IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 44, NO.

1, FEBRUARY 1997 19

Propulsion System Design of

Electric and Hybrid Vehicles
Mehrdad Ehsani, Fellow, IEEE, Khwaja M. Rahman, Student Member, IEEE, and Hamid A. Toliyat, Member, IEEE

Abstract— There is a growing interest in electric and hybrid- Environmental Protection Agency (EPA), conventional ICE
electric vehicles due to environmental concerns. Recent efforts vehicles currently contribute 40%–50% of ozone, 80%–90%
are directed toward developing an improved propulsion system
of carbon monoxide, and 50%–60% of air toxins found in
for electric and hybrid-electric vehicles applications. This paper
is aimed at developing the system design philosophies of electric urban areas [1]. Besides air pollution, the other main objection
and hybrid vehicle propulsion systems. The vehicles’ dynamics regarding ICE automobiles is their extremely low efficiency
are studied in an attempt to find an optimal torque-speed profile use of fossil fuel. Hence, the problem associated with ICE
for the electric propulsion system. This study reveals that the automobiles is threefold: environmental, economical, as well
vehicles’ operational constraints, such as initial acceleration and
grade, can be met with minimum power rating if the power train as political. These concerns have forced governments all
can be operated mostly in the constant power region. Several over the world to consider alternative vehicle concepts. The
examples are presented to demonstrate the importance of the California Air Resource Board (CARB) is among the few that
constant power operation. Operation of several candidate motors acted first through the declaration of the Clear Air Act of
in the constant power region are also examined. Their behaviors
are compared and conclusions are made. September 1990. This act requires that 52% of all vehicles sold
in that state be either low-emission vehicles (LEV’s)—48%,
Index Terms— Electric vehicle, hybrid electric vehicle, motor
ultralow-emission vehicles (ULEV’s)—2%, or zero-emission
drives, road vehicle electric propulsion, road vehicle propulsion.
vehicles (ZEV’s)—2%, by 1998 [2]. Similar measures are
being considered in other states and nations as well.
I. INTRODUCTION EV’s and hybrid-electric vehicles (HEV’s) offer the most
promising solutions to reduce vehicular emissions. EV’s con-
T HE CONCEPT of the electric vehicle (EV) was con-
ceived in the middle of the previous century. After the
introduction of the internal combustion engine (ICE), EV’s
stitute the only commonly known group of automobiles that
qualify as ZEV’s. These vehicles use an electric motor for
remained in existence side by side with the ICE for several propulsion and batteries as electrical-energy storage devices.
years. The energy density of gasoline is far more than what Although there have been significant advancements in motors,
the electrochemical battery could offer. Despite this fact, the power electronics, microelectronics, and microprocessor con-
EV continued to exist, especially in urban areas due to its trol of motor drives, the advancement in battery technology has
self-starting capability. However, soon after the introduction been relatively sluggish. Hence, the handicap of short range
of the electric starter for ICE’s early this century, despite associated with EV’s still remains. Given these technology
being energy-efficient and nonpolluting, the EV lost the battle limitations, the HEV seems to be the viable alternative to the
completely to the ICE due to its limited range and inferior ICE automobile at the present. HEV’s qualify as ULEV’s and
performance. Since then, the ICE has evolved, improved do not suffer from the range limitations imposed by EV’s.
in design, and received widespread acceptance and respect. These vehicles combine more than one energy source to propel
Although this essentially is the case, EV interest never perished the automobile. In heat engine/battery hybrid systems, the
completely and whenever there has been any crisis regarding mechanical power available from the heat engine is combined
the operation of ICE automobiles, we have seen a renewed with the electrical energy stored in a battery to propel the
interest in the EV. The early air quality concerns in the 1960’s vehicle. These systems also require an electric drivetrain to
and the energy crisis in the 1970’s have brought EV’s back convert electrical energy into mechanical energy, just like the
to the street again. However, the most recent environmental EV. Hybrid-electric systems can be broadly classified as series
awareness and energy concerns have imposed, for the first or parallel hybrid systems [3].
time since its introduction, a serious threat to the use of ICE In series hybrid systems, all the torque required to propel
automobiles. the vehicle is provided by an electric motor. On the other
The ICE automobile at the present is a major source of hand, in parallel hybrid systems the torque obtained from the
urban pollution. According to figures released by the U.S. heat engine is mechanically coupled to the torque produced
by an electric motor [3]. In the EV, the electric motor behaves
Manuscript received February 26, 1996; revised April 17, 1996. exactly in the same manner as in a series hybrid. Therefore,
The authors are with the Texas Applied Power Electronics Center, Depart- the torque and power requirements of the electric motor are
ment of Electrical Engineering, Texas A&M University, College Station, TX
77843-3128 USA. roughly equal for an EV and series hybrid, while they are
Publisher Item Identifier S 0278-0046(97)00069-5. lower for a parallel hybrid.
0278–0046/97$10.00  1997 IEEE
20 IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 44, NO. 1, FEBRUARY 1997

This paper presents the EV and HEV propulsion system 1) electric motor power rating;
design philosophies. The paper is organized as follows. Section 2) motor rated speed;
II describes the design constraints and the variables for EV and 3) motor maximum speed;
HEV systems. Design philosophies of EV and HEV propulsion 4) the extent of constant power speed range beyond the
systems are presented in Sections III and IV, respectively. Sec- rated speed;
tion V examines several of the most commonly used motors 5) gear ratio between motor shaft and the wheel shaft
for EV and HEV system design. Section VI presents some test (transmission).
data from the Texas A&M University, College Station, hybrid Designing an optimal torque-speed profile for the ICE is
vehicle. Summary and conclusions are presented in Section beyond the scope of this paper. However, assuming a typical
VII. ICE torque-speed profile, we will specify the required ICE
power by our design procedure. Therefore, the design variables
II. SPECIFICATIONS OF EV AND for the mechanical propulsion system are:
HEV PROPULSION SYSTEM DESIGN 1) ICE size;
2) gear ratio between ICE and the wheel shaft.
A. System Design Constraints As mentioned earlier, the main design objective is to find
the minimum drive weight, volume, and cost that will meet the
Vehicle operation consists of three main segments. These
design constraints with minimum power. EV system design is
are: 1) the initial acceleration; 2) cruising at vehicle rated
addressed first. The HEV system design is then presented as
speed; and 3) cruising at the maximum speed. These three
a modification of the EV system design.
operations provide the basic design constraints for the EV
and HEV drivetrain. A drivetrain capable of meeting these
constraints will function adequately in the other operational C. Road Load Characteristics
regimes. Refinements to these basic design constraints are The road load consists of rolling resistance ,
necessary for an actual commercial product, but those are aerodynamic drag , and climbing resistance [4]:
beyond the scope of this paper. The objective here is to meet
these constraints with minimum power. The variables defining (1)
the above design constraints are:
1) vehicle rated velocity, ; The rolling resistance is caused by the tire deformation
2) specified time to attain this velocity, ; on the road:
3) vehicle maximum velocity, ;
4) vehicle mass and other physical dimensions. (2)
1) Initial Acceleration: The initial acceleration force takes
where is the tire rolling resistance coefficient. It increases
the vehicle from standstill to its rated velocity, , in some
with vehicle velocity and also during vehicle turning ma-
specified time, seconds. This force is supplied entirely by
neuvers. Vehicle mass is represented by , and is the
the electric power train in an EV or series HEV. In a parallel
gravitational acceleration constant.
HEV, the acceleration force is supplied by the electric power
Aerodynamic drag, , is the viscous resistance of air acting
train in combination with the ICE power train.
upon the vehicle:
2) Cruising at Rated Vehicle Speed : The electric mo-
tor provides the necessary propulsion force at rated vehicle
(3)
speed in the EV and series HEV. On the other hand, the ICE of
the parallel HEV should be capable of delivering enough force,
where is the air density, is the aerodynamic drag
without any help from the electric power train, to overcome
coefficient, is the vehicle frontal area, is the vehicle speed,
road load and cruise at the rated vehicle speed on a grade of at
and is the head-wind velocity.
least 3%. In addition, there should be a margin of about 10%
The climbing resistance ( with positive operational sign)
power to charge the batterypack.
and the downgrade force ( with negative operational sign)
3) Cruising at Maximum Vehicle Speed: The maximum
is given by
cruising force is provided by the electric motor in the EV
and series HEV. In the parallel HEV, the electric motor and (4)
ICE should work in combination to provide the required force
to sustain the vehicle at its maximum velocity. where is the grade angle.
A typical road load characteristic as a function of vehicle
B. System Design Variables speed is shown in Fig. 1. The following assumptions are made
The main component of the EV is its electrical power train. in the plot:
However, in the HEV the propulsion system is a combination 1) velocity independent rolling resistance;
of the electric motor and ICE. The electric propulsion design 2) zero head-wind velocity;
variables are: 3) level ground.
EHSANI et al.: DESIGN OF ELECTRIC AND HYBRID VEHICLES 21

Fig. 1. Typical road load characteristics as a function of vehicle speed. Fig. 2. Torque-speed diagram of an electrical motor in terms of tractive force
and vehicular speed with gear size as the parameter.

These assumptions will be used in the analysis presented

in the following sections, unless otherwise specified. These is the profile of this envelope that is important in the motor
assumptions do not change the general trend of the solution drive selection and design.
and can be easily relaxed. In order to free up the motor speed from the vehicle speed,
The motive force available from the propulsion system is for design optimization, gearing between the motor shaft and
partially consumed in overcoming the road load, . The net the drive shaft is required. In our design, we will make the
force, , accelerates the vehicle (or decelerates when following assumptions.
exceeds ). The acceleration is given by 1) Single gear ratio transmission operation—power elec-
tronic control allows instantaneous matching of the
(5)
available motor torque with the required vehicle torque,
at any speed; therefore, multiple gearing in order to
where is the rotational inertia coefficient to compensate for
match the motor torque-speed to the vehicle torque-
the apparent increase in the vehicle’s mass due to the onboard
speed is no longer a necessity;
rotating mass.
2) Ideal loss free gear—without loss of generality, the gear
losses can be incorporated at the end of analysis.
III. EV SYSTEM DESIGN
The gear ratio and size will depend on the maximum motor
The main component of the EV drivetrain is its electric speed, maximum vehicle speed, and the wheel radius. Higher
motor. The electric motor in its normal mode of operation can maximum motor speed, relative to vehicle speed, means a
provide constant-rated torque up to its base or rated speed. higher gear ratio and a larger gear size. The selection criterion
At this speed, the motor reaches its rated power limit. The for the maximum motor speed will be further discussed later.
operation beyond the base speed up to the maximum speed The torque-speed diagram of a typical motor is drawn in
is limited to this constant power region. The range of the Fig. 2, but in terms of tractive force and vehicular speed for
constant power operation depends primarily on the particular different gear ratios. Notice the electric motor base speed and
motor type and its control strategy. However, some electric maximum speed, in terms of the vehicle speed, depend on the
motors digress from the constant power operation, beyond gear ratio. A design methodology based on the three regions
certain speed, and enter the natural mode before reaching of operation will now be presented.
the maximum speed. The maximum available torque in the
natural mode of operation decreases inversely with the square
of the speed. This range of operation is neglected in the A. Initial Acceleration
analysis presented in this section, unless otherwise specified. The force-velocity profile of a typical motor is redrawn in
It is assumed that the electric motor operates in the constant Fig. 3. In this figure, is the electric motor rated speed,
power region beyond the base speed and up to the maximum is the vehicle rated speed, and is the vehicle maximum
speed. Nevertheless, for some extremely high-speed motors speed. The motor maximum speed must correspond to this
the natural mode of operation is an appreciable part of its , after the gear ratio transformation. The figure also shows
total torque-speed profile. Inclusion of this natural mode for (the dashed curve) the force-velocity profile of the motor in the
such motors may result in a reduction of the total power natural mode. This mode of operation, however, is neglected
requirement. Of course, power electronic controls allow the unless otherwise specified.
motor to operate at any point in the torque-speed plane, below The range of operation for initial acceleration is – . For
the envelope defined by the mentioned limits. However, it now, we will focus our attention only on this interval. For
22 IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 44, NO. 1, FEBRUARY 1997

Fig. 3. Typical torque-speed profile of electric motor in terms of tractive Fig. 4. Acceleration power requirement as a function of motor rated speed.
force and vehicular speed. Continuous curve—resistanceless case, dashed curve—in the presence of road
load.

maximum acceleration, the motor operates in constant rated

force (torque), , up to the motor rated The left-hand side integral is broken into two parts: the –
speed, , and in constant power, , at speeds constant force operation and the – constant power
beyond the base speed, up to the vehicle rated speed, . operation:
Here, is the motor rated power. We assume .
The wisdom of this assumption will become clear shortly. The (7)
differential equation describing the performance of the system
is given by (5) and is repeated here for convenience:
Now solving for , we get
(8)
is the motive force available from the propulsion system
and is the running resistance (road load). The boundary For minimum motor power, differentiating with respect
conditions are: to and setting it to zero gives
at , vehicle velocity . (9)
at , vehicle velocity .
This establishes a theoretical limit for minimum motor
To gain insight, we will solve (5) under the most simplifying power. For , the electric motor operates entirely
assumptions. in the constant power region. Therefore, if the motor is
1) The vehicle is on a level ground. performing – in seconds in constant power alone, the
2) The rolling resistance is zero. power requirement is minimum. On the other hand, if the
3) Aerodynamic drag is zero. motor operates in the constant torque (force) region during
These assumptions will be relaxed later for a more realistic the entire to period, we will have . In this
solution. The above assumptions will result in a closed-form case, (8) shows that the power requirement is twice that of
solution for the motor rated power . The insight gained constant power operation. The solid line curve of Fig. 4 shows
from the closed-form solution is also valid for the more an example of the motor power requirements between these
practical design involving running resistances. two extremes. Of course, entire operation in constant power
With these simplifying assumptions, the governing differ- regime is not practically realizable. However, this theoretical
ential equation reduces to discussion demonstrates that longer constant power range of
operation will lower the motor power.
(assuming Having discussed the simplified resistanceless case, we now
This differential equation is solved with the previous bound- solve the more realistic case involving the running resistance.
ary conditions and the force-speed profile of Fig. 3. The The vehicle differential (5) can be solved under the same
differential equation is integrated within the acceleration in- boundary conditions as before with the presence of the running
terval of to in to s, in order to get a closed-form resistance . In this case, a closed-form solution is feasible.
solution for the rated power : However, the result is a transcendental equation involving
rated motor power , rated motor velocity , rated vehicle
(6) velocity , acceleration time , and all the other system
constants, e.g., vehicle mass , rolling resistance coefficient ,
EHSANI et al.: DESIGN OF ELECTRIC AND HYBRID VEHICLES 23

aerodynamic drag coefficient , etc. The resulting equation TABLE I

can be solved numerically for for a specific motor rated EV POWER REQUIREMENT AS A FUNCTION OF CONSTANT POWER RANGE
velocity , using any standard root-seeking method such as
the secant method [5].
Let us assume that it is desired to obtain for the
following case:
• 0–26.82 m/s (0–60 mi/h) in 10 s;
the entire constant power range for initial acceleration of the
• vehicle mass of 1450 kg;
vehicle. The operation beyond that would be in the natural
• rolling resistance coefficient of 0.013;
mode. This would allow a longer constant power operation
• aerodynamic drag coefficient of 0.29;
in the initial acceleration. Consequently, the motor power
• wheel radius of 0.2794 m (11 in);
requirement will be lower. This scheme will work provided
• level ground;
the motor has adequate torque in natural mode to meet the
• zero head-wind velocity.
constraints at the maximum vehicle speed. Otherwise, some
A plot of the resulting motor rated power versus motor rated part of the constant power operation has to be used for the
speed, in terms of vehicle speed, is shown in Fig. 4 (the dashed vehicle operation beyond the rated vehicle speed.
curve). Natural mode of motor operation is not the preferred mode
Examination of Fig. 4 (the dashed curve) results in the beyond the rated vehicle speed. Unfortunately, no control
following conclusions: algorithm presently exists to operate some high-speed motors
1) rated power versus curve shows the same general entirely in constant power beyond their base speed. However,
trend of the resistanceless case; the natural mode, if included, can lower the overall power
2) rated motor power requirement is minimum for contin- requirement. The speed at which the electric motor can enter
uous constant power operation ; the natural mode and still meet the power requirement at
3) rated motor power is roughly twice that of continuous maximum vehicle speed is obtained from
constant power operation for constant force (torque)
operation ; (11)
4) rated motor power remains close to its minimum up to
about 20 mi/h of rated motor speed and then grows Note that the initial acceleration power is also a function of
rapidly. (extended constant power range). Hence, and have
to be solved iteratively. Also, the gear ratio between the drive
B. Cruising at Rated Vehicle Velocity shaft and the motor shaft is to be determined by matching
with the motor speed at which it enters the natural mode. More
A power train capable of accelerating the vehicle to the
discussion about the natural mode of operation appears in
rated velocity, , will always have sufficient cruising power
Section VI. The rest of the analysis is done assuming constant
at this speed. Hence, the constraint of cruising at rated vehicle
power operation beyond the base speed up to the maximum
speed is automatically met for the case of the EV. Of course,
speed.
cruising range is another issue related to the battery design
The importance of extending the constant power speed
which is outside the scope of this paper. However, minimizing
range can be better understood by comparing the required
the power of the drive will help the battery size.
motor power for different constant power speed ranges (as
a multiple of its base speed). Table I shows an example of
C. Cruising at Maximum Vehicle Velocity
power requirement for several constant power ranges for the
The power requirement to cruise at maximum vehicle speed following case:
can be obtained as 1) maximum motor speed is 10 000 r/m;
2) maximum vehicle speed is 44.7 m/s (100 mi/h);
(10)
3) other system variables and constants are the same as the
Since aerodynamic drag dominates at high speeds, this previous example.
power requirement increases with the cube of maximum Here, the required gear ratio to match the maximum motor
vehicle velocity. If this vehicle power requirement is greater speed to the maximum vehicle speed for a wheel radius of
than the motor power calculated previously , 0.2794 m (11 in) is 1:6.55. The results of Table I suggest an
then will define the motor power rating. However, extended range of 4–6 times the base motor speed in order to
in general, will dominate , since modern vehicles significantly lower the motor power requirement.
are required to exhibit a high-acceleration performance. As Finally, we examine the effect of maximum motor speed
mentioned before, some extremely high-speed motors usually and the extended constant power range on the overall system
have three distinct modes of operation. The initial constant performance. In the context of EV/HEV design, we classify
torque operation, followed by a range of constant power motors with maximum speeds of less than 6000 r/m as
operation, then to the maximum speed in natural mode (see low-speed motors, those with speeds of 6000–10 000 r/m as
Fig. 3). For such a motor it may be advantageous to use medium-speed motors, and those with speeds of 10 000 r/m
24 IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 44, NO. 1, FEBRUARY 1997

Fig. 5. Rated motor shaft torque as a function of maximum motor speed.

Fig. 7. Family of ICE force-velocity profile with ICE size as the parameter.

range. It can be seen from the results of Table I that, after

a certain point, there is not any appreciable power reduction
with extended constant power range. Any further extension
of constant power range beyond this point will only adversely
affect the gearing and drive shaft appreciably, without reducing
the power requirement. This will set the upper limit of the
extended range of the constant power operation.
Overall, the EV drive system design philosophy can be
summarized as follows.
1) Power requirement for acceleration decreases as the
range of constant power operation increases, i.e., more
Fig. 6. Drive shaft torque as a function of extended constant power range. specifically, as the ratio of the vehicle rated speed to
motor rated speed increases.
2) The gear ratio between the electric motor and the drive
and beyond as high-speed motors. The power requirement is shaft is determined by the motor and vehicle maximum
not a function of the motor maximum speed. Motor maximum speeds.
speed only affects the gear size. However, maximum speed 3) Power requirement for cruising at the maximum vehicle
has a pronounced effect on the rated torque of the motor. An speed is obtained directly from the road resistance at
example of this is illustrated in the surface plot of Fig. 5. Low- maximum speed. In general, this power requirement will
speed motors with extended constant power speed range have a be lower than the initial acceleration power requirement.
much higher rated shaft torque. Consequently, they need more 4) High-speed motors would be more favorable for EV
iron to support this higher flux and torque. Furthermore, higher application, in general.
torque is associated with higher motor and power electronics
currents. This will also impact the power converter silicon IV. HEV SYSTEM DESIGN
size and conduction losses. Extended speed range, however,
is necessary for initial acceleration as well as for cruising In the series hybrid vehicle the electrical system design
intervals of operation. Therefore, the rated motor shaft torque is identical to that of the EV. The ICE size is specified for
can only be reduced through picking a high-speed motor. keeping the batteries charged. The parallel HEV system design
This would, however, affect the gear ratio. A good design philosophy, however, requires an extension of the EV system
is the result of a tradeoff between maximum motor speed and design philosophy. The gear ratio (single gear) between the
the gear size. However, this tends to be more in favor of ICE and the wheel shaft can be obtained by matching the
selecting a medium- or high-speed motor. For an extremely maximum speed of the ICE to the maximum speed of the
high-speed motor, a sophisticated gear arrangement might be wheel shaft. Following the same procedure as the EV system
necessary for speed reduction. Planetary gear arrangement [6] design, the variables for HEV system design are calculated to
could be the choice that is compact but allows high-speed fulfill the design constraints, as shown below.
reduction. Extended constant power range, on the other hand,
will increase drive shaft torque and stress on the gear, as can A. Cruising at Rated Vehicle Velocity
be seen in Fig. 6. Hence, another design trade-off is involved The ICE size is determined mainly by the vehicle cruising
between the gear stress and the extended constant power power requirement at its rated velocity. Fig. 7 shows an
EHSANI et al.: DESIGN OF ELECTRIC AND HYBRID VEHICLES 25

TABLE II
MOTOR DATA

TABLE III
RATED POWER AND CONVERTER VA REQUIREMENTS FOR
THE MOTORS OF TABLE II FOR A TYPICAL EV APPLICATION

Fig. 8. Acceleration power requirement as a function of vehicle speed at

which ICE is added.

example of the road load characteristics on a 3% grade, with

other constants remaining the same as before. The figure V. ELECTRIC PROPULSION SYSTEMS
also shows a series of force-velocity curves of the ICE FOR EV AND HEV DESIGN

(the throttle wide open) with the piston displacement as the An electric propulsion system is comprised of three main
variable. The correct ICE size can be determined from the elements: power electronic converter, motor, and its controller.
intersection of the road load curve with the ICE force-velocity This section is devoted to examining several of the most
profile at rated velocity, plus allowing a 10% margin for the commonly used motors and their control for EV and HEV
batterypack recharging. The exact amount of margin needed is propulsion. The importance of extended speed range under
the subject of a more complicated analysis involving vehicle motor constant power operation in an EV and HEV was
driving cycles, battery size, charge/discharge characteristics, established in the previous sections. This mode of operation
etc. is referred to as field weakening, from its origins in dc motor
drives. A detailed study of several commonly used motors for
B. Initial Acceleration EV and HEV propulsion application is presented in [7]. In this
section, we present a design example of several motors for the
The rated power to be delivered by the electric motor is constant power operation. This example will help clarify the
reduced in the case of the parallel HEV due to the mechanical capabilities of these motors for vehicle applications.
power available from the ICE. An example of the effect of ICE EV Data:
torque blending on the rated power requirement of the electric
• vehicle rated speed of 26.82 m/s (60 mi/h);
motor during initial acceleration is shown in Fig. 8. The figure
• required acceleration of 26.82 m/s in 10 s;
shows four different extended speed range operations of the
• vehicle maximum speed of 44.7 m/s (100 mi/h);
electric motor. The abscissa is the vehicle speed at which the
• vehicle mass of 1450 kg;
ICE torque is added. In all of the four cases, the ICE with
• rolling resistance coefficient of 0.013;
its low starting torque contributes little up to about the vehicle
• aerodynamic drag coefficient of 0.29;
speed of 20 mi/h. Therefore, this low-speed and low-efficiency
• frontal area of 2.13 m ;
operation of the ICE may be avoided without significantly
• wheel radius of 0.2794 m (11 in);
increasing the electric motor power requirement. It is important
• level ground;
to note that an extended constant power operation of the
• zero head wind.
electric motor is still a necessity to keep the power requirement
low (Fig. 8). The motor data are shown in Table II. The motor data
chosen are for the commercially available samples of these
motors. Clearly, more specific motors can be designed for
C. Cruising at Maximum Velocity vehicle applications, but such data were not available for this
At maximum vehicle velocity the power requirement is paper. Based on the vehicle data, the power requirement to
. This power is supplied by a combination of the cruise at the maximum speed is 41 kW. The motor power for
ICE and the electric motor. Once the ICE size is determined, acceleration and converter volt–ampere (VA) requirement for
the required electric motor power can be uniquely identified. each motor are shown in Table III.
As mentioned before, this power, in general, would be less The extended constant power range available from the
than the power requirement for the initial acceleration. induction motor clearly makes it highly favorable for vehicle
26 IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 44, NO. 1, FEBRUARY 1997

TABLE IV
COMPARISON OF GENERAL MOTORS EV, IMPACT, AND OUR DESIGNED EV

application. On the other hand, the limited constant power

range of the BLDC motor makes it appear inferior to the Fig. 9. Test data of motor power of Texas A&M University hybrid vehicle
(ELPH) for normal vehicle operation.
induction motor, despite its high power factor and high ef-
ficiency. The extremely high speed operation of the SRM and
its relatively longer constant power range helps it to overcome B. Texas A&M University Hybrid Vehicle
some of the difficulty associated with its lower power factor The Texas A&M University hybrid vehicle (ELPH) is a
operation. Furthermore, the SRM converter is simpler and parallel hybrid. The specifications of the ELPH prototype are
easier to control. as follows.
Performance:
VI. TEST DATA 1) 0–26.82 m/s (0–60 mi/h) in 15 s;
In this section, an EV and an HEV prototype are discussed. 2) top speed of 35.76 m/s (80 mi/h).
The actual design specifications of these vehicles are compared Dimensions:
with our theoretical design of these same vehicles, based on the 1) frontal area 2.13 m ;
ideas presented in this paper. The EV is the General Motors 2) drag coefficient of 0.29;
Corporation IMPACT car and the HEV is the Texas A&M 3) curb weight 1700 kg.
University ELPH car. ICE Specifications:
1) 500 cc Honda CX500 four-stroke engine;
A. General Motors EV IMPACT
2) peak power of 37 hp;
General Motors announced the first version of its EV, 3) maximum speed of 9500 r/m.
IMPACT, in January 1990. Over the years, there have been Electric Motor Specifications:
several modifications of the IMPACT. The following are the
1) DC series motor manufactured by Advanced DC Motor
most recent specifications of the IMPACT. We have included
Inc.;
only those features which are pertinent to this study.
2) rated power of 40 hp;
Performance:
3) maximum speed of 6000 r/m;
1) 0–26.82 m/s (0–60 mi/h) acceleration in 8.5 s; 4) maximum converter current limit of 400 A.
2) top speed of 35.76 m/s (80 mi/h).
Other Specifications:
Dimensions:
1) batterypack of 10 compact 12 V Electrosource Horizon
1) frontal area 2.2578 m ; batteries connected in series;
2) drag coefficient 0.19; 2) rolling resistance coefficient of the tire 0.013;
3) curb weight 1347.17 kg. 3) a specially-designed gear box, built at Texas A&M, to
Design Features: connect the ICE and the electric motor in parallel.
1) 102.16-kW three-phase induction motor; The ELPH prototype was not built based on the design
2) IGBT power inverter module—102 kW; guideline presented in this study, but on the available hardware
3) high-speed rated 205/50 R15 tires. at the time. The main objective was to provide us with some
Our propulsion system is to meet the same performance preliminary insights into the hybrid vehicle propulsion system.
specifications as that of the IMPACT. In light of the design Some performance data for this prototype are presented in
methodologies presented in Section III and the electric propul- Fig. 9. Here, motor power is plotted for a typical vehicle opera-
sion system performance analysis presented in Section V, we tion. Note that during acceleration the peak power delivered by
pick an induction motor with maximum speed of 14 000 r/m the motor is sometimes in excess of 50 kW. During cruising,
and the rated speed of 3500 r/m (extended constant power the necessary driving force is delivered by the ICE. Also
range of 1:4). A comparison of our design EV with that of the note that the propulsion system has no regeneration capability,
General Motors IMPACT is presented in Table IV. The cogent hence, motor power is zero while braking. A comparison of
result of this exercise is that our motor power rating for this the HEV designed based on the methodology of this paper and
vehicle is only 73 kW as compared to the 102-kW motor in the ELPH prototype is presented in Table V. Since the ELPH
the prototype. This demonstrates the importance of the design prototype has multigear transmission, a one-to-one comparison
approach presented in this paper. between our designed HEV (single gear transmission) and the
EHSANI et al.: DESIGN OF ELECTRIC AND HYBRID VEHICLES 27

TABLE V Mehrdad Ehsani (S’73–M’75–SM’84–F’96)

COMPARISON OF ELPH PROTOTYPE AND OUR DESIGNED HEV received the Ph.D. degree in electrical engineering
from the University of Wisconsin-Madison, in
1981.
Since 1981, he has been at Texas A&M
University, College Station, TX, where he is
currently a Professor of electrical engineering
and Director of Texas Applied Power Electronics
Center (TAPC). He is the author of more than
100 publications concerning pulsed-power supplies,
high-voltage engineering, power electronics, and
motor drives. He is the co-author of a book on converter circuits for
superconductive magnetic energy storage and a contributor to an IEEE guide
for self-commutated converters, as well as other monographs. He holds seven
U.S. patents. His current research work is in the areas of power electronics,
ELPH prototype is not possible. However, the rated power of motor drives, and hybrid vehicles and their control systems.
the designed HEV matches the actual peak power delivered Dr. Ehsani has been a member of the IEEE Power Electronics Society
AdCom, past Chairman of the PELS Educational Affairs Committee, past
by the ELPH prototype in the test run (Fig. 9). Chairman of the IEEE-IAS Industrial Power Converter Committee, and past
Detailed experimental design and evaluation of various Chairman of the IEEE Myron Zucker Student–Faculty Grant program. He was
motor drives for EV and HEV applications is presently under the General Chair of the 1990 IEEE Power Electronics Specialist Conference
and is an IEEE Industrial Electronics Society Distinguished Speaker. He was
way on the Stationary Test Bed Laboratory of the Texas the recipient of the Prize Paper Award in static power converters and motor
Applied Power Electronics Center, Texas A&M University. drives at the IEEE Industry Applications Society Annual Meetings in 1985,
1987, and 1992. In 1984, he was named the Outstanding Engineer of the
VII. CONCLUSIONS Year by the Brazos chapter of the Texas Society of Professional Engineers. In
1992, he was named the Halliburton Professor in the College of Engineering
A design methodology for EV and HEV propulsion systems at Texas A&M University. In 1994, he was also named the Dresser Industries
Professor at the same college. He is a Registered Professional Engineer in
is presented based on the vehicle dynamics. This methodology the State of Texas.
is aimed at finding the optimal torque-speed profile for the
electric power train. The design is to meet the operational con-
straints with minimum power requirement. The study reveals
that the extended constant power operation is important for
Khwaja M. Rahman (S’94) received the B.Sc.
both the initial acceleration and cruising intervals of operation. degree from Bangladesh Engineering University,
The more the motor can operate in constant power, the less Dhaka, Bangladesh, and the M.S. degree from Texas
the acceleration power requirement will be. A&M University, College Station, TX, both in elec-
trical engineering, in 1987 and 1992, respectively.
Several types of motors are studied in this context. It is He is currently working toward the Ph.D. degree at
concluded that the induction motor has clear advantages for Texas A&M University..
the EV and HEV at the present. A brushless dc motor must be From 1987 to 1990, he was with the Electrical
Engineering Department of Bangladesh Engineering
capable of high speeds to be competitive with the induction University as a Lecturer. His research interests
motor. The switched reluctance motor may be superior to both include adjustable speed drives, electric and hybrid-
electric propulsion systems, and microcomputer control of drives.
of these motors for vehicle application both in size and cost. Mr. Rahman is a member of Phi Kappa Phi.
However, more design and evaluation data is needed to verify
this possibility.
The design methodology of this paper was applied to an
actual EV and HEV to demonstrate its benefits. Clearly,
the detailed design of a vehicle propulsion system is more Hamid A. Toliyat (S’87–M’91) received the B.S.
degree from Sharif University of Technology,
complicated than in our examples. However, this methodology Tehran, Iran, in 1982, the M.S. degree from West
can serve as the foundation of the detailed design. Virginia University, Morgantown, WV, in 1986, and
the Ph.D. degree from the University of Wisconsin-
REFERENCES Madison, in 1991, all in electrical engineering.
Between 1982 and 1984, he worked for power
[1] A. Alison, “Searching for perfect fuel . . . in the clean air act,” in Proc. companies in Iran. In 1991, he joined the faculty of
Environ. Veh. Conf., Dearborn, MI, 1994, pp. 61–87. Ferdowsi University of Mashhad, Mashhad, Iran, as
[2] S. Barsony, “Infrastructure needs for EV and HEV,” in NIST Workshop an Assistant Professor of electrical engineering. In
Advanced Components for Electric Hybrid and Electric Veh., Gaithers- March 1994, he joined the Department of Electrical
burg, MD, 1993, pp. 14–23. Engineering, Texas A&M University, College Station, TX, in the same
[3] A. F. Burke, “Hybrid/electric vehicle design options and evaluations,” capacity. His main research interests and experience include variable speed
SAE Paper #920447, Feb. 1992. drives, analysis and design of electrical machines, fault diagnosis of electric
[4] Automotive Handbook. Germany: Robert Bosch Gmbh, 1986. machines, electric and hybrid vehicle traction systems, and power systems
[5] A. Ralston and P. Rabinowitz, A First Course in Numerical Analysis, and control. He is actively involved in presenting short courses in his
2nd ed. New York: McGraw-Hill, 1978. area of expertise to various corporations, including General Motors, Texas
[6] A. G. Erdman and G. N. Sandor, Mechanism Design: Analysis and Instruments, LeTourneau, etc.
Synthesis, Vol. I. Englewood Cliffs, NJ: Prentice-Hall, 1984. Dr. Toliyat is a member of the IEEE Industrial Applications, Industrial
[7] H. A. Toliyat, K. M. Rahman, and M. Ehsani, “Electric machines in Electronics, Power Electronics, and Power Engineering Societies. He also
electric and hybrid vehicle applications,” in Proc. 1995 Int. Conf. Power serves on several IEEE Committees and Subcommittees. He is a member
Electronics, Seoul, Korea. of Sigma Xi.