Modeling and Simulation of Plug-in Hybrid

Electric Powertrain System for Different Vehicular

Application

Rui Cheng Zuomin Dong

University of Victoria University of Victoria
Victoria, BC, Canada Victoria, BC, Canada
[email protected] [email protected]

Abstract—Plug-in hybrid electric vehicle (PHEV) presents the In recent years, applications of advanced hybrid
new trend of clean energy vehicle development due to their powertrain technology are extending from passenger cars
advantage of all electric driving, allowable external charging, toward high performance racing vehicles, such as Koenigsegg
lower emissions, and lower petroleum fuel consumption, Regera, Mercedes F1 W06 Hybrid, and Toyota TS040 Hybrid;
comparing to the maturing HEV. Applications of advanced plug-
as well as heavy duty trucks, marine vessels, and airplanes.
in hybrid powertrain system technology to different type of
vehicles lead to the need to identify the key characteristics of To make these extensions more effective, it is beneficial to
these different vehicular applications, and to develop more systematically examine the common and distinct design
systematic and effective powertrain design methods. objectives, functional features and design considerations of
these dramatically different hybrid vehicular applications.
In this work two different types of PHEV powertrain
architecture were investigated: a) pre-trans series-parallel multi- In this work, a vehicle model of the series-parallel plug-in
regime plug-in hybrid electric commercial vehicle (SPMR- hybrid electric powertrain and a parallel through road hybrid
PHEV), and b) post-trans parallel plug-in hybrid electric electric powertrain, which are the representatives of passenger
formula racing car. Model-based design (MBD) methods were vehicle and high performance racing car powertrains, were
used for powertrain system modelling of the two PHEV built for the investigation and comparison. MBD methods,
applications. With the powertrain system models developed which could greatly reduce the development times [2], were
using MATLAB Simulink and dSPACE Automotive Simulation used in the overall vehicle powertrain development process
Models (ASM), the powertrain configurations, control strategies (VDP) of the two vehicles, especially in the investigations to
and key features were investigated. The simulation results on
discover the differences and criteria of the two vehicular
standard driving cycles for each model using different rule-based
control strategies were compared. The differences and key
applications. This paper mainly focuses on the investigation
features of two types of vehicle in design and calibration were of powertrains architecture, the development of a quasi-static
presented and analyzed. model for passenger vehicle and a low-frequency model for
formula racing car, and the calibration and optimization for
Keywords—modeling; simulation; hybrid powertrain design; high efficiency control of the powertrain systems.
plug-in hybrid electric vehicle (PHEV); formula racing car

II. DESIGN OBJECTIVES

I. INTRODUCTION
The main purpose of this work is to investigate and
Environmental issues and energy concerns lead to discover the design differences and key features of two types
increased attention to fuel efficiency and emission reduction of vehicles, daily use passenger vehicle and high-performance
for regular and high performance vehicles in the past decades race vehicle. In developing a generic passenger vehicle, the
and particularly recent years. Vehicle electrification and main design objective is low life-cycle costs while satisfying
hybridization presents to be promising solutions to increase reasonable performance, drivability, safety, and utility
overall powertrain efficiency. With extensive research and requirements. Fuel efficiency, investment cost and longevity
development on hybrid powertrain technologies, the next are major concerns. For a performance car, on the other
generation PHEV emerges as a favorable technology solution hand, fast acceleration, deceleration and cornering are major
to greatly reduce petroleum consumption [1]. objectives. A good load transfer and handling capability is

978-1-4673-7637-2/15/$31.00 ©2015 IEEE

also needed for the high speed track. Sometimes, energy B. Parallel through Road Architecture
efficiency also plays a role in performance vehicle design for Fig.2 presents a parallel architecture that was implemented
superior endurance competition. into Formula Hybrid racing car for student team competition,
This study focused on the design and optimization of high UVic Formula Hybrid I. To achieve high power to weight
fuel efficiency for passenger vehicle, and high acceleration ratio and load transform, aluminum alloy tube frame,
performance for racing car. Powertrain architecture selection fiberglass body, and light engine and other components from
process, and vehicle powertrain performance modeling and light-weight sport motorcycles were chosen, including a 250cc
simulations using different measures, vehicle calibrations, and 37Hp KTM fuel engine couple with 5-speed transmission, 75-
optimizations are presented. 5 67Hp Z-Force electric machine, and a 96V 2.8 kWh Li-ion
battery pack.

III. POWERTRAIN ARCHITECTURES

A. SPMR-PHEV Architecture
The SPMR-PHEV powertrain architecture was selected
and implemented for a market-ready passenger car, UVic
EcoCAR2, as illustrated in Fig.1. The main design concerns
include the overall greenhouse gas emissions, fuel economy
and vehicle driveability. The powertrain consists of a 103kW
rear traction motor/generator, a 2.4L GM EcoTec LE9 engine,
a 105kW BAS motor acting as a supplement to the engine, and
a 16.2 kWh Li-ion battery pack serving as onboard energy
storage system (ESS) with plug-in charge capability .
This powertrain supports multiple regimes of operation,
including pure EV mode, as well as various series, parallel and
series-parallel hybrid operations, through front wheel drive
(FWD), rear wheel drive (RWD) and all wheel drive (AWD).
Engine start/stop function and creep torque were integrated to
Fig.2. Diagram of Parallel Architecture
achieve further fuel saving and better vehicle control for
safety. The multi-regimes architecture also provide a platform Different from a generic passenger car, this Formula
for our research on advanced hybrid controls to optimize the Hybrid I race car uses a rear-mount RWD, instead of AWD or
operation mode towards certain driving patterns to reduce FWD. The reason to choose RWD with all powertrain
energy consumptions, improve vehicle drivability and provide components mount in the back is to achieve better weight
various utilities. transfer during hard acceleration. Rear driving wheel will gain
more normal force that allows higher maximum propulsion
torque. Without complicated drive shaft joints at the front
wheels, the vehicle has a smaller turning radius. The even
weight distribution may also improve the vehicle handling
with more road grip during cornering.
The Formula Hybrid I car has a post-trans parallel through
road powertrain architecture in which the engine is coupled
with the electric machine after transmission to a different ratio
final drive. Different from a conventional passenger vehicle,
the major concerns of a race car is to reduce curb weight for
superior performance. Complex powertrain with extra weight
to gain powertrain efficiency and fuel economy falls to lower
priority. Normally, a parallel hybrid powertrain would gain
additional weight due to its added transmission, comparing
with the Trans-less series hybrid powertrain. However, in this
race car hybrid powertrain, a light small-ratio transmission is
coupled with a large electric motor to produce enough torque
during acceleration with minimum weight addition.
Meanwhile, the parallel hybrid powertrain avoids the inherent
Mechanical-to-Electrical and Electrical-to-Mechanical energy
conversion losses of the series or series-parallel hybrid
Fig.1. Diagram of SPMR-PHEV Architecture
architectures, enabling better fuel efficiency and range b) Driver Model
endurance. Both vehicle models used an identical forward-looking
driver model with a PID controller to activate acceleration and
IV. POWERTRAIN MODELING
braking commands to drive the vehicle plant model. This
The two different hybrid powertrain systems have been forward-looking driver model is capable of supporting optimal
developed and implemented using the model-based design power control and energy management with real-time traffic
method through MATLAB Simulink powertrain system condition and driving pattern considerations in our other
models with different design objectives and constraints. research.

A. Quasi-static Model for Passenger Vehicle c) Powertrain Model

A quasi-static vehicle model was built in MATLAB The proposed quasi-static SPMR-PHEV powertrain model
Simulink environment to investigate and evaluate the vehicle can be used for the characterization of basic vehicle
performance, fuel economy, emissions and tests high-level performance and dynamics. The component models of the
supervisory controller of the SPMR-PHEV powertrain. The hybrid powertrain including models of the E85 engine, belt-
quasi-static model [3-4] presented in Fig.3 reflects simplified alternator starter (BAS) motor/generator, belt system, energy
passenger vehicle multi-physics, which allows rapid storage system (ESS), rear traction motor, transmission, and
simulation tests and calibrations of advanced control vehicle longitudinal dynamics. The powertrain performance
algorisms and components sizing for vehicle powertrain model used backward calculation from wheel reaction back to
system design. For passenger hybrid vehicle design, the key each powertrain component. Component torque loss,
factors are fuel economy, powertrain efficiency and efficiency and operation map are incorporated using
drivability. The quasi-static model satisfied all design predefined look-up table. Detailed components data are
requirements and led to reduced development time. obtained from steady-state experimental testing to predict
vehicle fuel economy and powertrain efficiency.
Except for plant model and driver model, a high level
supervisory controller model was developed to interact with
the driver and plant models. Rule-based load-leveling energy
management strategy was used in this model to collect driver
pedal command and to generate torque request of each
components [5-6]. In this control scheme, the powertrain
system operates in the series-parallel mode to ensure the
engine running in its high efficiency area while meeting the
needed output power from the road loads. The BAS and rear
traction motor/generator operate according to the power
Fig.3. Quasi-static Vehicle Model in Simulink surplus or shortage from the road load variations. Once the
ESS reaches certain low level of the state of charge (SOC), the
a) Driving Schedule controller operates as normal series or parallel hybrid mode
Vehicle performance simulations of the SPMR-PHEV with the engine providing most of the torque to help the ESS
passenger vehicle model were carried out under consistent to re-gain its normal SOC.
conditions and time-dependent operational profile for fuel
consumption analysis, component sizing selection, and d) Model Validation
calibration of control methods. A standard daily driving cycle, To validate the vehicle physic model, comparisons of
US06 city cycle as shown in Fig.4, were selected in operation system component between real world testing data and
simulation. The vehicle target speed was directly obtained simulation were made. The experimental data was collected
from real world driving data from US EPA. from vehicle dynamometer tests in EV mode that runs on the
rear traction motor only. In the simulation, driver pedal data
were treated as control inputs to the vehicle plant model.
Vehicle supervisory controller remains the same to achieve
identical controller torque requests. Fig.5 illustrated the
measured EV mode rear traction motor torque and the torque
calculated from powertrain model simulation.

Fig.4. US06 City Driving Cycle

 vehicle model. The differences and critical criteria of Formula
Hybrid Vehicle’s powertrain model development and
simulation setup are discussed in the following sections.

a) Driving Schedule
Compare to the US06-City drive cycle used for passenger
vehicle fuel consumption prediction, the Formula Hybrid
Vehicle’s testing cycle presented more aggressive speed
changing with high power demand. The vehicle speed and
acceleration pattern was generated from competition data in
the past. A full driving cycle includes eleven laps as shown in
Fig.7. The dramatic varied speed keeps the vehicle
Fig.5. Traction Motor Torque Simulation Validation
continuously changing from accelerating to braking to test the
As shown above, the simulation results using quasi-static vehicle’s acceleration performance and stability.
vehicle plant model accurately predicted the motor torque
output with less than 5% of errors comparing to the
experiment results.

B. Low-frequency Model for Formula Hybrid Car

Selection on different types of powertrain system model is
mainly based on the type of vehicle application. For a race car,
such as the Formula Hybrid vehicle, the major concerns are
vehicle dynamics, acceleration, and braking performance. To
better capture and predict the dynamic responses of the
vehicle, a low-frequency vehicle powertrain dynamics model Fig.7. Formula Hybrid Testing Cycle
shown in Fig.6 was built using the dSPACE Automotive
Simulation Models (ASM) tool [7]. This model includes a b) Powertrain Model
detailed vehicle drivetrain with both longitudinal and lateral
The low-frequency powertrain dynamics model introduced
dynamics. With this low-frequency model, between 1 and 10
for the Formula Hybrid Vehicle was designed and modeled
Hz, it is possible to model and calibrate advanced engine start-
using dSPACE ASM modeling tools. High fidelity powertrain
stop control, transmission gear shifts control, and traction
component models were built, which includes a four stroke
control.
one cylinder engine, 5-speed sequential transmission with
controlled shifter pistons and friction clutch as well as traction
motor/generator and ESS. Components experimental testing
data were collected from sponsors, key features and physic
dynamics and kinematics like engine’s four stroke responses,
motor’s inertia and response time and transmission shifter’s
damping were modeled. In this way, the low-frequency high
fidelity powertrain model was able to simulate gear shifts,
engine power detune control, engine rev control and vehicle
traction control as well.
Besides vehicle high fidelity powertrain component
models, different control schemes for acceleration, autocross
and endurance tests were implemented. The launch control
maintains engine speed into a narrow area where the engine
could output maximum torque to the rear axle. Pre-shift
transmission control also greatly improves acceleration
performance, this control strategies allows shifter piston hold
in a pre-shifting status while any engine power dropping will
results in a rapid gear shift. To better serve the autocross test,
the traction control was also imbedded into the controller. This
Fig.6. Low-frequency Vehicle Plant Model in Simulink
traction control uses PID control to monitor wheel slip-ratio,
control output realized by engine retard ignition and
The low-frequency Formula Hybrid powertrain model decreasing throttle position as well as negative torque output
utilizes the same driver model in the quasi-static passenger from electric motor. The Rule-based load-leveling energy
management strategy was also used in this model to control .
powertrain system operation in either electric drive mode or u '(t ) = arg min ( J ( x, u , m f )) (3)
Pbat , Peng
hybrid mode to minimize total energy consumption during
endurance test. The electric traction motor/generator operates Pwhl = Pbat + Peng (4)
only in high power demand or start from a low speed zone.
The control scheme tuned motor engaging power and speed to Constrains,
maximize overall powertrain efficiency while maintaining a
high acceleration performance in the mean time.
SoCmin ≤ SoC (t ) ≤ SoCmax (1)
Tmot _ min ≤ Tmot (t ) ≤ Tmot _ max (2)
V. CALIBRATION AND OPTIMIZATION Teng _ min ≤ Teng (t ) ≤ Teng _ max (3)
A. Passenger car optimization Pbat _ min ≤ Pbat (t ) ≤ Pbat _ max (4)
For regular passenger vehicle controller design, the major
design objective is fuel economy. Within the allowable fuel Objective function,
consumption range, other objectives such as drivability, curb
weight and daily EV range also need to reach a sub-optimal Tf . .

value. This section explains the strategy optimization and J (t ) = ∫ (meng (t ) + m bat _ eq (t ))dt
0
design target calibration process for reducing fuel
consumption.
Where, Pbat is the ESS output power, Teng and Tmot are engine
a) Control Strategies and electric motor output torque. The objective function
Control strategies and algorithms used in hybrid vehicle minimizes the total fuel consumption from engine and electric
controller can affect extensively in petroleum fuel motor fuel equivalent at each time step.
consumption. To develop a proper strategy for vehicle
CHART I. FUEL CONSUMPTION IN MPG OF SEVEN ITERATION DRIVE CYCLES
supervisory controller requires control method optimizations.
In this work, two control strategies were taken into 50
comparison, rule-based load following strategy and equivalent
consumption minimization strategy (ECMS). 40

The rule-based control strategy designed for SPMR-PHEV 30

combines a local optimization method. In this strategy, engine Rule-based
operation points were maintained by BAS motor and the belt 20
ECMS
system. High engine efficiency was achieved by controlling
10
operation area as close as possible to the maximum efficiency
point. Engine fuel consumption data was collected from real 0
world engine bench tests. Meanwhile, the RTM meets the rest
of the power requirements at each loading conditions. When UDDS HWFET US06
the battery SOC drops to a low level, the vehicle will operate
in charge sustain mode as a normal hybrid electric vehicle
with the engine meeting most of the loads to obtain a constant Simulation result as shown in Chart I indicated the
SOC level. This rule-based strategy is suitable and flexible as robustness of the ECMS method. The total fuel consumption
it does not need pre-knowledge of a driving cycle. in the three drive cycles with ECMS method were lower than
with the rule-based load following strategy. A sub-global
Compare to the local minimum optimization, the ECMS optimal solution was found by searching minimum fuel
presents to be more robust and efficient in fuel consumption equivalent consumption in each time step. The optimal result
minimization. Because the fuel consumption is the only is consistently superior comparing to the rule-based strategy.
objective, ECMS is considered as a sub-global optimization
method, where two power sources will get combined and b) Charge Depleting (CD) Range Calibration
optimized in a minimum point at each time. The ECMS To further improve fuel economy, the optimization of CD
algorithms are shown as follows: range for PHEV with different engine and ESS control
variables is necessary. A parametric study by simulations
x = SoC (1) under ten iterations of UDDS drive cycle as shown in Table I.
which indicated how CD range affects total powertrain fuel
u = Pbat (2)
consumptions.
 TABLE I. CHARGE DEPLETING RANGE AND FUEL CONSUMPTION system continuously monitor vehicle slip ratio. If the
State of Charge to ESS regulation (kW) acceleration rate is not in optimal point, a controlled increment
Engine Power On (W) torque will be provided by electric motor to increase the slip
2.5 2 1.5
CD RANGE (km) 57.8 56.9 57.2 ratio until the acceleration reaches maximum at current state.
25000
FUEL CON(L/100km) 6.29 6.27 6.31 Further controlled slip ratio increment will results in a
40000
CD RANGE(km) 49.5 47.6 46.0 decrease of acceleration rate.
FUEL CON(L/100km) 5.93 5.87 5.82
CD RANGE(km) 44.0 44.7 44.6
55000
FUEL CON(L/100km) 5.77 5.78 5.78

Table I shows that by gradually increasing the engine

power on threshold value, the EV driving mode will operates
more often as long as RTM can sustain the load requirement.
As a result, the amount of CD range drops as a consequence
the fuel consumption can be reduced.

c) Engine Operation Controls Calibration

The SOC regulation controls the charging rate according to Fig.9. Traction Control System Model
the control error – the delta of current SOC to the desired
SOC. This control variable could change the engine operation Figure 10 shows the vehicle speed to time curves of the
point when the SOC reaches ESS charge sustaining (CS) level. simulation of launch acceleration from 0 to 100 km/h with
Formula Hybrid I powertrain system. The simulations were
done under 0.75u road condition, the blue curve is with
adaptive traction control and the red dot curve is without
traction control.

Fig.8. SOC Results in Seven Iterations of UDDS Drive Cycle Fig.10. Traction Control System Acceleration Test Simulation

As the parametric study shown in Figure 8, to acquire a b) Power to Weight Calibration

desire distance of CD range for daily uses and reduce fuel
consumption, engine operation control variables needed to be In general racing area, vehicle acceleration, deceleration,
well tuned and calibrated. cornering and vehicle controllability comprise about ninety-
eight percent of the performance. Vehicle gross weight affects
B. Performance car optimization the most of vehicle performance. Therefore, the weight
minimization for vehicle powertrain system became a major
a) Traction Control System part of the designer’s tasks [9].

The traction control technique is common used in auto The vehicle weight lightness will gain considerable
sport area. This control system helps to eliminate wheel advantages in auto sport acceleration event and endurance. In
spinning and loss of traction during vehicle launching [8]. this study, the power to weight calibrations simulated the
combination of two sets of motors and three sets of ESS
In this work, an adaptive traction control is modeled in scheme under two operation conditions. Simulation results of
Simulink environment as shown in Fig.9. This control scheme acceleration performance are shown in Table II and Table III.
is designed for acceleration event, and especially in vehicle
launching. During the launch process, the traction control
 The powertrain system models for both vehicles have been
fully developed and implemented in prototype vehicles, UVic
TABLE II. POWER TO WEIGHT RATIO STUDY (0-60MPH TIME) EcoCAR2 and UVic Formula Hybrid I with detailed
ESS powertrain component models and control system, and tested
Motor Power Capacity
2.4 3.4 4.4 using adequate driving cycles.
（hp） (kWh)
Weight (kg) 21.9 25.3 30.4 Design considerations on powertrain architecture, key
46
Acceleration
Time(sec)
5.6 5.2 4.8 hybrid powertrain components, and control strategies were
Acceleration discussed. Validation of powertrain system performance
67 5.7 5.3 4.8
Time(sec) model was made in simple operation condition. The different
design considerations and the results of the model simulations
TABLE III. POWER TO WEIGHT RATIO STUDY (75 METERS TIME)
were compared.
ESS
Motor Power Capacity The studies provide guidelines for further research in
2.4 3.4 4.4
（hp） (kWh) developing generic and application specific PHEV powertrain
Weight (kg) 21.9 25.3 30.4 system designs.
Acceleration
46 Time(sec)
5.2 5.2 5.2
Acceleration
67 Time(sec)
5.3 5.2 5.1 ACKNOWLEDGMENT
Assistances from UVic EcoCAR and Formula Hybrid
The electric motor used in this specific design is Zero sponsors and contributing team members and financial
motor Z-Force 75-5, capable of as much as 46 hp and 80 ft-lbs supports from NSERC, NRCan, US DOE, GM and others are
of torque. Parametric study was taken with Z-Force 75-7, gratefully acknowledged.
which provides 67 hp and 106 ft-lbs maximum power output.
With three sets of ESS varies from 2.4 kWh to 4.4 kWh, a
generic vehicle power to weight ratio simulation were done by
using acceleration event drive cycle. REFERENCES
[1]. J. Gonder and A. Simpson. “Measuring and Reporting Fuel Economy
The simulation results showed that the most ideal power to of Plug-in Hybrid Electric Vehicles”. NREL/CP-540-40377, Nov.
2006.
weight ratio for the acceleration event was realized by the
[2]. The MathWorks. Model Based Design for Embedded Systems, Sep.
combination of 46 hp Z-Force 75-5 and 3.4 kWh Li-ion ESS. 2008.
This combines both consistent acceleration time and keep the [3]. T. Markel, A. Brooker, T. Hendricks, V. Johnson, K. Kelly, B. Kramer,
investment as low as possible. An ideal longitudinal load M. O’Keefe, S. Sprik, K. Wipke. “ADVISOR: A Systems Analysis
transfer was achieved by balancing the front and rear wheel Tool for Advanced Vehicle Modeling,” Journal of Power Sources,
load and the rate of acceleration. This power to weight 110(2): 255-66, 2002.
calibration ensures the rear axle with enough torque to spin the [4]. G. Rizzoni, L. Guzzella, and B. M. Baumann, “Unified Modeling of
Hybrid Electric Vehicle Drivetrains,” IEEE-ASME Transactions on
wheel and approaches traction limited acceleration rate. As Mechatronics, 4(3): 246-257, 1999.
well, an even rear weight distribution, 46/54 front to rear, was
[5]. Harpreetsingh Banvait, Sohel Anwar, Yaobin Chen. “A Rule-Based
achieved to produce more forward traction from the rear Energy Management Strategy for Plug-in Hybrid Electric Vehicle
wheels. (PHEV)”, Proceedings of 2009 American Control Conference. 3938-
3943, 2009.
[6]. P. Tulpule, V. Marano, G. Rizzoni, “Effects of Different PHEV Control
Strategies on Vehicle Performance”, Proceedings of 2009 American
VI. SUMMARY Control Conference, 3950-3955, 2009.
In this work, model-based design method has been applied [7]. dSPACE ASM – Gasoline Engine Simulation Package, 2008; available
to develop the powertrain system models for two different from (accessed 2015): http://www.dspaceinc.com/ww/en/inc/home/
products/sw/automotive_simulation_models/asm_gasoline_engine_sim
types of PHEVs, market-ready passenger car and race sport ulation.cfm.
car. [8]. Xiaoxing Liu, Lianbing Li, Yoichi Hori, “Optimal Traction Control for
To address the different needs of design objectives, two EV utilizing Fast Torque Response of Electric Motor,” IEEE 2005, 0-
7803-9252-3/05.
different types of powertrain modeling methods were applied.
[9]. C. Smith, “Tune to Win”, Aero Publishers, Inc. 1978. ISBN 0-87938-
A quasi-static model was used to investigate and evaluate 071-3.
vehicle performance, fuel economy, emissions and tests high-
level supervisory controller of the passenger car with an
advanced SPMR-PHEV powertrain in MATLAB Simulink
environment. A low-frequency vehicle powertrain dynamics
model was used to study and evaluate the vehicle dynamics,
acceleration, and braking performance of the race car using
dSPACE ASM modeling tool on top of MATLAB Simulink
environment.