Acceleration-based Control Strategy and Design for Hybrid Electric Vehicle Auxiliary Energy Source 83

Acceleration-based Control Strategy andDesign for Hybrid Electric Vehicle Auxiliary

Energy Source

Aree Wangsupphaphol1 , Nik Rumzi Nik Idris2 , Awang Jusoh3 ,

Nik Din Muhamad4 , and Supanat Chamchuen5 , Non-members

ABSTRACT

The paper presents a new control strategy and de-sign for auxiliary energy source (AES) used in batteryhybrid electric vehicle (BHEV) based on the acceler-ation power. The control strategy takes actual speedand acceleration of the vehicle and system losses intoaccount for regulating the energy and power supplyto the propulsion load. The design of AES and its dy-namic control design are demonstrated. Cascade con-trol is availed in this work in order to control the ter-minal voltage and current of supercapacitors (SCs).The benefits of AES in which recapture of regener-ative braking energy are examined by the numericalsimulation and verified by a small scale experiment.The comparison of energy consumption and DC busvoltage regulation between pure battery and batterywith supercapacitors (BSCs) propulsion system de-clares the theoretical results and confirms the benefitsof the proposed method.

Keywords: Electric Vehicles, Control Strategy, DC-DC Power Converters, Supercapacitors, Batteries.

1. INTRODUCTION

BEV is the next era of vehicle subsequent the in-ternal combustion engine (ICE) vehicle or hybrid ve-hicles. Because of the environment problems causedby the ICE and the petroleum fuel absence drive theBEV, zero emission, reborn after disappearing fromthe commercial trade at the end of 19th century [1].ICE does not only poor in efficiency but also con-tributes the global warming by its heat generated,carbons and other hazardous gases. However, the en-ergy per weight of the ICE is lower than the specificenergy of batteries; comparing the same power out-put at the shaft of engine. Moreover, refueling timeis much faster than the battery charging period [2].

Manuscript received on September 30, 2014 ; revised onFebruary 1, 2015.Final manuscript received March 24, 2015.1,2,3,4 The authors are with Department of Electrical

Power Engineering, Faculty of Electrical Engineering, Uni-versiti Teknologi Malaysia, E-mail: awmr2@live.utm.my,nikrumzi@fke.utm.my, awang@fke.utm.my, nikd@fke.utm.my5 The author is with Department of Electrical Engineering,

Faculty of Engineering, King Mongkut’s University of Technol-ogy Thonburi, Thailand, E-mail: ssupanat45b@gmail.com

Fig.1: Configuration of BHEV power train.

These are significant matters among many designersto overwhelm the problems.

Lithium-ion based battery (LB) is one of an ap-propriate energy source for BEV. It has higher spe-cific energy and specific power than other batteries;nickel metal hydride, nickel cadmium and lead acidbattery. Normally, the particular energy can be upto 265 Wh/kg and the specific power is about 1000W/kg [3]. The amount of energy states the drivingrange while the peak power refers to the vehicle ac-celeration rate.

The acceleration of motor demands the high powerand rises the temperature of the battery. This phe-nomena enlarges the battery internal resistance, onthe other hand, it demands great ventilation. Other-wise, the battery can be damaged from overheating.Moreover, to design the battery capacity for peakload requires a large size of battery and absolutelyeffects to the whole vehicle cost and weight. The im-proper interfacing of battery with the high accelera-tion and deceleration causes the low utilization andeventually early failure [4]. Moreover, the batteryalso has a poor regenerative braking capture capabil-ity which reflects the low energy economy. One ex-ample is the existing LB using in BEV of this studycan absorb about 30 % of the peak power, so the im-plementation of AES is necessary for energy caption.

Supercapacitor energy storage is one of the bestAES assisted to the main energy source proved fromits safety concern [5]. The AES can be made of abank of supercapacitors or an automotive module if

84 ECTI TRANSACTIONS ON COMPUTER AND INFORMATION TECHNOLOGY VOL.9, NO.1 May 2015

Fig.2: Possible configurations of hybrid energy source.

compatible in size. Supercapacitor is an electrochem-ical double layer capacitor having very high specificarea of activated porous carbon for restoring signifi-cant amount of capacitance [6]. It has a specific en-ergy around 1-6 Wh/kg while having large amount ofspecific power about 1-14 kW/kg which is the mostcompromising solution for a hybrid energy sourcepresently.

There are several possible configurations of hy-bridizing battery and SCs. Figure 2(a) shows thedirect parallel relation between two energy sourcessupplying the propulsion unit. The terminal volt-age of SCs, vSCs, always follows the battery voltage,vbat; where the power flow is proportionally shareddepending on their internal resistances. This config-uration is easy to implement, but in reality, the SCsstored energy is lowly utilized because its state ofcharge (SOC) cannot be discharged greatly [4]. Fig-ure 2(b) shows the parallel connection of two energysources via a converter feeding to propulsion unit.This configuration maintains the bus voltage, vbus,and promotes the inverter efficiency. However, theSCs stored energy is utilized inefficiently because itsvoltage variation follows the battery voltage, and theconverter rated too large for supplying the maximumpower [7-8]. The reliability of this hybrid energysource mainly depends on the converter. To controlthe power of SCs, configuration in Fig. 2(c) is pro-posed by many researchers for connecting SCs to con-verter, parallel to the battery, [4], [9-16]. This schemeimproves the battery performance by decreasing thebattery peak power, reducing loss and rising the tem-perature. On the other hand, it requires a large sizeof the converter for providing the peak power. [17-

19] introduced configuration as shown in Fig. 2(d) toavoid using large size converter. This scheme sepa-rates the SCs into two banks, SCs 0 and SCs 1, andcontrols only one bank through a controllable con-verter. Thus, converter rates and losses are lowerthan the previous scheme, since its active componentsof the converter are half in size and the inductor is onethird smaller. However, the overall terminal of SCsvoltage is double the size of the vbus, and it requires adynamic balancing circuitry, which is expensive andcomplex. In term of the reliability, configurations inFig. 2(c) and (d) provide higher reliability than thatin Fig. 2(b) even though in the case of converter fail-ure, the vehicle is still available.

By modifying the power sources as shown in Fig.2(e), the power converter can be minimized [20]. Inthis case, the battery current supplies average powerto the load only when vSCs is higher than vbat, oth-erwise the battery power can flow through the diodeand discharges high power to the load directly. Themajor disadvantage of this scheme is the large varia-tion of vbus established by the SCs voltage. This be-havior concurrently generates high loss of the propul-sion inverter. However, this configuration improvesthe battery performance whenever it is not directlydischarged through the diode frequently, and it hasequal reliability as the two previous schemes. To over-come the problems of vbus variation, configurations inFig. 2(f)-(g) have been proposed by some authors [8],[21]. These schemes require a large size converter toprovide the dynamic power. Unavoidably, the cost,weight, and loss increase, although they are traded offwith the higher efficiency of the propulsion inverter.Besides, their power system reliability is lower than

Acceleration-based Control Strategy and Design for Hybrid Electric Vehicle Auxiliary Energy Source 85

those in Fig. 2(a) and Fig. 2 (c)-(e); where if oneof the converters malfunctions, the vehicle might beinoperative. To control the energy and power of eachsource completely, multi-converter is implemented,as shown in Fig. 2(h) [22-25]. This configurationpresents a steady vbus and protects the battery fromhigh pulsing power demand. Nevertheless, the demer-its are similarly found as in the schemes in Fig. 2(f)-(g). In electric vehicle application, the hybrid config-uration should be the most reliable, less complex, haslow weight, loss and cost. After much consideration,the topology in Fig. 2(c) had been selected to beevaluated in this study, particularly on its dynamiccascade control.

2. SYSTEM CONFIGURATION

2.1 Battery electric vehicle

BEV in this study is basically based on a city sa-loon vehicle modifying for the electric vehicle chal-lenge [8]. The vehicle has a curb weight ofMV = 1000kg with 2 front-wheel drive. The Differential Gear(DG) and 5-speed Gear (5-G) is simplified for thisstudy to be single gear ratio 1.4288, wheel diameter,rwh = 0.26 m, and frontal area Af= 2.098 m

2. Themaximum speed is limited at vV,max = 33.3 m/s, tak-ing acceleration from 0-100 km/h on the zero degreeground slope around 20 seconds (acceleration rate a=1.37 m/s2). The traction unit is comprised of a three-phase Brushless DC (BLDC) motor 52 HP, coupledto a three-phase inverter. The drive system has beenmade simpler by employing a DC motor (DCM) blockof SimPowerSystem in MATLAB Simulink, which hasthe specification as followed: 50 HP, 240 V, maximumangular speed, ωm,max= 183.2 rps, total motor inertiaJm = 0.2 kgm

2 and viscous friction coefficient, Bm=0.007032 Nms. The proposed power system is mainlycomposed of the battery 7.5 V/unit, 32 cells connec-tion in series with the equivalent internal resistance52 mΩ.

2.2 BHEV hybrid energy source

A hybrid power source is a combination of twoor more power sources in order to realize the opti-mum performance. One of the best solutions is ahybrid between LBs and SCs aforesaid [4]. Nonethe-less, there are a number of workable configurationsfor the hybridization. The selection for applicationis the most dependable, a few complicated, havinglow mass, low loss and expenditure. The non-isolatebidirectional DC-DC converter as shown in Fig. 3was implemented in this work.

2.3 Converter topology for BHEV

The determination of the converter in BHEV is toregulate the dynamic power forward and backwardbetween SCs and tractive load; in a normal case, itworks as a boost converter while feeding power and a

buck converter when recapturing power. The topol-ogy commonly used in an EV is a half bridge non-isolated bi-directional DC-DC converter, as shown inFig. 3. Among converter topologies, the half bridgeconverter has more benefits than Cuk and combinedSEPIC/Luo converters, which deals high efficiency,being most compact, having bottom cost and lessweight and easy to control. The converter requireshalf size of an inductor and other components com-pared to the Cuk and combined SEPIC/Luo convert-ers. Switching and conduction losses are lesser, sincethe lower number of switching components and in-ductor. However, half bridge converter needs largersize of an output capacitor than other topologies sothat keeping continuous output current [26].

2.4 Auxiliary energy source

The AES is made of a bank of SCs connecting tothe DC-DC converter. SCs can be a bank of an ele-mentary cell or an automotive module. As regarded,the dynamic power demanded during the accelera-tion has been considered to be provided by SCs andthe tractive power is intended to provide by LBs, sothe power from SCs is zero in cruising phase. Dur-ing braking, SCs is charged to their rated voltage bythe dynamic load and LBs power. As renowned, thesize of SCs is based on the acceleration power stage;translating mass and the top speed of the BEV aretaken into account. In order to evaluate the actualenergy availability, the mean efficiencies provided bythe manufacturers; SCs, ηSCs = 0.95, converter ef-ficiency, ηconv = 0.97, propulsion chopper efficiency,ηchop =0.97, DC motor efficiency, ηDCM = 0.91,and mechanical efficiency, ηmech = 0.98, had beentaken into consideration. The kinetic energy and en-ergy stored in the SCs were balanced with a definitetransformation by losses concurring to the followingequation:

(MV +MSCs)v2V,max=ηtotCSCs

[V 2SCs,max−(12VSCs,max)

2]

(1)where ηtot = ηSCsηconvηchopηDCMηmech. SCs ter-

minal voltage is chosen to vary between rated termi-nal voltage, VSCs,max and half of its rated voltage inorder to utilize 75 % of energy content. This pro-duces the optimized capacity and weight of the aux-iliary energy source. In order to calculate the SCscapacitance, CSCs, in (1) can be transformed to:

CSCs =(MV +MSCs)v

2V,max

ηtot

(34V

2SCs,max

) (2)The weight of SCs is initially approximated to be

zero but subject to change in final calculation. Therated voltage of SCs is selected to achieve the maxi-mum voltage gain lower than 3 [9]. To find the capac-itance of a single cell, CSC,cell , The CSCs are taken

86 ECTI TRANSACTIONS ON COMPUTER AND INFORMATION TECHNOLOGY VOL.9, NO.1 May 2015

Fig.3: Power train configuration of the proposed hybrid energy source.

into account as in the following equation:

CSCs,cell =CSCs ·Ns

Np(3)

where Ns is the number of series connection and Npis the number of parallel connection. The Ns can berealized by dividing the SCs terminal voltage with thevoltage rating of CSC,cell available. Np is a selectionof an integer number to achieve CSC,cell appropri-ately to the manufacturer data. In this study, thecalculation of CSCs produced 49 Farad, where SCsterminal rated voltage of 200 V was chosen. Thiswas made by a parallel connection of series CSC,cell= 2000 Farad, 74 cells, 2.7 V/cell. However, the con-nection produced CSCs = 54 Farad with internal re-sistance 26 mΩ and 53 kg. These increased the ca-pacitance by 10% and weight by 5%. The capacitanceand weight were final calculated to ensure the vehicleperformance and found that it was maintained.

To evaluate the rated power of SCs in acceleration,PSCs,acc, maximum acceleration rate a=1.37 ms

(−2)was implemented to design the converter capacity asthe following equation:

PSCs,acc =MtotvV,maxa

ηconvηchopηDCMηmech= 57.1 kW (4)

where Mtot = MV +MSCs is the total vehicle mass.After that, the current was evaluated by using thefollowing equation:

ISCs,acc =PSCs,acc

VSCs,max/2= 571 A (5)

where ISCs,acc is accelerated current of SCs. Fromthe calculation, the current of each series distributes

by 286 A which is endorsed by the maximum peakcurrent of 1600 A, according to manufacturer specifi-cation.

3. MATHEMATICAL MODEL OF THEPHYSICAL SYSTEM

3.1 Model of propulsion load

Power demanded by the propulsion chopper, Pchop,can be developed mathematically as the followingequations:

Pchop =

{Pd/ηchopηDCM ; Pd > 0PdηchopηDCM ; Pd < 0

(6)

Pd = Ptl + Pdl (7)

where Pd is the driving power, Ptl is the tractive loadpower and Pdl is the dynamic load power at motorshaft. To obtain the tractive load power, vehicle dy-namic resistances have to be derived by using (8),where Ftr is the tractive resistance, Frr is the rollingresistance generated between tires and road surfaces,Far is aerodynamic resistance dragging the vehiclemotion to move into the air, Fgr is the grading re-sistance that downgrades the force of vehicle, Twh isthe load torque at wheel, Teq is the equivalent loadtorque transferred through the single gear ratio G.

Acceleration-based Control Strategy and Design for Hybrid Electric Vehicle Auxiliary Energy Source 87

Ftr = Frr + Far + FgrFrr = µrrMtotg cosαFar = 0.5ρAfCdv

2V,max

Fgr = Mtotg sinαTwh = Ftrrwh;Teq = Twhηmech/G

G = ωm,max/ωwh,maxPtl = Teqωm,max = 18.38 kW

(8)

Any coefficient values for calculation were as fol-lows: rolling resistance µrr =0.0048, Air density ρ= 1.25 kgm−3, aerodynamic drag Cd = 0.353, grav-ity acceleration rate g = 9.8 ms−2, and road gradingangle α = 0 degree. The dynamic power, Pdl, wasreacted by SCs, which diverse to the angular accel-eration rate of the propulsion load, according to thefollowing equations:

Pdl = Tdlωm.max = 48 kW

Tdl = Jeqdωmdt

= 262 N ·mJeq = Jm +Mtotηmech

(rwhG

)2= 34.2 kg ·m−2

(9)

where Tdl is the dynamic load torque, Jeq is the equiv-alent moment of inertia referred to the motor shaft.By solving the equation, Pdl = 48.31 kW, the SCssize for the acceleration power can be approximatedby:

PSCs,acc = Pdl/ηconvηchopηDCM = 56.1kW (10)

3.2 Model of auxiliary energy source and bat-tery system

The equivalent circuit of SCs can be modeled asthe RC series circuit as shown in Fig. 3. The equiva-lent electric circuit model is evaluated by the follow-ing equations:

iSCs(t) = −CSCsduSCs

dtVSCs(t) = uSCs − rSCsiSCs

uSCs(0) = VSCs,max

(11)

where uSCs is the internal voltage of SCs, rSCs is theinternal resistance of SCs and vSCs is the terminalvoltage of SCs and iSCs is SCs current .

In application, bus voltage, vbus, varies to the in-ternal battery voltage, vbat, and the battery current,ibat. The bus voltage is changed by SOC and thevoltage drop across the equivalent battery resistance,rbat, as in the following equations:

{vbat ∝ SOC

vbus = vbat − rbatibat(12)

4. CONTROL STRATEGY AND DESIGN

4.1 Control strategy

The main objective of the control strategy is toreduce pulse discharge of the battery and recapturemuch more regenerative energy. The energy conserva-tion perspective has been implemented in this study.In order to manage the energy, the auxiliary energysource terminal voltage is varied to the speed of ve-hicle or, on the other hand, the kinetic energy, as inthe following equation:

uSCs,ref ∼=

√V 2SCs,max −

Mtotv2refηtotCSCs

(13)

With the purpose of power control, SCs currentand voltage are controlled simultaneously. The fun-damental relation in (11) is availed to control the cur-rent by replacing with (13). This results in referencecurrent control relating to the speed and accelerationof the vehicle as shown in the following equation:

iSCs,ref=−CSCsduSCs,refdt ∼=MtotvV,ref

ηtotVSCs,max√√√√√1− Mtotv2V,refηtotCSCsv

2SCs,max

dvV,refdt

(14)Similarly, the strategy provided the ability to sup-

ply and capture power according to the accelerationor deceleration and the conversion of energies betweenthe two storages.

4.2 Converter design

Cascade control is utilized to control the drivingpower by regulating duty cycle current of the innerloop received from the outer voltage control loop. Tocontrol the outer and inner loop, the linearization ofconverters state equation has to be derived by us-ing state space averaging technique. The system ma-trices, A1,2, B1,2 and C1,2, can be achieved by usingKVL in the on and off stage of the boost converter[7] as shown in the following equations:

A1=

[−(rL+rSCs)

L0

0 − 1C(R+rC )

], B1=

[1L

0

], C1=[1 0]

; system matrices during on stage

A2=

[−(Rrc+R(rL+rSCs))

L(R+rC )−R

L(R+rC )

RC(R+rC )

−1C(R+rC )

], B2=

[1L

0

], C2=[1 0]

; system matrices during off stage(15)

where R is the equivalent resistance of maximumpower supply by SCs, 2.1 Ω, L is the converter in-ductance 0.1 mH, rL is the equivalent inductive resis-tance 20 mΩ, C is the converter capacitance 10 mFand rC is the equivalent capacitive resistance 25 mΩ.Equation (16) presents the system matrices of on andoff interval, which are used for linearizing the system

88 ECTI TRANSACTIONS ON COMPUTER AND INFORMATION TECHNOLOGY VOL.9, NO.1 May 2015

at the quiescent operating point. To obtain the trans-fer function of duty ratio, D, to the SCs current, thestate equation can be settled by the following equa-tion:

ISCs(s)

D(s)= Cs(sl −As)−1[(A1 −A2)X +

(B1 −B2)U ] + (C1 − C2)X (16)

where X is the equilibrium state vector and U isthe equilibrium input vector. Equation (16) can betransformed to the standard equation of MATLABtransient response analysis for finding the transferfunction as following:

ISCs(s)

D(s)= Cs(sl −As)−1Bs + Es (17)

where As, Bs, Cs and Es are the group of matrix asdescribed by the following equations:

As = A1D +A2(1−D)

B2 = (A1 −A2)X + (B1 −B2)UCs = C1D + C2(1−D)

Es = (C1 − C2)X

(18)

Thereby, the duty ratio to inductor current canbe derived. The outer loop, SCs current to SCsvoltage, is the pure integrator of inversion of ca-pacitance. Subsequently, the PI controller of MAT-LAB/Simulink is implemented and tuned for the sta-bility as shown in Fig. 4.

Fig.4: Cascade control for SCs.

After receiving the proper PI controllers, they havebeen simulated with the real scale parameter of theBHEV as shown in Fig. 5 in order to observe thebehavior of the proposed system.

5. TEST BENCH AND SIMULATION RE-SULTS

5.1 Test bench

The electromechanical hardware as shown in Fig.6 consists of rechargeable lead-acid batteries, SCs,DC motor with flywheel and dSPACE controller cardDS1104 was set up for the simulation and experiment.The energy sources were the hybridized between bat-teries 12 V/cell connected in series for 7 units andSC 25 F, 2.7 V/cell connected in series for 30 unitsthrough a controllable bidirectional dc-dc converter.The separated field excite DC motor connected with

Fig.5: MATLAB/Simulink block diagram for theBHEV power train system.

flywheels were used as the active load. The parame-ters of the hardware are shown in Table 1.

5.2 Simulation results

The small-scale experiment was simulated in orderto validate the proposed control technique by using

Fig.6: Front view of the experimental hardware.

Table 1: Experimental Hardware Parameters.

Lead-acid batteryRated/Floating voltage [V] 12/13.5Maximum discharge/charge [A] 105/2.1Capacity [Ah] 7Internal resistance [mΩ] 23

SupercapacitorRated voltage [V] 2.7Capacitance [F] 25Internal resistance [mΩ] 25

Bidirectional dc-dc converterInductor/ESR [mH/mΩ] 5/500Output capacitor/ESR [mF/mΩ] 10/25IGBT modules Fuji 2MBI100TA-060

DC motor and loadRate voltage/current [V/A] 120/3.3Rated power [W] 250Rated speed [rpm] 3000Weight/inertial (with flywheel) [kg/kgm2] 14/0.03

Acceleration-based Control Strategy and Design for Hybrid Electric Vehicle Auxiliary Energy Source 89

numerical simulation. The simulation is not only forpreliminary observation but also for safety reason be-fore the experiment, so the switching model of 20 kHzas same as in experiment was used as the same in ex-periment.

The maximum acceleration driving cycle, whichis composed of the acceleration, cruising and brak-ing, was supplied to the controller of SCs to generatethe reference voltage and current. According to thedriving cycle, the motor was started at t = 6 s andsprinted up to the maximum speed of 183 rps at t= 13 s, after that the motor speed was maintainedfor 5 s and decelerated until stoppage from t = 18-25 s as shown in Fig. 7(a) where the reference andactual speed were completely superimposed. Powersharing diagram can be observed in Fig. 7(b) wherebattery provided a small amount of power in acceler-ation, while SCs supplied larger amount of 200 W forthe motor acceleration. In cruising phase, the batteryresponded for the tractive power alone at 60 W whileSCs was silent. At the starting of braking phase, thebattery supplied high peak power of 150 W togetherwith the regenerative braking power 100 W to chargeSCs. DC bus voltage of pure battery and BSCs ex-periments were compared to observe the effectivenessof the proposed

technique. They can be observed in Fig. 7(c),where voltage regulation of 0.5% and 0.34% were pro-vide by the pure battery and BSCs respectively, andthese clearly show that the proposed technique re-duces the battery voltage varying in propulsion ap-plication.

Energy consumption comparison for the drivingcycle between pure battery and BSCs is reported inFig. 7(d) where the end of cycle, BSCs and purebattery consumed 0.285 W and 0.305 W respectively,and this can be considered that the energy was saved6

6. EXPERIMENTAL RESULTS

The electromechanical experiment is used for veri-fying the simulation results presented previously. Theagreement of the experimental results along with thesimulation as aforesaid is shown as in Fig. 8(a)-(d).In Fig. 8(a), the actual speed is completely overlaidthe reference speed that accelerating from t = 6-13 sto the motor speed of 183 rps, cruising for 5 s, andbraking between t = 18 - 25 s, as same in the simu-lation. Power sharing between the battery and SCsis shown in Fig. 8(b). At t = 2 s, the battery wasconnected to the power system, so it can be seen thatthe transient battery power was pre-charged to out-put capacitor of bidirectional converter. At t = 6s, the motor was accelerated where the battery pro-vided a little amount of power at the early stage, butit was inserted by the inclining SCs power till it waszero at the end of acceleration phase. The SCs sup-plied power until reaching the maximum at 180 W,

Fig.7: Simulation results: (a) reference and actualspeed (b) batteries, SCs, and motor output power, (c)DC bus voltage of pure battery and BSCs, and (d)energy consumption of pure battery and BSCs.

90 ECTI TRANSACTIONS ON COMPUTER AND INFORMATION TECHNOLOGY VOL.9, NO.1 May 2015

Fig.8: Experimental results: (a) reference and ac-tual speed (b) batteries, SCs, and motor output power,(c) DC bus voltage of pure battery and BSCs, and (d)energy consumption of pure battery and BSCs.

and it was zero in cruising phase at t = 13 s wherethe battery fed tractive power 60 W until the end ofthis phase. In braking phase starting at t = 18 s, themotor acted as a generator supplying power 70 W tothe SCs with the battery power 150 W for chargingthe SCs fully. The SCs charging losses in any com-ponents can be observed from the power summationin this stage, and it has the same behavior as in thesimulation. This clearly shows that the result in Fig.8(b) accompanies with the result in simulation. Fig.8(c) shows the comparison of DC bus voltage regula-tion between the pure battery and BSCs where theyare 2.5% and 2.17% respectively, so the advantage ofBSCs is the decrease of voltage regulation by 0.33%.Energy consumption for the driving cycle was pre-sented in Fig. 8(d) where the BSCs consume almostthe same as pure battery. This because the toleranceand non-linearity of the hardware, so the profit interm of energy consumption was not achieved.

7. CONCLUSIONS AND FUTURE WORK

This paper presents a new control strategy and areliable DC-DC converter for SCs auxiliary energysource using in BHEV, fore reducing the energy con-sumption of the battery and improve the DC voltageregulation. The proposed strategy is to control SOCand current of SCs according to the speed and ac-celeration of the vehicle. The simulation successfullyproved the benefits of the BSCs over the pure batteryby saving energy of 6% and reducing of DC bus volt-age regulation by 0.16%. The experimental resultsverify the workability of the control strategy wherethe proposed system reduced voltage regulation by0.33%. However, the energy consumption betweenpure battery and BSCs in the experiment showed thesame amount at the end of driving cycle tested. Thiscan be improved by the further optimization the realsystem for a specific application.

ACKNOWLEDGEMENT

The author would like to thanks for finance sup-ported by the PhD. Merit Scholarship of Islamic De-velopment Bank (IDB) for conducting the research atUTM-Proton future drive.

References

[1] M. Ehsani, Y. Gao, and A. Emadi, Modern elec-tric, hybrid electric, and fuel cell vehicles: fun-damentals, theory, and design. CRC press, 2009,pp. 13-15.

[2] L. James and L. John, Electric Vehicle Technol-ogy Explained, John Wiley and Sons, Inc., NewYork, 2003, pp. 1-5.

[3] Lithium-ion battery. Availablehttp://en.wikipedia.org/wiki/Lithium-ion battery

Acceleration-based Control Strategy and Design for Hybrid Electric Vehicle Auxiliary Energy Source 91

[4] S. Pay and Y. Baghzouz, “Effectiveness ofbattery-supercapacitor combination in electricvehicles,” in Power Tech Conference Proceed-ings, 2003 IEEE Bologna, 2003, vol. 3, pp. 1-6.

[5] G. Zorpette, “Super charged,” Spectrum, IEEE,vol. 42, No. 1, 2005, pp. 32-37.

[6] M. Steiner, M. Klohr, and S. Pagiela, “Energystorage system with ultracaps on board of rail-way vehicles,” in Power Electronics and Applica-tions, 2007 European Conference on, 2007, pp.1-10.

[7] S. M. Lukic, S. G. Wirasingha, F. Rodriguez,J. Cao, and A. Emadi, “Power management ofan ultracapacitor/battery hybrid energy storagesystem in an hev,” in Vehicle Power and Propul-sion Conference, 2006. VPPC’06. IEEE, 2006,pp. 1-6.

[8] I. Aharon and A. Kuperman, “Topologicaloverview of powertrains for battery-powered ve-hicles with range extenders,” Power Electronics,IEEE Transactions on, vol. 26, No. 3, pp. 868-876, 2011.

[9] D. Iannuzzi and P. Tricoli, “Speed-based state-of-charge tracking control for metro trains withonboard supercapacitors,” Power Electronics,IEEE Transactions on, vol. 27, No. 4, pp. 2129-2140, 2012.

[10] M. Ortzar, J. Moreno, and J. Dixon,“Ultracapacitor-based auxiliary energy sys-tem for an electric vehicle: Implementationand evaluation,” Industrial Electronics, IEEETransactions on, vol. 54, No. 4, pp. 2147-2156,2007.

[11] N. JInrui, S. Fengchun, and R. Qinglian, “Astudy of energy management system of electricvehicles,” in Vehicle Power and Propulsion Con-ference, 2006. VPPC’06. IEEE, 2006, pp. 1-6.

[12] A. Jarushi and N. Schofield, “Battery and super-capacitor combination for a series hybrid electricvehicle,” in Power Electronics, Machines andDrives (PEMD 2010), 5th IET InternationalConference on, 2010, pp. 1-6.

[13] M. Ortuzar, J. Dixon, and J. Moreno, “Design,construction and performance of a buck-boostconverter for an ultracapacitor-based auxiliaryenergy system for electric vehicles,” in IndustrialElectronics Society, 2003. IECON’03. The 29thAnnual Conference of the IEEE, 2003, vol. 3, pp.2889-2894.

[14] B. Arnet and L. Haines, “High power DC-to-DC converter for supercapacitors,” in ElectricMachines and Drives Conference, 2001. IEMDC2001. IEEE International, 2001, pp. 985-990.

[15] L. Gao, R. A. Dougal, and S. Liu, “Powerenhancement of an actively controlled bat-tery/ultracapacitor hybrid,” Power Electronics,IEEE Transactions on, vol. 20, No. 1, pp. 236-243, 2005.

[16] J. W. Dixon, M. Ortzar, and E. Wiechmann,“Regenerative braking for an electric vehicleusing ultracapacitors and a buck-boost con-verter,” in 17th Electric Vehicle Symposium(EVS17),(Canada), 2000, pp. 1-11.

[17] G. Guidi, T. Undeland, and Y. Hori, ”An op-timized converter for battery-supercapacitor in-terface,” in Power Electronics Specialists Con-ference, 2007. PESC 2007. IEEE, 2007, pp. 2976-2981.

[18] G. Guidi, T. M. Undeland, and Y. Hon, “Aninterface converter with reduced VA ratingsfor battery-supercapacitor mixed systems,” inPower Conversion Conference-Nagoya, 2007.PCC’07, 2007, pp. 936-941.

[19] G. Samson, T. M. Undeland, O. Ulleberg,and P. Vie, “Optimal load sharing strategyin a hybrid power system based on pv/fuelcell/battery/supercapacitor,” in Clean ElectricalPower, 2009 International Conference on, 2009,pp. 141-146.

[20] J. Cao and A. Emadi, “A new bat-tery/ultracapacitor hybrid energy storagesystem for electric, hybrid, and plug-in hybridelectric vehicles,” Power Electronics, IEEETransactions on, vol. 27, No. 1, pp. 122-132,2012.

[21] S. F. Tie and C. W. Tan, “A review of energysources and energy management system in elec-tric vehicles,” Renewable and Sustainable EnergyReviews, vol. 20, pp. 82-102, 2013.

[22] J. Wong, N. Idris, M. Anwari, and T. Taufik,“A parallel energy-sharing control for fuel cell-battery-ultracapacitor hybrid vehicle,” in EnergyConversion Congress and Exposition (ECCE),2011 IEEE, 2011, pp. 2923-2929.

[23] L. Solero, A. Lidozzi, and J. A. Pomilio, “Designof multiple-input power converter for hybrid ve-hicles,” Power Electronics, IEEE Transactionson, vol. 20, No. 5, pp. 1007-1016, 2005.

[24] A. Florescu, H. Turker, S. Bacha, and E. Vinot,“Energy management system for hybrid electricvehicle: Real-time validation of the VEHLIBdedicated library,” in Vehicle Power and Propul-sion Conference (VPPC), 2011 IEEE, 2011, pp.1-6.

[25] M. Jain, C. Desai, and S. S. Williamson, “Ge-netic algorithm based optimal powertrain com-ponent sizing and control strategy design for afuel cell hybrid electric bus,” in Vehicle Powerand Propulsion Conference, 2009. VPPC’09.IEEE, 2009, pp. 980-985.

[26] R. M. Schupbach and J. C. Balda, “ComparingDC-DC converters for power management in hy-brid electric vehicles,” in Electric Machines andDrives Conference, 2003. IEMDC’03. IEEE In-ternational, 2003, vol. 3, pp. 1369-1374.

92 ECTI TRANSACTIONS ON COMPUTER AND INFORMATION TECHNOLOGY VOL.9, NO.1 May 2015

Aree Wangsupphaphol was born inBangkok, Thailand, in 1975. He re-ceived B.Eng. and M.Eng. de-grees in electrical engineering from KingMongkut’s Institute of Technology Lad-krabang, Thailand, in 1999 and 2007 re-spectively. He becomes a Special Lec-turer in the Rail Transportation Engi-neering, Department of Mechanical En-gineering, King Mongkut’s Institute ofTechnology Ladkrabang in 2014. His re-

search interests include energy management system, electricvehicle applications, and control of power electronics systems.

Nik Rumzi Nik Idris received theB.Eng. degree in electrical engineer-ing from the University of Wollongong,Australia, in 1989, the M.Sc. degreein power electronics from Bradford Uni-versity, West Yorkshire, U.K., in 1993,and the Ph.D. degree from UniversitiTeknologi Malaysia, Malaysia, in 2000.He is an Associate Professor at the Fac-ulty of Electrical Engineering, UniversitiTeknologi Malaysia. His research inter-

ests include ac drive systems and DSP applications in powerelectronic systems. Dr. Idris is a senior member of the IEEEand the Chair for the Power Electronics Chapter of the IEEEMalaysia Section for 2014-2015.

Awang Jusoh was born in Tereng-ganu, Malaysia in 1964. He receivedhis B.Eng. degree from Brighton Poly-technic, U.K., in 1988. He obtainedhis M.Sc. and Ph.D. degrees fromthe University of Birmingham, U.K., in1995 and 2004, respectively. He is cur-rently with the Department of PowerEngineering, Faculty of Electrical Engi-neering, Universiti Teknologi Malaysia,Malaysia. His research interests are in

DC-DC converter, renewable energy, and control of power elec-tronics systems.

Nik Din Muhamad received B.Eng.(Hons.) and M.Eng. degrees inelectrical engineering from UniversitiTeknologi Malaysia, Malaysia, in 1988and 1999 respectively. He is currently asenior lecturer with Universiti TeknologiMalaysia, Malaysia. His research in-terests include modeling and control ofpower electronic systems.

Supanat Chamchuen was born inKalasin, Thailand, in 1992. He is study-ing B.Eng. degree in electrical engi-neering at King Mongkut’s Universityof Technology Thonburi, Thailand. Hewas an internship student at the UTM-PROTON Future Drives Laboratory,Universiti Teknologi Malaysia, Malaysiaduring June-July 2014. His research in-terests are in electric motor drives andrailway electrification.