INTELLIGENT ENERGY MANAGEMENT IN A TWO POWER-BUS VEHICLE SYSTEM 1
Report Documentation Page Form Approved OMB No. 0704-0188 Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE 09 AUG 2011 2. REPORT TYPE N/A 3. DATES COVERED - 4. TITLE AND SUBTITLE Intelligent Energy Management in a Two Power-Bus Vehicle system 6. AUTHOR(S) M. Abul Masrur 5a. CONTRACT NUMBER 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) US Army RDECOM-TARDEC 6501 E 11 Mile Rd Warren, MI 48397-5000, USA 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) US Army RDECOM-TARDEC 6501 E 11 Mile Rd Warren, MI 48397-5000, USA 8. PERFORMING ORGANIZATION REPORT NUMBER 22202 10. SPONSOR/MONITOR S ACRONYM(S) TACOM/TARDEC/RDECOM 11. SPONSOR/MONITOR S REPORT NUMBER(S) 22202 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release, distribution unlimited 13. SUPPLEMENTARY NOTES Presented at the 2011 NDIA Vehicles Systems Engineering and Technology Symposium 9-11 August 2011, Dearborn, Michigan, USA, The original document contains color images. 14. ABSTRACT 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT SAR a. REPORT unclassified b. ABSTRACT unclassified c. THIS PAGE unclassified 18. NUMBER OF PAGES 35 19a. NAME OF RESPONSIBLE PERSON Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18
INTELLIGENT ENERGY MANAGEMENT IN A TWO POWER-BUS VEHICLE SYSTEM Zhihang Chen 1, Yi L. Murphey 1, Zheng Chen 1, Abul Masrur 2, Chris Mi 1 1 College of Electrical and Computer Science University of Michigan-Dearborn Dearborn, Michigan, USA, 48128 2 US Army RDECOM-TARDEC Warren Michigan 48397-5000 yilu@umich.edu, corresponding author 2
Intelligent Vehicle Power Management Vehicle power management has been an active research area in recent years Most approaches were developed based on mathematical models, human expertise, or knowledge derived from simulation data. Control strategy of a military vehicle is more complicated than commercial vehicles Multiple power sources the complex configuration and operation modes, heavy weight multiple functions, which cause big load fluctuation engaging weapons, turning on sensors, silent watch, etc. Our research: Cognitive Intelligent Vehicle Power Management Intelligent power control based on machine learning, optimization and human intelligence 3
In this presentation We present our research in optimizing power flow in a vehicle power system that employs multiple power sources. focusing on a vehicle power system architecture that is used in vehicles such as Mine Resistant Ambush Protected (MRAP) vehicle Developing algorithms for intelligent energy control Using a commercial simulation software to model the vehicle system for experiments Constructing a lab hardware setup to verify the energy management algorithms 4
MRAP Power System Simulation, Optimization and Hardware Implementation 5
Building three MRAP systems Full scale simulation using a commercial software Using Stryker as a vehicle model Constructing the power components with the same sizes as in SPEC Simulation of Hardware MRAP system Using the same sizes of power components as in hardware setup Hardware Implementation of MRAP Scaled down version due to the available hardware 6
Intelligent Power Management in a Simulated MRAP Vehicle 7
Simulation Environment Simulation program: Vehicle Model: Stryker model Power system: Build based on specification of MRAP power system 8
MRAP Power System Specification D-power system P l1 Automotive Loads 1 Battery 1 f e Engine 1 P e P m1 Alternator 1 P a1 P b1 P s1 Power Storage 1 P loss1 Power Loss 1 Interswitch Battery 2 Hydaulic System P P m2 a 2 Alternator 2 P b2 P s2 P loss2 Power Storage 2 Power Loss 2 C-power system P l 2 Automotive Loads 2 9
Model Overview Simulated MRAP Power System Engine Fuel Inter. Switch Loads Clean Alt. CleanBus Clean Batt. Dirty Batt. Dirty Alt. DirtyBus 10
Intelligent Power Management Algorithms Only optimize the power to or from dirty battery, D_P s Will optimize both C_P s and D_P s in next phase Three algorithms: Applying DP optimization to a given drive cycle Applying optimal power setting found by DP to online controller Training a neural network (NN) for online controller 11
Dirty and Clean Loads used in Experiments 4500 Dirty Load Power 4500 Clean Load Power 4000 4000 3500 3500 Power (W) 3000 2500 Power (W) 3000 2500 2000 2000 1500 1500 1000 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec) 1000 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec) 12
Designed Drive Cycle A drive cycle in which vehicle is moving 30% of time and 70% of time is idle The speed profile is constructed based on two standard drive cycles for heavy trucks WVU_Inter and WVU_City drive cycle; 30 25 Vehicle Speed('mph') 20 15 10 5 0 0 2000 4000 6000 8000 10000 12000 Time('s') Experiment 13
Performances for the 6000 second drive cycle Cycle Segment (sec) Benchma rk Fuel (kg) DP Fuel (kg) Online Fuel (kg) Online Fuel (kg) w/ SOC Correction DP Savings (%) Online Open Loop DP Savings (%) 0-2000 1.6920 1.5614 1.6030 1.6100 7.72 4.8 2001-4000 1.8919 1.753 1.7585 1.7660 7.34 6.65 4001-6000 2.1115 1.9706 1.9846 1.9921 6.67 5.65 14
Neural Network Online Implementation Neural networks are trained to behave like DP, outputting optimal Ps (Dirty bus battery power). Online implementation contains two neural networks One is trained for interstate drive cycle, another for city drive cycle When the vehicle is not in idle mode: determine the current roadway type Call the neural network trained on the current roadway type 15
Performances DP and NN for the first 2000 second drive cycle controller Fuel Consumed (kg) Fuel w/ SOC correction Software based controller 1.6920 Savings (%) DP 1.5614 7.72 Online DP 1.6030 1.6100 4.85 Online NN 1.5961 1.6074 5.00 16
MRAP Power System Hardware Implementation and Experiments 17
Hardware Implementation of MRAP Power System It is a scale-down version due to the hardware components currently available at the authors power electronics labs. Using Electric Motor and Generator with a reduced ratio to replace the real MRAP engine and alternator Control algorithm: Commercial software based control algorithm and Algorithm based on DP developed by the authors 18
System Configuration Dirty Bus Branch Engine Alternator Bus 28V Battery Load Dirty Bus Branch Induction Motor Field Controlled DC generator Bus 28V Battery Load Realy Realy Hydraulic systems Clean Bus Branch Alternator Bus 28V Battery Load Clean Bus Branch Permanent Magnetic DC Motor Field Controlled DC generator Bus 28V Battery Load 19
Hardware System Implementation Mechanical Connection A/D Current Sensor 28V Battery DC Power Supply (208V, 20A) Power Inverter Induction Motor Field Controlled DC generator Dirty Bus Branch Current System Switch PWM Driven Board D/A Control Speed Encoder Dspace CLP1104 DC Power Supply D/A Control I/O Control D/A Control A/D Voltage Current A/D Relay Current Electronic Load RS-232 Clean Bus Branch Current Speed Encoder DC Power Supply A/D Voltage Current A/D Current Sensor DC Power Supply (125V, 3A) Permanent Magnetic DC Motor Field Controlled DC generator 28V Battery Mechanical Connection Relay Controlled Load 20
System Components Dirty Bus Branch Clean Bus Branch Original System Demo System Original System Demo System Engine Induction Motor Hydraulic System Permanent DC Motor Alternator Field Controlled DC Generator Alternator Field Controlled DC Generator Battery Lithium Battery Battery Lithium Battery Load Electronic Load Load Resistance Load with Relay Control 21
Parameters of Platform Induction Motor( Engine): Rated Voltage: 208v Rated Current:10.8A Rated Power: 3 H.P Rated Speed: 1800 rpm 3 phase, winding wounded rotor Dirty power system DC Generator(Generator) Rated Voltage: 125v(90v) Rated Current: 19A Rated Power: 3 H.P Rated Speed: 1800 rpm Battery: Type: Lithium Battery Rated Voltage: 12.8v(per unit) Capacity: 40Ah(per unit) Total: Rated Voltage: 76.8v(*6) Capacity: 40Ah Power Supply Maximum Capacity : 600v/20A, Inverter: Applied Power System Products Maximum Capacity : 600v/100A, Maximum Frequency: 20KHz 22
Parameters of Platform PM DC Motor: Rated Voltage: 125 Rated Current:3.2A Rated Power: 1/3 H.P Rated Speed: 1750 rpm Clean power system DC Generator Rated Voltage: 125v Rated Current: 3.0A Rated Power: 1/3 H.P Rated Speed: 1800 rpm Battery: Type: Lithium Battery Rated Voltage: 12.8v(per unit) Capacity: 40Ah(per unit) Total: Rated Voltage: 25.6v Capacity: 40Ah Power Supply Maximum Capacity : 600v/10A Relay: Maximum Current: 200A 23
Control Strategy Dirty Bus branch Induction Motor: VVVF Control; Generator: Field Voltage Control. Proportional Integral Control (PI) Control Clean Bus branch PM DC Motor: Voltage Control. PI Control Generator: Field Voltage Control. PI Control System Control Algorithm: (baseline controller) Target: SOC=0.7; control target: SOC=0.7 using PI Control Not allowed to exceed the maximum current of alternator. 24
Hardware Configuration Dirty Bus Motor and Alternator Clean Bus Load Clean Bus Motor and Alternator Dirty Bus Load 25
Hardware Configuration Real time environment tool Power Inverter PC Clean Bus Battery Dirty Bus Battery Current and Voltage Sensor 26
Measurable data Dirty Load Branches: Generator output current Battery current Bus Voltage Power supply output voltage Power supply output current Induction motor speed Clean Load Branches: Battery current Bus voltage Speed of PM DC motor Power supply output voltage Power supply output current Generator output current 27
Calculated Data Dirty Bus Branch: SOC of Battery; Torque of Induction Motor; Equivalent Fuel Consumption Clean Bus Branch: SOC of Battery; Torque of PMDC Motor. 28
Real-time Environment Data Acquisition Based on the real time environment tool. 29
Fuel Consumption Obtain the DC bus current get the power and torque of clean bus branch and dirty bus branch Add vehicle motion power Look up the engine fuel map to get the fuel consumption. Hydraulic system efficiency: 75%; Vehicle Motion Power Dirty Bus Branch Power SUM Fuel Map Fuel Rate Ingeration Fuel Consumption Clean Bus Branch Power Hydraulic Efficiency 75% 30
Designed Drive Cycle 30 Based on the WVU_Inter and WVU_City Drive Cycle; 70% idle. 25 Vehicle Speed('mph') 20 15 10 5 0 0 2000 4000 6000 8000 10000 12000 Time('s') Experiment 31
System Performances Algorithm Fuel Consumption Improving Software based Control Algorithm Dynamic Programming Control 3.5336kg 3.352kg 5.14% Online Dynamic Programming Control 3.419kg 3.24% 32
Performance Summary of Intelligent Power Controller in MRAP Power System Simulated MRAP Power System Embedded in software Stryker Model controller Fuel Consumed (kg) Fuel w/ SOC correction Savings (%) Software tool based controller 1.6920 DP 1.5614 7.72 Online DP 1.6030 1.6100 4.85 Online NN 1.5961 1.6074 5.00 Hardware Implementation of MRAP Power System Control ler Fuel Consumption Saving (%) Software tool based controller 3.5336kg Offline Dynamic Programming 3.352kg 5.14% Online Dynamic Programming 3.419kg 3.24% Simulated hardware MRAP Power System Controller Fuel Consumption Saving (%) Software tool based controller 3.453kg Offline Dynamic Programming 3.27kg 5.51% Online Dynamic Programming 3.336kg 3.39% 33
Video demo if time permits 34
Conclusion An intelligent power controller for a two power-bus system in vehicle systems is presented. Based on the simulations, and experimental results implemented in the lab setup, it can be concluded that the intelligent controller developed by the authors can improve fuel consumption through online vehicular power management in a real time environment. In the simulated vehicular system, this controller saved about 5% fuel. In a lab setup environment, the controller saved about 3.2% fuel. The tools developed by the authors and reported in this paper can be used to save significant cost and development efforts by the manufacturers prior to any production level activities involving such vehicular systems. 35