1. EXECUTIVE SUMMARIES ...................................................................................................................................
6
2. PUBLICATIONS AND REPORTS RESULTING FROM ONR SUPPORT ...................................................
83
3. PERSONNEL STATISTICS DURING THIS REPORTING PERIOD ............................................................
85
3.1. Total No of Co-PI: 8 ...........................................................................................................
85
3.2. Total No of students working on the grant ........................................................................
86

CHRYSSOSTOMIDIS GROUP:
Figure 1: Notional ship modeled in Paramarine with machinery spaces and zones delineated.
Figure 2: Notional ship machinery layout
Figure 3. A ring bus architecture modeled in Simulink [The MathWorks]. This example connects four generators (grey) to four zonal loads (orange) and four bus-connected loads
(orange). The switches are green when closed, red when open. In this image, the bow and stern disconnects are open, the energy storage module is disconnected, and one generator is disconnected.
KARNIADAKIS GROUP:
Figure 1: Overall MVDC (750V) System Diagram (Purple Team design).
Figure 2: (Left) DC Bus Voltage and (Right) DC inverter current when an additional resistive
9kW load is added in parallel with IM.
Figure 3: (Left) DC rectifier current and (Right) IM electrical torque when an additional resistive 9kW load is added in parallel with IM.
Figure 4: (Left) DC Bus Voltage and (Right) DC inverter current when an additional resistive
9kW load is added in parallel.
Figure 5: (Left) IM electrical torque and (Right) IM slip-frequency when an additional resistive
9kW load is added in parallel.
Figure 6: (Left) DC Bus Voltage and (Right) DC inverter current with smaller filter capacitors.
Figure 7: (Left) IM electrical torque and (Right) IM slip frequency with smaller capacitors.
Figure 8: (Left) Rectifier DC currents and (Right) DC bus voltage in the case with no droop control (Kd = 0) and with droop control (Kd = 0.59)
Figure 9: Sensitivity analysis showing (Left) variation of the DC rectifier current and (Right) DC bus voltage.
KIRTLEY GROUP:
Figure 1. Contra-Rotating Propeller Set
Figure 2. Lifting Line Lattice of bound and free, trailing vortices

Figures 3-4-5. Circulation Distributions given by the Lerbs and the Variational methods for various thrust loadings
Figure 6. Optimum Circulation Distributions
Figure 7. Axial induced velocities far downstream
Figure 8. Tangential induced velocities far downstream
Figure 9. Efficiency as a function of thrust loading with viscous forces considered
Figure 10. Efficiency as a function of thrust loading with viscous forces neglected
Figure 11. Efficiency as a function of advance ratio when viscous forces are considered
Figure 12. Efficiency of Contra-rotating propellers as a function of advance ratio
Figure 13: FEA Results for a 50lbf Applied Load
Figure 14: Roller Bearing Assembly Including Drive Shaft, Slip Rings and Brush Block
Figure 15: K089300-7Y Torque Speed Curve. Dashed line represents continuous operation; solid line represents peak output for intermittent operation.
Figure 16: Output Power Capability. Dashed line represents continuous operation; solid line represents peak output for intermittent operation.
Figure17: Assembled Test Fixture
Figure 18: Complete Assembly Explode
Figure 19: Test Fixture Internals
Figure 20: Encoder Assembly
Figure 21: Test Fixture Internals Exploded
Figure 22: Disassembled motor.
Figure 23: Rotor model
LEEB GROUP:
Figure 1: Multi-bus AC and DC power distribution
Figure 2: Block Diagram
Figure 3: Solid model
Figure 4: Short-circuit test results
Figure 5: Comparison of simulated and measured generator short-circuit current.
TRIANTAFYLLOU GROUP:

Figure 1: DDG 1000 Scale Model
Figure 2: LQR Controlled Turbine Model
Figure 3: Open and Closed Loop Transfer Function Singular Values
Figure 4: DDG 1000 Scale Model
Figure 5: Heading and Rudder Angle for Zig-Zag Maneuver
Figure 6: Dieudonné Spiral Stability Results for a 200 second and 400 second intervals.
Figure 7: Comparison of Surge Power for ASSET and AES Models
Figure 8: Execution of Heading Step Command and Associated Power Transience
Figure 9: Propeller Optimization Studies

1. E
XECUTIVE
S
UMMARIES
FACULTY RESPONSIBLE FOR THIS RESEARCH: Chrys Chryssostomidis
FACULTY, RESEARCHERS, AND STAFF CONTRIBUTING TO THIS RESEARCH:
Matthew Angle, Julie Chalfant, Chrys Chryssostomidis, David Hanthorn, James Kirtley
Architectural Model to Enable Power System Tradeoff Studies
We continue the development of an overall architectural model for an all-electric ship using a physics-based simulation environment to perform fully-integrated simulation of electrical, hydrodynamic, thermal, and structural components of the ship operating in a seaway. The goal of this architectural model is to develop an early-stage design tool capable of performing tradeoff studies on concepts such as AC vs. DC distribution, frequency and voltage level, inclusion of reduction gears, energy and power management options, and effect of arrangements and topology. The results of the studies will be presented in standard metrics including cost, weight, volume, efficiency/fuel consumption, reliability and survivability. We will specifically look at the hull, mechanical and electrical (HM&E) systems that support the ship and its missions; specifically, the electrical generation and distribution system, propulsion equipment, fresh- and saltwater pumping and distribution, control systems, and structural components.
We have previously reported the creation of a basic design tool which uses hullform, drive train particulars, and operational employment of the vessel to determine resistance, powering and fuel usage. Specific advancements reported herein include:
- Notional Ship modeled in Paramarine
- Electrical Load Analysis and Prioritization
- In-zone Distribution Equipment Modeling
- Metrics Development
^ Survivability Metric
^ Graph Theory Applied to the Survivability Metric
FACULTY RESPONSIBLE FOR THIS RESEARCH: Franz Hover
FACULTY, RESEARCHERS, AND STAFF CONTRIBUTING TO THIS RESEARCH:
Franz Hover, Joshua Taylor

We are developing efficient functions that characterize the robustness of large-scale power systems. Exhaustive, simulation-based approaches to robust design can scale poorly, and this motivates new tools.
We approach the problem by modeling the distribution system as a graph with power flows at each vertex, and formulate specialized eigenvalue problems using variants of the graph
Laplacian matrix. Because eigenvalues are easy to compute (n3), they provide quantities wellsuited for use in contingency analysis, i.e. determining functionality after damage, and optimization. Drawing on spectral graph theory, we use these metrics to write practical upper and lower bounds for an underlying, NP-hard robustness parameter.
FACULTY RESPONSIBLE FOR THIS RESEARCH: George Karniadakis
FACULTY, RESEARCHERS, AND STAFF CONTRIBUTING TO THIS RESEARCH:
George Karniadakis, Mirjana Milosevic Marden
The main activity in this quarter was the development and implementation of Matlab models for the Purple Team design. The main challenges to overcome (new compared to previous work on the AC system) were the modeling of appropriate controllers for the synchronous generator and induction motors as well as the detailed modeling of rectifiers. The first model we developed included the generation system assuming a hypothetical resistive load. We then modeled the propulsion system assuming an ideal DC voltage source. After verifying and validating both submodels we combined them and performed extensive tests on the end-to-end model. The results we include here compare against the results obtained by the Purdue team. In addition, we include results that test the stability of the end-to-end system in the presence of two generators, and also a sensitivity study result using polynomial chaos.
FACULTY RESPONSIBLE FOR THIS RESEARCH: James Kirtley
FACULTY, RESEARCHERS, AND STAFF CONTRIBUTING TO THIS RESEARCH:
Julie Chalfant, Chrys Chryssostomidis, Steve Englebretson, Jerod Ketcham, James Kirtley

Tools for the early-stage design of propeller and motor combinations for electric-drive ships are of interest, as are the more detailed design and testing of promising candidates. Four closely related research projects pertaining to the design, construction and testing of various propellermotor combinations for electric drive vessels are described below. Ship propulsion induction motor design software has been developed to accomplish both early-stage general design and more detailed analysis for later stage design. OpenProp is an open-source propeller design code; recent work contributes to expanding its capabilities to design contra-rotating propellers. A propeller and turbine test fixture is being developed for use in verification testing of ducted propellers and turbines designed using OpenProp, and a proof-of-concept construction of a contra-rotating has begun.
A contra-rotating propulsion system may be able to take advantage of an increase in relative motor speed to significantly reduce the size and weight of the drive motor while minimizing increase in propeller speed and the potential efficiency reduction or propeller damage from cavitation. Furthermore, significant increases in propeller efficiency on the order of 5-10 percent have been reported for contra-rotating compared to single propellers. In fact, the preliminary overall ship architectural model described in Prof. Chryssostomidis’ section of this report was used to study the effect of a contra-rotating propeller and motor combination on a notional Navy destroyer [Chalfant et al. 2009]; this study demonstrated significant savings in weight and annual fuel usage.
One goal of this research is to design a contra-rotating propeller using the capabilities of
OpenProp to design and produce, using 3D printing, a candidate contra-rotating propeller, then test the propeller-motor combination using the proof-of-concept motor that is constructed in this work in the MIT test facilities.
FACULTY RESPONSIBLE FOR THIS RESEARCH: Steven B. Leeb
FACULTY, RESEARCHERS, AND STAFF CONTRIBUTING TO THIS RESEARCH:
Steven B. Leeb

During the performance period, we have continued a multi-pronged development of analytical and hardware demonstrations directed at determining the feasibility and desirability of different proposed shipboard power distribution systems, including MVDC. We have completed an initial design and preliminary construction for our laboratory scale MVDC hardware test bed. This test bed will be used to explore possibilities for zonal protection, condition-based monitoring, and control on MVDC systems. It will also be used to consider the desirability of a hybrid AC and
DC power distribution system, or MVADC power.
Our work has been invaluably informed by our on-going interaction with the USCG Cutters
ESCANABA and SENECA. During the reporting period, we have continued our interaction with the crews of these ships, especially USCGC ESCANABA. These interactions provide valuable data about current practices in ship power system control, load sizing and rating, diagnostic conditions in mission critical loads, and opportunities for zonal protection and condition-based maintenance. Field data is used to guide our MV(A)DC analysis and test bed construction.
FACULTY RESPONSIBLE FOR THIS RESEARCH: Michael Triantafyllou
FACULTY, RESEARCHERS, AND STAFF CONTRIBUTING TO THIS RESEARCH:
Michael Triantafyllou
In accordance with our project plans, we have continued the simulation, validation, and analysis of our electric ship model. Work was especially focused on the gas turbine and hydrodynamic elements to be coupled to high-order integrated power system models being developed elsewhere in the AES community. Finally, we continued developing control protocols to simulate the response of the ship under transient conditions.

Technical Details by Task or Project:
• Key Accomplishments
CHRYSSOSTOMIDIS GROUP
Architectural Model to Enable Power System Tradeoff Studies
Key Accomplishments:
Architectural Model to Enable Power System Tradeoff Studies: Key accomplishments this quarter include the specification of the notional ship and modeling thereof in Paramarine, the analysis and prioritization of the electrical load on the notional ship, the initial modeling of inzone power conversion equipment, and the development of pertinent metrics, especially the survivability metric.
Technical Detail:
Architectural Model to Enable Power System Tradeoff Studies:
We describe here the further development of an overall architectural model for the physics-based simulation of an all-electric warship in an operational environment. This simulation integrates electrical, mechanical, structural, thermal and hydrodynamic components to accomplish trade-off studies against a standard set of metrics for evaluation.
Notional Ship:
As in previous work, the notional ship is an electric-drive small surface combatant based upon
SFI-2 as described in [Webster et al., 2007]. This notional ship was modified from SFI-2 to include the specific concepts shown in Table 1. The vessel is an all-electric surface combatant of approximately 10,000 tons with primary mission areas of anti-submarine warfare (ASW), antisurface warfare (ASUW) and anti-air warfare (AAW).
-----------------------------------------------------------------------------------------------------------
Table 1 – Notional Ship Requirements

All-electric ship: all systems requiring power are electric and serviced by an Integrated Power
System (IPS)
Power system architecture: Integrated Power System (IPS) and zonal distribution consistent with
Next-Generation IPS (NGIPS) architecture
Size: ~10k tons displacement
Weapons: Rail gun, VLS, point defense system, electronic attack
Sensors/Comms: Integrated communications suite based on anticipated future capabilities, advanced sensing suite (imagery, electronic surveillance, etc.), air and missile defense radar suite, sonar suite
Cmd&Ctrl/Computing: Integrated command and control, total ship computing
Aviation: Accommodate 2 helicopters or 3 vertical take-off unmanned aerial vehicles
-----------------------------------------------------------------------------------------------------------
The ship is modeled in Paramarine version 6.1 created by Graphics Research Corporation
(GRC). Paramarine is a commercial product which provides integrated Naval Architecture
Design and Analysis; details can be found at http://www.qinetiq.com/home_grc/products.html.
We chose Paramarine as the modeling component of this architectural tool due to its modular structure, physics-based design components, and ease of exporting and importing information.
New advanced modeling components developed through the overall architectural model will be overlaid upon the Paramarine capability.
The basic ship layout and hull shape were taken from a notional low-level design of a balanced
Frigate and scaled up from 6,000 tons to 10,000 tons. An initial estimate of required installed power was created by combining a maximum speed of 30 knots, requiring 81 MW of installed power, with a future estimated electrical loading of 37 MW based on the predicted value for future margined ship service loads from [Webster et al., 2007]. The total required power of 118
MW was met with six gas turbines: two 36 MW, two 19.5 MW and two 4.5 MW gas turbines, yielding a surplus of 2MW. Design margins are included in the 37MW ship service power estimate. Figure 1 shows the notional ship with machinery spaces and zones delineated. Figure
2 shows the machinery space layout including gas turbines and generators, propulsion motors and shafts.

The ship is divided into 6 zones for survivability purposes as shown in Figure 1. Support services will be designed so that distribution branching occurs within a zone and only major cross-connection capability crosses a zone boundary. Isolation devices are located at the zonal boundaries to permit continued operation of systems in adjacent un-damaged zones. Redundant equipment is located in separate zones if possible. While 6 zones were chosen as a starting point, the number and location of zones is a subject for potential future tradeoff studies.
The separation of zones occurs at the transverse watertight bulkheads between each of the machinery spaces. This placed at least one Power Generation Module (PGM) in each zone with the exception of zone 1, which is forward of all the machinery spaces. There is the possibility of placing an Energy Storage Module (ESM) in this zone to provide an additional power source in the event of a casualty. Since cableways run fore and aft the length of the machinery spaces, the decision was made to extend each zone vertically into the superstructure instead of separating the superstructure into a separate zone. This also allows the vital equipment contained in the superstructure to be divided into multiple separate zones.
Figure 1: Notional ship modeled in Paramarine with machinery spaces and zones delineated.

Figure 2: Notional ship machinery layout
One major advantage to designing the ship with an Integrated Propulsion System (IPS) is the flexibility the designer has in placing the Power Generation Modules (PGM) and Propulsion
Motor Modules (PMM). Gas turbines were chosen because of the high power levels required and their advantage in power density over diesel engines. During the initial design and layout of the IPS, this flexibility became significantly constrained by the placement of intake and exhaust stacks from the gas turbines. The exhaust could not be routed forward of the superstructure due to the obscured view from the bridge, hazard to the crew and seawater ingestion. Because the ship supports helicopter operations, the exhaust could not be routed aft because it would interfere with helicopter take-offs and landings. This led to a layout similar to that of current ships in the fleet.
There are options to be considered in future design iterations that could remove some of these restrictions. One would be to move the bridge to a more forward part of the ship allowing the gas turbines to be placed farther forward in the ship. Using diesels for smaller power loads would allow more flexibility in the air duct routing. Lastly an option to consider is placing some of the gas turbines aft where their exhaust would still interfere with flight ops but impose restrictions on the concept of operations (ConOps) so that the aft gas turbines would be restricted from operating when flight operations are in progress. These possibilities would remove some of the constraints allowing the designer more flexibility in locations and layout. In particular this would allow for more separation between PGMs resulting in a more survivable design.

The placement of the PMMs was also restricted within the ship due to the length of shafting required to obtain an acceptable shaft angle. This forced the PMMs farther forward with only one compartment of separation. If podded propulsion were used in future iterations this restriction would be lifted.
Electrical Load Analysis and Prioritization:
We have developed a prioritized listing of electrical loads; this listing is needed for many aspects of the project, such as load shedding, electrical component sizing, cooling system design, and survivability calculations.
The initial step was to develop a prioritized general categorization of electrical loads, assuming the ship was in a damaged state. This prioritized list indicates the electrical loads that should be serviced in order of priority for the ship to survive the damage, continue to operate, and meet its mission. This prioritized list is shown in Table 2.
This prioritized list was then further broken down into specific electrical loads assigned to zones with an associated heat generation factor and cooling medium. Load usage was estimated for various battle conditions and external temperatures. Several sources of data were used to compile this high-level list of electrical loads for the notional ship. BMT Syntek provided a load analysis for ESRDC use which included typical loads divided into five zones, and provided heat conversion factors and cooling media for each load. The BMT Syntek list did not include any values for advanced sensors or weapons and did not include propulsion loads. Unclassified copies of electrical load analysis charts for DDG 51, DD 963 and FFG 7 class ships compiled by
Gibbs and Cox and by Ingalls Shipbuilding were consulted for typical ship service loads as well.
To ensure the resulting list is unclassified, electrical analysis was maintained at a high level, not delving into specific pieces of equipment or their specific location. A sample of the load breakdown by zone and power requirements for the first three items in the electrical load priority list is shown in Table 3.
Using this loading scheme, the total electrical load, divided into zones and into type of power required (60 Hz AC, 400 Hz AC, and DC) is listed in Table 2. Note that ‘direct-fed DC’ is used

to denote items that are fed directly from the bus through their own dedicated transformer, rectifier or drive instead of being fed from a load center. These direct-fed loads include the propulsion motors, radar and rail gun.
Energy storage is to be included in further research.
-----------------------------------------------------------------------------------------------------------
Table 2: Electrical Load Priority List
1. Power generation
For each generator that is operational, supply power to the associated auxiliaries such as lube oil, vent fans, fuel, starter, control panel, alarms, etc. Activate the engineering control system.
2. Damage Control
Fire fighting, dewatering, lighting, electrical receptacles (for powering damage control equipment such as submersible pumps, blowers), basic ventilation, interior communications, countermeasure washdown. Emergency medical support. Potable water distribution.
3. Basic Mobility
Propulsion to achieve 4 knots (steerageway). One motor and drive, associated propulsion sensors, control system. One steering system (hydraulics, controls).
4. Self Defense
SeaRAM/CIWS or equivalent (self-contained system that requires only electrical power and chill water to operate). Electronic warfare sensor (such as SLQ-32 or basic InTop) and missile decoy such as NULKA or chaff. Degaussing. Chill water pump, A/C compressor and ASW pump.
5. Exterior Communications
Full external communications suite including IFF, TACAN. Navigation system to include basic close-in radar, gyro. Assume that the cooling online already can also support this equipment.
6. Helicopter and Boat Recovery
Basic helicopter systems required for recovery of helicopters only, not full flight operations; e.g., landing lights, RAST, winches, control station. Power to boat davit.
7. Increase speed to 10 knots.
8. Basic Offense
Sensors, Weapons, Ship Control System. Assume a low-power level of radar is available with degraded performance either reducing range, resolution, or both. Bring full hull-mounted sonar capability online. Energize ship control system including weapons control and sensor integration

in all intact zones. For weapons, energize one of any redundant systems such as gun, torpedo,
VLS. Only energize a weapon if supporting sensors are available; e.g., do not energize torpedoes if sonar not available, do not energize VLS if sufficient radar not available. Increase chill water capacity by adding pump, compressor, and ASW pump. Add remaining ventilation systems.
9. Full Flight Operations
All flight equipment not already energized, including refueling capability and fuel transfer.
10. Increase Offense
Increase radar to medium power level. Increase cooling capacity to support. Activate the remaining weapons systems, excluding high-energy weapons.
11. Increase speed to 20 knots.
12. Miscellaneous supporting mechanical services.
Fan cooling units for all spaces. Air compressors. Battery chargers for small boats, helo, gyro.
Potable water equipment including distillers, reverse osmosis units, brominator, etc.
13. Full Offense
Increase radar to full power level. Provide sufficient cooling. Energize high-energy weapons
(rail gun, laser). Activate towed-array sonar, if applicable.
14. Increase speed to 25 knots.
15. Non-Vital Loads
All non-vital loads. Examples include galley equipment, heat, water heaters, laundry, stores handling, and miscellaneous machinery systems such as fuel transfer, air compressors, anti-icing, lube oil heating, sewage treatment.
16. Increase speed to maximum possible.
----------------------------------------------------------------------------------------------------------

Table 3: Electrical requirements by zone and category
Battle Power Requirement, kW
60 400 Dire
Hz
AC
Hz
AC DC ct
DC TOT equipment
POWER GENERATION
PGM 1 (med) AUXILIARIES
PGM 2 (large) AUXILIARIES
PGM 3 (small) AUXILIARIES
PGM 4 (small) AUXILIARIES
PGM 5 (large) AUXILIARIES zone
PGM 6 (med) AUXILIARIES
ENG CTRL SYS zone 1 (no MR)
ENG CTRL SYS zone 2 (AMR)
ENG CTRL SYS zone 3 (MMR)
ENG CTRL SYS zone 4 (AMR)
ENG CTRL SYS zone 5 (MMR)
ENG CTRL SYS zone 6 (AMR, eng control) 6
DAMAGE CONTROL
Fire Pump (two of 6 on line)
Fire Pump
1
2
3
4
5
6
1
2
2
3
4
4
5
Fire Pump
Fire Pump
Fire Pump
Fire Pump receptacles, IC, vent, water coolers receptacles, IC, vent, water coolers receptacles, IC, vent, water coolers receptacles, IC, vent, water coolers receptacles, IC, vent, water coolers receptacles, IC, vent, water coolers
IC 400 Hz equipment interior and exterior lighting interior and exterior lighting interior and exterior lighting interior and exterior lighting interior and exterior lighting interior and exterior lighting
Medical Equipment
BASIC MOBILITY steering gear pump, steering control system steering gear pump, steering control system various propulsion sensors, power panels motors and motor drive various propulsion sensors, power panels
6
1
2
3
4
5
5
6
3
4
3
4
5
6
1
6
1
2
6
6
3
3
5
100
130
60
60
130
100
3
7
10
7
10
25
0
12
12
12
12
12
12
0
60
0
60
0
2
50
50
5
0
16
10
10
20
10
20
10
160
50
50
5
160
0
100
130
60
60
130
100
3
7
10
7
10
25
0
12
12
12
12
12
12
0
60
0
60
0
16
10
10
20
10
20
10
2

motors and motor drive 5 0
-----------------------------------------------------------------------------------------------------------
0
Table 4: Electrical load in kW by zone and power type
Zone
3
4
5
1
2
6
Total
60 Hz
AC
429
542
1,488
413
424
731
4,027
400 Hz
AC
55
0
4
0
0
16
75
DC
46
88
50
46
66
88
384
Subtotal Direct-fed
530
630
1,541
459
490
835
4,486
DC
20,000
0
44,765
0
35,130
0
99,895
Total
20,530
630
46,306
459
35,620
835
104,381
In-zone Distribution Equipment Modeling:
Included in the architectural model is the basic power conversion and distribution equipment including location, connectivity, weight and volume. As an initial starting position for modeling inzone power distribution and conversion, off-the-shelf equipment performance numbers for power conversion equipment are scaled based on the expected size of the loads in the zone to determine approximate weight and volume. MVDC is an emerging technology, so equipment at an appropriate voltage level is not commercially available off-the-shelf; we used equipment that performs the same function at different voltage levels as a basis. Information was obtained from
Dennis Mahoney and Ed Wagner of Satcon and pertains to the equipment planned for DDG 1000.
The PCM-1 is the primary DC-DC converter and converts power from bus voltage down to a more usable 800V. The PCM-1 contains several smaller, line-replaceable units called SSCMs, each rated at 100 kW. The PCM-1 feeds the PCM-2, which is an inverter converting from the DC output of the PCM-1 to 450V 60Hz AC supplied to loads in the zone. It contains line-replaceable units called
SSIC modules.

Detailed weights and sizes of the PCMs, SSICs and SSCMs are presented in Table 5. Note that the specific power (power per unit weight) of the PCM-1 is half of the specific power of the SSCM and the power density (power per unit volume) is about a tenth that of the SSCM. Similarly, the specific power and power density for the PCM-2 is significantly lower than that of the SSIC. The numbers for the PCMs include main disconnects, relays, diodes, cooling system, man-machine interface, and other hardware in addition to the SSCM/SSIC functionality. The DDG 1000 bus voltage is 1kV.
For a conceptual zonal power distribution design with an MVDC bus, the same basic architecture is used. Voltage is stepped down from the 10kV distribution bus before being converted in to 60 or
400 Hz AC. This means that the DC-DC converters convert all power that is used for DC and both types of AC loads. Thus the total power for each zone is divided by the specific power of the PCM-
1 to get the mass for the DC/DC converter. Similarly, the mass for the 60 Hz and 400 Hz inverters are obtained by dividing the power for each subsection by the specific power for that converter.
The 400Hz inverter should be significantly lighter than the 60Hz converter since the size of the magnetic components in the converter scale inversely with frequency. The output filter and the inductor inside it are the driving forces behind the mass of the converter. The filter inductor value varies inversely with frequency. The number of turns varies linearly with the inductor value. Flux varies linearly with number of turns, assuming current is constant. Core size varies linearly with flux to avoid saturation. Core size drives the weight of the inductor. Following this logic, the dominant piece of the converter should be 400/60, or 6.67 times smaller than the 60 Hz inverter.
The amount of 400 Hz power is insignificant in comparison to the amount of other types of power, so sizing of the inverter has reduced significance. For the purpose of obtaining round numbers, the
400 Hz inverter is assumed to be ¼ the mass and volume of the 60 Hz inverter.
The results of equipment scaling are shown in Tables 6 and 7. Table 6 presents the mass of the scaled equipment, and table 7 presents scaled volumes, based upon the power requirements shown in Table 4.
-----------------------------------------------------------------------------------------------------------

Table 5: Existing equipment power, mass and volume
Converter Function Power
(kW)
Weight
(kg)
Specific
Power
Dimensions
(m x m x m)
Volume
(m^3)
Power
Density
SSCM
SSIM
DC/DC
DC/AC
100
90
PCM-1 DC/DC 900
272
166
4495
(kW/kg)
0.37
0.54
0.20
(kW/m^3)
0.20
0.48 0.91 0.090
1115
0.20
0.48 0.91 0.090
1004
2.80
1.22 1.92 6.559
137
PCM-2 DC/AC 540 7600 0.07
2.30
1.22 1.92 5.388
100
-----------------------------------------------------------------------------------------------------------
Table 6: Scaled mass in kg of power conversion equipment in each zone. Note that all power is first converted to 800V DC.
PCM-2 PCM-2 PCM-1
2
3
Zone
1
60hz
6,038 modified
400Hz
194
7,628 0
20,942 14
DC
2,647
3,147
7,696
Subtot
8,878
10,775
28,653
4
5
5,813
5,967
0
0
2,292
2,447
8,105
8,415
6
Total
10,288
56,676
56
264
4,170
22,400
14,515
79,340
-----------------------------------------------------------------------------------------------------------
Table 7: Scaled volume in m3 of power conversion equipment in each zone.
PCM-2 PCM-2 PCM-1
Zone
1
2
3
4
5
6
Total
60hz
4.3
5.4
14.8
4.1
4.2
7.3
40.2 modified
400Hz
0.1
0.0
0.0
0.0
0.0
0.0
0.2
DC
3.9
4.6
11.2
3.3
3.6
6.1
32.7
Subtot
8.3
10.0
26.1
7.5
7.8
13.4
73.0
Metrics Development:
The following set of metrics is selected to provide a consistent background for analysis of tradeoff variants:

- weight
- volume
- efficiency/fuel consumption
- cost
- reliability
- survivability
The weight, volume, efficiency and fuel consumption metrics are well defined and described below.
The cost and reliability metrics are still being developed; initial thoughts are presented herein.
During this reporting period, significant study has gone into the survivability metric; this research is presented below as well.
The weight and volume metrics are defined simply as the change in weight and volume occupied by the equipment in each tradeoff as compared to the baseline.
The metric for efficiency is annual fuel consumption. Efficiency of the systems as installed will be traced back to total load placed on the engines in each of several battle conditions and ship speeds.
The amount of fuel required annually is then determined using a speed-condition profile that delineates the percent of time spent at various speeds and battle conditions, the engines’ specific fuel consumption (SFC) at each loading, and an engine usage profile specifying numbers and types of engines online and load sharing between engines at each speed-condition state.
We plan to use a twofold measure of cost: first, as a measure of operating cost, the annual fuel usage is translated to dollars using the current bunker price of marine diesel oil. Second, we hope to use a purchase price for the equipment selected as a measure of construction cost; however, we are still working to obtain relevant purchase price data for some equipment. This may prove quite difficult due to the cutting-edge nature of some of the technology.
The cost metric is ripe for study; there are many factors of cost that are not yet incorporated into this metric. For example, installation and maintenance costs associated with how tightly the equipment is packed into the space and how easily the cable can be routed, maintenance costs associated with

reliability of the equipment, training costs for Navy personnel to operate and maintain the equipment, and disposal costs related to the amount of hazardous waste incorporated.
The reliability metric is yet to be fully defined. It will most likely be some combination of meantime-between-failure data along with the quality of service metric as described in [Doerry and
Clayton, 2005].
Survivability Metric:
The current method for analyzing survivability of a ship is accomplished using a the mission set of the ship and a series of potential threats. The ship’s ability to respond to those threats, including the ability to defeat or evade the threats before impact, the ability to withstand the impact, and the ability to recover from the impact, are combined into a measure of survivability involving Design
Threat Outcomes that range from completely unaffected to total loss of the ship and personnel. See, for example, [Doerry 2007]. This extensive survivability analysis can take weeks to properly calculate; in the early stages of design, the ship may have gone through several iterations in that amount of time, rendering the survivability analysis obsolete before its completion. In addition, this metric is a measure of the survivability of the ship as a whole including the effectiveness of a multitude of factors such as weapons systems, personnel performance, and ship design.
We seek a metric to rapidly measure the impact of support system design on the overall survivability of the vessel. Thus, when comparing various electrical distribution systems, cooling methodologies, or other hull, mechanical and electrical (HM&E) systems, we can accurately compare the survivability of the ship using the given system. The survivability metric measures the relative survivability of ship variants when subjected to damage.
The new survivability metric is constructed as follows:
Given a set of loads, establish a weighted prioritized list for servicing the loads. The list of loads is dynamically prioritized such that a load that has been demolished by the imposed damage is removed from the list, as there is no reason at that point to allocate resources to it. These loads are then grouped into tiers indicating levels of survivability that establish expected capability of the ship following an attack.

Next, impose damage upon the ship. Initially, damage will be imposed in manually selected positions and will consist of a center of impact and a radius of destruction. All equipment within the radius is assumed destroyed. Service systems are traced back from the damage to the first intact isolation point; all services between the isolation points and the damage are assumed lost. Future research may include a stochastic imposition of damage simulating a series of threat weapon encounters at random areas of the ship.
The survivability score is twofold to measure two distinct issues. The first metric is the highest tier of survivability that can be achieved while simultaneously fulfilling all lower tiers. The second metric is a total load value that indicates the maximum value of all loads that can be serviced, proceeding in the order of the dynamically prioritized list, even if some higher priority loads cannot be serviced due to connectivity problems.
This method can be used for any type of service or a combination of services: electrical power, chill water, seawater, ventilation air. It can also be tailored as desired by the user to increase or decrease the fidelity at which survivability is analyzed by changing the detail incorporated into the dynamically prioritized list. As an example, we present a thorough analysis of the survivability metric for analyzing the electrical distribution system on the notional ship described above.
Survivability Metric as applied to the Electrical Distribution System:
The survivability metric analysis described above requires a prioritized list of loads, a tiered survivability system, a set of sources and a distribution system. The prioritized list of electrical loads for the notional ship is based upon the electrical load priority list shown in Table 2 above.
Each of the items in the priority list is broken down into more specific loads which have an associated power requirement and zone location. The breakdown for the first three line items is shown in Table 3.
Levels of survivability were grouped into tiers that establish expected capability of the ship following damage. The survivability tiers developed for the notional ship are shown in Table 7; capabilities refer to the electrical load priority list shown in Table 2. For the ship to meet any

survivability tier, it must meet the minimum requirements of that tier and all lower tiers. For example, a ship that meets tier E must achieve 4 knots with steering, have one operational selfdefense system, two operational firepumps, and one generator with associated auxiliaries, and must have sufficient power to service all of those loads simultaneously.
The first two survivability tiers shown in Table 8 will be determined by analyzing the ship’s conditions following impact to determine if there is sufficient buoyancy to remain afloat. If there is not then the outcome will fall into either condition A or B. A notional distribution of personnel throughout the ship will be assumed based on general quarters at the time of impact. All personnel in the affected spaces will be lost and only 80% of those remaining will escape successfully.
The second measure of survivability is calculated as follows. Given a damage scenario, determine which generators are operational, which loads are operational, and the connectivity between the generators and loads. Then service all possible loads in a manner to achieve the highest weighted score, given the priority listing of loads, an example of which is shown in Table 3. For example, if the two medium-sized generators survive, then a total of 39 MW of power is available to service all undamaged loads; however, integrity of the distribution bus must be examined to determine connectivity between the generators and remaining loads. Then, the power is assigned to the loads for which connectivity exists in priority order until all power is allocated. Note that this is an exercise for calculating the survivability metric and is not a proposed method of power control or load shedding.
This survivability metric differs from that suggested in [Doerry 2007]. For example, we do not place the same emphasis on primary mission areas and do not include such factors as repair and restoration capabilities onboard or in-theater. While Doerry’s metric gives a more complete picture of ship survivability, it is significantly more difficult to calculate. Our goal in this research is to construct a metric that can be used in the early stages of design to rapidly assess the impact of decisions on the survivability of the ship, thus allowing survivability to be a factor in the decision making early in the design of the ship.
-----------------------------------------------------------------------------------------------------------

Table 8: Proposed Survivability Tiers
A Ship likely to be lost with the loss of over 25% embarked personnel
B Ship likely to be lost with the loss 25% or under embarked personnel
Does not meet Power Generation and Damage Control - Ship's status is likely to continue
C to degrade
Meets Power Generation and Damage Control- Ship is able to combat existing damage but
D
E is vulnerable to further threats.
Minimum requirements: one operational generator with associated auxiliaries, two firepumps.
Meets Basic Mobility and Self Defense - Is able to sustain itself against the enemy.
Minimum requirements: achieve 4 knots with steering, one self-defense system (either a
F
CIWS/SeaRAM or a sensor and associated missile decoy system).
Meets Exterior Communications and Helicopter and Boat Recovery - Basic functionality without offense.
Minimum requirements: one operational mode of exterior communications, close-in radar, able to recover helicopter and boat.
Meets Increased Speed and Basic Offense - Can perform at least one primary mission
G
H
Minimum requirements: operate at 10 knots, perform one primary mission.
Meets Full Flight Operations and Increased Offense - Can perform most or all primary missions
Minimum requirements: flight operations, two primary missions
Meets Increased Speed, Mechanical Services and Full Offense - As far as a military asset, unaffected I
Minimum requirements: power to all loads at this level and below
Complete functionality - all loads are fulfilled
J
Minimum requirements: fulfill all loads, including non-vital loads
-----------------------------------------------------------------------------------------------------------
Graph Theory Applied to the Survivability Metric :
As a step toward automating the calculation of the survivability metric, research exploring the application of graph theory concepts to a ring bus power flow example was conducted. The goal of this research is to answer the following question:
Given:
- m sources (generators) of known capacity, and

- n sinks (loads) of known size, with
- known connectivity between sources and sinks, and
- a weighted priority of satisfying each load, how should the sources be allocated to maximize the total weight (priority) of the satisfied loads?
The following additional notes apply as well:
- Not all sources are connected to all sinks.
- There is insufficient total power generated to satisfy all the loads.
- A single generator can simultaneously power more than one load.
- A single load can be satisfied using power from more than one generator.
- There are more individual loads than individual generators.
- Power cannot flow through loads.
- Power cannot flow back into the generators.
The diagram showing the network is found in Figure 3. This figure is based upon the ring bus architecture of the notional MVDC system model in RTDS at FSU CAPS.
The use of four algorithms is described here:
1. Multihop connectivity [Miller, 2001]- calculates the number of ‘hops’ from one node to each other node in an undirected graph, given an adjacency matrix, Aij. An undirected graph allows power to flow in either direction through a link; this means that for our example a load cannot be simultaneously linked to the bus through more than one link, or the total power of the network would be allowed to flow through the load to another load. This algorithm can be used to determine the length of the shortest path in number of nodes traversed from any one point in the network to any other point in the network. To determine full connectivity in our sample network, the algorithm must be run twice; once with each load connected to either the port or starboard bus and a second time with the load connected to the opposite bus.

Figure 3. A ring bus architecture modeled in Simulink [The MathWorks]. This example connects four generators (grey) to four zonal loads (orange) and four bus-connected loads (orange). The switches are green when closed, red when open. In this image, the bow and stern disconnects are open, the energy storage module is disconnected, and one generator is disconnected.
2. Dijkstra’s algorithm [Cormen, 2002]- calculates the lowest weight path from a single node to each other node in a directed weighted graph. Returns the weight (cost) of the path and the sequence of nodes in the lowest cost path. This allows the determination of shortest path dependent upon some characteristic of the path such as total resistance or length of cable. This algorithm eliminates the need to run twice as required in the multihop connectivity example, since the graph is directed (in a directed graph, flow can be specified to be one-way in either direction or two-way in any link between nodes).
3. Edmonds-Karp algorithm [Cormen, 2002] - calculates maximum possible flow in a network represented as a directed graph with capacities associated with each leg. The algorithm determines the maximum possible flow through the network and the flow in each identified path. This prevents you from overloading any section of the network, given the capacity of the cabling connecting that section of the network. However, this algorithm does not prioritize which load is filled first; it

maximizes the flow in the shortest path found using a breadth-first search, then seeks the next shortest path with capacity available.
4. Linear programming [The MathWorks]- using the LINPROG command in the MATLAB optimization toolbox, this approach finds the minimum of f*x subject to a set of linear constraints such that Ax ≤ b and lb ≤ x ≤ ub, where fi is the weight/priority of filling load xi, lb and ub are the lower and upper bounds on x, and A is the set of constraints imposed such that the total power supplied by a generator is less than or equal to the capacity of the generator, the total power supplied to a load is less than or equal to the size of the load, and a generator only supplies power to a load for which connectivity exists between the generator and the load. At present, the path followed by the power is not determined, and the losses therein are not calculated. Dijkstra’s algorithm is run first to determine connectivity between a generator and load.
REFERENCES
Cormen, Leiserson, Rivest, Stein, Introduction to Algorithms. MIT Press, Cambridge, MA, 2002.
N. H. Doerry, Designing Electrical Power Systems for Survivability and Quality of Service, Naval
Engineers Journal, 119:2, pages 25-34, 2007.
N. H. Doerry, Next Generation Integrated Power System (NGIPS) Technology Development
Roadmap. Ser 05D/349, Naval Sea Systems Command, Washington Navy Yard, DC, November
2007. Approved for public release, distribution unlimited.
N. H. Doerry and D. H. Clayton, Shipboard Electrical Power Quality of Service, Proceedings of the
IEEE Electric Ship Technologies Symposium, Philadelphia, PA, July 25-27, 2005.
L. E. Miller, Multihop Connectivity of Arbitrary Networks, Wireless Communication Technologies
Group, NIST, 29 March 2001.
The MathWorks, MATLAB and Simulink for Technical Computing, Natick, MA.
J. S. Webster, H. Fireman, D. A. Allen, A. J. Mackenna, and J. C. Hootman. Alternative Propulsion
Methods for Surface Combatants and Amphibious Warfare Ships. Transactions of the Society of
Naval Architects and Marine Engineers, 115, 2007.
HOVER:
Key Accomplishments:

- Theoretical development of a Laplacian matrix for flow networks.
- Explicit bounds relating the eigenvalues to a key robustness constraint.
Technical Detail:
We model the electric distribution system as a network with a flow source or sink at each node.
Specific eigenvalues, based on the power-normalized Laplacian, can be used to bound standard constraints in grid design. Specifically, for any set of vertices X and its complement X, we guarantee that the ratio of the flow from X to X, to the capacity of the lines from X to X, is less than one. Intuitively, this is a functionality constraint ensuring that all demand can be served; however, it can also be formulated in terms of robustness because it captures the overcapacity of any cut in the network. This is an NP-hard calculation on its own. Using spectral graph theory we can bound it with eigenvalues of the aforementioned matrix, and they are easy to compute. Eigenvalues also have a clean structure in that they are convex and analytically differentiable, making them ideally suited for nonlinear optimization. Using such an approach, we can optimally design for robustness at a fundamental level. Given edge removal (damage), the amount of power deliverable by the remaining distribution system is maximized.
Our work derives from a central result in spectral graph theory, the Cheeger constant where X is any partition of the graph, and |C(X,X)| is the number of edges connecting X with its complement. Vol(X) is the sum of degrees of the vertices in X.
A key result from Fan Chung bounds h to the algebraic connectivity 1 (smallest non-zero eigenvalue of the Laplacian):
A flow-based equivalent of the Cheeger constant h is:

where |C(X,X)| is the aggregate capacity of links connecting X and its complement, and |S(X)| is the absolute value of power consumed in X. Like h, q is NP-hard to compute. We developed two new theorems bounding q. The first is
Here, P is a specialized eigenvalue derived from the Laplacian and the loads. The number of nodes in the matrix is n. Our second theorem is
Here, pv is the power at a vertex v, and  is a second specific eigenvalue. In both theorems, the upper and lower bounds are efficient to compute.
We have tested the bounds on flow networks for which exact solutions of q are available; representative results are given in the Figure below, showing the insights gained.

The top plot shows two networks and cuts: a poorly designed network, in which a large amount of flow must cross a small cut (all supplies on one side and all demands on the other), and a balanced network in which the flow is better distributed. The lower plots show q and the eigenvalue bounds for growing network size. The upper of these corresponds to the unbalanced network, and q and the bounds decrease as 1/n. The lower plot corresponds to the balanced network; q and the bounds remain constant with network size, indicating scalable robustness.
KARNIADAKIS GROUP
Key Accomplishments:
We report next on the main results we obtained in developing and validating the Purple Team design. We do this in stages (in 5 different case studies) as we show below using various benchmarks. The MIT models are very similar to Purdue models except for the rectifier and some details of the controller if the SM.

Topology and System Components: The overall diagram of the simulated system is given in Figure
1. The Generation System consists of a 59kW wound rotor synchronous machine (SG) interfaced through a three-phase diode passive rectifier to the 750V DC bus. The propulsion drive consists of a
4-pole, 460V, 50Hp, 60Hz induction motor.
SG
E fd
Three-phase
Diode Rectifier r
Ldc1
L dc1
C dc1 r
Ldc2
L dc2 r
Cdc2
C dc2
DC Voltage
Control
I dc
V dc
* r
L
L
+ r pCdc
V dc
-
R shunt
R
D
C in r
ESR
Three-phase
IGBT Inverter
IM
Switch signals
Hysteresis
Current Control
I abc
T e
* ω r
Induction Motor
Torque Controller
I abc
*
Figure 1: Overall MVDC (750V) System Diagram (Purple Team design).
Study I: Step Change in the Resistive Load: An additional resistive load of approximately 9kW
(60.2
Ω
) is added in parallel with the induction motor at 10 seconds. The induction motor is started at 4 seconds with torque command of 100Nm. The comparison of simulation results between the
MIT and Purdue models for the DC bus voltage, rectifier and inverter DC current, and IM electrical torque are given in Figures 2, 3. From the results it can be seen that adding the additional load does not influence the controller of the IM; only the generation current (rectifier DC current) increases and the DC bus voltage decreases slightly from its reference value (750V) since we use the voltage droop control method for the DC bus voltage control. Our results correspond closely to the Purdue hardware measurements.

Figure 2: (Left) DC Bus Voltage and (Right) DC inverter current when an additional resistive 9kW load is added in parallel with IM.
Figure 3: (Left) DC rectifier current and (Right) IM electrical torque when an additional resistive
9kW load is added in parallel with IM.
Study II: Step Change in the Torque Command: The IM is started at 4 seconds at Te* = 100Nm, at
10 seconds the commanded torque is changed to Te* = 125Nm, and at 18 seconds it is changed to
Te* = 200Nm. The comparison of simulation results between MIT and Purdue models for the DC bus voltage, inverter DC current, IM electrical torque and slip-frequency are given in Figures 4, 5.
Figure 4: (Left) DC Bus Voltage and (Right) DC inverter current when an additional resistive 9kW load is added in parallel.

Figure 5: (Left) IM electrical torque and (Right) IM slip-frequency when an additional resistive
9kW load is added in parallel.
Study III: Smaller Filter Capacitors: This study aims to investigate the stability of the system. The
IM is started at 4 seconds with Te* = 200Nm, capacitances Cdc2 and Cin have values 100µF. The comparison of simulation results between MIT and Purdue models for the DC bus voltage, inverter
DC current, IM electrical torque, and slip frequency are given in Figures 6, 7.
Figure 6: (Left) DC Bus Voltage and (Right) DC inverter current with smaller filter capacitors.

Figure 7: (Left) IM electrical torque and (Right) IM slip frequency with smaller capacitors.
Study IV: Simulation Results with Two Synchronous Machines: Simulation results for a system that has two synchronous machines as generators are given in this section. The two SMs are identical.
Simulation results are given for the cases with and without droop voltage control. Simulation results in the case where both SM have the same droop constants (Kd1 = Kd2 = Kd) for DC rectifier current and DC bus voltage are given in Figure 8.
Figure 8: (Left) Rectifier DC currents and (Right) DC bus voltage in the case with no droop control
(Kd = 0) and with droop control (Kd = 0.59)
Study V: Stochastic Modeling of the Purple Team Design: In preliminary work we have demonstrated the application of polynomial chaos to investigate sensitivity of the Purple team design. Specifically, we consider case study 1 with Rshunt being a stochastic variable with

uncertainty levels of 20% and 50% of its mean value. We examine the DC current and voltage in the bus as shown in Figure 9. The results show that during transients the “error bars” or
“confidence intervals” are very large whereas in steady states these error bars are small. The IM controller seems to be robust so variations in the resistive load do not affect the torque. For large variations the deterministic mean and the stochastic mean are different (see DC current plot).
Figure 9: Sensitivity analysis showing (Left) variation of the DC rectifier current and (Right) DC bus voltage.
KIRTLEY GROUP
Key Accomplishments
- Contra-Rotating Propeller Design: Development of procedure for determining optimum radial circulation distributions for contra-rotating propellers within the framework of vortex-lattice lifting line method.
- Ship Propulsion Induction Motors: The development of a general motor evaluation program to be used with existing propeller design software to jointly optimize designs of motor-driven propellers and the refinement of existing detailed motor design and analysis software to better account for the differences between the megawatt scale propulsion machines and smaller, more traditional induction motors.
- Propeller/Turbine Test Fixture: Design of a propeller/turbine test fixture, manufacture of all mechanical components, assembly of electrical components, and manufacture of a turbine in preparation for testing.
- Contra-Rotating Induction Motor Proof-of-Concept: Selection, testing and modeling of appropriate equipment to be re-configured as a contra-rotating motor.

Technical Details
Contra-Rotating Propeller Design
Introduction
Contra-rotating propellers (CRP) are propulsor configurations known to offer higher efficiencies in comparison with equivalent single propellers. Moreover, further gains can be achieved by taking advantage of the fact that, by sharing total thrust on two propellers of approximately the same diameter as the single propeller, it becomes easier to control cavitation and its effects.
Several cases have been reported where the implementation of CRP systems on ships has led to significant energy savings without compromising the reliability of the propulsion systems. Fuel savings of more than 10% have been confirmed, exclusively due to improvements in the propulsive efficiency of the CRP sets compared to conventional propellers. Moreover, the advent of electric propulsion is expected to further facilitate the use and improve the performance of CRP sets in the future by eliminating the need for heavy and complex shafting and gearing systems.
In this work a method for designing contra-rotating propellers is developed by studying two different approaches for determining optimum circulation distributions in the context of Lifting-line theory. Comparisons of numerical predictions of efficiency for CRP and single propellers as a function of thrust loading and advance coefficient are presented.
CRP –Basic Considerations:
A Contra-rotating propeller is defined to consist of two, coaxial, open propellers positioned a short distance apart and rotating in opposite directions. In order for the design theory to be as general as possible, certain characteristics of each propeller are allowed to be chosen independently (diameter, rpm, number of blades, radial distribution of loading, chord length, etc.). The hub diameter is assumed to be constant. Slipstream contraction is not taken into account and therefore the two propellers are assumed to have the same blade diameter.

Figure 3. Contra-Rotating Propeller Set
It would be desirable for the presented theory to produce a hydrodynamic propeller design that meets a specified propulsion requirement, i.e. a specified thrust or power at a given rpm and ship speed (or speed of advance). It is known that the hydrodynamic design of a propeller is accomplished in two steps. One first establishes the radial and chordwise distribution of circulation over the blades that will produce the desired total thrust, subject to efficiency considerations. In the second step, one finds the shape of the blade that will produce this described distribution of circulation.
This implies that the present theory provides a means to:
1) Predict time-averaged or steady loadings and forces for selected values of propeller characteristics during the preliminary design phase.
2) Determine blade geometry, specifically pitch and camber distributions, required to develop the design loadings.
Before discussing the detailed implementation of the CRP design theory, it is necessary to describe the vortex lattice lifting line theory as applied to the case of the single propeller design.

Propeller Vortex Lattice Lifting Line Theory:
In spite of the development of elaborate lifting surface design and analysis methods, as well as the introduction of surface panel codes, lifting line theory still plays an essential role in propeller design and particularly in the preliminary design stage. The reason is that the most reliable prediction of the relationship between the radial distribution of circulation and the resulting thrust and torque comes from a Treffz plane analysis, which follows directly from lifting line theory. In addition, lifting line theory provides a variety of necessary essential inputs to the propeller design process thus permitting parametric studies to be performed in order to determine an optimum design from the point of view of efficiency, cavitation, strength and cost.
In the framework of vortex lattice lifting line theory the propeller blades are represented by straight, radial lifting lines with the blades having equal angular spacing and identical loading. The inflow to the propeller disk is assumed to vary radially but is constant in the circumferential sense. Since all blades have the same circulation distribution, one blade is designated as the key blade. The span of the key blade is divided into M panels extending from the hub root to the blade tip. The radial distribution of bound circulation Γ(r) is approximated by a set of M vortex elements of constant (but not identical) strength extending from rv(m) to rv(m+1). A discrete trailing (free) vortex line is shed at each of the panel boundaries, with a strength equal to the difference in strengths of the adjacent bound vortices. Alternatively, the vortex system can be thought to comprise a set of M horseshoe vortex elements, each consisting from a bound vortex segment and two free vortex lines which are of helical shape as we will see later. Taking into account all blades, each horseshoe element actually represents a set of Z identical elements of equal strength, one emanating from each blade. The velocity field induced at the lifting line by this system of vorticity is computed using the efficient asymptotic formulas developed by Wrench (1957).

Figure 4. Lifting Line Lattice of bound and free, trailing vortices
Vortex elements shed by the propeller blade rotating about a fixed point at angular velocity ω in a stream Ua are in principle convected by the resultant relative velocity composed of Ua, rω plus the axial, tangential and radial components (self-induced velocity components) induced at the shed element by all members of the vortex array. Thus the trajectory of vortices shed from any radial blade element is not a true helix as the induced velocities vary with distance from the propeller.
Only in the ultimate wake (some two- three diameters downstream) a true helical pattern is achieved as the axial inductions achieve their asymptotic values and the radial component vanishes.
Moreover, as the vortices act on each other the sheet of vorticity shed from all blade elements as in the flow abaft wings is unsteady and wraps up into two concentrated vortices, a straight one streaming aft of the hub and one inboard of the tip radius.
Simplifications which are often applied to the lifting line theory involve the geometry of the propeller wake to be purely helical, with a pitch at each radius determined either by the undisturbed inflow in the lightly loaded case (linear theory) or by the induced flow at the lifting line in the moderately loaded case. In the present application of the CRP design theory the moderately loaded model is implemented. It is worth noting that the lightly loaded propeller is analogous to the wing where the trajectories of the trailing vortices are assumed to be independent of the wing loading. In addition, the trailing vortex rolling-up process is neglected due to the extremely large computational

burden and the fact that the precise details of the deformed trailing vortex wake are not critical in determining the flow at the blades.
Criteria for Optimum Distribution of Circulation:
The lifting line theory is the basis for propeller design since it provides the radial distribution of loading or circulation. This distribution is obtained by use of criteria for optimum efficiency or modifications of such a distribution, for example to reduce the hub or the tip loading, avoid cavitation, high vibratory forces and noise, etc.
Betz (1919) first derived the optimum circulation distribution criterion for propellers operating in uniform wake by using Munk’s ‘displacement law’ that stated that the total force on a lifting surface is unchanged if an element of bound circulation is displaced in a streamwise direction. His result suggested that the ultimate forms of the vortices far downstream for an optimum circulation distribution are true helices and is expressed as tan β i
= tan β l *
where l *
is a dimensionless constant depending on the required thrust produced by the propeller.
The condition for non-uniform or wake-adapted inflow was given by Lerbs (1952) by extending
Betz’s work after including the thrust deduction and the wake fractions in his computations.
Several other criteria were developed afterwards but all gave a distribution of the hydrodynamic pitch angle tan
β i of the form tan
where k is an unknown factor related to the required thrust and F a function depending on the optimum criterion.
A different procedure for determining optimum circulation distributions has been developed by
Kerwin, Coney and Hsin (1986). Instead of deriving optimum criteria corresponding to Betz or
Lerbs a numerical version of their derivation using calculus of variations and Lagrange multipliers, but working with the unknown circulations, is used. An auxiliary function
H Q λ (
T T r
)
is formed and its partial derivatives with respect to the unknown m circulation values and the
Lagrange multiplier are set to zero. The resulting system of m+1 equations is linearized by assuming that the Lagrange multiplier is known where it forms quadratic terms with the circulations

and solved for the m+1 unknowns. The solution yields the optimum circulation distribution and the value for the Lagrange multiplier lambda.
Interestingly, both methods yield similar results and in the limit of light loading the variational optimum approaches the Lerbs optimum as will be shown later.
Lifting Line Methods for CRP Design:
The first lifting line method for predicting optimum circulation distributions for CR propellers with equal number of blades was outlined by Lerbs (1955). His method was an extension of his lifting line method for SR propellers with the inclusion of the mutual interaction velocities as well as the self induced velocities. Lerbs in his theory first assumed that the axial distance between the fore and aft propellers was zero which he later corrected for the actual spacing. Thus he was able to design a so-called ‘equivalent’ propeller which produced one half the total thrust. He also determined the interaction velocities between the two propellers by using weighting factors applied to the selfinduced velocities.
Morgan (1960) derived once again Lerbs’ theory for CR propellers with any combination of number of blades for both the wake-adapted and the free-running cases. In a complimentary paper, Morgan and Wrench (1965) rederived the integro-differential equation for the equivalent circulation distribution of a set of CRP, and most importantly, derived efficient and accurate formulas for the evaluation of the velocity induction factors.
CRP theory has been developed since that time as a logical extension of the foregoing concepts underlying the classical vortex theory for single rotation (SR) propellers, but several additional approximations have to be introduced.
The mutual interactions between forward and aft propellers give rise to time dependent flow and forces. In particular, the aft propeller blades rotate through the vortex sheets in the slipstream of the forward propeller. The forward propeller is also subjected to the circumferentially varying flow disturbance generated by the aft propeller. The theory for time-average or steady forces rests on a fundamental approximation. The onset flow to each propeller is divided into a circumferential

average component (which may vary radially and axially) and components which are periodic
(harmonics). It is assumed that the average component of velocity produces the steady forces on the propeller while the periodic components produce alternating forces with zero mean. Thus, the forward and aft propellers are regarded as SR propellers operating in steady, axisymmetric flows in which the onset flow to each propeller is modified to include the average axial, radial, and tangential components of the velocity field induced by the other propeller. This necessarily involves an iterative procedure in which the loadings and induced velocities of each propeller are successively determined until a converged solution is reached.
Two different methods for determining optimum circulation distributions for multi component propulsors can be identified in general. The first one, referred to from now on as the ‘Uncoupled method’, treats the components of the contra-rotating propeller set as if they were SR propellers.
Optimum circulation distribution criteria for SR propellers can therefore be utilized in order to obtain the solution for the CRP set. The second one, referred to as the ‘Coupled’ method, is the extension of the variational optimization approach developed by Kerwin, et al to the case of twocomponent propulsor. A detailed description of both methods follows.
‘Uncoupled Method’
As already mentioned this method uncouples the circulation distributions for the forward and aft propellers by eliminating the requirement of designing an equivalent propeller. Either Lerbs’ or
Kerwin’s variational optimization methods can be used for setting up the system of equations for the bound circulation values on the two propellers. In the present work the variational optimization is implemented so that there is consistency in the way the circulation values are determined in both methods (‘Uncoupled’ - ‘Coupled’) and the results of the comparison capture the differences exclusively due to the way the CRP set is treated (a combination of two SR propellers or an integrated propulsor with two components), even though the implementation of Lerbs’ method instead is expected to yield similar results (as already mentioned). The coupling between the ‘two
SR propellers’ is provided entirely by the interaction velocities between them and the resulting equations are subjected to two constraints, the total required thrust produced by the set and the torque ratio between the elements of the set.

This system of equations is solved using an iterative scheme where one equation is solved at a time as if it was solved for a conventional SR propeller. The interaction velocities induced on the forward propeller by the aft are initially assumed to be zero and the linear system of equations for the unknown circulations and the Lagrange multiplier for the forward propeller are solved. Once the solution for the forward propeller is obtained, the interaction velocities induced by the forward propeller on the aft one are computed. These interaction velocities are then added to the onset flow for the aft propeller and the linear system of equations for that propeller is then solved as if it were a
SR propeller. The interaction velocities induced on the forward by the aft component are then computed and the new circulation values on the forward component for the updated onset flow are then determined. This iterative procedure is repeated until convergence for the forward and aft circulation distributions is achieved.
The fact that the induced velocities due to the aft propeller acting on the forward one are generally small, especially as the axial separation increases, insures that this iterative scheme converges very quickly. In order to compute the circumferential mean interaction velocities the generalized actuator disk theory developed by Hough and Ordway (1965) is used in both the ‘Uncoupled’ and the
‘Coupled’ methods. Besides, Hsin (1986) as part of his masters degree project compared three different methods for computing the circumferential average induced velocity and found out
Hough’s and Ordway’s method to be the most computationally efficient.
The matching of the specified total thrust and torque ratio is accomplished by first using an initial guess with the total thrust being equally divided into the two components and then applying
Newton’s method to find the thrust ratio which produces the required torque ratio.
‘Coupled Method’
This method was developed by Kerwin, Coney and Hsin (1986) and is an extension of the variational optimization approach for single propeller design. Following Coney, in the case of two propulsor components, the goal is to find the discrete circulation values
Γ
1
( ) ( )
, Γ
2
( ) ( )
such that the total power
P = ω
1
⋅ Q
1
+ ω
2
⋅ Q
2 , absorbed by the propulsor is minimized. The propulsor is additionally required to develop a prescribed thrust
T r . In addition, two component propulsors are often constrained to have a specific division of torque

between the two components. Therefore, a torque ratio, conditions are used to form an auxiliary function, H.
H = ( ω
1
⋅ Q
1
+ ω
2
⋅ Q
2
) + λ
T
( +
2
− T r
) + λ
Q
( −
2
)
1 , is also specified. These three
After expressing the thrust and torque of the individual components of the CRP set in terms of the circulation values the partial derivatives of H with respect to the unknown circulation values and the
Lagrange multipliers are taken and set equal to zero. Kerwin, et al. solved the resulting system of non-linear equations by freezing the Lagrange multipliers where they formed quadratic terms with the circulation values. In this formulation, like in the variational optimization for SR propellers,
Kerwin et al. had two nested problems to solve. The inner problem was the determination of the optimum circulation distributions for the forward and aft propellers; it was solved for fixed value of the hydrodynamic pitch angles, tan
β i , on the forward and aft propellers. The outer problem was that of determining the appropriate distributions of tan
β i for both propellers; it was solved by selecting an initial distribution of tan
β i , usually, tan
β
, and solving the inner problem for the optimum circulation distributions. The induced velocities due to these circulation distributions were then used to determine new tan
β i distributions, which were used to solve again the inner problem for the optimum circulation distributions. This process was repeated as many times as necessary until the tan
β i and the circulation distributions had converged.
Results:
Before presenting the results for the two different CRP design methods, a comparison between
Lerbs’ method and the variational optimization method for SR propeller design is performed.
Optimum circulation distributions plots obtained from these two methods for a single four bladed propeller operating in uniform wake are given below. Figures 3-5 indicate that the variational optimum approaches the Lerbs optimum as the thrust loading becomes lighter.

C T = 2 . 0 7
0 . 1 4
0 . 1 2
0 . 1
0 . 0 8
0 . 0 6
0 . 0 4
0 . 0 2
0
0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7
N o n - d i m e n s i o n a l r a d i i ( r / R )
0 . 8 0 . 9 1

C T = 0 . 6 9
0 . 0 4
0 . 0 3 5
0 . 0 3
0 . 0 2 5
0 . 0 2
0 . 0 1 5
0 . 0 1
0 . 0 0 5
0
0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7
N o n - d i m e n s i o n a l r a d i i ( r / R )
0 . 8 0 . 9 1
C T = 0 . 3 4 5
0 . 0 2
0 . 0 1 8
0 . 0 1 6
0 . 0 1 4
0 . 0 1 2
0 . 0 1
0 . 0 0 8
0 . 0 0 6
0 . 0 0 4
0 . 0 0 2
0
0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7
N o n - d i m e n s i o n a l r a d i i ( r / R )
0 . 8 0 . 9 1
Figures 3-4-5. Circulation Distributions given by the Lerbs and the Variational methods for various thrust loadings
A comparison between the results obtained by the ‘Coupled’ and the ‘Uncoupled’ methods is performed next. Figures of optimum circulation distributions, axial and tangential induced velocities

for a pair of free-running contra-rotating propellers are presented. The propellers are rotating at the same speed at an advance coefficient of Js=0.89 and are required to develop a thrust corresponding to Ct=0.69. Both propellers of the set are four bladed and are required to absorb equal torque, thus the torque ratio is set to one. The axial separation distance between the components is equal to a quarter of the propeller diameter. Viscous forces are neglected. Figure 6 gives the circulation distributions for the forward and the aft propellers.
Also shown is an optimum circulation distribution for a single eight bladed propeller operating at the same advance ratio and required to generate the same thrust. Each blade of the SR propeller produces approximately the same thrust as the blades of the CRP set. Constant radial spacing with
15 panels has been implemented for all of the results presented.
0 . 0 2
0 . 0 1 8
C i r c u l a t i o n D i s t r i b u t i o n c o m p a r i s o n f o r C o u p l e d / U n c o u p l e d m e t h o d s
U n c o u p l e d / f o r w a r d
U n c o u p l e d / a f t
C o u p l e d / f o r w a r d
C o u p l e d / a f t
S i n g l e p r o p e l l e r
0 . 0 1 6
0 . 0 1 4
0 . 0 1 2
0 . 0 1
0 . 0 0 8
0 . 0 0 6
0 . 0 0 4
0 . 2 0 . 3 0 . 4 0 . 5 0 . 6
N o n - d i m e n s i o n a l r a d i i ( R C )
0 . 7
Figure 6. Optimum Circulation Distributions
0 . 8 0 . 9 1
The results for the ‘Coupled’ CRP Design method are characterized by an inward shift in the circulation distributions for both the forward and the aft propellers of the set (given by lines with asterisks) compared to the SR propeller. On the other hand, from the application of the ‘Uncoupled’

method we obtain an inward shifted circulation distribution for the aft propeller (represented by the solid red line) while the circulation distribution for the forward one is relatively unaffected.
However, the efficiencies obtained by the two methods are comparable (83.4% versus 83.25%). The fact that the two methods specify different circulation distributions as the optima can be attributed to the different optimization routines used. In the ‘Coupled’ method the inward shifted distributions are obtained under the assumption of uniform onset flow, whereas in the ‘Uncoupled’ method the shift of the circulation distributions (always compared to the SR propeller) is due to the modified onset flows, mainly by the interaction velocities. The fact that the interaction velocities on the forward propeller induced by the aft are small provides an explanation for the relatively unaffected circulation distribution on the forward propeller.
Before attempting to explain the efficiency increase of the contra-rotating propulsor configurations a short reference to the energy losses related to the propeller action is necessary. The hydrodynamic energy losses associated with the action of a propeller are due to increases in the kinetic energy of the water passing through the propeller disk area. These losses have two components, axial and rotational. The development of thrust results from axial acceleration of the water and causes an increase in the axial kinetic energy, while the shaft torque is transferred to the water causing induced rotational velocities and a rotational energy loss. Taken together, these two losses result in the ideal efficiency of the propeller. Moreover, in an ideal fluid the efficiency will approach 100% as the propeller diameter approaches infinity since, as the mass flow increases, the increase in speed for a required increase in momentum becomes smaller. In a real fluid with viscous drag losses there will be an optimum, finite diameter at which the efficiency will have a maximum value for a given thrust and advance coefficient. Inversely, within the limitations of hull/propeller clearances, for a given diameter and thrust loading there will be an optimum advance coefficient for which the highest efficiency will be realized. In the current work, the presentation of efficiency plots for a wide range of thrust loadings and advance coefficients for both CR and SR propellers will show the agreement with the aforementioned theoretical concepts.
Returning to the comparison between the two optimization methods, the axial and tangential induced velocities far downstrean by the same propulsor configurations are given in Figures 7-8.
Some physical insight into the increased efficiency of the CRP pairs over that of the SR propeller

(efficiency of the 8-bladed SR propulsor is 80.9%) can be obtained by comparing these velocity components to the one corresponding to the ideal case of the actuator disk.
A x i a l i n d u c e d v e l o c i t i e s f a r d o w n s t r e a m / C o u p l e d m e t h o d
0 . 4
0 . 3 5
0 . 3
0 . 2 5
0 . 2
0 . 1 5
0 . 1
0 . 0 5 f o r w a r d p r o p e l l e r a f t p r o p e l l e r t o t a l C R P a c t u a t o r d i s k s i n g l e p r o p e l l e r
0
0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7
N o n - d i m e n s i o n a l r a d i i ( R C )
0 . 8 0 . 9 1
Figure 7. Axial induced velocities far downstream
0 . 4
A x i a l i n d u c e d v e l o c i t i e s f a r d o w n s t r e a m / U n c o u p l e d m e t h o d
0 . 3 5
0 . 3
0 . 2 5
0 . 2
0 . 1 5
0 . 1
0 . 0 5
0
0 . 2 0 . 3 0 . 4 0 . 5 f o r w a r d p r o p e l l e r a f t p r o p e l l e r t o t a l C R P s i n g l e p r o p e l l e r a c t u a t o r d i s k
0 . 6 0 . 7
N o n - d i m e n s i o n a l r a d i i ( R C )
0 . 8 0 . 9 1
The total axial induced velocities of the contra-rotating pairs are more uniform and less concentrated toward the tip than that of the single propeller. This observation is connected with the increased efficiency. These velocities are also closer to the constant value predicted from actuator disk theory. In the limit of light loading we expect the circumferential mean axial induced velocities to approach that of the actuator disk.
The fact that the two methods yield slightly different results in terms of the hydrodynamic efficiency of the CRP sets can partly be attributed to the different forms of the total axial induced velocities. An axial velocity distribution which is closer to the ideal can merely account for the slightly better efficiency of the ‘Coupled’ method.

0 . 0 5
0
- 0 . 0 5
- 0 . 1
0 . 2 5
0 . 2
0 . 1 5
0 . 1
- 0 . 1 5
- 0 . 2
0 . 2
T a n g e n t i a l i n d u c e d v e l o c i t i e s f a r d o w n s t r e a m / C o u p l e d m e t h o d
0 . 3 f o r w a r d p r o p e l l e r a f t p r o p e l l e r t o t a l C R P s i n g l e p r o p e l l e r
0 . 4 0 . 5 0 . 6 0 . 7
N o n - d i m e n s i o n a l r a d i i ( R C )
0 . 8 0 . 9 1
Figure 8. Tangential induced velocities far downstream
0 . 0 5
- 0 . 0 5
0
0 . 2 5
0 . 2
T a n g e n t i a l i n d u c e d v e l o c i t i e s f a r d o w n s t r e a m / U n c o u p l e d m e t h o d
0 . 1 5
0 . 1
- 0 . 1
- 0 . 1 5
- 0 . 2
0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7
N o n - d i m e n s i o n a l r a d i i ( R C ) f o r w a r d p r o p e l l e r a f t p r o p e l l e r t o t a l C R P s i n g l e p r o p e l l e r
0 . 8 0 . 9 1
Likewise, the almost complete cancellation of the tangential induced velocities far downstream of the CR propeller planes is another fact that explains the efficiency gains over the conventional single propeller. The difference in mean tangential (swirl) velocities for a CRP set compared to a
SR propeller is significant and is the primary reason for its better hydrodynamic efficiency. It is worth noting that according to the generalized actuator disk theory, the induced tangential velocity is proportional to the bound circulation values and the number of blades and independent of the axial distance downstream of the propeller plane. Since the ‘Coupled’ method gives similar circulation distributions on both components, the total swirl velocity (represented by the green dashed line) is closer to zero than that predicted by the ‘Uncoupled’ method. This fact also accounts for the slightly better efficiency of the former method.
Until this point, only results corresponding to a unique combination of thrust loading and advance coefficient have been reported. Values of propeller efficiency covering a wide range of thrust loadings and advance ratios are therefore necessary for understanding the mechanisms governing the action of contra-rotating propellers and the potential benefits over conventional single component propulsors. The calculated efficiencies obtained by both methods for the same CRP sets with torque ratio equal to unity are shown in Figure 9.

0 . 7
0 . 6 5
0 . 6
0 . 5 5
0 . 5
0 . 8
0 . 7 5
E f f i c i e c y c o m p a r i s o n a t J s = 0 . 8 9 f o r v i s c o u s c a s e ( C D = 0 . 0 0 8 5 )
C R C o u p l e d ( C o n e y )
C R U n c o u p l e d
S R - 8 b l a d e d
S R - 4 b l a d e d
0 . 4 5
0 . 2 0 . 4 0 . 6 0 . 8 1 1 . 2 1 . 4
T h r u s t L o a d i n g ( C T )
1 . 6 1 . 8 2 2 . 2
Figure 9. Efficiency as a function of thrust loading with viscous forces considered
In this case viscous forces are taken into account by introducing a constant drag coefficient value of
0.0085. Also included are the results for a 4-bladed and an 8-bladed single propeller. All propulsors have the same diameter as before and the same chord distribution, given in Table 1, and are required to operate at an advance coefficient of 0.89. However, the chord lengths have been halved for the eight-bladed configurations (the CRP pair and the 8-bladed SR propeller) so that the expanded area ratio is the same among all.
Table 1. Chord Distribution r/R
0.2
0.3
0.4
0.5
c/D
0.174
0.202
0.229
0.275
r/R
0.7
0.8
0.9
0.95
c/D
0.347
0.334
0.280
0.218

0.6
0.7
0.312
0.337
0.98
1.0
0.144
0
All of the curves in Figure 9 clearly demonstrate that there is an optimum thrust loading value for which propulsor configurations reach their maximum efficiency. Furthermore, another common characteristic is that the efficiency increases as the thrust loading decreases up to a maximum value and then, as the thrust loading further decreases, the efficiency degrades. Comparing to Figure 10 which gives the efficiency for the inviscid case, we conclude that for lightly loaded propellers the frictional losses constitute a significant part of the total losses, but they become less significant at higher thrust loadings. All of the propulsor alternatives show the trend of decreased efficiency with increased thrust loading, something that also describes the actuator disk. However, the percent difference in efficiency between that of the actuator disk and the single propellers increases with increasing thrust. The CRP configurations seem to be following curves of lower decrease rates which effectively makes them even more efficient compared to SR propellers as the thrust loading increases. Both the ‘Coupled’ and the ‘Uncoupled’ methods yield similar results over a wide range of thrust. The small difference can mainly be attributed to the better cancellation of the swirl velocities obtained by the ‘Coupled’ method.

0 . 9 5
0 . 9
0 . 8 5
0 . 8
0 . 7 5
0 . 7
0 . 6 5
0 . 6
0 . 5 5
E f f i c i e n c y c o m p a r i s o n a t J s = 0 . 8 9 f o r i n v i s c i d c a s e
C R C o u p l e d
C R U n c o u p l e d
S R - 8 b l a d e d
S R - 4 b l a d e d i d e a l ( a c t u a t o r d i s k )
0 . 5
0 . 4 0 . 6 0 . 8 1 1 . 2 1 . 4 1 . 6
T h r u s t L o a d i n g ( C t )
1 . 8 2 2 . 2
Figure 10. Efficiency as a function of thrust loading with viscous forces neglected
2 . 4
Figure 11 shows the real efficiency (including viscous drag) of the same propulsor configurations as a function of the advance coefficient for a constant thrust loading of Ct=0.69. As the advance coefficient Js approaches zero the rotational speed of the propellers increase and the efficiency decreases due to high frictional losses.

0 . 6
0 . 5
0 . 8
0 . 7
0 . 4
0 . 3
0 . 2
C R C o u p l e d ( C o n e y )
C R U n c o u p l e d
S R - 8 b l a d e d
S R - 4 b l a d e d
0 . 1
0 . 2 0 . 4 0 . 6 0 . 8 1 1 . 2 1 . 4 1 . 6
A d v a n c e C o e f f i c i e n t ( J s = V s
/ n ⋅ D )
1 . 8 2 2 . 2
Figure 11. Efficiency as a function of advance ratio when viscous forces are considered
Figure 12 gives the results for the CRP sets for both the viscous and the inviscid cases. In the limit of zero Js the propellers behave as an actuator disk. Once again the efficiency gain of the CRP sets increases as the rotational speed decreases. The hydrodynamic losses in this regime become dominant since the mass flow through the propeller disk decreases driving up the induced axial and rotational velocities.

E f f i c i e c y c o m p a r i s o n f o r C T = 0 . 6 9
0 . 9
0 . 8
0 . 7
0 . 6
0 . 5
0 . 4
C R C o u p l e d ( I n v i s c i d )
C R U n c o u p l e d ( I n v i s c i d )
C R C o u p l e d ( V i s c o u s )
C R U n c o u p l e d ( V i s c o u s )
A c t u a t o r D i s k
0 . 3
0 . 2 0 . 4 0 . 6 0 . 8 1 1 . 2
A d v a n c e C o e f f i c i e n t ( J s = V
1 . 4 s
/ n ⋅ D )
1 . 6 1 . 8 2
Figure 12. Efficiency of Contra-rotating propellers as a function of advance ratio
2 . 2
Propeller/Turbine Test Fixture
Introduction and Motivation:
Since 2006, graduate students at the Massachusetts Institute of Technology (MIT) have been developing an open source propeller and turbine design and analysis tool under the supervision of
Professor Richard Kimball. The tool is a set of open source MATLAB scripts published under the
GNU license which are capable of performing design and analysis studies for open and ducted propellers. This suite of MATLAB scripts is called OpenProp. OpenProp propeller design capabilities include performing parametric studies of propellers using various propeller diameters, number of blades and rotation speeds. Propeller analysis features include performing off-design analysis and cavitation analysis. Additional features of OpenProp that are being developed include a check of propeller blade thickness against the American Bureau of Shipping (ABS) steel vessel rules for minimum blade thickness and a module for the design of contra-rotating propellers. The portion of the code suite which designs ducted propellers has been validated against the US Navy’s

PLL code with excellent correlation. The axial turbine design and analysis capabilities have been recently added but no validation of the OpenProp’s performance predictions has been performed. In order to perform experimental validation of OpenProp’s predictions it has become necessary to design and construct a propeller and turbine design test fixture.
MIT has a tow tank and water tunnel that are capable of performing propeller and axial turbine testing. The test fixture described in this paper is primarily designed for use in the water tunnel, however it can be used in the MIT tow tank. The advantage of testing in the water tunnel is that a set of propeller curves can be quickly generated and flow visualization can be easily performed.
The disadvantages of the water tunnel are that the relatively small test section limits the diameter of propellers that can be tested and the maximum allowable size of the test fixture. The advantage of testing in the MIT tow tank are that larger diameter propellers and turbines can be tested and a propeller bollard pull measurement can more easily be made but the finite tow tank length limits the length of the data runs and requires that the tow carriage be cycled between finish and start points, which add times to the test program. The ability to test in both the tow tank and water tunnel has been included by attaching a stand pipe of 4 feet to the test fixture, this allows the test fixture to be attached to the tow carriage of the tow tank and the dynamometer of the water tunnel.
Design Decisions
Sensor:
The test fixture design was developed around a force and torque sensor that is on loan to Professor
Kimball from the US Navy. This sensor contains two sets of strain gages which allow the sensor to simultaneously measure thrust and torque. An overarching design goal was to keep the sensor as the limiting component of the test fixture. The first step in the design process was to determine the maximum torque and thrust capabilities of the sensor. One end of the sensor is engraved with 50lbf which was assumed to be the thrust limit of the sensor but no markings were found that could be construed as a maximum torque limit. In order to confirm the maximum thrust limit of the sensor and determine a maximum torque limit, a three dimensional model of the sensor was made in
SolidWorks in order to calculate the material volume of the sensor and perform finite element analysis (FEA). From the calculated material volume and a mass measurement of the sensor it was determined that the sensor was manufactured from titanium. FEA shows that the maximum

allowable thrust is 100 lbf and the maximum allowable torque is 12 ft-lbf to prevent plastic deformation of the sensor. A picture of the sensor showing the color bands associated with the thrust FEA is shown in the figure below. The deformation scale necessary to make the sensor deflections visible is such that there appears to be an interference between the parts; this interference does not actually exist and is merely a carryover of the large deformation scale that is used. The results of the FEA show that the engraved number of 50 lbf is a reasonable thrust limit with a factor of safety of 2.
Figure 23: FEA Results for a 50lbf Applied Load
Propeller and Turbine Use:
Because of the desire to use the test fixture to test both propellers and turbines, it is necessary for the test fixture to be able to handle axial loads in two directions. In the test fixture design this is accomplished using two tapered roller bearings which are secured to the test fixture drive shaft using a finely threaded axle nut which will allow the axial movement of the drive shaft to be minimized. A view of the tapered roller bearing assembly is shown in the figure below.

Figure 14: Roller Bearing Assembly Including Drive Shaft, Slip Rings and Brush Block
Maximum Diameter:
The desire to use the test fixture in the water tunnel where the test section measures 18in x 18in limits the maximum diameter of the test fixture to approximately 4inches. Trolling motors of approximately 3.5inches in diameter have been successfully used in the water tunnel to conduct flow visualization experiments. It was desired to use a standard off the shelf motor for ease of installation, however the required motor performance meant that a standard motor was too large.
Motor:
In selecting a motor, it was desirable to select a motor such that the sensor remained the limiting component in the design. Therefore a motor capable of torque in excess of 16N-m (12ft-lbf) was selected. A test fixture using a standard motor required a test fixture 6 inches in diameter. In order to minimize test fixture diameter, a kit motor was selected. The specific motor selected was the
Parker Frameless Kit Motor K089300.
The K089 series motor has a maximum diameter of 3.5 inches which allows the motor to be attached inside a tube with a maximum outside diameter of 4 inches. The K089300 was selected because it is the highest torque motor listed in the catalogue for this series. Parker frameless kit motors generate additional torque in a given series by increasing the length of the stator windings and rotor. A kit motor has the following advantages 1. Able to fit in a tighter package 2. Use of one shaft for both the motor and driveshaft which eliminates the need for a shaft coupling between motor shaft and propeller drive shaft 3. Stator windings are in direct contact with test fixture housing which make for efficient heat transfer out of the stator windings and into the fluid

surrounding the test fixture. The disadvantages of using a kit motor are: 1. Attachment of the motor into the fixture required that additional holes had to be drilled into the motor housing which meant an increase in the probability of a leak into the fixture. 2. Kit motors do not come with a high resolution angle encoder installed like the standard off the shelf motors.
Calculated performance curves are shown below. Dashed line represents continuous operation; solid line represents peak output for intermittent operation.
6
4
2
0
12
10
8
20
18
16
14
0
Simulated Torque Speed Curve for a K089300-7Y at 170VDC
500 1000
Speed (RPM)
1500 2000
Figure 15: K089300-7Y Torque Speed Curve. Dashed line represents continuous operation; solid line represents peak output for intermittent operation.
Output Power
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0 500 1000
Speed RPM
1500 2000
Figure 16: Output Power Capability. Dashed line represents continuous operation; solid line represents peak output for intermittent operation.

Material selected for the construction of the test fixture was 303 and 304 Stainless Steel. These stainless alloys were selected for their ease of machining and corrosion resistance in the fresh water environment of the water tunnel and tow tank. The nose fairing is made from ABS plastic in a 3D printer.
Views of the test fixture are shown below. The test fixture has been designed to accommodate a high resolution angle encoder which is circled in figure 7. Figure 8 shows the encoder assembly by itself. A high resolution angle encoder would allow the test fixture to be used to validate assumptions that are currently made when conducting studies of dynamic propeller performance.
This encoder is not necessary to conduct steady state tests and has not been purchased for use in the test fixture at this time.
Figure17: Assembled Test Fixture

Figure 18: Complete Assembly Explode
Figure 19: Test Fixture Internals
Figure 20: Encoder Assembly

Figure 21: Test Fixture Internals Exploded
Electrical:
Students at the University of Maine selected a Parker MPP series motor and a Copley Xenus controller for a cross flow turbine test fixture. Because of the success that the University of Maine students have had with their equipment, it was decided to use similar components. The Copley
Xenus controller provides significant flexibility to the test fixture. Most importantly, it has the capability to act as a motor controller or as a load controller when the motor is being used as a generator. The controller also exhibits low noise characteristics.
Data signals from the strain gages are amplified using amplifiers mounted inside the sensor.
Amplifying the data signal near the source prevents the data signal from being lost in the noise that is picked up through the data signal transmission path. The amplifiers being used are specifically designed for the motor sports industry for the amplification of strain gage voltages measured from strain gages mounted in a wheatstone bridge.
The motor listed above is a DC brushless motor. A DC brushless motor was selected because it is a synchronous machine which works well with the the controller selected. Also, the absence of brushes will preclude carbon dust build up within the test fixture.
All electrical power and data signal wires are being passed out of the fixture through the large hole that is visible in figures 5 and 6 above. Because the power and data signal wires are in such close proximity, both sets of wires are shielded.

Contra-Rotating Induction Motor Proof-of-Concept
One task begun this quarter is the construction of a demonstration and proof-of-concept contrarotating induction motor and propeller. Once constructed, this motor could be tested in conjunction with a contra-rotating propeller designed using OpenProp and produced using 3D printing technology available in the lab.
A Westinghouse Life-Line T AC motor was selected to use as a contra-rotating motor. This motor is 3 HP, 1730 RPM, 230V, and operates on 3 phase, 60 Hz power. The motor was first connected and run to make sure it was operational, then disassembled, measured, and modeled in Solid Works.
The disassembled motor can be seen in Figure 22; the Solid Works model of the rotor can be seen in
Figure 23.
Figure 22: Disassembled motor.

Figure 23: Rotor model
A Variable Frequency Drive (VFD) was ordered to power the motor; the selected drive is a
YASKAWA GPD315/V7 3.7 kW (5 HP). Input required is 200-230V, 3 Phase 50/60 Hz, 24A.
Output is 0-230V, 0-400 Hz, 17.5A
The existing motor housing must be removed and a new mounting and support structure must be constructed including a new and longer double layer/concentric shaft, bearings now for the rotating
“stator,” and a slip ring assembly for the stator winding. The rotor and ‘stator’ will be individually tied to concentric contra-rotating shafts that drive a contra-rotating propeller.
References
Norbert Doerry, Next Generation Integrated Power System: NGIPS TechnologyDevelopment
Roadmap, Washington Navy Yard DC: Naval Sea Systems Command.
LEEB GROUP
Key Accomplishments:
Design of mechanical generation framework for power system hardware emulator
Construction of mechanical generation framework for power system hardware emulator

Field interviews with USCGC ESCANABA crew to examine power system, protection gear, opportunities for fault detection, mitigation, and condition based maintenance.
Technical Detail:
Ship designers working now and in the near future face the “curse of dimensionality.” Growing arrays of options and solutions are available for nearly every aspect of a ship design: materials, control and automation systems, weapons systems, power sources, power distribution systems, propulsion, hull design options, and more. Choices in any one dimension affect options in many or all dimensions, increasing the challenge of designing a vessel that provides peak capability, survivability, and maintainability at the lowest reasonable life-cycle cost.
Since at least World War II, there has been a steady technological pressure increasing not only the performance but also the complexity of many weapons systems. Combat aircraft are illustrative, benefiting from advances in materials, avionics, propulsion, and airframe design. These advances and their associated performance enhancements have “changed the calculus of strike from sending multiple aircraft to destroy a single target to sending one aircraft to destroy multiple targets” ([1], in reference to JDAM). These benefits come with a price. Field “repair” has essentially become a matter of replacement. Problems in key systems are ultimately resolved with depot-level maintenance, and field service depends either on the availability of spare modules or the ability to operate the system absent a malfunctioning subsystem. Generally, this trade-off is tolerated because there is no choice. Ceding the “high ground” afforded by advanced avionics and propulsion systems, for example, would open an immediate advantage for a motivated, technologically sophisticated opponent.
Future naval combat vessels stand on the threshold of this complexity/capability trade-off. In some cases, systems like advanced radar require a level of performance that may only be achieved with sophisticated “depot-repairable” modules. Field replacement of modules keeps such systems operating.
Increased demand for electrical power in projected future naval combatants with electric drive, advanced sensors, and electrically actuated weapons and launch systems motivates the

consideration of new power systems architectures [2]. This consideration includes potentially extensive use of DC power distribution, i.e., a medium voltage direct current or MVDC power system [2]. Development of MVDC power systems for naval combatants is posited to provide many operational benefits, and will entail technological risk.
Practical MVDC power systems are likely to require mission-critical modules or subsystems that are at most “line replaceable,” i.e., subsystems that can be replaced in the field but repaired only at a depot-level facility. Many assumptions about future field conditions and operational scenarios are implicit in any decision to implement full MVDC power distribution on combat vessels. Loss of some key components in an MVDC power system could cripple the operational capability of a ship.
Recovery of a damaged ship with an MVDC power system may require replacement modules, possibly from off-board the ship, implying a commanding access to the damaged vessel and its environs, including a de facto requirement for air transport and air supremacy.
Are we comfortable with a combat vessel whose “heart beat” increasingly depends on modules that can only be replaced, but not serviced, in the field? What will such a vessel cost to construct? If full
MVDC is the only means to achieve peak combat capability and survivability, we are obligated to consider developing and deploying this type of power system. However, if equivalent combat capabilities can be achieved using power systems with less technological risk and lower cost, we are equally obligated to consider other options.
We are exploring the anticipated benefits and open concerns and questions associated with proposed future power systems, including an MVDC power system architecture. We are also examining alternatives for prime movers and variable speed electrical propulsion drives that might make a hybrid power system, with both AC and DC power, i.e., an “MVADC” system, attractive.
To test and explore these ideas, we have begun construction of a hardware test bed during the performance period, with custom features for testing new ideas for propulsion, zonal protection, and condition-based maintenance on different kinds of power systems. A test platform with two 2000 watt synchronous generators is being constructed to topologically emulate part of a U.S. Navy DDG
51 FLT IIA class ship electric plant. The test bed is reconfigurable to permit experiments with DC

distribution. An overview is shown by the circuit fragment in Figure 1. This test bed will be used to evaluate the feasibility of improving the fault isolation capabilities of the MFM with NILM implementation. It will be used to study new ideas for ship propulsion. It will also be used to explore new possibilities for CBM.
Figure 1: Multi-bus AC and DC power distribution
During the performance period, we have laid out the structure of a two machine generating system capable of driving the Multi-bus power distribution test bed. A block diagram overview of the
Assembly
12V Power
Supply
Main Power
Supplies
Power to Cooling Fans
DC
Motors
Loads
Generators
Distribution
Board

Figure 2: Block Diagram
Figure 3: Solid model
To be able to compare the bench generators to any Navy electrical generator it was first necessary to classify the characteristics of the test bed generator. Preliminary tests were accomplished during the performance period to characterize the test bed generators:
DC test: Armature Resistance (R1).

Open Circuit Test: The Magnetization Characteristic and Field Resistance.
Short Circuit Test: Synchronous D-Axis Reactance (Xd).
The specifications of the generators are shown below in Table 1. This generator is a salient pole
synchronous wound field machine.
Power (W) 5000
Voltage (phase to phase), (VRMS) 207.85
Current (phase to neutral), (ARMS) 13.89
Frequency (Hz) 60
Poles
Power factor
4
0.8
Table 1: Bench Generator Specifications
Figure 4 below shows typical results obtained for the test bed generator short circuit tests. These results show the tachometer results for the 1.5 HP motor/prime mover, emulating, for example, a gas turbine prime mover. The tachometer is a Hall effect sensor that was placed in close proximity to the gears on the uncoupled side of the motor. The output of the tachometer was used to capture the speed profile of the generator during the short circuit transient.
- 5 0
0
- 1 0 0
- 1 5 0
- 2 0 0
0
2 0 0
1 5 0
1 0 0
5 0
1 0 2 0
I
T a c h o m e t e r
S W I T C H
φ
3 0 4 0 5 0
T i m e ( m s )
6 0 7 0 8 0 9 0 1 0 0

Figure 4: Short-circuit test results
Also during the performance period, we have developed full simulation models and validated fitted parameters of the test bed generator using these models, with excellent fidelity to observed laboratory data. A comparison of simulated and measured waveforms is shown in Figure 5, which compares simulated, measured, and “classical circuit” predictions of the short circuit current.
S h o r t C i r c u i t C u r r e n t
2 0 0
1 5 0
1 0 0
I
φ
I
φ n
F L U K E n
C l a s s i c a l
I
φ n
S i m u l a t i o n
- 1 0 0
- 1 5 0
- 2 0 0
0
5 0
- 5 0
0
5 1 0 1 5 2 0 2 5
T i m e ( m s )
3 0 3 5 4 0 4 5 5 0

Figure 5: Comparison of simulated and measured generator short-circuit current.
Estimated parameters for the generator used to conduct the simulation are shown below in Table 2:
Pbase (Watts) 5000 xkdl(Ω) .0628
p.u.
Vbase (V)line-line RMS 207.8
xq(Ω) p.u.
.770
Ibase (A)RMS 13.9
xd(Ω) p.u.
1.007
pf
Pole Pairs
H (sec) ra (Ω) p.u.
xl (Ω) p.u.
0.8
2
.190
.056
xkq(Ω) p.u.
xd’(Ω) p.u.
.837
xq’’(Ω) p.u. .150
.197
xd’’(Ω) p.u. .094
.0500
xf(Ω) p.u.
1.13
xaq (Ω) p.u.
rkq (Ω) p.u.
xkql (Ω) p.u.
xad (Ω) p.u.
rf (Ω) p.u.
xfl(Ω) p.u.
rkd(Ω) p.u.
.720
xkd(Ω) p.u.
1.020
.0317
tdo' (sec) .409
.116
.957
tqo'’ (sec) tdo'’ (sec)
.070
.030
.0126
.173
.0185
td' (sec) td'’ (sec)
.045
.015
Table 2: Bench Generator Parameters

The structure of the test bed, models for ship board loads, and valuable comparative data for our power systems comparisons underway have all been invaluably informed by our interactions with US Coast Guard crews on “Famous” class cutters, as well as our field experiments at the
USN LBES facility in Philadelphia.
Bibliography:
Air Force Magazine, “The Testers from Tucson,” December 2009, pg. 48.
CAPT Norbert Doerry. Next Generation Integrated Power System: NGIPS technology development roadmap. Technical report, Naval Sea Systems Command, November 2007.
Naval Sea Systems Command. Technical Manual Organizational Level Maintenance LM2500
Propulsion Gas Turbine Module Description and Operation, S9234-AD-MMO- 010/LM2500, revision 3 edition, May 2000.
David Wipf and Eugene Parker, Jr., Demonstration and Assessment of High Speed Relay
Algorithm, Technical Report, Barron Associates, June 1997.
Robert Balog and Philip Krein, “Bus selection in multibus DC power systems,” IEEE Electric
Ship Technologies Symposium, May 2007.
M. Bash, R. R. Chan, J. Crider, C. Harianto, J. Lian, J. Neely, S. D. Pekarek, S. D. Sudhoff and
N. Vaks, “A Medium Voltage DC Testbed for Ship Power System Research,” IEEE Electric
Ship Technologies Symposium, April 2009.
TRIANTAFYLLOU:
The primary modeling development in progress is the integration of accurate and reliable ship seakeeping equations, which when supplemented with our propeller model will allow for the exact quantification of propeller and electric loading in random sea-states. Moreover, it will allow the prediction and approximation of potentially hazardous transients, such as the propellerout-of-water scenario. We attacked this problem by adopting a very flexible program with a long history of reliable predictions, know as the fivedegree-of freedom MIT program (MIT-5D) as the first calculation step of ship motions in rough

seas. In addition, we have set out to enhance its capabilities to (a) calculate any ship hull form and not just “Lewis-type” forms -- this is essential for ships with bulbous bows or large sonar domes, or novel type sections; and (b) to calculate wave drift forces and moments which are essential for maneuvering calculations. Our own 5 degree of freedom model will allow for integration into the existing AES model with maximal flexibility and speed. In essence, we will run MIT5D as a subroutine to update the relatively slow seakeeping dynamics as necessary for accurate results.
Several smaller model augmentations have been researched and fully implemented this quarter. First, we have rescaled the former AES model to meet the power demands of the
DDG1000. The new model includes four LM2500 turbines (84MW source power) accounting for two power plants; split system architecture is employed to isolate the plants. Each propeller is driven by a hypothetical 38MW induction motor model. A schematic is shown in Figure 1.
Figure 1: DDG 1000 Scale Model
Uncoupled turbine, generator, and motor controllers have proven effective for simulation in ideal conditions but more sophisticated feedback control, perhaps cross-connected control, is an area of research that will benefit the IPS robustness.
Other developments include addition of rudder added mass/damping to maneuvering equations and a PD heading control (via rudder actuation) designed to give optimal, first-order
74

heading response. The heading control design was based on the observation that the yaw dynamics of the ship were approximately first-order resulting in a heading transfer function of
1/s(τs+1). Using PD control, a zero placed between the two poles, at 0 > z > τ, creates first order dynamics as controller gain goes to infinity. Some small integral control could be added to counteract zero order forcing terms inherent in the empirically derived maneuvering equations.
The last modeling pursuit we will mention here is the implementation of a high-order linear turbofan model with LQG/LTR control [Kappos, 1983] which may be manipulated to be used as a turboshaft; this would accomplish a very accurate prime mover model with thermodynamics, hydraulics, and sensor dynamics all considered. The closed-loop diagram for the system is shown in Figure 2.
Figure 2: LQR Controlled Turbine Model
75

Figure 3: Open and Closed Loop Transfer Function Singular Values
The full order linear model, G(s), is 23 states and the linear model used for controller design (F,G,H,D) is 13 states, attained from modal decomposition and residue matrix comparison. Both models are given in [Kappos]. The Kalman filter gain matrix is attained with loop-shaping techniques and the LQR gain matrix following from loop-transfer recovery techniques. The open and closed-loop transfer function singular values are shown in Figure 3 and demonstrate the feasibility of the controller.
Early simulation with our AES maneuvering model has been more pragmatic than exploratory. We have looked at acceleration, deceleration, and execution of heading commands, including zigzag and Dieudonné spiral proxies. Figures 4-6 show results for some example trials. Characteristics worth noting include 1) the incapacity of the power system to maintain 30 knots during the ramp heading command due to increased drag, 2) the 150˚ phase lag for the zigzag test, and 3) the minor instability of the ship indicated by the gaps in the Dieudonné spiral results.
76

Figure 4: DDG 1000 Scale Model
Figure 5: Heading and Rudder Angle for Zig-Zag Maneuver
77

Figure 6: Dieudonné Spiral Stability Results for a 200 second and 400 second intervals.
Results have compared well with practice including steady-state results close to those achieved with ASSET for the DDG1000 as shown in Table 1. Results are especially near the point of linearization for our maneuvering equations.
Speed (knots)
20
22
24
26
28
30.2
Figure 7: Comparison of Surge Power for ASSET and AES Models
ASSET (MW )
9
12.5
17.1
22.8
30.4
43.6
Model (MW)
12.1
15.9
20.4
25.7
31.9
40.1
Figure 8: Execution of Heading Step Command and Associated Power Transience
These trials serve to validate our model but also to demonstrate the importance of this capability to the AES community. As an example of interest, our model predicts up to 10% power system transience during the execution of a simple heading step change because of the changes in flow to the propeller. Figure 7 shows the maneuver and the change in power delivered
78

to the induction motor to maintain ship speed. (Also note the immediate sway away from the turn before the yaw begins to dominate the dynamics.)
Other analysis with our model includes propeller optimization studies to ascertain pitchratios optimal (in the sense of power conservation) at different ship steady states. Figure 8 shows the results of two such trials: the first is for pure surge and the second is for sharp turning
(25˚ rudder angle). The independent variables are the ship target speed and the propeller pitch ratio. The dependent variable is efficiency defined as the amount of induction motor power output that reaches the ship surge power. Both plots demonstrate local minima that could be informative for early ship design.
Figure 9: Propeller Optimization Studies
• Plans for the Next Quarter
79

CHRYSSOSTOMIDIS GROUP:
Implement survivability metric calculation on notional ship using manually selected damage locations.
Continue architectural model development, increasing accuracy of weight and volume information.
Refine and apply full range of metrics to the notional ship.
HOVER GROUP:
We will continue with the theoretical development of the key robustness parameter q, and on its completion so as to formulate optimization problems for the layout of distribution systems.
KARNIADAKIS GROUP:
We plan to perform systematic sensitivity analysis of the Purple team design allowing large variations in the key parameters of the system, hence investigating its nonlinear stability. Here we will apply the new ANOVA theory we developed that allows us to handle efficiently with polynomial chaos several hundred of uncertain parameters simultaneously for a meaningful global sensitivity analysis. We also plan to work on developing models for an end-to-end simulator of the Red Team design (full scale). To this end, we will need to address the design of a robust controller for the Induction Motor. In previous work with the AC system we used a 15phase controller, which is robust but quite complex to implement. We will investigate other possibilities including the maximum torque per amp control strategy employed in the Purple team design with adaptive slip frequency.
KIRTLEY GROUP:
Contra-Rotating Propeller Design
Extend the single propeller performance analysis method of OpenProp to contra-rotating propellers for obtaining performance curves.
Propeller/Turbine Test Fixture
Complete mechanical and electrical assembly of all components and begin testing.
Contra-Rotating Induction Motor Proof-of-Concept
80

LEEB GROUP:
During the first quarter of 2010, we intend to further complete construction of our test bed, and begin exploration of the impact of electric propulsion drive on the power system. We will also make two presentations related to our work at the IEEE APEC conference. The first regards an
FPGA controller/monitor. This FPGA controller will be used to operated the generators in our test bed. It will also be used as a core for our new CBM monitor based on the NILM. The second explores techniques for tracking motor operation through the ship power system for control and monitoring using slot harmonics.
We will meet with the crew of USCGC ESCANABA during Q1 2010 as they return to port.
Experiments for diagnostic indicators and power system metrics will be conducted in the field.
During the Q2 of 2010, we intend to make an ASNEDAY publication of our preliminary comparative observations on MVDC versus MVADC power systems, again working closely with our USCG collaborators.
TRIANTAFYLLOU GROUP:
Over the next quarter, we plan to continue the development of our models through extensive testing as well as model improvements. The MIT-5D program will be tested extensively as a subroutine and the turbine model will be implemented. Also, stability analysis of the Purdue
Test Bed model will be a focus for our team in the upcoming quarter. We will attempt to develop techniques for instability identification and write a program for the analysis, allowing the user the freedom to investigate stability about different state arrangements and for different loading scenarios, in both the frequency and time domain. The three conditions of immediate concern are i) loading in parallel with propeller, 2) induction motor start-up and
3) power electronics capacitance perturbations.
81

82

2. P
UBLICATIONS AND
R
EPORTS
R
ESULTING FROM
ONR S
UPPORT
• Journals
• Book or Chapters
• Technical Report / Presentations
CHRYSSOSTOMIDIS GROUP:
J. S. Chalfant and C. Chryssostomidis, Toward the Development of an Integrated Electric Ship
Evaluation Tool, Proceedings of the 2009 Grand Challenge in Modeling and Simulation, GCMS
'09, Istanbul, Turkey, July 13-16, 2009.
B. Epps, J. S. Chalfant, R. Kimball, A. Techet, K. Flood, and C. Chryssostomidis, OpenProp: An
Open-Source Parametric Design and Analysis Tool for Propellers, Proceedings of the 2009
Grand Challenges in Modeling and Simulation, GCMS '09, Istanbul, Turkey, July 13-16, 2009.
J. Herbst, A. Gattozzi, J. Uglum, J. S. Chalfant, C. Chryssostomidis, J. Langston, M. Steurer and
M. Andrus, Modeling and Simulation of Electric Propulsion Concepts for a Multimodal
Prototype Demonstrator, Proceedings of the 2009 Grand Challenges in Modeling and
Simulation, GCMS '09, Istanbul, Turkey, July 13-16, 2009.
KIRTLEY GROUP:
S.C. Englebretson, J.L. Kirtley, Jr, and C. Chryssostomidis, Optimization of Direct Drive
Induction Motors for Electric Ship Propulsion with High Speed Propellers, Proceedings of the
2009 Grand Challenges in Modeling and Simulation, GCMS '09, Istanbul, Turkey, July 13-16,
2009.
S. Englebretson, Induction Machine Stray Loss from Inter-bar Currents, Ph.D. Thesis,
Department of Electrical Engineering and Computer Science, Massachusetts Institute of
Technology, September, 2009.
83

LEEB GROUP:
Ramsey, J.S., S.B. Leeb, T. Denucci, J. Paris, M. Obar, R. Cox, C. Laughman, T. McCoy,
“Shipboard Applications of Non-Intrusive Load Monitoring,” American Society for Naval
Engineers Reconfiguration and Survivability Symposium, Orlando, FL, February 2005.
Greene, W.C., J.S. Ramsey, S.B. Leeb, T. Denucci, J. Paris, M. Obar, R. Cox, C. Laughman, T.
McCoy, “Non-Intrusive Monitoring for Condition-Based Maintenance,” American Society for
Naval Engineers Reconfiguration and Survivability Symposium, Orlando, FL, February 2005.
DeNucci, T, R. Cox, S.B. Leeb, J. Paris, T.J. McCoy, C.R. Laughman, W.C. Greene, “Diagnostic
Indicators for Shipboard Systems using Non-Intrusive Load Monitoring,” IEEE Electric Ship
Technologies Symposium, Philadelphia, PA, July 2005.
Cox, R.W., S.B. Leeb, S.R. Shaw, L.K. Norford, “Transient Event Detection for Nonintrusive
Load Monitoring and Demand Side Management using Voltage Distortion,” Applied Power
Electronics Conference, Dallas, TX, March 2006.
Cox, R.W., J.P. Mosman, D. McKay, S.B. Leeb, T. McCoy, “Diagnostic Indicators for
Shipboard Cycling Systems Using Non-Intrusive Load Monitoring,” American Society for Naval
Engineers Day 2006, Arlington, VA, June 2006.
Cox, R., P. Bennett, D. McKay, J. Paris, S.B. Leeb, “Using the Non-Intrusive Load Monitor for
Shipboard Supervisory Control,” IEEE Electric Ship Technologies Symposium, Arlington, VA,
May 2007.
Cox, R.W., M. Piber, G. Mitchell, P. Bennett, J. Paris, W. Wichakool, S.B. Leeb, “Improving
Shipboard Maintenance Practices Using Non-Intrusive Load Monitoring,” ASNE Intelligent
Ships Symposium VII, Philadelphia, PA, May 2007.
Wichakool, W., A. Avestruz, R.W. Cox, S.B. Leeb, “Resolving Power Consumption of Variable
Power Electronic Loads Using Nonintrusive Monitoring,” IEEE Power Electronics Specialists
Conference, Orlando, FL, June 2007.
G. Mitchell, R.W. Cox, M. Piber, , P. Bennett, J. Paris, W. Wichakool, S.B. Leeb, “Shipboard
Fluid System Diagnostic Indicators Using Nonintrusive Load Monitoring,” ASNE Day 2007,
Arlington, VA June, 2007.
84

Laughman, C.R., R. LaFoy, W. Wichakool, P. Armstrong, S.B. Leeb, et.al., “Electrical and
Mechanical Methods for Detecting Liquid Slugging in Reciprocating Compressors,”
International Compressor Engineering Conference, Purdue University, West Lafayette, IN,
C1437, July 14-17 2008.
Proper, E., Cox, R., Leeb, S., Douglas, K., Paris, J., Wichakool, W., Foulks, L., Jones, R.,
Branch, P., Fuller, A., Leghorn, J., Elkins, G., “Field Demonstration of a Real-Time Non-
Intrusive Monitoring System for Condition-Based Maintenance,” Electric Ship Design
Symposium, National Harbor, Maryland, February 2009.
Paris, J., Remscrim, Z., Douglas, K., Leeb, S., Cox, R., et.al., “Scalability of Non-Intrusive Load
Monitoring for Shipboard Applications,” ASNE Day 2009, April, 2009.
Douglas, K., “Shipboard Aggregate Power Monitoring,” ME Ph.D., MIT, 2009.
Elkins, G., “Hardware Model of a Shipboard Generator,” ME Ph.D., MIT, 2009.
Leghorn, J., “Modeling for Ship Power System Emulation,” ME Ph.D., MIT, 2009.
TRIANTAFYLLOU GROUP:
1. P. Prempraneerach, J. Kirtley, C. Chryssostomidis, M.S. Triantafyllou , & G.E. Karniadakis,
2008, “Design of the All-Electric Ship: Focus on Integrated Power System coupled to
Hydrodynamics”, Electric Ship Design Symposium 2009 (SNAME/ASNE), National Harbor,
MD.
** Outgrowth of supervised student research
• Inventions
• Honors
3. P
ERSONNEL
S
TATISTICS DURING THIS
R
EPORTING
P
ERIOD
3.1.
Total No of Co-PI: 8
• No of Woman Co-PI: 2
• No of Minority Co-PI
85

3.2.
Total No of students working on the grant
• Total Grad: 15 o Woman Grad Students: 3 o Minority Grad Students:
• Total Under Grad: 2 o Woman Under Grad Students: 1 o Minority Under Grad Students
• Total Post Docs: 1 o Woman Post Docs: 1 o Minority Post Docs
• No of degrees Granted during this report Period: None
86