 Trial 8 voltage measured at selected loads for ACDS .. 56 Trial 8 voltage measured at selected loads for ACDS .. 57 Voltage and current mea 0 58 Generator 1S voltage before and after transient for

PAGE 12

xii 59 Generator 1S voltage before and after transient for 1 60 Trial 9 voltage measured at selected loads for ACDS. 1 61 Trial 9 voltage measured at selected loads for 62 63 ..

PAGE 13

1 CHAPTER I INTRODUCTION Background The United States Coast Guard was founded as the Revenue Marine on 4 August 1790 with the explicit purpose of enforcing tariff and trade laws to finance the creation of a new nation [1] From its humble origins with a f leet of ten cutters, the Coast Saving Service, the Lighthouse Service, the Steamboat Inspection Service, and the Bureau of Marine Inspection and Navigation [1]. Today the Coast Guard operates a fleet of 90 cutters and patrol craft in addition to over 1800 boats [2] in support of the eleven statutory missions displayed in Table 1: Table 1 : Statutory m issions of the U.S. Coast Guard [3] As the requirements for its cutter fleet. s feature sophisticated command, control, communications, computers, intelligence, surveillance and reconnaissance (C4ISR) systems and wea pons systems requiring seamless interoperability with other Department of Homeland Security (DHS) and Department of Defense (DOD) assets [4]. T hese C4ISR systems when coupled with the ever expanding use of technology for personal and professional matters a re driving up the electrical load requirements for Coast Guard cutters. The U.S. Naval Sea Services Command describes the importance :

PAGE 14

2 s fighting and functional effectiveness. Electric power trains [and] elevate [s] gun turrets and missile launchers; operate [s] [s] auxiliaries; provide [s] light; and power [s] interior communication, weapons control, radio, radar, sonar, and missile systems. A ship without electric power is useless as a fighting or supporting unit and is almost totally d e fenseless against enemy attack [6] The electric distribution system on board a ship must be able to provide six basic f unctions: 1. Control: The functional element of the power system that coordinates the other functional elements. [7 ][ 8 ] 2. Power Generation: Converts fuel into electrical power. In the Coast Guard, this is primarily accomplished with a generator utilizing eith er a diesel engine or gas turbine as the prime mover. [7][8] 3. Power Distribution: The ability of the system to move the power generated to the functional loads that will utilize the power. The distribution system consists of cables, switches, and fault pro tection equipment. [7][8] 4. Power Conversion: The ability of the system to convert the power generated into an acceptable form for each load. Three types of shipboard power are defined by the NSTM 320: [6][7][8] Table 2: Electric p ower s ystem c haracteristi cs at the user i nterface [6] 5. Energy Storage: The ability of a system to store energy that is generated to be used during a loss of primary generation capability [7][8]. 6. Utilization: The ultimate purpose of the distribution system, the delivery of the generated power to the user. Typical loads consist of lighting,

PAGE 15

3 communications and navigation gear, weapons systems, motors for pumps, galley equipment and personal use by the crew. [7][8] These basic functions apply to all ships from the tugboat in the harbor to the cruise ship sailing around the world. The cutters of the U.S. Coast Guard, however, are expected to continue performing in extremely hazardous conditions up to and including fighting a war. Therefore, the cutters must be designed to co ntinue to operate even after taking heavy damage [7]. To this end, the loads on a cutter are characterized based on their importa its mission. These loads are broken down into vital, semi vital and non vital categori es [7] Vital loads will be connected to two sources of power through an automatic b us transfer (ABT) switch. The ABT will automatically switch the power source feeding the load when the primary power source sources, but the bus transfer switch will be operated manually [7]. Semi vital and non vital loads are fed by one power source and are only characterized differently for simplicity in the event load shedding becomes necessary [7]. The ultimate goal being to [9]. Two major architecture categories have emerged to meet these survivability requirements: conventiona l, or radial, architecture and a zonal architecture. Today, every U.S. Coast Guard cutter employs a radial ACDS architecture where a generator provides 450 VAC 60 Hz Type I then distributes the power to a load center which distributes the power to the associated equipment. can be fed by its own generator, or through the connection of bus ties, by any other generator [9] [10] The Type I power provided by the generators ar e also used to run a motor generator set that produces 400 Hz Type II

PAGE 16

4 or Type III power. Conversely, the zonal architecture consists of two power distribution busses that run the length of the ship. The two busses should be separated port and starboard a nd vertically as far as possible to provide maximum survivability [10]. Each generator is connected directly to the main bus and feeds its power to the bus. The ship is then broken down into zones utilizing existing water tight boundaries. Each zone wou ld contain two load centers to feed the associated equipment in that zone only [10]. Figure 1 provides a comparison of a radial and zonal distribution architecture for a typical ship. Figure 1: Comparison of zonal and r adial d istribution a rchitecture [1 0]. Scope of Research Over the last 25 years, the U.S. Navy has conducted multiple studies into the benefits associated with a Direct Current Zonal Electrical Distribution System (DC ZEDS ) for the next generation of war fighting ships and detailed numerou s benefits that will be

PAGE 17

5 expounded upon in Chapter 5 of this Thesis. However, no U.S. Coast Guard project has examined the implications of this technology specifically for Coast Guard use In the late 90s, t he Coast Guard embarked on a $21.1 billion acqu isition progr am designed to recapitalize aging cutters and prepare the surface fleet for continued operations in the twenty first century [2][4]. As of January 2015, the Coast Guard received delivery of four National Security Cutters, with four more in de velopment and ten Fast Response Cutters with 24 more in development [4]. The requirements for these vessels are well defined and locked in to their respective acquisition contracts. Additionally, the Coast Guard is planning for the acquisition of up to 2 5 Offshore Patrol Cutters (OPC). A preliminary design contract w as awarded in February 2014 to three different vendors to design the OPC then evaluate the proposed designs and award a final contract to one vendor to construct the vessels [11]. The requirements for the OPC are still generally defined and are being interpreted differently by each of the preliminary design contractors; thus, t he OPC program could benefit most from an analysis of potential elec trical distribution architectures. Medium Endurance Cutters (WMEC) and accomplish the same mission set This Thesis ystem its missions, and design requirements and use it as a case study for a hypothetical DCZED allowing for a direct comparison between the system architectures based on the actual characteristics of a well defined, actively serving system. The Coast G fleet consists of 13 cutters built between 1979 and 1989. cs are shown in Table 3

PAGE 18

6 p rinciple c haracteristics [5] Ser vice Diesel Generators and an Emergency Dies el Generator connected to an AC radial distribution system (ACDS) that feeds 51 load centers throughout the ship [12] Fig ure 2 s implified l ine d iagram

PAGE 19

7 Objectives The objectives of this Thesis are as follows: 1. Examine the existing electrical distribution system of the Famous Class Medium Endurance Cutter and outline the current load requirements. 2. Utilize Matlab Simulink modeling software to create a virtual model of the Famous Class Cutter's ACDS. Simulate common casualties in a variety of plant and the requirements outlined by Coast Guard policy 3. Design a DC ZED to meet the requirements outlined by the current load configuration and future requirements outlined in the NSC and OPC acquisition documents. Simulate the same casualties in the same plant configurations as the ACDS simulations. 4. Compare and evaluate the DC ZED model against the model for the e xisting AC configuration using the following criteria: a. Transient Response b. Weight c. Standardization d. Survivability Organization of Thesis and the design requirements that were met during its construction. It will then discuss building the model of the ACDS and the development of the simulation trials that both models were tested under. Chapter 3 will outline the design requirements for the DC ZED and the Simulink model co nstruction. Finally, it will discuss the simulation trials and problems encountered. Chapter 4 will examine the simulation results for the transient study of the ACDS and DCDS trials. Chapter 5 will examine and compare the weight, survivability and stand ardization potential of the DCZED system. Chapter 6 outlines the analysis conclusions and recommendations for the OPC acquisition and future work

PAGE 20

8 CHAPTER II AC DISTRIBUTION ARCHITECTURE AND MODELING As mentioned previous ly, Coast Guard vessels are designed to continue to operate despite faults, failures and damage. In order to accomplish this feat, Coast Guard vessel design is governed by numerous publications produced by the Department of Homeland Security and Departmen t of Defense. The electric plant of a military ship must meet the requirements set forth in the Naval Ships Technical Manual Chapter 300, Electric Plant General Chapter 310 Electric Power Generators and Conversion and Chapter 320 Electric Power Distribut ion Systems Those requirements will be discussed in depth in this chapter. Floating Ground Coast Guard e lectrical d istribution s ystems are ungrounded systems in that there rd a ship, system will provide limited current during a single phase to ground fault thus allowing equipment to continue to operate throughout the fault condition [13]. Thi s is most easily seen at any 115 V receptacle where both prongs have a potential of approximately 60 Vrms relative to ground as seen in Figure 3. Figure 3: Ashore and shipboard receptacle p otential [13]

PAGE 21

9 A shipboard ungrounded system will contain s tray resistance and capacitance relative to ground caused by electrical equipment and cables in the system. Inherent system resistance is the parallel sum of the inherent resistance caused by insulation around generator, cable and load components. Simila rly, the inherent system capacitance is the parallel sum of the inherent capacitance of generators, cables and loads relative to ground [13]. While these are not physical components, the inherent system resistance and capacitances can create a return path between the ground and the conductor if a sailor comes into contact with a live conductor as in Figure 4. Figure 4: Realistic s hipboard u ngrounded s ystem [13] Beyond the survivability benefits of an ungrounded system, avoiding ground currents helps to prevent and minimize electrolytic [14] for any type of corrosion to occur, four components must be present to create a galvanic cell: an anode, a cathode, a metallic path that allows electron flow, and an electrolytic path that allows ionic flow [14]. Figure 5 illustrates the galvanic cell as a basic lead acid battery:

PAGE 22

1 0 Figure 5: Simple galvanic cell in a lead acid battery On board the WMEC, the electrolyte is created by the sea water upon which the ship sails The anode and cathode can be the ship s hull and any underwater fitting (propellers, rudders, transducer etc. ). The only missing component is the electron flow circuit path is often high enough to restrict the current to a level insufficient to activate circuit In this case, the ground must be found and isolated as soon as possible. Cable Insulation In order to reduce the shock and grounding ris k associated with the ungrounded system, all cables must be properly insulated based on compatibility with other insulators and the operating environment of the cable or equipment [13]. Cables approved for use on board Coast Guard cutters are listed in M ilitary Standard 242J [31] The cables

PAGE 23

11 Appendix A. Power Generation Type I and Type II. Type I power is produced by a 450 volt 60 hertz three phase machine with either a Y connected or delta connected stator [25]. The generator is excited by a direct current revolving field type rotor with a salient pole. The DC current is created through a rotating rect ifier connected to the exciter armature [26]. The prime mover is directly connected to the generator rotor. Three generator parameters must be controlled: frequency, voltage and load division [25] : Frequency : Frequency depends on the speed of the prime m over and is controlled by a mechanical hydraulic governor that is set by the generator watch stander through an adjustment potentiometer Voltage : Voltage is controlled through a voltage regulator. The voltage regulator al voltage and ensure it does not exceed a preset value durin g operation or load changes Load Division : The division of the electrical load between the generators operating in parallel is controlle These characteristics are constantly monitored by watch standers through voltage, current, power switchboards and the prime movers. Generator Protection Overcurrent protection is provided through a circuit break er connecting the generator to its service switchboard. Significant damage can occur when generators are

PAGE 24

12 operated in parallel if one generator takes the load and forces the other into motoring mode. To prevent this occurrence, the generator circuit break er is equipped with reverse power protection which trips the generator breaker if it senses a reversal of the power flow [25]. Load Characteristics The electric load of the 270 WMEC consists of a wide variety of equipment that is used (or not) during a w ide variety of situations. The largest loads on the distribution system consist of purely resistive loads such as heaters, and highly inductive loads such as the motor generator sets that produce 400 Hz Type II power [20]. The power factor of the system is tightly controlled by the generator watch stander to maintain a .8 lagging power factor. Thus, the overall system tends to act as an inductive load. ACDS Simulink Model In this thesis, the SimPower Systems toolbox of the MATLAB Simulink software suit e has been utilized to construct the model and run the simulations. Synchronous Generator Electrical Model [15] For ease of analysis, each pole, perfectly round rotor, synchronous machine driven by a diesel engine. The electrical frequency produced by a synchronous machine is related to the mechanical rotational speed of the rotor as seen in ( 1 ) : (1) Where n m denotes the mechanical speed of the rotor, and P is the number of poles in the machine. Thus, to produce the standard 60 Hz frequency, the generator must spin at a constant 3600 RPM The mechanical speed is controll ed by a governor that is

PAGE 25

13 manually controlled by the generator watch stander. The internal phase voltage produced by the generator is given by (2) : (2) where N c represents the number of turns of wire in the stator coil, is the flux in the machine and f is the frequency of rotation. The terminal phase voltage of the ge nerator V will differ from the i nternal voltage E A due to the following factors [15]: 1. The armature reaction in the air gap magnetic field 2. The self induc tance of the armature coils 3. The resistance of the armature coils The armature reaction induces a second voltage in the stator coils 90 behind and direct ly proportional to the current I A Thus, the armature reaction can be modeled as an inductor in series with the voltage E A The self inductance and resistance of the armature coils are similarly proportional to the armature current and can be modeled by an inductor and resistor in series wit h the armature reaction. The 3 phase equivalent circuit of the sy nchronous generator is thus given in Figure 6 : Figure 6: Three Phase e quivalent c ircuit m odel of s ynchronous g enerator.

PAGE 26

14 The voltage of each phase can then be calculated using (3) : (3) The output active and reactive power measured at the terminals of the generator can be expressed as in (4) and (5) respectively : (4) (5) The output power of the generator is directly related to the input power provided to the generator by the diesel engine prime mover, but mechanical and electrical losses must be accounted for. The input mechanical power is calcu lated by (6) : (6) From this power, the machine s uffers friction and windage losses, core losses and other stray losses that reduces the mechanical power delivered to the generator After the mechanical power is converted, the generator windings will produce additional copper losses that further reduce the output power calculated as in ( 4 ) [15]. Generator Excitation [19] not well known due to the age and lack of publicly available documentation. The DC1A excitation model is utilized to represent a field controlled dc commutator exciter with a continuously acting voltage regulator thanks to its wide implementation throughout the Electrical Engineering indu stry and simplicity of use and is modeled by the following block diagram: Figure 7 : Type DC 1A DC commutator exciter block diagram [19]

PAGE 27

15 T he terminal voltage transducer and load compensator output ( V c ) is subtracted from a reference voltage ( V ref ) while the stabilizing feedback ( V f ) is subtracted and the system stabilizing signal ( V s ) is added to create an error signal. The major time constant ( T A ) and gain ( K A ) incorporate limits representing saturation or power supply limitations. The regulator output, ( V r ) controls the exciter and allows for a self excited shunt field when K e = 0 which accou nts for the shunt field rheostat setting, and is automatically calculated by MATLAB V x accounts fo r saturation in the exciter The complete generator model implemented in Simulink can be seen in Figure 8 Figure 8 s ynchronous g enerator Load Model directly from [20] In this thesis, the load centers were characterized as either purely resistive or as RL loads: 1. Purel y resistive loads. These loads consist primarily of lighting and heating circuits that contain little or no inductive responses. 2. WMEC. It contains rotating equipment like the p umps for hydraulic, lubrication, and damage control systems. 20 07 that analyzed the load draw during in port and underway conditions

PAGE 28

16 [20] These measurements formed the basis of the analysis. In this work however, t he data set is incomplete in that it only contain s the measured active power for each load center. Reactive power on board USCG ships is tightly controlled by the Generator Watch Stander to mainta in power factor = .8 Lagging. As such, all reactive powers in the model is assumed using the following relations : (11) (12) (13) Solving (11) for and then plugging the result into (12) yields: (1 4 ) Modeling all 51 load centers created an unpalatable Simulink model that took weeks to run to completion. As such, the load centers were combined into a single RL load based on the switchboard fe e d ing it, and the geographic location within the ship. The nine resulting loads are summarized in Table 4. Table 4: Model c onsolidated l oad s pecifications [20] At no time would the ship actually be running at maximum capacity. The numbers in Table 4 r epresent the maximum possible load that could be reached. Additionally, five loads were eliminated from the study due to their infrequent use. The loads that were eliminated are shown in Table 5.

PAGE 29

17 Table 5: Loads e liminated from m odel c onsideration Every load was fitted with a circuit breaker that would allow it to be turned on or off depending on the parameters of the simulation. The load model created for the Simulink model can be seen in Figure 9 Additional resistors were placed in parallel wi th the RL loads to eliminate e rrors associated with having an inductor and a synchronous machine connected in series (through the ground) Figure 9 : Simulink implementation of load Line Model The transmission line model used in this thesis comes from Be Power System Analysis [21] The model uses three elements to capture the dynamics of a transmission line. A resistor captures the heat caused by current circulation A n inductor models the magnetic field T he shunt capacitive effect c aused by the potential relative to the ground beneath the transmission line is modeled as a capacitor. All distributed parameters were drawn from the data sheets for cables as identified in Appendix A. Figure 1 0 shows the per phase equivalent circuit of a transmission line.

PAGE 30

18 Figure 1 0 : Transmission l ine equivalent circuit [21] E xamining a section of the entire transmission line of length dx and applying Kirchhoff s Voltage Law and Current Law yields the following equations: (1 5 ) (1 6 ) Solving ( 1 5 ) and ( 1 6 ) yields : (1 7 ) (1 8 ) Where Z c represents the characteristic impedance of the line and is the propagation constant: (1 9 ) ( 20 ) all shunt admittances were neglected [21 ] Additionally, the resistive elements of the t ransmission line would only increase the active power used and is captured within the load model Transmission line lengths were calculated using the following assumptions:

PAGE 31

19 1. th ree 2. The emergency diesel generator terminals are located at frame 197 on deck three 3. The fore/aft length of a transmission line is the difference between the frames between the power source and the load center. 4. The athwartships length of a transmission li ne is 9.5 feet each if the source/load is port or starboard of the centerline or 19 feet if the load/source is on the centerline. 5. There are 8 feet vertically between decks. 6. The distance of each load in a geographic region of a ship were averaged together t o determine the distance. 7. All cables feeding a composite load are assumed to be TSGU 30, to provide the greatest dynamic effects. Figure 1 1 shows the final transmission line model utilized for the Simulink model: Figure 1 1 : Simulink implementation of tra nsmission line Detailed calculations used to determine the transmission line values can be found in Appendix B.

PAGE 32

20 Simulink Simulation U sing the ACDS Model The ACDS model was tested against three common transients in three plant configurations. Plant Confi gurations Manual and unit level standing orders. The Engineering Department on board the US CGC MOHAWK (WMEC 913) provided valuable insight to determine the requirements for the plan t co nfigurations to test: 1. Underway Steaming This is the most common plant configuration when the and both bus ties are closed so the two generators are sharing the load. T he Emergency Diesel Generator is placed in standby and will start automatically if it senses a loss of power at its switchboard [23][24]. 2. General Quarters This plant configuration is set during high risk situations such as operating near shore or dur ing battle conditions. All three generators are online and both bus ties are open so each generator only supplies the loads on its own switchboard. This ensures maximum survivability in case of damage [23][24]. 3. Anchor This plant configuration is set during low risk situations such as general steaming at night when the load is low, or while at anchor. Only one generator, either the 1S or 2S, is online and supplying all the electrical demand. The offline ships service generator is typically cold iron, meaning it would take several minutes before it could supply power to a load. The Emergency Diesel

PAGE 33

21 Generator is placed in standby and will start automatically if it senses a loss of power at its switchboard [23][24] Table 6 summarizes the load during ea ch of these plant configurations in k W : Table 6: Total p ower ( k W) d uring e ach p lant c onfiguration [20][24] Transient Sources The transients chose n to test against are either common for the system, or notably severe. 1. Loss of Generator: This tr ansient is tested by allowing the system to reach a steady state, and then opening a generator circuit breaker, disconnecting it from the distribution network. The loss of a generator could occur for reasons as complex as, mechanical failure or as simple as a mistake by a watch stander. During this type of event, if another generator is online, it would take the load. If another generator is not already online, the emergency generator would start up to restore power. 2. Starting a Load: The electrical l throughout the day as machinery is brought online and taken off line in the normal course of duty. This transient is tested by allowing the system to reach a steady state and then closing the circuit breaker to the aft 2S switchboard loads causing a sudden increase of 30 40 k W. 3. Battle Damage. Potentially the most disastrous transient the electric distribution system could face. This transient is tested by allowing the system to reach a

PAGE 34

22 steady state and then op ening the circuit breakers for all the forward loads causing a sudden drop in the load between 30 300 k W (depending on configuration) This analysis seeks only to examine the transients caused by this sudden shift in the load, and as such, the operation o f automatic bus transfers and/ or system reconfiguration to restore power go beyond the scope of this analysis and are not considered

PAGE 35

23 CHAPTER III DC DISTRIBUTION ARCHITECTURE AND MODELING DCDS Design The zonal electric distribution system in this study was constructed within the following constraints that were used in a similar U.S. Navy study [10] : 1. Electrical zones are delineated based on existing watertight boundaries on the zones between watertight bulkheads had a lo ad smaller than 50 kW they were co mbined with adjacent zones to create a larger load. 2. The maximum electrical load within a zone should not exceed the trip coil setting of the largest circuit breaker currently available within government st ock 3. The e lectrical load of a zone should permit the use of standard ized distributi on equipment. In order to create a direct comparison between the distribution architectures the following assumptions were used: 1. Generation equipment (prime movers, generators, c ontrols, and switchboards) remain in their original locations on the vessel. 2. Distribution feeders after the zonal load center remain as originally constructed. 3. Shore power and casualty power systems are not affected 4. Vital loads are fed from al ternate buses within the same zone but bus transfer equipment is the same as the ACDS model. Utilizing these assumptions and rules as outlined produces a distribution system architecture as seen in figure 1 2 :

PAGE 36

24 Figure 1 2 : 270 WMEC proposed DC Zonal Electri cal Distribution System The total load (expressed in kW) found in each zone is outlined in Table 7. Full calculations can be found in A ppendix C Table 7: Proposed zonal load breakdown Modeling of DCDS Generator Model The ultimate purpose of this res earch is to provide a direct comparison between hypothetical DC distribution system that is designed under the assumptions and constraints listed in Chapter 3.1. To this end, the ACDS model outlined in Chapter 2.2 and 2.3 is kept intact as much as possible for use in the DCDS model. The Generator

PAGE 37

25 model is taken directly from the ACDS Simulink model and only the addition of power electronic converters between each generator and the DC bus changes the nature of the distribution system. Voltage Source Converter Model [27] [30] three generators into a DC voltage/current for transmission throughout the ship on the DC bus. Diode and thyristor rectifiers have been popular but can inject harmonic or reactive currents into the bus. To avoid this phenomenon, a pulse width modulated (PWM) voltage source converter (VSC) is used to convert the AC voltage/c urrent from the generators to DC for transmission on the bus and will be im plemented at each generator as seen in Figure 13 The VSC is controlled using the vector control technique which transfers the three phase AC quantities onto a synchronously rotati ng reference frame (abc to dq) and controls it as if it were a DC quantity. By aligning the d axis of the reference frame with the voltage vector, the resulting current components will directly control the active and reactive power flow through the conver ter allowing for decoupled control of the power components [28] Additionally, the DC components integrate easily with a proportional integrator (PI) controller which ensures a fast rise time, a bounded input bounded output stability, and no error once th e transient has passed and the resulting quantity has reached steady state The VSC model for thi s project is shown in Figure 1 3

PAGE 38

26 Figure 1 3 : VSC b ased r ectifier for DC s hipboard p ower s ystem a pplication [ 30 ] The VSC is connected to a balanced three pha se system. The voltage and currents of the system are expressed as ( 21 ) ( 22 ) Where v a v b and v c are the voltages produced in each phase at the source and, i a i b and i c are the phase currents at the input of the VSC. The VSC can then be m odeled by the following set of differential equations: (2 3 ) (2 4 ) (2 5 ) (2 6 )

PAGE 39

27 Where R is the series resistance in each phase, L is the inductance of the VSC input filter, C is the capacitance of the DC Link capacitor, i load is the current fed to the load, and V dc is the voltage at the DC link. S a S b and S c are the switching functions of the IGBTs on each of the p hase legs. Note that the legs for each phase are independent of the others, though the two switches in the phase operate opposite one another. That is, when S 1 is on, S 4 is off. Similarly, when S 3 is on S 6 is off and when S 5 is on, S 2 is off. Applying the Park transform to the VSC model yields the equations in the dq domain: (2 7 ) (2 8 ) (2 9 ) Where i d and i q are components of the input currents, v d and v q are the components of the input voltage, is the angular frequency of the rotating reference frame and S d and S q a re the relative components of the switching functions. The power transferred through the VSC can be found by ( 30 ) ( 31 ) As mentioned earlier, when the d axis of the reference frame is aligned with the v d vector, the v q vector will be equal to 0 and the power equations simplify to ( 32 ) (3 3 )

PAGE 40

28 Thus, the active power is proportional to the direct component of the input current, and the reactive power is proportional to the quadrature component of the input curr ent allowing for decoupled control of the active and reactive power. Voltage Source Converter Controller [27] The controller for the VSC consists of three PI controllers operating in the DQ domain. One for the DC link voltage, one for the direct current component and the last for the quadrature current component. Solving equations 24 and 25 for the voltage components yields (3 4 ) (3 5 ) V q =0 since t he d axis of the reference frame is aligned with the v d vector. The switching functions are replaced by (3 6 ) (3 7 ) The voltage equations each contain a cross coupling term that depends on the opposite current from that which is b eing controlled. The controller takes this into account by adding in the cross coupled terms after the PI controller. These controller equations result in the control block diagram seen in Figure 1 4 :

PAGE 41

29 Figure 1 4 : Voltage s ource c onverter control ler block diagram [27] The current PI controllers have a transfer function of the form [2 9 ] (3 8 ) Where K p represents the proportional gain of the controller and K i represents the integral gain. The transfer function of the VSC is of the form (3 9 ) This results in a controlled system with the transfer function ( 40 ) t the following design goals [29 ]: state error 2. A bounded input results in a bounded output (BIBO stability)

PAGE 42

30 3 The system has a large bandwidth To meet these goals, the following rules must be enforced [2 9 ] : 1. T(0) = 2. Gain Margin must be larger than 6 dB. 3. Phase Margin must be larger than 45. 4. The Phase Inversion Frequency must be larger than the Gain Crossover Frequency. These four rules result in the following gains for the direct and quadrature current controllers: ( 41 ) ( 42 ) Similarly, the gains for the DC Link voltage PI controller are as follows (4 3 ) (4 4 ) Where is the damping ratio and n is the undamped natural frequency both of which can be manipulated to control the maximum overshoot and settling time. V d is the peak of the supply voltage and C is the DC L ink capacitance. Load Model The DCDS model proved to be extremely volatile in the MATLAB Simulink environment In order to eliminate numerous MATLAB Simulink errors, the DCDS load

PAGE 43

31 model was simplified to a larger degree than the ACDS model. The forward amidships and aft loads fed by each generator in the ACDS model were combined into a single RL load with their transmission line parameters. The resistive load is calculated using equations 45 47. (45) (46) (47) The inductive part of the load is calculated using the transmission line parameters as in Appendix D. The DCDS load is shown in Figure 15. Figure 15: Simulink implementation of DCDS loads (Underway Steaming and Anchor configuration). To capture the isolated nature of the system in Gener al Quarters configuration, a separate model was created in which each load is connected to its generator only. The Simulink implementation of the GQ configuration is shown in Figure 16.

PAGE 44

32 Figure 16: Simulink implementation of DCDS loads (General Quarters configuration). Simulink S imulation U sing the D CDS M odel To provide a direct comparison between the ACDS and the DCDS models, the DCDS model is tested using the same plant configurations and transients as outlined in Chapter 2.3 which results in nine tra ils. Plant Configurations 1. Underway Steaming 2. General Quarters 3. Anchor Transient Sources 1. Loss of Generator 2. Starting a Large Load 3. Battle Damage

PAGE 45

33 CHAPTER IV SIMULATION RESULTS Each Simulink model was tested with three transients under th ree plant configurations resulting in nine total trials. In this chapter the results of each simulation trial are presented side by side in order to more clearly analyze the differences between the two simulations The graphs presented are narrowed aroun d the time the transient was introduced and the time scale was limited to the extent of the transient recovery. The configurations are evaluated based on the design requirements outlined in NSTM 320 [6]. 1. Nominal voltage is 440 V. 2. Average of the three phase line to line voltages must not exceed 5% of nominal. 3. Maximum voltage transient is 16% of nominal. 4. Voltage transient recovery takes less than 2 seconds. 5. Voltage spike does not exceed 2500 V. Trial 1: Loss of a Generator While Under way Trial 1 tests the effect of a generator circuit breaker being tripped open while the system is configured for general underway steaming with two generators online and splitting the load evenly. Figure 1 7 shows the transient inducing condition of the g enerator tripping offline. The ACDS transient was in troduced six seconds into the simulation while the DCDS transient was in troduced one second into the simulation

PAGE 46

34 Figure 1 7 : Voltage (left) and current (right) for generator 1S (top) and ge nerator 2S (bottom) Figure 1 8 : Generator 1S voltage before and after the t ransient introduction for ACDS.

PAGE 47

35 Figure 19: Generator 2S voltage before and after the transient introduction for DCDS. The ACDS generator suffered a 2.1 % voltage drop when the 1S generator was tripped offline, then recovered to nominal voltage after 1.8 seconds. The DCDS generators suffered a 2804 V spike when the generator was disconnected, but did not experience any change in the voltage. Fig ure 20: Trial 1 Voltage meas ured at selected loads for ACDS.

PAGE 48

36 Figure 21 : Trial 1 Voltage measured at selected loads for DC DS. The ACDS loads suffered a 2.1% voltage drop when the generator tripped offline then took 1.7 seconds to recover. The DC DS load voltage oscillated for .64 seconds before returning to nominal. The oscillation peaked 20.9% below nominal value. Table 8: Trial 1 ACDS and DCDS comparison summary The loss of a generator causes a significant transient to be introduced to the system. The introduction of power electronic converters allowed the loads to recover three times faster than when the generator supplies the loads directly and insulated the generator from the transient However, the DCDS system transient was ten tim es greater than the ACDS system transient and oscillated Additionally, the DCDS generator suffered from a voltage spike nearly twice as high as allowed by NSTM 320.

PAGE 49

37 Trial 2: Starting a Large Load While Underway Trial 2 tests the effect of a large load b eing started while the system is configured for general underway steaming with two generators online and splitting the load evenly. Figure 20 shows the transient inducing condition. The ACDS transient was induced six seconds into the simulation while the DCDS transient was induced one second into the simulation. Figure 2 2 : Voltage (left) and current (right) measured at selected loads during Trial 2. Figure 2 3: Trial 2 Generator 1S and 2S voltage before and after transient for ACDS.

PAGE 50

38 The ACDS generator voltage drops .16% from nominal before recovering after 1.9 seconds. The DCDS generator voltage remained steady throughout the transient seen on the loads. Figure 24: Trial 2 Generator 1S and 2S voltage before and after tran sient for DCDS. Figure 2 5 : Trial 2 v oltage measured at selected loads for ACDS

PAGE 51

39 Figure 26: Trial 2 voltage measured at selected loads for DCDS The existing ACDS load voltage dropped .1% when the aft load breaker was closed before recovering after 1.9 seconds. The DCDS load voltage oscillated to a peak of 31.7% from nominal when the new load was started, and came back to nominal after 47 seconds. Table 9: Trial 2 ACDS and DCDS comparison summary When the load was started, t he ACDS system remained fairly steady suffering a tiny fraction of a voltage change. The DCDS conversely, saw an oscillating voltage that lasted .47 seconds before returning to nominal. Despite the significant oscillation, the recovery time of the DCDS w as four times faster.

PAGE 52

40 Trial 3: Battle Damage While Underway Trial 3 is meant to simulate the system experiencing damage by losing the large forward load while the system is configured for general underway steaming with two generators online and splitting the load evenly. Figure 2 3 shows the load change to introduce the transient. The ACDS transient was induced six seconds into the simulation while the DCDS transient was induced one second into the simulation. Figure 2 7 : Voltage (left) and curren t (right) measured at selected loads during Trial 3. Figure 2 8 : Trial 3 Generator 1S and 2S voltage befo re and after transient for ACDS.

PAGE 53

41 Figure 29: Trial 3 Generator 1S and 2S voltage before and after transient for DCDS. The ACDS generators see a voltage increase of 1.1% when the forward load is lost and recovers back to nominal voltage after 1.87 seconds. The DCDS generators voltage remained steady throughout the period of time that the loads experienced the transient. Figure 30 : Trial 3 v oltage measured at selected loads for ACDS.

PAGE 54

42 Figure 31: Trial 3 voltage measured at selected loads for DCDS. The ACDS loads voltage increased 1. 1 % when the forward load circuit breaker opened and recovered after 2 seconds. The DCDS loa d voltage spiked at 10.8 % above nominal before recovering after 1 2 seconds. Table 10: Trial 3 ACDS and DCDS comparison summary The sudden loss of a load places a large strain on a generator. The kinetic energy stored in the rotor cause s the speed of the rotor to increase when the load is removed. The introduction of power electronics allowed for a lar ger voltage transient, but also recovered significantly faster Additionally, the DCDS generator experienced a large voltage spike that was not seen on the ACDS simulation.

PAGE 55

43 T rial 4: Loss of a Generator During General Quarters Trial 4 is designed to simulate the loss of a generator during a General Quarters situation where all three generators are online, but the bus tie circuit breakers a re open and each generator is only supplying the loads attached to its own switchboard. The configuration change to induce the transient is shown in Figure 26 The ACDS transient was induced six seconds into the simulation while the DCDS transient was in duced one second into the simulation. Figure 32 : Voltage (left) and curren t (right) for generator 1S, generator 2S, and generator 1E during Trial 4.

PAGE 56

44 Figure 33 : Trial 4 Generator 1S (top) and 2S (middle) and 1E (bottom) voltage befo re and after transient for ACDS. Figure 34: Trial 4 Generator 1S (top) and 2S (middle) and 1E (bottom) voltage before and after transient for DCDS.

PAGE 57

45 The generators did not suffer from any transient, as they were isolated from the lost generator b y th e open bus tie circuit breakers. Figure 35 : Trial 4 v oltage measured at selected loads for ACDS Figure 36: Trial 4 voltage measured at selected loads for DCDS.

PAGE 58

46 On both the ACDS and DCDS simulations, the unaffected loads did no t experience the transient as they were isolated from the transient by the open bus tie circuit breakers The loa ds attached to the 1S generator lost all power and would need to be restored manually by closing one of the bus tie circuit breakers. Table 11: Trial 4 ACDS and DCDS comparison summary The General Quarters configuration is designed to isolate the system from potentially disastrous transients during the most volatile situations the ship may encounter The simulation demonstrates this conc ept works as designed and protected the unaffected generators and loads from the transient caused by the total loss of a generator. The aft loads did experience a voltage spike on the ACDS simulation, however, it is caused by the shunt resistances built i nto the bus tie circuit breakers Simulink model and would not appear in an actual system. Trial 5: Starting a Large Load During General Quarters Trial 5 simulates the transient caused by a large load starting when the distribution system is set up in the General Quarters configuration. The load change to introduce the transient is shown in Figure 2 9 The ACDS transient was induced six seconds into the simulation while the DCDS transient was induced one second into the simulation.

PAGE 59

47 Figure 37 : Voltage (left) and current (right) measured at selected loads during Trial 5. Figure 38 : Trial 5 Generator 1S 2S and 1E voltage before and after transient for ACDS.

PAGE 60

48 Figure 39: Trial 5 Generator 1S, 2S, and 1E voltage before and after transient for DCDS. In the both configuration s when the load is started, the transient is isolated to Generator 2S as both bus tie breakers are open. The AC DS g enerator 2S voltage dropped .3% when the load was started. It then took 1.76 seconds for th e voltage to return to nominal. The DCDS generator voltage remained steady throughout the transient on the aft loads. Figure 40 : Trial 5 v oltage meas ured at selected loads for A C DS (right set).

PAGE 61

49 Figure 41: Trial 5 voltage measured at selected loads for DCDS (right set). As on the generator side of the system s the loads are isolated and there is no transient except on the load that is started. The AC load transi ent maxes out .31% low before climbing to nominal voltage in 1.76 seconds. The DC system voltage drops 13% before recovering after .15 seconds. Table 12: Trial 5 ACDS and DCDS comparison summary The General Quarters configuration isolates the unaffected sections of the system leaving only a transient on the load that is started. The DCDS system allowed the load to reach nominal voltage eleven time s faster than the ACDS system despite suffering a larger overshoot.

PAGE 62

50 Trial 6: Battle Damage During General Quarters Trial 6 is meant to simulate the system experiencing damag e by losing the entire forward load while all three generators are online and supplying the loads only attached to their own switchboards The load change to introduce the transient is seen in Figure 3 2 The ACDS transient was induced six seconds into th e simulation while the DCDS transient was induced one second into the simulation. Figure 4 2 : Voltage (left) and current (right) measured at selected loads during Trial 5. Figure 4 3 : Trial 6 Generator 1S 2S and 1E voltage befo re and af ter transient for ACDS.

PAGE 63

51 Figure 4 4 : Trial 6 Generator 1S, 2S, and 1E voltage before and after transient for DCDS. The ACDS system generator model shows a transient in all three generators despite the isolation provided by the General Quarters con figuration. This is likely caused by the shunt resistances around the bus tie circuit breakers. Generator 1S voltage maxes out at 3.5% above nominal. Generator 2S maxes out at 1.9% above nominal. Generator 1E maxes out at 2.8% above nominal. All three generators take 1.7 seconds to recover from the transient. On the DCDS generators, the isolation is intact and no transient is seen in any of the generator voltage waveforms. Figure 45 : Trial 6 Vo ltage measured at selected loads for ACDS

PAGE 64

52 Figure 46: Trial 6 Voltage measured at selected loads for DCDS. The ACDS amidships load voltage showed an unexpected transient which peak s at 3.5% of nominal before settling in 1.78 seconds. The aft loads peak at 1.9% above nominal and settled after 1 .78 seconds. T he DCDS loads were completely isolated from the transient due to the open bus tie circuit breake rs, but the affected DC Bus saw an 18.5% overshoot before settling in .4 seconds. Table 13: Trial 6 ACDS and DCDS comparison summary The ACDS system suffered a large voltage overshoot and despite the isolation provided by the open bus tie circuit breakers, the transient was seen in every generator and load waveform. This is most likely caused by the shunt resistances in the Simulink circu it breaker model. The DCDS model was more realistic in that the sections were isolated and only the affected section of the DC Bus saw a transient. The overshoot on

PAGE 65

53 the DC Bus was three times larger than the transient on the ACDS, but recovered over four times faster. Trial 7: Loss of a Generator While at Anchor Trial 7 simulates the situation when a generator is lost while the ship is configured to be at anchor with one generator online and supplying a limited load. When the generator is lost, the sh ip loses all power and the emergency generator comes online to take the full load. The configuration change to introduce the transient is seen in Figure 3 5 The ACDS transient was induced six seconds into the simulation while the DCDS transient was induc ed at the beginning of the simulation. Both emergency generators come online .5 seconds later Figure 47 : Voltage (left) and current (right) for generator 1S and generator 1E during Trial 7.

PAGE 66

54 Figure 48 : Voltage at generator 1E for A CDS during startup Figure 49: Voltage at generator 1E for DCDS during startup. When the ACDS generator 1E breaker closes, the system experiences a 2239 V spike before experiencing a 3.3% voltage drop finally stabilizing after 1.5 seconds. Whe n the DCDS 1E generator comes online, the voltage starts 1.3 % below nominal which s tabilizes after 3 seconds.

PAGE 67

55 Figure 50 : Trial 7 v oltage measu red at selected loads for ACDS. Figure 51: Trial 7 voltage (left) and current (right) m easured at selected loads for DCDS.

PAGE 68

56 The loads on both the ACDS and DCDS startup from no voltage. When the breaker closes, t he ACDS loads experienced a 2284 V spike before dropping to 3.4% below nominal before recovering after 2 seconds. The DCDS loads e xperienced a 6.1% voltage overshoot which recovered after .25 seconds. Table 14: Trial 7 ACDS and DCDS comparison summary Trial 7 is unique in that it is the only trial in which the system is completely de energized and the transient is caused by t he startup of a generator. The DCDS system provided full voltage to the loads 5 times faster than the ACDS and did not experience a voltage spike when the 1E generator started up. Trial 8: Starting a Large Load While at Anchor Trial 8 simulates the s ituation when a large load is started during anchor configuration and only one generator is online supplying all the loads. Figure 3 7 shows the transient introduced. The ACDS transient was induced six seconds into the simulation while the DCDS transient w as induced one second into the simulation. Figure 52 : Voltage (left) and current (right) measured at selected loads during Trial 8.

PAGE 69

57 Figure 53 : Generator 1S voltage before and after transient for ACDS. Figure 54: Generator 1S voltage before and after transient for DCDS. The ACDS generator suffers a .2% voltage drop when the load is started and then recovers in 2 seconds. The DCDS generato r did not show any voltage drop during the time the loads showed oscillating voltage.

PAGE 70

58 Figure 55 : Trial 8 Voltage meas ured at selected loads for ACDS. Figure 56: Trial 8 Voltage measured at selected loads for DCDS.

PAGE 71

59 On the load side of the ACDS system, the existing loads suffered a .2% voltage drop when the load starte d and recovered after 1.5 seconds. Similarly, the load that came online started .2% below nominal then reached nominal voltage after 1.5 seconds. The DCDS loads saw a voltage oscillation when the load came online which lasted .64 seconds and maxed out 23. 6% larger than nominal Table 15: Trial 8 ACDS and DCDS comparison summary The voltage transient on the ACDS system was minimal at only .2%, but took nearly 2 seconds for the transient to recover. The DCDS system boasted a significantly faster res ponse time, approximately 4 times faster but suffered from an oscillating voltage on the DC Bus. Trial 9: Battle Damage While at Anchor Trial 9 is meant to simulate battle damage by suddenly opening the circuit breaker to every forward load and removing a large load from the system. The load change to introduce the transient is shown in Figure 40. The ACDS transient was induced six seconds into the simulation while the DCDS transient was induced one second into the simulation.

PAGE 72

60 Figure 5 7 : Voltage (left) and current (right) measured at selected loads during Trial 9. Figure 58 : Generator 1S voltage before and after transient for ACDS.

PAGE 73

61 Figure 59: Generator 1S voltage before and after transient for DCDS. The ACDS g ener ator voltage peaked at 2.3% above nominal before settling after 2 seconds. The DCDS generator voltage maintained its voltage throughout the period that the loads displayed the transient. Figure 60 : Trial 9 v oltage measured at selected loads for ACDS

PAGE 74

62 Figure 61 : Trial 9 voltage measured at selected loads for DCDS. The ACDS loads peaked at 2.8% above nominal after the forward loads drop offline before recovering after 1.7 seconds. The DCDS load voltage briefly peaked 12.4 % above nominal before finally recover ing 25 seconds after the load was lost. Table 16: Trial 9 ACDS and DCDS comparison summary When the load was lost, the remaining ACDS loads saw a 2.3% voltage increase then recovered after 2 seconds. The DCDS generator did not experience any transient, but the other loads saw the voltage overshoot but within requirements.

PAGE 75

63 Simulation Results Analysis Consistently across all the trials, the DCDS simulation provided significantly faster recovery times th an the AC D S simu lation. Without the power electronic governor is a mechanical c ontrol system, which has a response time limited by the mechanical components Thus, irrelevant of the t ype of transient introduced, or the configuration of the system, the ACDS generators recovery was limited to between 1.5 and two seconds. Without changing the machinery installed in the system, there is no way to increase the recovery speed of the generat or. The DCDS power converters, on the other hand, respond to transients significantly faster thanks to the high switching frequency of the transistors This allows the PI controllers to identify changes and respond at a rate beyond the capability of th e generator governor. The DCDS sim ulation recovered an average of nearly five times faster than the ACDS. Recovery time was the greatest advantage identified by the simulation trials, in other areas however, the DCDS simulation fell short. When the tran sients were introduced, the ACDS system suffered a voltage change of 1.5% on average. The DCDS system, on the other hand, saw a voltage change ten times larger, 14.7 V on average. Much of the volatility of the DC bus voltage could be minimized by introdu cing an energy storage mechanism such as a battery bank or flywheel, on the DC bus, to provide additional constant power during the transient. The voltage on the DCDS simulations exhibited excessive noise in the voltage waveforms compared to the ACDS s imulation waveforms. That noise is created by the switching action on the IGBT and could be eliminated with the installation of additional filters on the power converter.

PAGE 76

64 Unsurprisingly, in both the ACDS and DCDS simulations, when more generators were o nline and sharing the load, the transients tended to be smaller thanks to the kinetic energy stored in the generator rotor. While rotor inertia allows the generator to governor responds slowly to changes in the load. Conversely, while the power electronics allow for significantly faster recovery, the lack of inertia allows a much larger overshoot which could potentially be controlled with additional energy storage on t he bus.

PAGE 77

65 CHAPTER V DISTRIBUTION SYSTEM COMPARISON Thus far, this thesis has focused on a transient analysis of the existing AC However there are additional adv antages to a zonal distribution system that are worth discussion. Weight Benefits cables to distribute the power from the ship service switchboards to the load centers throug switchboards located in either the engine room or emergency generator room, there are many redundant cables running from fore to aft (longitudinally) that are eliminated with a dis tribution bus. A permanent distribution bus c onsists of only two cables running longitudinally thus eliminating all the extra cables that are ordinarily needed to distribute the electrical power. To analyze the weight difference, the following assumptio ns were made: 1. A DC Bus would opera te at 1000 kV with 2 conductors and would require a DSGU 9 cable [31] 2. The location of the generators and load centers remains the same as the ACDS configuration. 3. All longitudinal cables feeding individual load centers are removed and the DC bus cables are added. 4. All athwartships and vertical cables are left in place to feed from the distribution bus to the load canter. Following these assumptions, over 4500 lbs worth of cables could be removed from the ship b y using a distribution bus Detailed calculations can be found in Appendix D

PAGE 78

66 In 1993, Naval Engineers Journal published a study conducted by Chester Petry and Jay Rumburg who compared the radial AC electrical distribution systems on board 51 class destroyers to a zonal AC distribution system In addition to cable weight reductions, that study found additional weight savings by removing the following equipment [10]: Power distribution switchboard sections Power distribution load centers Th e removed equipment was replaced with smaller more efficient equipment [10]: Main distribution bus feeder switchboards Two p ower distribution load centers in each new electrical zone Load center to main distribution bus connection boxes The overall weigh t change (in long tons) is summarize d in Table 17 : Table 17 : DDG 51 weight comparison between radial and zonal architectures [10] A more complete weight comparison taking into account switchgear and foundations, was not comple ted as part of this thesis due to a lack of available information for the DDG 51 class. These weight reductions translate to significant cost savings earned through more efficient propulsion and a reduction in fuel usage.

PAGE 79

67 Standardization Benefits The benefits of standardization have already been realized during the Coast Bruce Jone you can take a rescue swimmer from Savannah and stick him on a helicopter from Houston with a pilot from Detroit and a flight mech [sic] from San Francisco, and these guys have n ever met before and they can go out and fly for six hours and rescue 80 benefit of standard equipment and training [34]. The zonal architecture is extremely flexible and c an be expanded or contracted to fit any size of vessel. The Office of Naval Research has continued the work to develop a standard Power Electronic Building Block (PEBB) that can change any electrical power input into any desired (Type I, II or III) output [33]. These PEBBs can not only replace the motor generators currently employed on Coast Guard cutters, but in coordination with a DC distribution system, could be employed across multiple ship classes. As the PEBBs can change any type of power into any o ther, generation equipment could then be standardized and only scaled for larger demand on larger cutters. This allows for shipbuilders to stock a large number of the same parts which could then be employed across multiple construction projects and the Co ast Guard to reduce its logistics footprint by stocking parts for a smaller variety of equipment. Standardizing the equipment employed across all Coast Guard cutter classes would also simplify training and qualification requirements for personnel. This o pens a wide range of options for personnel assignments, especially for short term or emergent personnel needs at a unit where a member assigned to any other ship would be knowledgeable enough to operate the equipment.

PAGE 80

68 Utilizing a zonal electrical distrib ution system would provide numerous advantages during the acquisition as well At present, with a radial architecture, electrical equipment can not be installed during construction until after the hull is complete ly assembled Only then can cables be run throughout the ship from switchboards to loads [10] The ZED concept keeps all distribution cabling and equipment isolated to a single zone while the zones are connected to each other only through the distribution bus. This allows for shipbuilders to use a modular construction technique where the entire zone could be completely outfitted and tested before zone assembly as in Figure 43 [32] Figure 62 : Advantage of z onal c onstruction [32] Outfitting an individual zone pro vides greater accessibility for workers running cables, testing equipment, and troubleshooting problems before the zone is joined to the rest of the ship [32]. Additionally, since they do not rely on each other, multiple zones could be built concurrently, and then joined together all of which reduce the overall cost and period of performance to build the ship. Similarly, as the ship ages and receives equipment upgrades, a radial architecture would simplify the upgrade as power cables would only need to be run within a watertight zone rather than from the generator [10].

PAGE 81

69 From reducing construction costs and timelines, to simplifying training and support, the zonal electrical distribution architecture could revolutionize cutter acquisition and sustainmen areas. Survivability Benefits On October 12, 20 00, while refueling in Aden, Yemen, the USS Cole was attacked by two al Qaeda terrorists carrying a bomb in a small boat. The detonation ne Figure 63 : Damage to USS Cole [37] Despite the significant damage and flooding the ship suffered, the USS Cole did not sink w which exploited of the ship, mission critical systems, and the crew to perform assigned warfare missions, and of the protection pr

PAGE 82

70 accidents such as fire, flooding, groundings, etc. Survivability is bro ken down into three disciplines: Sus ceptibility : A measure of the capability of the ship, mission critical systems, and crew to avoid and or defeat an attack and is a function of operational tactics, signature reduction, countermeasures, and self defense system effectiveness. Vulnerability : A measure of the capability of the ship, mission critical systems, and crew to withstand the initial damage effects from conventional, Chemical Biological or Radiological (CBR), or asymmetric threat weapons or accidents, and to continue to perform assign ed primary warfare missions, and protect the crew from serious injury or death. Recoverability : A measure of the capability of the ship and crew, after initial damage effects, whatever the cause, to take emergency action to contain and control damage, pre vent loss of a damaged ship, minimize personnel casualties, and restore and sustain primary mission capabilities [35] A zonal electrical distribution system drastically increases survivability by decreasing vulnerability and increasing recoverability. The first step to combating any casualty and what saved the USS Cole, is to contain it by setting boundaries. A boundary can be set at any watertight (for flooding emergencies) or fume tight (for fire or CBR emergencies) bulkhead [36]. By reducing the n umber of longitudinal cables that repeatedly penetrate watertight bulkheads to two (the distribution bus) those boundaries are more effective for stopping flooding and fire from spreading, allowing the ship to continue its primary mission despite the damag e that has been suffered. In the case of the USS Cole, the flooding boundaries ensured that the damaged section of the ship was the only section that suffered flooding, not nearly enough to overcome her innate buoyancy and cause her to sink. Additionall y, a zonal electrical distribution system increases recoverability by minimizing the damage that would need to be repaired to continue functioning. With two busses running the length of the ship, separated port and starboard and separated vertically as mu ch as practical, the chance of both busses being damaged

PAGE 83

71 simultaneously is minimized. If one bus is damaged, there is a second bus within the watertight compartment that can provide power to necessary equipment without breaching watertight boundaries with casualty power cables This allows the ship to maintain its high state of readiness without compromise. Repair to the damaged bus is also simplified as casualty power can be rigged around the damaged section, restoring power to the rest of the bus quick ly and efficiently. survivability compared with a radial distribution system by reducing penetrations through watertight boundaries and drastically simplifying casualty power routes. The attack on the USS Cole vividly demonstrated that ensuring the integrity of boundaries and preventing the spread of damage is the surest way to keep the ship in fighting shape.

PAGE 84

72 CHAPTER VI CONCLUSIONS AND RECOMMENDATIONS Throughout the work completed in this thesis, the purported benefits of a direct current zonal electrical distribution system have been tested and analyzed. When compared with an existing alternating current radial distribution system the benefits are striking. The intro duction of power electronic converters to rectify the current and voltage from the generators not only provide insulation to the generators but increases the speed at which the system can recover. Additionally, switching from a radial type distribution s redundant cabling, foundations, and transformers, which ultimately leads to significant fuel savings over the life of the ship. Unfortunately, the work on this thesis is incomplet e. While the wide range of computer models developed as part of this thesis create a solid basis for future deeper mathematical analysis of each trial, t he DCDS model was drastically simplified to eliminate errors encountered in the MATLAB Simulink softwa re and meet necessary timelines. An inverter model needs to be developed and incorporated on the load side of the DCDS model and the trials tested again. The creation of a droop controller for the generator models would ensure the load is more evenly div ided between the online generators. Additionally, incorporating power converters to produce Type II and Type III power which could ultimately replace motor generator sets currently employed on cutters is a necessary addition to the simulation. The ben efits demonstrated by the work on this thesis suggest that taxpay er funds would be saved in the acquisition phase by reducing the amount of work and materials needed to install the system. Additional expendit ure would be

PAGE 85

73 avoided dur ing the sustainment phase with a reduced weight, increasing engine fuel efficiency. Despite these benefits, there are significant challenges that still need to be addressed on a DC based system. A fault on the DC bus allows for a very large fault current in a short amount of time. If not addressed quickly, the large fault current can cause significant damage to sensitive electronics. Additionally, there were significantly larger overshoots on the DCDS transients, though tuning of the power electronics co ntrollers could reduce the observed overshoot. Finally, DC grids are inherently unstable due to the dynamics of the power electronic interfaces and the low frequency oscillation when multiple converters are connected together Th ough these challenges rem ain, the purported benefits of a DC ZED show that these challenges are worth solving. In an era of constrained budgets and increasingly complex operating environments, the Coast Guard must find sustainable ways to reduce operating costs. A DC ZED system re However significant challenges remain. Fault conditions and stability issues need to be further studied and mitigation strategies developed before a DCZED can be incorporated into Coast Guard use. Beyond the work considered for the Offshore Patrol Cutter, the Coast Guard, and the American taxpayer would be best served by considering a zonal distribution system for all future cutter acquisitions.

PAGE 86

74 REFERENCES [1] "U.S. Coast Guard History." U.S. Coast Guard History: Brief U.S. Coast Guard, 16 Oct. 2014. Web. 01 July 2015. [2] United States. Cong. Congressional Research Service. Coast Guard Cutter Procuremen t Background and Issues for Congress By Ronald O'Rourke. Cong. Rept. R42567. Washington, DC: Congressional Research Service, Library of Congress, 2015. U.S. Coast Guard Cutters & Craft Index U.S. Coast Guard, 9 Apr. 2015. Web. 1 July 2015. [3] "Missions ." USCG: About Us U.S. Coast Guard, 20 Mar. 2014. Web. 01 July 2015. [4] "Recapitalization." U.S. Coast Guard Acquisition Programs U.S. Coast Guard, Nov. 2014. Web. 1 July 2015. [5] "Famous Cutter Class 270 Foot Medium Endurance Cutter (WMEC)." Global Security GlobalSecurity.org, 2015. Web. 01 July 2015. [6] United States. Department of Defense. Naval Sea Systems Command. NSTM 320: Electric Power Distribution Systems Naval Sea Systems Command, 1 Mar. 2005. Web. 14 Nov. 2014. [7] Amy, John V. "Consid erations in the Design of Naval Electric Power Systems." IEEE Power Engineering Society Summer Meeting, (2002): IEEE Xplore Web. 14 Jan. 2015. [8] Doerry, Norbert H., and James C. Davis. "Integrated Power System for Marine Applications." Naval Engineers Journal 106.3 (1994): 77 90. Web. 15 Jan. 2015. [9] Ciezki, John G., and Robert W. Ashton. "Selection and Stability Issues Associated with a Navy Shipboard DC Zonal Electric Distribution System." IEEE Transactions on Power Delivery IEEE Trans. Power Deliv ery 15.2 (2000): 665 69. IEEE Xplore Web. 21 Oct. 2014. [10] Petry, Chester R., and Jay W. Rumburg. "Zonal Electrical Distribution Systems: An Affordable Architecture for the Future." Naval Engineers Journal 105.3 (1993): 45 51. NEJ Online Web. 14 Jan. 2015. [11] "Offshore Patrol Cutter." OFFSHORE PATROL CUTTER (n.d.): n. pag. Offshore Patrol Cutter U.S. Coast Guard Acquisition Directorate, May 2015. Web. 24 Aug. 2015. [12] Electrical One Naval Engineering Division Dwg No 905 WMEC 320 014 F.

PAGE 87

75 [13] United States. Department of Defense. Naval Sea Systems Command. NSTM 300: Electric Plant General Naval Sea Systems Command, 1 May 2012. Web. 14 Jan. 2015. [14] Collier, Everett. The Boatowner' s Guide to Corrosion: A Complete Reference for Boatowners and Marine Professionals Camden, Me.: International Marine/McGraw Hill, 2006. Print. [15] Chapman, Stephen J. "Chapter 5 Synchronous Machines." Electric Machinery and Power System Fundamentals Bo ston: McGraw Hill, 2002. 192 264. Print. [16] Mohan, Ned. "10.6 Synchronous Generators." Electric Machines and Drives: A First Course Hoboken, NJ: Wiley, 2012. 187 90. Print. [17] Generation." Power Generation, Operation, and Control Third ed. Hoboken, NJ: John Wiley & Sons, 2014. 468 500. Print. [18] MATLAB Simulink Program documentation. Model and Dynami cs of the Synchronous Machine Vers. R2015a. Mathworks, 2015. Web. 2 Sept. 2015. [19] "IEEE Std 421.5 2005 IEEE Recommended Practice for Excitation System Models for Power System Stability Studies," IEEE Std 421.5 2005 (Revision of IEEE Std 421.5 1992), pp. 0_1 85, 2006. [20] Load & Power Office of Naval Engineering Dwg No 901 WMEC 311 001 G. [21] Bergen, Arthur R., and Vijay Vittal. Power Systems Analysis Second ed. Upper Saddle River, NJ: Pren tice Hall, 2000. Print. [22] United States of America. Department of Homeland Security. United States Coast Guard. Naval Engineering Manual Washington DC: Office of Naval Engineering, 2011. Print. COMDTINST M9000.6F. [23] Sowers, Leigh G. Personal inter view. 20 Mar. 2015. [24] United States of America. Department of Homeland Security. U.S. Coast Guard. Standing Orders for the Engineering Department USCGC MOHAWK By A. B. Morrison. N.p.: n.p. 2014. Print. MOHAWKINST 9000.1 G.

PAGE 88

76 [25] United States. Departm ent of Defense. Naval Sea Systems Command. NSTM 310: Electric Power Generators and Conversion. Naval Sea Systems Command, 1 Mar 2005. Web. 14 Sep. 2015. [26] United States. U.S. Coast Guard. Office of Naval Engineering. Technical Manual Generator Section SSG N.p.: n.p., 1974. Print. TP 2820B. [27] Teleke, Sercan, Subhashish Bhattacharya, and Mesut Baran. "Enhanced Control of Voltage Source Converters for DC Shipboard Power Systems." Naval Engineers Journal 122.1 (2010): 81 91. Wiley Online Library. Web. 8 Dec. 2015. [28] Sul, Seung Ki. Control of Electric Machine Drive System. Hoboken, NJ: Wiley IEEE, 2011. Print. [29] Voland, Gerard. Control Systems Modeling and Analysis. Englewood Cliffs, NJ: Prentice Hall, 1986. Print. [30] Teleke, Sercan, Subhashis h Bhattacharya, and Mesut E. Baran. "A Novel PWM Voltage Source Converter for a DC Zonal Shipboard Power System." Proc. of 33rd Annual Conference of the IEEE Industrial Electronics Society, Taipei, Taiwan. N.p.: n.p., n.d. N. pag. IEEE Xplore. Web. 12 Oct. 2015. [31] United States of America. Department of Defense. Department of the Navy. Military Standard: Electronic Equipment Parts Selected Standards Wire and Cable. 242J ed. N.p.: n.p., 1986. Everyspec. Everyspec LLC. Web. 31 Mar. 2015. [32] Cecere, Mic hael L., III, Jack Abbott, Michael L. Bosworth, and Tracy Joseph Valsi. "Commonality Based Naval Ship Design, Production, and Support." Journal of Ship Production 11.1 (1995): 1 14.SNAME. Society of Naval Architects and Marine Engineers, Feb. 1995. Web. 20 Jan. 2016. [33] Piff, Joseph C. "Power Electronics Building Blocks (PEBB) Program: PEBB Bringing a Whole New Perspective to Power Control and Distribution." Program Manager 1 Mar. 2003: 46 54. DAU Publications and Catalogs. Web. 20 Jan. 2016. [34] Unite d States of America. U.S. Coast Guard. Historian's Office. The U.S. Coast Guard & Hurricane Katrina. By Scott Price. U.S. Coast Guard, Feb. 2006. Web. 20 Jan. 2016. [35] United States of America. Department of the Navy. Office of the Chief of Naval Operat ions. Department of the Navy Issuances. Department of Defense, 13 Sept. 2012. Web. 25 Jan. 2016.

PAGE 89

77 [36] United States of America. U.S. Coast Guard. Office of the Commandant. Machinery Space Firefighting Doctrine for Class Bravo Fires. Washington DC: U.S. Co ast Guard, 2009. COMDTINST M9555.1B.CG 612 Directives and Publications Division. U.S. Coast Guard, 23 Nov. 2009. Web. 25 Jan. 2016. [37] Evans, Mark L. "Cole (DDG 67)." Cole (DDG 67). Naval History and Heritage Command, 30 June 2015. Web. 25 Jan. 2016.

PAGE 90

78 APPENDIX A

PAGE 91

79 APPENDIX B TRANSMISSION LINE CALCULATIONS

PAGE 92

80 APPENDIX C DC ZED LOAD DISTRIBUTION CALCULATIONS

PAGE 93

81

PAGE 94

82

PAGE 95

83

PAGE 96

84

PAGE 97

85

PAGE 98

86

PAGE 99

87 APPENDIX D DC ZED TRANSMISSION LINE CALCULATIONS

PAGE 100

88 APPENDI X E MATLAB CODE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %Simulations Code %MSEE Thesis %Started: 17 Aug 15 %Last Edited: 17 Aug 15 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% clear all clc % Run Simulation %Run Trial 1: Tripping Generator CB While Underway %Set Distribution System Configuration Parameters ACDS_270_Parameters_Underway_Steaming DCDS_270_TL_Parameters Tele ke_Converter_Parameters %Set Trial Parameters WMEC_Model_Test1_Open_Gen_CB Time_Start=tic; tt=sim( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( 'DCDS_Trial_1_Variables' ); %Run Trial 2: Start Large Load while underway clear all Teleke_Converter_Parameters %Set Distribution System Configuration Parameters ACDS_270_Parameters_Underway_Steaming DCDS_270_TL_Parameters %Set Trial Parameters WMEC_Model_Test2_Start_Large_Load Time_Sta rt=tic; tt=sim( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( 'DCDS_Trial_2_Variables' ); %Run Trial 3: Battle Damage while Underway clear all Teleke_Converter_Parameters %Set Distribution System Configura tion Parameters ACDS_270_Parameters_Underway_Steaming DCDS_270_TL_Parameters %Set Trial Parameters WMEC_Model_Test3_Battle_Damage Time_Start=tic;

PAGE 101

89 tt=sim( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/ (3600) save( 'DCDS_Trial_3_Variables' ); %Run Trial 4: Tripping Gen CB While at GQ clear all %Set Distribution System Configuration Parameters ACDS_270_Parameters_General_Quarters DCDS_270_TL_Parameters Teleke_Converter_Parameters %Set Tri al Parameters WMEC_Model_Test1_Open_Gen_CB %Run Simulation Time_Start=tic; tt=sim( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( 'DCDS_Trial_4_Variables' ); %Run Trial 5: Starting Large Load W hile at GQ clear all %Set Distribution System Configuration Parameters ACDS_270_Parameters_General_Quarters DCDS_270_TL_Parameters Teleke_Converter_Parameters %Set Trial Parameters WMEC_Model_Test2_Start_Large_Load W2=6000; %Run Simulation Time_Start=tic; tt=sim( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( 'DCDS_Trial_5_Variables' ); %Run Trial 6: Battle Damage While at GQ clear all %Set Distribution System Configura tion Parameters ACDS_270_Parameters_General_Quarters DCDS_270_TL_Parameters Teleke_Converter_Parameters %Set Trial Parameters WMEC_Model_Test3_Battle_Damage %Run Simulation

PAGE 102

90 Time_Start=tic; tt=sim( 'TELEKE_Converter_RLoad' ); Time_ Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( 'DCDS_Trial_6_Variables' ); %Run Trial 7: Tripping Gen CB While at Anchor clear all %Set Distribution System Configuration Parameters ACDS_270_Parameters_Anchor DCDS_270_TL_Paramet ers Teleke_Converter_Parameters %Set Trial Parameters WMEC_Model_Test4_Deadship CB1S=0; CB1E=1; %Run Simulation Time_Start=tic; tt=sim( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( DCDS_Trial_7_Variables' ); %Run Trial 8: Starting Large Load While at Anchor clear all %Set Distribution System Configuration Parameters ACDS_270_Parameters_Anchor DCDS_270_TL_Parameters Teleke_Converter_Parameters %Set Trial Parameters WMEC_Model_Test2_Start_Large_Load %Run Simulation Time_Start=tic; tt=sim( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( 'DCDS_Trial_8_Variables' ); %Run Trial 9: Battle Damage While at Anchor c lear all %Set Distribution System Configuration Parameters ACDS_270_Parameters_Anchor DCDS_270_TL_Parameters Teleke_Converter_Parameters %Set Trial Parameters WMEC_Model_Test3_Battle_Damage %Run Simulation

PAGE 103

91 Time_Start=tic; tt=s im( 'TELEKE_Converter_RLoad' ); Time_Elapsed = toc(Time_Start); Time_Hours = Time_Elapsed/(3600) save( 'DCDS_Trial_9_Variables' ); %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %M SEE Thesis %Started: 17 Aug 15 %Last Edited: 17 Aug 15 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Plots %Generator Voltage %Generator Measurements figure(1) %Generator 1S V Subplot subplot(3,2,1) hold on plot(tt,Vabc1S_ Out, 'g' ) grid ylabel( 'Gen 1S' ) %axis([5.75 7 250 450]) legend( 'Vabc' ) title( 'Gen 1S Voltage' ) %Generator 1S I Subplot subplot(3,2,2) plot(tt,Iabc1S_Out, 'r' ) grid %axis([.4 .45 550 550]) legend( 'Iabc' ) title( 'Gen 1S Current' ) %Generator 2S V Subplot subplo t(3,2,3) hold on plot(tt,Vabc2S_Out, 'g' ) grid ylabel( 'Gen 2S' ) %axis([.49 .52 450 450]) legend( 'Vabc' ) title( 'Gen 2S Voltage' ) %Generator 2S I Subplot subplot(3,2,4) plot(tt,Iabc2S_Out, 'r' ) grid %axis([.4 .45 550 550]) legend( 'Iabc' ) title( 'Gen 2S Curren t' ) %Generator 1E V Subplot subplot(3,2,5) hold on

PAGE 104

92 plot(tt,Vabc1E_Out, 'g' ) grid ylabel( 'Gen 1E' ) %axis([.4 .45 450 450]) legend( 'Vabc' ) title( 'Gen 1E Voltage' ) %Generator 1E I Subplot subplot(3,2,6) plot(tt,Iabc1E_Out, 'r' ) grid %axis([.4 .45 550 550]) le gend( 'Iabc' ) title( 'Gen 1E Current' ) %Power Graph figure(2) subplot(5,1,1) hold on plot(tt,P_1S_Out, 'g' ) %plot(tt,Q_1S_Out/1000,'r') legend( 'P' 'Q' ) xlabel( 'Gen 1S' ) grid %axis([.3 .5 0 500]) subplot(5,1,2) hold on plot(tt,P_2S_Out, 'g' ) %plot(tt,Q_2S_Ou t/1000,'r') legend( 'P' 'Q' ) xlabel( 'Gen 2S' ) grid %axis([.3 .5 0 500]) subplot(5,1,3) hold on plot(tt,P_1E_Out, 'g' ) %plot(tt,Q_1E_Out/1000,'r') legend( 'P' 'Q' ) xlabel( 'Gen 1E' ) grid %axis([.3 .5 0 500]) subplot(5,1,4) hold on plot(tt,(P_1E_Out+P_2S_Out+P _1S_Out), 'g' ) %plot(tt,Q_1E_Out/1000) legend( 'P' 'Q' ) xlabel( 'Total Power' ) grid on %axis([.3 .5 0 1200]) % subplot(5,1,5) % plot(tt,PLoads) % grid on figure(3) subplot(3,1,1) plot(tt,Wm_1S) grid on

PAGE 105

93 title( 'Generator Speed (pu)' ) ylabel( 'Gen 1S' ) %axis([0 7 .99 1.015]) subplot(3,1,2) plot(tt,Wm_2S) grid on ylabel( 'Gen 2S' ) %axis([0 7 .995 1.005]) subplot(3,1,3) plot(tt,Wm_1E) grid on ylabel( 'Gen 1E' ) xlabel( 'Time (s)' ) legend( '1S' '2S' '1E' ) %axis([0 7 .995 1.005]) figure(4) subplot(3,2,1) plot(tt,V_1SF) grid on axis([6.25 8 420 460]); ylabel( 'Voltage (V)' ) title( ) subplot(3,2,2) plot(tt,I_1SF) grid on ylabel( 'Current (A)' ) axis([6.25 8 10 13]); subplot(3,2,3) plot(tt,V_1EA) grid on ylabel( 'Voltage (V)' ) axis([6.25 8 420 460]); title( 'Amids hips Loads' ) subplot(3,2,4) plot(tt,I_1EA) grid on ylabel( 'Current (A)' ) axis([6.25 8 5 8]); subplot(3,2,5) plot(tt,V_2SAFT) grid on ylabel( 'Voltage (V)' ) axis([6.25 8 420 460]); title( 'Aft Loads' ) subplot(3,2,6) plot(tt,I_2SAFT) grid on ylabel( 'Current (A )' ) axis([6.25 8 10 15]); figure(5) %Generator 1S V Subplot subplot(3,1,1) hold on plot(tt,Vabc1S_Out, 'g' )

PAGE 106

94 grid on ylabel( 'Gen 1S' ) %axis([5.5 8 430 450]) legend( 'Vabc' ) title( 'Gen 1S Voltage' ) %Generator 2S V Subplot subplot(3,1,2) hold on plot(tt,Vabc2 S_Out, 'g' ) grid on ylabel( 'Gen 2S' ) %axis([5.5 8 430 450]) legend( 'Vabc' ) title( 'Gen 2S Voltage' ) Generator 1E V Subplot subplot(3,1,3) hold on plot(tt,Vabc1E_Out, 'g' ) grid on ylabel( 'Gen 1E' ) axis([5.5 8 430 460]) legend( 'Vabc' ) title( 'Gen 1E Voltage' ) % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %Shipboard Anchor Parameters %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %60 Hz Generato r Parameters %Generator 1S V_Gen_1S=450; %Peak Line to Line Voltage Phi_Gen_1S=0; %Phase A Phi F_gen_1S=60; %Frequency Produced by Generator in Hertz PF = .8; %System Power Factor %Generator 2S V_Gen_2S=450; %Peak Line to Line Voltage Phi_Gen_2S=0; %Phase A Phi F_gen_2S=60; %Frequency Produced by Generator in Hertz %Generator 1E V_Gen_1E=450; %Peak Line to Line Voltage Phi_Gen_1E=0; %Phase A Phi F_gen_1E=60; %Frequency Produced by Generator in Hertz %Circuit Breaker Controls (0 Open, 1 Close d) %Generators CB1S = 1; %Circuit Breaker between Generator and Loads CB1E = 0; %Circuit Breaker between Generator and Loads CB2S = 0; %Circuit Breaker between Generator and Loads

PAGE 107

95 %Bus Tie BT1S2S = 1; BT1E2S = 1; %1S Loads CBF1S = 1; %Forward 1S Generator Loads P1SF_Max = 202.1e3; P1SF = .03676*P1SF_Max; CBA1S = 1; %Amidships 1S Generator Loads P1SA_Max = 130.8e3; P1SA = .05726*P1SA_Max; CBAft1S = 1; %Aft 1S Generator Loads P1SAft_Max = 51.2e3; P1SAft = .1355*P1 SAft_Max; %2S Loads CBF2S = 1; %Forward 2S Generator Loads P2SF_Max = 202.1e3; P2SF = .28981*P2SF_Max; CBA2S = 1; %Amidships 2S Generator Loads P2SA_Max = 130.8e3; P2SA = .27365*P2SA_Max; CBAft2S = 1; %Aft 2S Generator Loads P2 SAft_Max = 51.2e3; P2SAft = .17512*P2SAft_Max; %1E Loads CBF1E = 1; %Forward 1E Generator Loads P1EF_Max = 202.1e3; P1EF = .09231*P1EF_Max; CBA1E = 1; %Amidships 1E Generator Loads P1EA_Max = 130.8e3; P1EA = .03337*P1EA_Max; CBAft1 E = 1; %Aft 1E Generator Loads P1EAft_Max = 51.2e3; P1EAft = .03438*P1EAft_Max; Pload = P1EA+P1EAft+P1EF+P1SA+P1SAft+P1SF+P2SA+P2SAft+P2SF; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Di stribution System Model %Shipboard General Quarters Parameters %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %60 Hz Generator Parameters %Generator 1S V_Gen_1S=450; %Peak Line to Line Voltage Phi_Gen_1S=0; %Phase A Phi F_gen_1S=60; %Frequency Produced by Generator in Hertz PF = .8; %System Power Factor

PAGE 108

96 %Generator 2S V_Gen_2S=450; %Peak Line to Line Voltage Phi_Gen_2S=0; %Phase A Phi F_gen_2S=60; %Frequency Produced by Generator in Hertz %Generator 1E V_Gen_ 1E=450; %Peak Line to Line Voltage Phi_Gen_1E=0; %Phase A Phi F_gen_1E=60; %Frequency Produced by Generator in Hertz %Circuit Breaker Controls (0 Open, 1 Closed) %Generators CB1S = 1; %Circuit Breaker between Generator and Loads CB1E = 1; %Circuit Br eaker between Generator and Loads CB2S = 1; %Circuit Breaker between Generator and Loads %Bus Tie BT1S2S = 0; BT1E2S = 0; %1S Loads CBF1S = 1; %Forward 1S Generator Loads P1SF_Max = 202.1e3; P1SF = .68699*P1SF_Max; CBA1S = 1; %Amidships 1S Gene rator Loads P1SA_Max = 130.8e3; P1SA = .88547*P1SA_Max; CBAft1S = 1; %Aft 1S Generator Loads P1SAft_Max = 51.2e3; P1SAft = .37624*P1SAft_Max; %2S Loads CBF2S = 1; %Forward 2S Generator Loads P2SF_Max = 202.1e3; P2SF = .37376*P 2SF_Max; CBA2S = 1; %Amidships 2S Generator Loads P2SA_Max = 130.8e3; P2SA = .86083*P2SA_Max; CBAft2S = 1; %Aft 2S Generator Loads P2SAft_Max = 51.2e3; P2SAft = .24412*P2SAft_Max; %1E Loads CBF1E = 1; %Forward 1E Generator Loads P1 EF_Max = 202.1e3; P1EF = .67107*P1EF_Max; CBA1E = 1; %Amidships 1E Generator Loads P1EA_Max = 130.8e3; P1EA = .93325*P1EA_Max; CBAft1E = 1; %Aft 1E Generator Loads P1EAft_Max = 51.2e3; P1EAft = .275*P1EAft_Max; Pload = P1EA+P1EAft+ P1EF+P1SA+P1SAft+P1SF+P2SA+P2SAft+P2SF;

PAGE 109

97 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %Shipboard Underway Steaming Parameters %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %60 Hz Generator Parameters %Generator 1S V_Gen_1S=450; %Peak Line to Line Voltage Phi_Gen_1S=0; %Phase A Phi F_gen_1S=60; %Frequency Produced by Generator in Hertz PF = .8; %System Power Factor %Generator 2S V_G en_2S=450; %Peak Line to Line Voltage Phi_Gen_2S=0; %Phase A Phi F_gen_2S=60; %Frequency Produced by Generator in Hertz %Generator 1E V_Gen_1E=450; %Peak Line to Line Voltage Phi_Gen_1E=0; %Phase A Phi F_gen_1E=60; %Frequency Produced by Generator in Hertz %Circuit Breaker Controls (0 Open, 1 Closed) %Generators CB1S = 1; %Circuit Breaker between Generator and Loads CB1E = 0; %Circuit Breaker between Generator and Loads CB2S = 1; %Circuit Breaker between Generator and Loads %Bus Tie BT1S2S = 1; BT1E2S = 1; %1S Loads CBF1S = 1; %Forward 1S Generator Loads P1SF_Max = 202.1e3; P1SF = .13305*P1SF_Max; CBA1S = 1; %Amidships 1S Generator Loads P1SA_Max = 130.8e3; P1SA = .219*P1SA_Max; CBAft1S = 1; %Aft 1S Generator Loads P1SAft_Ma x = 51.2e3; P1SAft = .17188*P1SAft_Max; %2S Loads CBF2S = 1; %Forward 2S Generator Loads P2SF_Max = 202.1e3; P2SF = .13305*P2SF_Max; CBA2S = 1; %Amidships 2S Generator Loads P2SA_Max = 130.8e3; P2SA = .219*P2SA_Max;

PAGE 110

98 CBAft2S = 1 ; %Aft 2S Generator Loads P2SAft_Max = 51.2e3; P2SAft = .17188*P2SAft_Max; %1E Loads CBF1E = 1; %Forward 1E Generator Loads P1EF_Max = 202.1e3; P1EF = .13305*P1EF_Max; CBA1E = 1; %Amidships 1E Generator Loads P1EA_Max = 130.8e3; P1EA = .219*P1EA_Max; CBAft1E = 1; %Aft 1E Generator Loads P1EAft_Max = 51.2e3; P1EAft = .17188*P1EAft_Max; Pload = P1EA+P1EAft+P1EF+P1SA+P1SAft+P1SF+P2SA+P2SAft+P2SF; Qload = (Pload/PF)*sin(acos(PF)); %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %Open Generator Circuit Breaker Test %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Generator Circuit Breaker Steps %Gene rator 1S CB1S_Step_Time = 1.5; CB1S_Init_Val = 0; CB1S_Fin_Val = 1; %Generator 2S CB2S_Step_Time = 1; CB2S_Init_Val = 0; CB2S_Fin_Val = 0; %Generator 1E CB1E_Step_Time = 1; CB1E_Init_Val = 0; CB1E_Fin_Val = 0; %Start Large Load Steps %Aft 2S Loads CBAft2S = 1; AFT2S_Step_Time = 1; AFT2S_Init_Val = 0; AFT2S_Fin_Val = 0; %Battle Damage Steps Battle_Damage_Step_Time = 1; Battle_Damage_Init_Val = 0; Battle_Damage_Fin_Val = 0 ;

PAGE 111

99 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %Start Large Load Test %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Generator Circui t Breaker Steps %Generator 1S CB1S_Step_Time = 1; CB1S_Init_Val = 0; CB1S_Fin_Val = 0; %Generator 2S CB2S_Step_Time = 1; CB2S_Init_Val = 0; CB2S_Fin_Val = 0; %Generator 1E CB1E_Step_Time = 1; CB1E_Init_Val = 0; CB1E_Fin_Val = 0; %Start Large Load Steps %Aft 2S Loads CBAft2S = 0; AFT2S_Step_Time = 1; AFT2S_Init_Val = 0; AFT2S_Fin_Val = 1; %Battle Damage Steps Battle_Damage_Step_Time = 1; Battle_Damage_Init_Val = 0; B attle_Damage_Fin_Val = 0; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %Battle Damage Test %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % %Generator Circuit Breaker Steps %Generator 1S CB1S_Step_Time = 1; CB1S_Init_Val = 0; CB1S_Fin_Val = 0; %Generator 2S CB2S_Step_Time = 1; CB2S_Init_Val = 0; CB2S_Fin_Val = 0;

PAGE 112

100 %Generator 1E CB1E_Step_Time = 1; CB1E_Init_Val = 0; CB1E_Fin_Val = 0; %Start Large Load Steps %Aft 2S Loads CBAft2S = 1; AFT2S_Step_Time = 1; AFT2S_Init_Val = 0; AFT2S_Fin_Val = 0; %Battle Damage Steps Battle_Damage_Step_Time = 1; Battle_Damag e_Init_Val = 0; Battle_Damage_Fin_Val = 1; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %Trial 7: Deadship Test %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%% %Generator Circuit Breaker Steps %Generator 1S CB1S_Step_Time = .5; CB1S_Init_Val = 0; CB1S_Fin_Val = 1; %Generator 2S CB2S_Step_Time = .5; CB2S_Init_Val = 0; CB2S_Fin_Val = 0; %Gener ator 1E CB1E_Step_Time = 1; CB1E_Init_Val = 0; CB1E_Fin_Val = 1; %Start Large Load Steps %Aft 2S Loads CBAft2S = 1; AFT2S_Step_Time = .5; AFT2S_Init_Val = 0; AFT2S_Fin_Val = 0; %Battle Damage Steps Battle_Damage_S tep_Time = 4; Battle_Damage_Init_Val = 0; Battle_Damage_Fin_Val = 0;

PAGE 113

101 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %ACDS Transmission Line Parameters %MSEE Thesis %%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Rp = [.5284 .5284]; Lp = [9.43e 5 9.43e 5]; Cp = [1.605e 7 1.605e 7]; %Cable Lengths TL1SFL = 2.273E 6; TL1SAL = 1.585E 6; TL1SAFTL = 3.463E 6; TL2SFL = 2.541E 6; TL2SAL = 1.592E 6; TL2SAFTL= 3.432E 6; TL1EFL = 4.424E 6; TL1EAL = 2.91E 6; TL1EAFTL = 1.186E 6; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %DCDS Transmission Line Parameters %MSEE Thesis % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Rp = [.5284 .5284]; Lp = [9.43e 5 9.43e 5]; Cp = [1.605e 7 1.605e 7]; %Cable Lengths TL1SFL = 1.291e 6; TL1SAL = 1.291E 6; TL1SAFTL = 1.291E 6; TL2SFL = 1.291E 6; TL2SAL = 1.291E 6; TL2SAFTL= 1.291E 6; TL1EFL = 9.847E 6; TL1EAL = 4.674E 6; TL1EAFTL = 9.847E 6; TLDCBus = 6.438E 6;

PAGE 114

102 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Keoni Hutton %270' WMEC AC Distribution System Model %DCDS Power Electro nic Converter Parameters %MSEE Thesis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% R=.1; %L=.0151; L=.005; C=500e 6; Rload=32.23; Lload=1e 4; Fp=60; Ts=1e 4; SR = 500; %PWM Generator Switching Ratio Vg = 1000; DC_bus_star=750 ; Iq_star=0; d=.707; wn=339.46; wL=2*pi*Fp*L; Kpi=(2*pi*L)/(100*Ts); Kii=(2*pi*R)/(100*Ts); Kpv=1.5*(C*2*DC_bus_star*d*wn)/(3*DC_bus_star); Kiv=1.5*(C*DC_bus_star*wn*wn)/(3*DC_bus_star);