Modeling and Simulation of an Energy Management System for Plug-in Hybrid Electric Vehicles

Salisa Abdul Rahman1,2, Nong Zhang1and Jianguo Zhu1

School of Electrical, Mechanical and Mechatronic Systems, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia. 2 Department of Physical Sciences, Faculty of Science and Technology, Universiti Malaysia Terengganu, 21030 Kuala Terengganu, Terengganu Darul Iman. Salisa.AbdulRahman@student.uts.edu.au
Abstract- This paper presents modeling, control strategy and simulation results of an energy management strategy (EMS) for a specific Plug-in Hybrid Electric Vehicle (PHEV). A good control strategy is required among components, such as the Energy Storage System, an Electric Motor, a Power Control Unit, and an Internal Combustion Engine in order to ensure that the vehicle achieve an improvement in energy efficiency, reduction in emissions, fuel use, weight and cost. In this work, firstly, through a power flow analysis, the vehicle components are sized to meet the expected power and energy requirements of a typical 5passenger car. Then, the model is tested numerically in the MATLAB/SIMULINK environment using the Highway Fuel Economy Test drive cycle with the proposed EMS. The accuracy of the model is verified by a comparison between the simulation results from a specific PHEV code and Advanced Vehicle Simulator (ADVISOR) Hybrid Electric Vehicle (HEV) model. The simulation results of the two different HEV models are compared and the pros and cons discussed.
1

and desired vehicle speeds are used as the inputs to determine the required drivetrain torque, speed and power [4]. Figures 2, 3 and 4 show the ADVISOR screens of input, simulation setup and results, respectively. The ADVISOR model has been validates and used as a benchmark of reference model. The ADVISOR model vehicle contains two separate EMs which are used as the motor and generator, respectively, and no Ultracapacitor in the ESS. The proposed PHEV, however, has only one EM which functions as either a motor or generator at a time, specified by the special energy management scheme, and an Ultracapacitor bank for fast charging and discharging during the regenerative braking and fast acceleration. To simulate the proposed PHEV, we have derived a model and compiled a PHEV code. This code is verified by comparing the simulation results of the ADVISOR model HEV.

I.

INTRODUCTION

Figure 1 shows a block diagram of the proposed Plug-in Hybrid Electric Vehicle (PHEV) configuration, consists of an Energy Storage System (ESS), a Power Control Unit (PCU), an Electric Motor (EM) and an Internal Combustion Engine (ICE) [1,2]. A proposed Energy Management Strategy (EMS) is applied to the PHEV code to ensure the vehicle achieve the target driving performance without sacrificing the optimum operating condition. II. ADVISOR Advanced Vehicle Simulator (ADVISOR) is a software based on MATLAB/SIMULINK, originally developed by the U. S. Department of Energy (DOE) and the National Renewable Energy Laboratory (NREL), to simulate and analyze light and heavy vehicles, including hybrid and fuel cell vehicles [3]. It allows the user to perform rapid analysis of the performance, emissions and fuel economy of conventional, electric and hybrid vehicles. ADVISOR was initially developed as an analysis tool, rather than a detailed design tool. Its components are created as a quasi-stated model, and therefore, it cannot be used to perform dynamic analysis. ADVISOR utilizes a backward looking vehicle simulation architecture, in which the required

ESS

PCU

ICE Wheel Gear Reduction

EM Wheel

Figure 1. Block diagram of PHEV configuration

III. VEHICLE MODELING The vehicle type selected is an average 5-passenger sedan, which is the majority of the vehicles on road. Table 1 lists the parameters of a typical vehicle of this type [5]. In the simulation, the air density is chosen as 1.18 kg/m3, and the gravitational acceleration 9.81 m/s2.

2008 Australasian Universities Power Engineering Conference (AUPEC'08)

Paper P-274 Page 1

TABLE I PARAMETERS OF AN AVERAGE 5-PASSENGER SEDAN

Name Aerodynamic drag coefficient Coefficient of rolling resistance Frontal area Wheel radius Vehicle mass

Value 0.29 0.01 2.52 0.291 1255

Units m2 m kg

Figure 2. ADVISOR vehicle input screen

The development of the vehicle models begins with calculations of vehicle energy and power requirements for typical driving conditions based on the parameters and target specifications of the vehicle. The size and capacity of each vehicle component are then determined through a power flow analysis accordingly to meet these requirements. Combining the constitutive equations of all components, we obtain a mathematical model of the vehicle. The vehicle performance for a given EMS and driving cycle is simulated in the MATLAB/SIMULINK environment. Figure 5 illustrates the overall structure of the PHEV model in MATLAB/SIMULINK [6-8].

Figure 5. Overall structure of the PHEV model in MATLAB/SIMULINK Figure 3. ADVISOR simulation setup screen

IV. ENERGY MANAGEMENT STRATEGY The EMS is responsible for deciding in which mode that the vehicle is operating. Figure 6 shows various operation modes of the proposed EMS to control the distribution of power amongst the components, which are mechanical braking, regenerative braking, motor only, engine recharge, engine and motor assist and engine only mode according to the vehicle power demand in acceleration and deceleration and the ESS SOC level [911]. V. SIMULATION RESULTS A standard U.S. EPA (Environmental Protection Agency) drive cycle for highway driving is simulated by ADVISOR and the PHEV code. Figure 7 illustrates the time history of the HWFET drive cycle.

Figure 4. ADVISOR results screen

2008 Australasian Universities Power Engineering Conference (AUPEC'08)

Paper P-274 Page 2

Deceleration High SOC Moderate SOC Low SOC Mechanical braking mode

Acceleration Motor only mode Motor only or Engine recharge mode Engine recharge mode Moderate Power Demand of Vehicle Engine and Motor assist mode

Regenerative braking mode

Engine only mode High

Negative

Figure 6. EMS modes of operations

Figure 8. Vehicle speed in HWFET cycle (Blue: PHEV, Red: ADVISOR)

Figure 7. Highway Fuel Economy (HWFET)

The actual speeds and forces for the HWFET drive cycle are presented in Figure 8 and 9, respectively. There is a close match between the results, as expected. Figures 10, 11 and 12 show the simulated results of ESS current, voltage and output power for the HWFET drive cycle. The peak currents are due to the high power demand to achieve fast vehicle accelerations during the respective periods. The negative values on the graph represent the regenerative braking events during the hard braking periods in the cycle. In the ESS voltage graph, the voltage increases during recharging from regenerative braking and decreases during high current discharge when the power demand from EM is at peak. The PHEV code for the HWFET drive cycle exhibits values lower for the ESS current than that of ADVISOR model. Such phenomenon is due to the power consumption of the vehicle under different EMS, and therefore can be accepted with a reasonable explanation.

Figure 9. Vehicle force in HWFET drive cycle (Blue: PHEV, Red: ADVISOR)

Figure 10. ESS current in HWFET cycle (Blue: PHEV, Red: ADVISOR)

2008 Australasian Universities Power Engineering Conference (AUPEC'08)

Paper P-274 Page 3

Figure 11. ESS voltage in HWFET cycle (Blue: PHEV, Red: ADVISOR)

Figure 12. ESS output power in HWFET cycle (Blue: PHEV, Red: ADVISOR)

Figure 13 illustrates the State-Of-Charge (SOC) of the ESS system for the HWFET drive cycle. The overall trends of the energy consumption and generation of the two models match reasonably well. However, there is some discrepancy between the ESS SOC results of the PHEV and ADVISOR codes. This is because the PHEV code has a better EMS and can capture more regenerative braking energy. The EM speed, torque and power of the PHEV and ADVISOR codes for the HWFET drive cycle are included in Figures 14, 15 and 16. As shown in the simulation results, when the vehicle accelerates, the required Motor/Generator torque increases quickly, and when the vehicle reaches the relatively stable highway velocity level, a much smaller torque is required o overcome the resistance and air drag to the vehicle. The average power demand from the Motor/Generator is 8 kW at the highway velocity level and the peak power demand is 24 kW during the acceleration. The speed, torque and power results from the two codes match reasonably well. However, the PHEV code exhibits lower torque and input power than the ADVISOR code. Figure 17 and 18 depict the wheel speed and torque requirement for the HWFET drive cycle simulated by two codes. The maximum wheel torque, 600Nm, occurs when the vehicle is accelerating from standstill to the highway speed. The required torque then reduces since the HWFET drive cycle only consists of mild accelerations and decelerations. The overall results and trends match very closely. However, it is noted that there are several differences between the magnitudes of the wheel torque. It can be attributed to difference between the two strategies.

Figure 13. ESS SOC in HWFET drive cycle (Blue: PHEV, Red: ADVISOR)

Figure 14. EM speed in HWFET drive cycle (Blue: PHEV, Red: ADVISOR)

2008 Australasian Universities Power Engineering Conference (AUPEC'08)

Paper P-274 Page 4

Figure 15. EM torque in HWFET drive cycle (Blue: PHEV, Red: ADVISOR)

Figure 17. Wheel speed in HWFET cycle (Blue: PHEV, Red: ADVISOR)

Figure 16. EM input power in HWFET cycle (Blue: PHEV, Red: ADVISOR)

Figure 18. Wheel torque in HWFET cycle (Blue: PHEV, Red: ADVISOR)

Figure 17 and 18 depict the wheel speed and torque requirement for the HWFET drive cycle for two codes. The maximum wheel torque, 600Nm, occurs when the vehicle is accelerating from standstill to the highway speed. The required torque then reduces since the HWFET drive cycle only consists of mild accelerations and decelerations. The overall results and trends match very closely. However, it is noted that there are several differences between the magnitudes of the wheel torque. It can be attributed to difference between the two strategies. Figure 19 plots the acquired and required speeds of the HWFET drive cycle. It can be seen that acquired and required speeds agree reasonably well. The PHEV code followed the required drive cycle speed very well for the standard drive cycle used to simulate highway driving.
Figure 19. HWFET cycle (Green: required speed, Magenta: acquired speed)

2008 Australasian Universities Power Engineering Conference (AUPEC'08)

Paper P-274 Page 5

VI. CONCLUSION The results of the vehicle subsystems in terms of ESS current, voltage, output power and SOC, Motor/Generator speed and torque, vehicle speed and force and wheel speed and torque are within reasonable and expected range of actual typical behavior of these subsystems. The components of the vehicle subsystems are correctly sized as the vehicle is capable of achieving performance to a target velocity. In combination with previous discussion, it can be concluded that results of the PHEV code are correct ACKNOWLEDGEMENT This research is part of the work performed by the Centre of Intelligent Mechatronic Systems, University of Technology, Sydney, Australia. REFERENCES
[1] [2] B. K. Powell, K. E. Bailey and S. R. Cikanek, Dynamic Modeling and Control of Hybrid Electric Vehicle Powertrain Systems, IEEE Control Systems, 0272 - 1708, 1998. L. B. Karen, E. Mehrdad and K. Preyas, A Matlab-Based Modeling and Simulation Package for Electric and Hybrid Electric Vehicle Design,

IEEE Transactions on Vehicular Technology., vol. 48., No. 6, 1999, pp.1770-1778. [3] B. K. Wipke, ADVISOR 2.1: A User-friendly Advanced Powertrain Simulation using a Combined Backward/forward Approach, IEEE Transactions on Vehicular Technology, vol. 48, 1999, pp 1751-1761. [4] T. Markel, ADVISOR: A Systems Analysis Tool for advanced Vehicle Modeling, Journal of Power Sources 110, 2002, pp. 255-266. [5] W. Johanna, Modelling of Components for Conventional Car and Hybrid Electric Vehicle in Modelica, ME Thesis, Linkoping Universitet, 2004. [6] L. Z. Yuliang, Modeling and Simulation of Hybrid Electric Vehicles, ME Thesis, University of Victoria, 2007. [7] J. K. Young, Integrated Modeling and Hardware-In-The-Loop Study for Systematic Evaluation of Hydraulic Hybrid Propulsion Options, PhD Thesis, University of Michigan, 2008. [8] L. Jinming, Modelling, Configuraiton and Control Optimization of Power Split Hybrid Vehicles, PhD Thesis, University of Michigan, 2007. [9] J. Nashat, A. K. Naim and S. Mutasim, A Rule Based Energy Management Strategy for Series Hybrid Vehicle, Proceedings of the American Control Conference, Albuquerque, New Mexico, 1997, pp. 689-693. [10] T. N. Csaba, Investigation and Simulation of the Planetary Combination Hybrid Electric Vehicle, ME Thesis, University of West Virginia, 2000. [11] W. Maria, Energy Management of Hybrid Electric Vehicles, ME Thesis, University of Waterloo, 2006.

2008 Australasian Universities Power Engineering Conference (AUPEC'08)

Paper P-274 Page 6