Investigation on Chassis Dynamometer with Capability to Test Regenerative Braking Function

International Journal of Power Electronics and Drive System (IJPEDS)
Vol. 6, No. 3, September 2015, pp. 657~664
ISSN: 2088-8694  657
Journal homepage: http://iaesjournal.com/online/index.php/IJPEDS
Investigation on Chassis Dynamometer with Capability to Test
Regenerative Braking Function
Nurul Azwa Othman*, Hamdan Daniyal
Faculty of Electrical & Electronics Engineering, Universiti Malaysia Pahang, Malaysia
Article Info ABSTRACT
Article history:
Received Apr 28, 2015
Revised Jul 29, 2015
Accepted Aug 20, 2015
An investigation-based approach to a bidirectional power flow method for
testing regenerative braking function on a chassis dynamometer is presented.
The requirements and specifications for capability to test regenerative
braking function of Electric Vehicle (EV) emulated by using a bidirectional
chassis dynamometer are discussed. The dynamometer emulates road load
conditions during testing, and regenerative braking is able to test their
function while the vehicle is in deceleration condition. Performances of
power requirement are illustrated and translated into sequence diagram. It is
shown that the proposed topology is particularly advantageous in generating
and regenerating power for energy consumption. The overview of
conventional chassis dynamometer and the proposed chassis dynamometer is
compared to investigate the parameter in the development of regenerative
braking test.
Keyword:
Chassis dynamometer
Electric vehicle
Energy consumption
Power performance
Regenerative braking
Copyright © 2015 Institute of Advanced Engineering and Science.
All rights reserved.
Corresponding Author:
Nurul Azwa Othman,
Faculty of Electrical & Electronics Engineering,
Universiti Malaysia Pahang,
26600 Pekan, Pahang, Malaysia.
Email: nurulazwa_othman@yahoo.com
1. INTRODUCTION
The increasing world population soars the demand for fossil fuels in electricity energy generation
[1], which subsequently influences the growth of Energy Efficient Vehicle (EEV). EEV is defined as vehicles
that meet a set of defined specifications in terms of carbon emission level (g/km) and fuel consumption
(l/100km) [2], which include electric vehicle (EV) and hybrid electric vehicle (HEV) technologies. EV and
HEV consists of a very different powertrain compared to that of a conventional automobile, hence vehicle
performance test should be applied not only to the engine, but also to the complete powertrains [3]. Vehicle
performance must be measured using chassis dynamometer, a device for determining the overall delivered
torque and power of a vehicle from its wheel’s propulsion. The performance produced by the whole
powertrain (the engine, motor and transmission) is collectively measured by the continuous readings of the
rotational speed (rpm) and torque on the wheel(s).
In recent years, the usage of chassis dynamometer for performance testing of EV has rapidly
increased as a consequence of the development of EVs. Typically, chassis dynamometer utilizes either eddy
current brake or DC generator brake as the power absorber. However, the actual use of eddy-current
dynamometer has been rather limited due to its accuracies that produce imperfect results [4]. DC generator
brake, on the other hand, converts the mechanical energy from the Vehicle under Test (VUT) to electrical
energy. The energy from the DC generator is then stored, utilized or dissipated as heat using a variable load.
The main advantage of DC generator brake over eddy current brake is its capability to emulate the changes in
the dynamic load, hence test the VUT’s dynamic performance. Thus, DC generator brake is more appropriate
for testing EEV.

 ISSN: 2088-8694
IJPEDS Vol. 6, No. 3, September 2015 : 657 – 664
658
The challenge in emulating road conditions on a dynamometer is caused by the differences in the
characteristics of road load and dynamometer’s load, which produce inaccuracies in the performance results.
To measure the performance of the VUT, the force and the inertia on the vehicle and the dynamometer drum
must be considered. Force is known as the tractive effort and is commonly measured on a drum-type
dynamometer by a load cell indirectly connected to the rollers [3]. Thus, speed and force are two important
parameters for power performance test of the VUT.
Typically, emulating road condition in a chassis dynamometer would enable the test on a specific
real road tracks, called drive cycles, be carried out [4-5]. Vehicle performance test conducted on conventional
chassis dynamometer focuses on power and torque developed by the VUT in a range of speed [3] but little
attentions are being paid to the VUT performance during deceleration [4]. This is because in traditional
internal combustion engine (ICE) vehicles, the deceleration is just a state where the vehicle is slowing down
with the vehicle’s momentum. However, the situation is different with EV and HEV as deceleration usually
involved regenerative braking [8-9].
In this paper, attempting to overcome the problem, the possibility of a new design of chassis
dynamometer that emulates the VUT’s deceleration inertia, and consequently assess the regenerative braking
function, is investigated. This investigation considers the use of chassis dynamometer for three functions; (1)
determining power performance, (2) emulating road condition, and (3) testing the regenerative braking.
Section II describes the chassis dynamometer, which includes operation and power performance. Section III
proposes a solution of a new design of chassis dynamometer for EV with regenerative braking testing
capability. Section IV discusses the classifications of chassis dynamometer and identifies further research to
improve performance.
2. PRINCIPLE OF THE CHASSIS DYNAMOMETER
2.1. Operation of Conventional Chassis Dynamometer
Performance of a VUT can be measured through simultaneous operation of load variation and
parameter measurement. Load variation refers to the process of manipulating dynamometer drum’s inertia to
emulate road load condition, while parameter measurement is about the power, torque and speed
measurements using various sensors; mostly installed at the load cell. Chassis dynamometer implicates an
absorbing power output from the test vehicle’s engine to allow different loads to be applied on drum for
various testing procedures [3].
A conventional dynamometer operates by absorbing the power and energy produces from the
vehicle and dissipates the absorbed power usually as heat. The absorbing power flow is shown in Figure 1,
where the flow originated from the VUT on a chassis dynamometer drum to the load. Conventional chassis
dynamometer is focused on measuring power delivered by vehicle that is dependent on road condition.
Several chassis dynamometer that could emulate the physical system configuration, inertia and friction
calibration techniques, and control software have been developed; however the testing is applicable for
Internal Combustion Engine (ICE) vehicles.
2.2. Power Requirement/Driving Resistance Simulation of Chassis Dynamometer Testing
This section investigates the limitation of a conventional chassis dynamometer. The direction of
force applied during chassis dynamometer testing is illustrated in Figure 1(a). The inertia from the vehicle
and the force on the chassis dynamometer assist in the emulation of actual road condition on chassis
dynamometer drum. As different vehicles have different mass, inertia simulation must be employed to
provide realistic loading during transient proceedings [7]. The force acting on the vehicle can be calculated
using equations (6) to (8), where is the weight of the vehicle to be calculated, with acceleration and
gravitational forces taken into account. The gravitational force is equal to the net force, expressed in Equation
(3), because the directions of the two forces are in parallel.
F = m a (1)
F = m g (2)
F = F (3)

IJPEDS ISSN: 2088-8694 
Investigation on Chassis Dynamometer with Capability to Test Regenerative Braking Function (Nurul AO)
659
Equation (4) expresses the friction force, , in which its direction is opposite to the applied force as
shown in Figure 1. Friction force is due to the traction of the vehicle, while applied force is due to the
propulsion of the vehicle.
F = μF (4)
Since is derived based on frictional force, then the values for vehicle force, , or applied
force, , can be defined using equations (5):
F = F − F (5)
Based on equations above, the total road-load of the vehicle is calculated using Equation (8), and
then road condition emulated by the wheels is then delivered to the chassis dynamometer. Torque is
generated to provide realistic load and consists of forces that are constants, , or dependent on
acceleration, , or vehicle velocity, . The total force applied on wheels is translated as in Equation (8). The
acceleration force can be expressed as = , and the constant force = . Setting, = + [7],
the total load can be expressed as:
F = F + F + F (6)
F = A + Bv + Cv (7)
F ( ) = A + Bv + Cv + M
dv
dt
(8)
The chassis dynamometer emulates the total road load to make vehicle experiences road condition, and
consequently, power performance of the vehicle can be measured. The instantaneous dynamometer power is
defined as in Equation (9):
P = τω = F. V (9)
where is the power produced by the vehicle and is the associated torque. The power required must be
considered to determine how much torque should be added when conducting the test [4]. From equations (10)
and (11), the parameters of chassis dynamometer, including drum radius, distance of load cell and moment of
inertia, can be determined from the structure, while angular frequency, and force, is obtained from
measurement method.
P = m ×
∂(V)
∂t
+ F × V (10)
P = F ( ) × V (11)
As a case study, the scope of force calculation is sufficient for conventional chassis dynamometer
test but is limited when testing the performance of EV. On the other hand, the bidirectional chassis
dynamometer is proposed as a new method to test the capability of regenerative braking function of EV.
3. PROPOSED BIDIRECTIONAL CHASSIS DYNAMOMETER
3.1. Chassis Dynamometer for Regenerative Braking
The working principle of bidirectional chassis dynamometer with the associated devices is
illustrated in Figure 1(b). It should be noted that the system is similar to the conventional chassis
dynamometer in Figure 1(a). The only difference between the two systems is conventional chassis
dynamometer utilize Power Absorption Unit (PAU) to dissipate the power from VUT, whereas the proposed
system use a new subsystem called Power Absorption and Delivery Unit (PADU). PADU consists of a DC
machine, a bidirectional DC-DC converter and a power sink/source. Unlike PAU which can only absorb
VUT’s energy, PADU can operates in two modes; Mode 1 – absorbing energy from VUT and Mode 2 –
delivering energy to VUT.

 ISSN: 2088-8694
660
Mode 1 is when VUT is in normal motoring condition. Once VUT starts accelerating, the rotating
wheels force the drum to turn. Due to higher starting torque of the DC machine which is directly coupled to
the shaft of the drum, the VUT need to provide additional power for relatively hard accelerations [8]. The
rotational force put DC machine in generating mode, where it convert the mechanical energy into electrical
energy. DC-DC converter transfers the electrical energy into another form of electrical energy, suitable for
storage in the power source/sink. In practical application, this power source/sink might be an energy storage
system like battery and ultracapacitor or, it might be a bidirectional grid-tied inverter.
Mode 2 is exclusive to HEV or EV as the VUT, operating in regenerative braking mode. For most
HEV and EV, regenerative braking happens when acceleration pedal is released. In this mode, HEV and EV
harvest the mechanical energy from the rotating wheels to charge the main battery pack, thus slowing down
the vehicle. In the proposed bidirectional chassis dynamometer, this Mode 2 demands an opposite energy
transfer as compared to Mode 1. In this condition, power source/sink transfers energy to DC machine through
the bidirectional DC-DC converter. Upon receiving the energy, DC machine starts operates as motor and
rotates the drum. This subsequently rotates the wheel of VUT, hence allowing the VUT to test its
regenerative braking operation.
The main purpose of PADU in the proposed system is to vary the system’s inertia to closely
resemble the real VUT’s inertia. By controlling the power transfer in magnitude, the system is able to control
the inertia of the drum, make it lighter or heavier to turn. With bidirectional capability, the system is able to
control the direction of the power transfer, make the drum turns even without the power from the VUT. This
variation of the drum’s inertia emulates the real VUT’s inertia during both motoring mode and regenerative
braking mode of VUT. Moreover, with precise control of the inertia’s variation, the system is capable to
emulate road condition and is capable of testing the regenerative braking of HEV and EV.
In order to simulate similar loads on the drum of the chassis dynamometer as on real road
conditions, the road load forces have to be determined by executing a coast-down testing [9]. The testing is
used to determine the target coefficients in Equation 8. These target coefficients (A, B and C) are unique for
each vehicle under certain operating condition [7], [10-11]. The target coefficients are needed in the
calculation of the total road force on the dynamometer, as expressed in the equation.
a) Conventional chassis dynamometer b) Bidirectional chassis dynamometer
Figure 1. Comparison of proposed and conventional chassis dynamometer
4. SIMULATION AND RESULT
4.1. Simulation parameter
The proposed bidirectional chassis dynamometer is simulated to prove its working concept, as
shown in Figure 2. The VUT is resembled by a DC motor as prime mover in the simulation, directly coupled
to the DC machine on the chassis dynamometer side. The absence of drum in the simulation simplifies the
speed conversion, where it is assumed that the speed of the PADU’s DC machine is same as the speed of
VUT’s wheel. The availability of half bridge bidirectional DC-DC converter is used to allow power flow in
both directions, whether in Mode 1 or Mode 2. In Mode 1, the energy from the DC machine (operates as
generator) is stored in the battery at the last station. This stored energy is reused for Mode 2 (regenerating
braking test), where the DC machine operates as motor. The main focus of the bidirectional chassis
Chassis Dyno
Vehicle Under Test
(VUT)
Driver’s
Aid Screen
DC
Machine DC/DC
Converter
Power
Source/Sink
Drum/
Roller
Chassis Dyno
Vehicle Under Test
(VUT) Driver’s
Aid Screen
LoadPower
Absorber
Unit (PAU)
Drum/
Roller

661
dynamometer is to investigate the testing capability of regenerative braking function. In the proposed
method, bidirectional DC-DC converter power flow is used to test regenerative braking performance of the
dynamometer.
In real-world application, the operation of the bidirectional chassis dynamometer starts when the
driver engaged the acceleration pedal to accelerate the VUT, according to the speed trace display on a
Driver’s Aid Screen. To simulate this, a predefined variation of voltage is supplied to the DC motor, as
shown in the left side of Figure 2. The predefined voltage shape is chosen to directly resemble the real-world
speed trace. As a result, the speed of the simulated VUT is identical to the speed trace on Driver’s Aid
Screen.
DC motor coupled with PADU’s DC machine yields a variable voltage supply to the half bridge
bidirectional DC-DC converter. Half bridge topology has been chosen because less complexity and simple
for bidirectional power transfer. This topology is using hysteresis current control method for controlling
power transfer of bidirectional chassis dynamometer where the inductor, IL as control signals of current
changes. The direction of current flows from PADU’s DC machine to the Battery is in Buck Mode while
current flows from battery to PADU’s DC machine are in Boost Mode. In Buck Mode the bidirectional DC-
DC converter transfers power from PADU’s DC machine to the Battery. Meanwhile power is transfer from
the Battery to the PADU’s DC machine when PADU’s DC machine demands an energy to turn as motor.
Summary of bidirectional chassis dynamometer mode with acceleration pedal variation is tabulated in
Table 1.
Figure 2. Half-Bridge bidirectional DC-DC converter topology on chassis dynamometer
Table 1. Classifications of mode operation for each section according to the acceleration pedal level
Acceleration
Pedal
Mode Mode 1 Mode 1 Mode 1 Mode 1 Mode 2
PADU Absorbing power from
VUT
Absorbing power
from VUT
Absorbing power
from VUT
Absorbing power
from VUT
Delivering power to
VUT
VUT Motoring Motoring Motoring Motoring Regenerative braking
DC Machine Generator Generator Generator Generator Motor
DC-DC
converter
Buck Buck Buck Buck Boost
Power
source/sink
Sink Sink Sink Sink Source
4.2. Results of Performance Test Sequence
Simulation performance of the bidirectional chassis dynamometer with PADU is carried out using
PSIM. The inductor, resistance and IGBT are not considered in this simulation. For the proposed half bridge
bidirectional DC-DC converter topology input voltage is consider as variable from PADU’s DC machine
while the Battery voltage is fixed to 30V. Figure 3 shows two performances test sequence is available, which
is Mode 1 and Mode 2. The slope of the speed and current (power) of VUT and PADU is the main test
Power Absorption & Delivery Unit (PADU) in Bidirectional Chassis dynamometerVehicle Under Testing
(VUT)
0

 ISSN: 2088-8694
662
parameter of the bidirectional chassis dynamometer. Mode 1 operation is labeled as A in Figure 3. Mode 1 in
region A indicates when the acceleration pedal is pressed and the brake pedal is disengaged. During this,
speed from DC motor prime mover is increased consequent from level acceleration pedal pressed. Regarding
the Speed performance in Figure 3, it should be noted that Ireference is take part as reference at the bidirectional
DC-DC converter controller for control current flows during Mode 1 and Mode 2 from the Speed trace.
Meanwhile, the Speed response during Mode 1 is simultaneous with Ireference. This simulation is observing the
power performance of bidirectional chassis dynamometer, however current performance is analyzed for
perceive the power flows. Mode 1 still occurs at A1 as delays before achieve to Mode 2.
As expected, IDCmachine shows the performance of current is opposite the results from Ireference. This
occurs when DC machine turn as generator resulting IDCmachine in negative value while VUT’s in motoring
mode. DC machine emulates VUT’s speed thus generating input voltage for bidirectional DC-DC converter.
IDCmachine flows through half bridge bidirectional DC-DC converter to be stored at the Battery right side of
PADU. The Battery is charging during interval A in Mode 1. On the other hand, Mode 2 performance is
shown at interval B (trace) where the performance is exactly as Mode 1; slope is opposite between IDCmachine
and Ireference. Mode 2 starts when the acceleration pedal is released. After Ireference signal is approaching zero to
negative values, Mode 2 is activate. During this situation, the Battery is discharging and power transfer is
change opposite Mode 1. According to Figure 3, IBattery plotted at positive value to transfer current from the
Battery to PADU’s DC machine. IDCmachine approaches to zero earlier than Ireference resulting from Speed
performance to test regenerative braking from VUT.
It was observed that, for bidirectional chassis dynamometer, the performance of VUT is exactly as
that of the chassis dynamometer. The main purpose of this investigation is to emulate road condition during
testing of EV with the ability to test regenerative braking of EV.
Figure 3. Simulation results during motoring and regenerative braking at controller with enlarged figure at
regenerative braking interval.
5. DISCUSSION
As presented above, the performance of Mode 1 and Mode 2 of bidirectional chassis dynamometer
power transfer is to vary the capability of regenerative braking of EV at VUT as well as the road load
condition. Hence, it can be concluded that the performance of the bidirectional DC-DC converter is important
to controlled DC machine emulates VUT performance with bi-direction power transfer. Also, it is observed
that drive cycle is change with Ireference to measured power performance is applicable and variation between
Ireference and IDCmachine in bidirectional DC-DC converter is satisfactory. On the other hand, the bidirectional
chassis dynamometer significant to be perform than the conventional chassis dynamometer for measuring
others than ICE due to the capability to testing regenerative braking which available in HEV and EV.
Based on the investigation, the study is classified as conventional chassis dynamometer and
bidirectional chassis dynamometer. The testing method for both types of chassis dynamometer type is listed

663
in Table 1; included in the table are the testing method parameters for accelerating test and braking test. Also,
the chassis dynamometer achieves the power performance during regenerative braking test as a result of
using the bidirectional dc-dc converter.
Table 2. Classification methods of conventional and bidirectional chassis dynamometer test
Conventional dynamometer Bidirectional chassis dynamometer
Testing Methods Constant vehicle (horizontal speed) test Constant vehicle (horizontal speed) test
Power/torque measurement Power/torque measurement
- Regenerative braking test
Acceleration Test Load or flywheels attached works to
emulate dynamometer speed similar to
road condition.
Current direction is change by bidirectional
dc-dc converter to emulate dynamometer
speed similar to road condition.
Inertia Inertia from drum and flywheel. Inertia from drum, flywheel and additional
force from positive-torque from DC
machine.
Despite the advantages offered by bidirectional chassis dynamometer, further research is needed to
improve performance and overall control and management of the chassis dynamometer. In addition, to
improve braking on dynamometer drum, precautionary measure is needed while conducting the regenerative
braking test.
6. CONCLUSION
This paper has investigated certain practical challenges that are associated with the regenerative
braking of a modern chassis dynamometer. In the study, the results obtained has showed that the performance
of the chassis dynamometer present a significant sequence due to the motoring and regenerative braking
mode covered on bidirectional dc-dc converter control. The power performance is influenced by the force
applied. The application presented in this paper proposes that significant benefits may be achieved by
utilizing chassis dynamometer with bidirectional dc-dc converter as controller for regenerative braking tested
capability. The model based on the bidirectional dc-dc converter is an effective and practical method for
testing the capability of chassis dynamometer in investigating the regenerative braking from EV present in
the conventional chassis dynamometer. Further research is required in attempting to increase the capability
and performance of bidirectional chassis dynamometer.
ACKNOWLEDGEMENTS
The financial support provided by The Malaysian Ministry of Higher Education (MOHE) and
Universiti Malaysia Pahang (UMP) is gratefully acknowledged. The authors also gratefully acknowledge the
support of RDU130143 grant.
REFERENCES
[1] T. Boon Kai, “Design of a Multiple Input Bidirectional Isolated DC-DC Converter for a Battery Ultracapacitor
Hybrid Energy Storage System of an Electric Vehicle”, Universiti Tenaga Nasional, 2014.
[2] Y.D.S. Mustapa Mohamed, “Sidang Media Dasar Automotif Nasional (NAP) 2014”, in National Automotive
Principle, 2014.
[3] C.D. Bennetts, “Chassis Dynamometer Software, Inertia Determination and Recalibration”, University of
Canterbury, 2002.
[4] J. Jirawattanasomkul and S. Koetniyom, “Design and Development of Road Load Conditions for Chassis
Dynamometer”, in International Conference on Production, Materials and Automobile Engineering
(ICPMAE’2012), 2000, pp. 153–157.
[5] C.C. Hua and S.J. Kao, “Design and Implementation of a Regenerative Braking System for Electric Bicycles based
on DSP”, in 2011 6th IEEE Conference on Industrial Electronics and Applications, 2011, pp. 703–707.
[6] L. Chu, L. Yao, J. Chen, L. Chao, J. Guo, Y. Zhang, and M. Liu, “Integrative Braking Control System for Electric
Vehicles”, in 2011 IEEE Vehicle Power and Propulsion Conference, 2011, pp. 1–5.
[7] C. Matthews, P.B. Dickinson, and A.T. Shenton, “Chassis Dynamometer Torque Control: A Robust Control
Methodology”, SAE Int., vol. 2, no. 1, pp. 263–270, Apr. 2009.
[8] K.J. Kelly, M. Mihalie, and M. Zolot, “Battery Usage and Thermal Performance of the Toyota Prius and Honda
Insight during Chassis Dynamometer Testing”, in The Seventeenth Annual Battery Conference on Applications and
Advances, 2002, pp. 247–252.

 ISSN: 2088-8694
664
[9] A. Mayyas, R. Prucka, P. Pisu, and I. Haque, “Chassis Dynamometer as a Development Platform for Vehicle
Hardware In-the-Loop ‘VHiL’”, SAE Int., vol. 6, no. 1, pp. 257–267, May 2013.
[10] F. Maodong, P. Lei, Q. Weigao, Q. Kongjian, and G. Junhua, “Study on resistance setting method for chassis
dynamometer test of Hybrid Electric Bus”, Vehicle Power and Propulsion Conference (VPPC), 2010 IEEE. pp. 1–
4, 2010.
[11] L. Jeongwoo and N. Douglas J., “Rotating Inertia Impact on Propulsion and Regenerative Braking for Electric
Motor Driven Vehicles”, in 2005 IEEE Vehicle Power and Propulsion Conference, 2005, pp. 308–314.
BIBLIOGRAPHIES OF AUTHORS
Nurul Azwa Othman was born in Pahang, Malaysia, in 1988. She received her Bachelor
Degree in Electrical Engineering (Power Systems) from Universiti Malaysia Pahang (UMP),
Malaysia, in 2012. She is currently pursuing her postgraduate degree at the Fakulti Kejuruteraan
Elektrik & Elektronik (FKEE), UMP, at the same time serving as Graduate Research Assistant
for Sustainable Energy & Power Electronics Research Cluster (SuPER), UMP. Her research
interests include power electronics in general, and specifically in automotive application.
Hamdan Daniyal (S'07–M’11) received the B.Eng. degree in electrical & electronics (2002)
from Universiti Teknologi Malaysia and the M.Eng. degree in mechatronics (2004) from Kolej
Universiti Teknologi Tun Hussein Onn. He received the Ph.D. degree in digital current control
for power electronics converter from The University of Western Australia in 2011.
In 2002, he worked as an electrical and electronics engineer at Smart Ind., a switched mode
power supply manufacturing company. Later in 2003, he joined Universiti Malaysia Pahang
(formerly known as KUKTEM) as a lecturer. After finished his Ph.D. study, he became one of
the key person in Sustainable Energy & Power Electronics Research (SuPER) group, UMP. His
research interest include switching strategy, nonlinear control and digital control in power
electronics applications such as renewable energy, electric vehicle, battery management, power
quality and active power filters.
Dr. Hamdan Daniyal is a member of IEEE Power Electronics Society (PELS) and IEEE
Industrial Electronics Society (IES).

Investigation on Chassis Dynamometer with Capability to Test Regenerative Braking Function

More Related Content

What's hot (19)

Similar to Investigation on Chassis Dynamometer with Capability to Test Regenerative Braking Function (20)

More from IJPEDS-IAES (20)

Recently uploaded (20)

Investigation on Chassis Dynamometer with Capability to Test Regenerative Braking Function