A MATLAB Simulink Model for Toyota Prius 2004 based on DOE reports

Hiva Nasiri Ahmad Radan Maziar Parizadeh Abbas Ghayebloo Power Electronics Laboratory K.N.Toosi University of Technology Seyed Khandan Bridge, 1431714191 Tehran, Iran Tel.: +98 21 84062408, Fax: +98 21 88462066 hiva.nasiri@gmail.com, Radan@kntu.ac.ir, parizade.maziar@gmail.com abbas.ghayebloo@gmail.com http://www.kntu.ac.ir/

Keywords
Hybrid Electric Vehicle (HEV), Modeling, Simulation, Energy Management System

Abstract
In this paper, a dynamic model for Toyota Prius 2004 hybrid electric vehicle (HEV) is presented. The model is based on MATLAB Simulink HEV model which is modified in such a way that dynamic response of model can be verified by experimental results given in DOE reports for Toyota Prius 2004. The newly developed model can help the designers to enhance the performances of subsystems.

1. Introduction
Environmental pollution and energy crisis have been the most concern of automotive industry for the past four decades. Consequently, major automotive companies have developed the technology of hybrid electric vehicle (HEV) to achieve better fuel economy and lower emissions through optimizing the vehicle performances and engine operation points[1], [2]. These improvements have been achieved by adding some electrical components such as electrical storage devices and electric motors to power and torque path of conventional automobiles with internal combustion engine (ICE) [3]. In order to design and test different components and strategies in HEVs to achieve better performances, several computer modeling and simulations have to be used to examine and compare the performance of these vehicles. Based on the goals of application, these models can be divided into two categories [4]: Models for designing stages and evaluating high level operating strategies that include long-term analysis such as SIMPLEV from the DOEs Idaho National Laboratory[5], ADVISOR from the DOEs National Renewable Energy Laboratory [6] and PSAT from Argonne National Research Laboratories [7] belong to the first category. But the second ones analyzing the interactions between subsystems and their design are those which model the subsystems in details and are used to address the dynamic behavior of HEV subsystem. V-Elph developed at Texas A&M University [8], PSIMbased model from Illinois Institute of Technology [9] are two dynamic simulators that use lower level of HEVs for studying detailed performance issues. In this paper, authors present a dynamic model for well-known hybrid vehicle whose precise test results reported by DOE can validate its model with an acceptable accuracy. For this purpose the second generation of Toyota hybrid system (THS II) of 2004 Prius considering the information given in U.S. Department of Energy (DOE) reports [10] was selected (Table I). This vehicle has one of famous HEV drive train and earns many awards for its design such as Best Engineered Vehicle for 2004 chosen by readers and editors of Automotive Engineering International (AEI) [11]. The model presented here is based on the MATLAB Simulink model for HEV [12] with several improvements in various subsystems and modifications in model specifications (Fig. 1). In the following section, at first, different parts of model like energy management, ICE, electric subsystem are described and the necessary modifications of original MATLAB Simulink model have been discussed. Then, in section 3, some simulation results using the newly developed model are presented and compared with the experimental results of DOE reports to validate the model. Finally, some conclusions have been made in section 4.

Table I: Toyota Prius 2004 components

Subassembly Vehicle Maximum vehicle speed Engine Planetary gear Specifications Description Weight Electric mode Hybrid mode Max power Ratio (ring, planet, sun) Max power Maximum speed Maximum torque Max power Maximum speed Maximum torque NiMh module number Nominal energy Nominal voltage Value 1360kg 60 km/h 160 km/h 57kw@5000rpm 2.6 (78/23/30) 50 kW 6000 rpm 400 Nm (01200 rpm) 30 kW 10000 rpm 160 Nm 28 1.3 kWh 201.6 V

Electrical motor Electrical generator

battery

2. Developed Simulink Model

The model presented here is based on MATLAB Simulink model for HEV with several changes in various subsystems and modifying model specifications (Fig. 1). The major changes are in energy management subsystem and ICE. We will describe each subsystem individually.
[Wice] 1+2.6 Gain2 [Wmot]
Accelerator Accelerator Motor torque Motor torque ref (Nm)

[Wgen]

[T mot]

2.6 Gain3

Accel3

[Wice]

ICE Speed Gen torque

[T gen]
Generator peed ref (rpm)

Accelerator

[T mot] [Wmot] [Tgen] [Img]

T orque Power

Driv e torque (ref erence, measured) Electrical Power (Motor, Generator, Battery )

[Wgen]

Gen Speed

Car

Throttle ICE

ICE Throttle

[Wgen] [Vmg]
Car speed (km/h)

[Wmot]

Motor Speed

[Batt]

Batt

DC Link Control

Electrical Subsystem
Speed Sensor2

v Speed Sensor1

[Wmot]

speed1

Energy Management Subsystem

Vehicle Dynamics
Accel2

Internal Combustion Engine

[Wice]

Hybrid Electric Vehicle (HEV) Power Train Using Batte ry M odel

pow ergui Discrete, Ts = 6e-005 s.

The 'Ts' parameter used in this model is set to 6e-5 by the Model Properties Callbacks
[Img] [Batt]

[Vmg]

Power Subsystem

Fig.1: MATLAB Simulink model for Toyota Prius 2004

Energy Management System: Energy management is mainly based on vehicle speed, subsystems (engine, motor and generator) speed, input acceleration (Drive Cycle), and Batterys variables (voltage, current and State of Charge (SOC)). This subsystem contains a main controller for defining the amount of torque that each torque provider (ICE, Motor/Generators (MG)) should produce and is divided into three parts: battery management, hybrid management and ICE speed controller (Fig. 2). Battery management system receives batterys data values from energy storage subsystem to limit the range of SOC between 40% and 80% and to specify the amount of receiving or sending power for battery.

Fig. 2: Energy Management Subsystem

In a hybrid management system (Fig. 3) the amount of required torque for engine, traction motor and generator are specified with the help of subsystems rotating speed and demand power as well as the torque defined by the amount of acceleration and brake pedal from drive cycle. At first, state of hybridization is determined by demanded power, vehicle speed, and batterys SOC. In elementary model the instant of hybridization is not specified by vehicle speed, but according to DOE report, any increase in speed more than 24 km/h should activate ICE [13,17], therefore speed condition is added to this function. Another improvement in Simulink model is about motor reference torque during the brake time when the brake power exceeds the power capacity limit of battery. In our model, additional power is sent to a mechanical brake instead of motor and battery pack. In addition A DC link voltage controller has been added to the original model that provides the ability of dc link voltage controlling in different vehicle cycling modes. This has not been considered in MATLAB original model. ICE speed controller produces reference torque for ICE from its speed that achieved from efficiency map. Note that the efficiency map in MATLAB Simulink model doesnt match with original map that this problem has been corrected in new model.

Fig. 3: Hybrid Management Subsystem Electrical Subsystem

This subsystem contains motor and generator and planetary gear (Fig. 4). Motor has field weakening controller and its characteristics are like torque-speed data provided by DOE report [14].

This block is based on the AC6 IPMSM drive block Motor Enable 1 1 i1
s + + i -

demux
motor i_a speed
Stator current (A) Rotor speed (rpm) Electromagnetic Torque (Nm)

PMSM Motor Drive

[Imot] Goto1

Torque Ref

Ctrl Torque [measured, ref erence]

T SIm/SDL1 1 Ring

Motor speed
Te

Motor Drive 50kW

[Imot] 5 Vmg DC Voltage Link

V motor

[Igen]

1 Img

Carrier 2

SIm/SDL2 T

Gain2 1+2.6 -2.6 Gain T SIm/SDL3

Gen Enable 1 [Igen] i2

s + + i -

This block is based on the AC6 drive block

demux1
motor i_a speed
Stator current (A)

PMSM Generator Drive

3 Torque ref 4 Generator Speed

Te

Rotor speed (rpm) Electromagnetic Torque (Nm)

Goto

Ctrl Torque [measured, ref erence]

Generator Drive 30kW

Fig. 4: Electrical subsystem containing torque relationships model for planetary gear

Power Subsystem

In this model, as depicted in Fig. 5, battery and DC-DC converter are placed in one block so that control strategy or its whole system could be changed and replaced with new systems. An additional ability to control DC-DC link voltage has been added to optimize switching losses, since it exists in real Prius 2004 model. With this ability in the future, various DC voltage link strategies could be investigated. So the difference between Prius 2004 Simulink model and MATLAB model in this case is providing an additional ability to control DC Voltage link and also packing the whole blocks together so that new strategies could be preceded conveniently.
2 DC Link Control
DC Link Control +Vdc Bus

1
iout v out v ref iref +

Img

Img + motor - motor + generator Vmg

+ m

+ battery -Vdc Bus

Meas VI3

2 Vmg

_ Battery

- battery

DC/DC Converter

Imot Vdc m battery Igen

- generator

Subsystem

1 Batt

Fig. 5: Energy storage subsystem including DC-DC converter ICE

We have used Toyota Prius 2004s experimental characteristic for designing ICE. Torque-speed characteristic of the ICE is based on [15]. In energy management for controlling the ICE we need speed-power lookup table to estimate the amount of power needed to drive the vehicle so by modifying this table we have produced a speed-power lookup table. The lookup table is extracted from Fig. 6[16]. We have extracted the maximum torque limitation of ICE but we try to use ICE at its maximum efficiency. In Fig. 6 the highest line indicates the maximum torque limitation and the line under it shows the most efficient working area.

i1

Electrical measurements

i -

Fig. 6: ICE torque-speed lookup table

Planetary Gear

Planetary gear block of original MATLAB HEV model is mechanical and a little confusing so based on [17] we have designed a new and simple model which satisfies both torque and speed requirements (Fig. 7). There is another model for torque connection between generator, traction motor and vehicle differential which is not depicted here.
[Wice] 1+2.6 Gain2 [Wmot] 2.6 Gain3 [Wgen]

Fig. 7: Speed coupling model for planetary gear. Vehicle Dynamics

Since in DOE report there is no information on vehicle conditions, the parameters used in this model are based on both Toyota company reports and MATLAB default HEV model. Table II contains some of the parameters used as vehicle dynamics. Table II: Vehicle Dynamics
Description Mass(kg) Frontal Area(m^2) Drag coefficient CG height from ground (m) Value 1360 2.57 0.26 0.5

3. Simulation Results
For evaluating the model, first, the data of experimental results given in DOE report are extracted. Since there is no detail of road condition, through a procedure, the road and its conditions are rebuilt. The road conditions should be extracted for every drive test, indeed. After finding road conditions, simulations have been performed using the same drive cycle and road conditions as in DOE reports. Fig. 8 and 9 elaborate the results. Fig (8-a) indicates the driver commands used for these simulations. The positive and negative amounts are considered as acceleration and brake signals respectively. In Fig (6-b) the experimental results of DOE report, namely the vehicles speed, are compared with the simulation results obtained using the developed model by Simulink. The digitized one corresponding to the DOE results is completely consistent with the smooth one corresponding to the simulation results for the same driver command pattern. The last figure in this series indicates the battery SOC. Fig. 9 shows the results for another driver command pattern which just consists of only acceleration time and had no brake.

Vehicle Drive Cycle 0.8 0.6 Acceleration(positive amount) 0.4 Acceleration(per unit) 0.2 0 -0.2 Brake (negative amount) -0.4 -0.6 -0.8

a)

10

20

30 Time(s)

40

50

60

Vehicle speed diagram. Speed VS Time 80 Simulation output (smooth) 70 60 Speed(Km/h) 50 40 30 20 10 0 0 10 20 30 Time(s)
Battery SOC VS Time 63.5 63 State of Charge (%) 62.5 62 61.5 61 60.5 60 59.5 0 10 20 30 Time(s) 40 50 60

b)

DOE result (digitized)

40

50

60

c)

Fig. 8: First evaluation result output compared with DOE report (Fig A3 in DOE report). a) Drive cycle, b) vehicle speed and c) battery SOC

Vehicle's Drive Cycle 0.8 0.6 Acceleration(per unit) 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 0 5 10 15 Time(s)
Vehicle speed VS time 80 Simulink result(smooth graph) 70 Vehicle Speed(km/h) 60 50 40 30 20 10 0 0 5 10 15 Time(s)
Battery SOC VS Time 63.5 63 Battery State of Charge 62.5 62 61.5 61 60.5 60 59.5 0 5 10 15 Time(s) 20 25 30 35

a)

20

25

30

35

b)

DOE report(digitized graph)

20

25

30

35

c)

Fig. 9: Second evaluation result compared with DOE report (Fig 3.7 in DOE report). a) Drive cycle, b) vehicle speed and c) battery SOC

4. Conclusion
In this paper a dynamic MATLAB Simulink model based on the experimental data of a real Hybrid Electric Vehicle has been developed. The reliable reference data of Toyota Prius 2004 has been elicited from the DOE reports and the model is developed based on them. The simulation results using newly developed model are consistent with those of DOE reports and show the validity of model. The model could be used as a base model for developing various strategies on hybrid electric vehicles to obtain reliable and accurate results.

5. References
[1] M. Ehsani, Y. Gao, and J. M. Miller, Hybrid Electric Vehicles: Architecture and Motor Drives, in Proc. IEEE, vol. 95 ,NO. 4, pp. 719728, April 2007 [2] C. C. Chan, The state of the art of electric and hybrid vehicles, Proc. IEEE, vol. 95, NO. 4, pp. 704718, April 2007 [3] Z. Q. Zhu, and David Howe, Electrical Machines and Drives for Electric, Hybrid, and Fuel Cell Vehicles, Proc. IEEE, Vol. 95, No. 4, pp. 749765, April 2007 [4] M. Amrhein, and P. T. Krein, Dynamic Simulation for Analysis of Hybrid Electric Vehicle System and subsystem Interactions,Including Power Electronics IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 54, NO. 3, pp. 725836, MAY 2005 [5] G. Cole, Simple Electric Vehicle Simulation (SIMPLEV) v3.1: DOE Idaho National Eng. Lab. [6] K. B. Wipke, M. R. Cuddy, and S. D. Burch, ADVISOR 2.1: A user-friendly advanced powertrain simulation using a combined backward/forward approach, IEEE Trans. Veh. Technol., vol. 48, no. 6, pp. 17511761, Nov. 1999. [7] Argonne National Lab. PSAT (Powertrain System Analysis Toolkit) [Online]. Available: http://www.transportation.anl.gov/software/PSAT/index.html] [8] K. L. Butler, M. Ehsani, and P. Kamath, A MATLAB-based modeling and simulation package for electric and hybrid electric vehicle design, IEEE Trans. Veh. Technol., vol. 48, no. 6, pp. 17701778, Nov. 1999. [9] S. Onoda and A. Emadi, PSIM-based modeling of automotive power systems: Conventional, electric and hybrid electric vehicles, IEEE Trans. Veh. Technol., vol. 53, no. 2, pp. 390400, Mar. 2004. [10] U. S. Dept. Energy, Evaluation of 2004 Toyota Prius Hybrid Electric Drive SystemTech. Rep. ORNL/TM2006/423, May 2006. [Online] Available at: http://inspire.ornl.gov/OriginalDocument/f38948a5-d6a24f80-8cc9-34b77e3862a3 [11] Toyota Prius Awarded Title of "Best Engineered Vehicle for 2004", SAE International. [online]. Available at: http://www.sae.org/news/releases/prius2004.htm [12] MATLAB Simulink Software, Hybrid Electric Vehicle model Version 2008 and 2009 [13]Toyota Hybrid System course 071, section 1 [online]. Available at: http://www.autoshop101.com/forms/Hybrid01.pdf [14] MATLAB Simulink Software, Battery- Implement generic battery model Version 2008 and 2009 [15] U. S. Dept. Energy, Report on Toyota/Prius Motor Torque Capability, Torque Property, No-Load Back EMF, and Mechanical Losses Tech. Rep. ORNL/TM-2004/185, Feb 2004.[Online]. Available at: http://www.ornl.gov/~webworks/cppr/y2001/rpt/121119.pdf [16] K. Muta, M. Yamazaki and J.Tokieda, Development of New-Generation Hybrid System THS II Drastic improvement of Power Performance and Fuel Economy Tech. paper series, 2004-01-0064, Detroit, MI, March 8-11, 2004 [17] M. Ehsani, Y. Gao, A. Emadi, Modern Electric, Hybrid Electric and Fuel cell vehicles, Fundamentals, Theory and Design, 2nd Ed. CRC Press, 2010.