Speed Control Simulation of the Electric Vehicle Driving Motor

Transcription

1 Availal onlin at Vol. 13, No. 7, Novmr 2017, pp DOI: /ijp p Spd Control Simulation of th Elctric Vhicl Driving Motor Wanmin Li a,, *, Mnglu Gu, Lulu Wi a School of Automoil Enginring, Lanzhou Institut of Tchnology, Lanzhou , China School of Automoil, Chang an Univrsity, Xi an , China Astract In ordr to raliz prcis spd control of driving motor, an adaptiv fuzzy PID control stratgy for motors was stalishd asd on th xisting proportional intgral drivativ (PID) control thory. Th motor spd control modl is uilt y simplifying th paramtrs of a rushlss DC motor using th Sim Powr Systms toolox in MATLAB/Simulink nvironmnt, which involvs th simulation of motor spd control including low spd, high spd, and road ump situations in city traffic nvironmnt. Rsults show that th tim of th adaptiv fuzzy PID control is 0.08s at low spd, th adjustmnt tim of th convntional PID control is 0.22s, and th adjustmnt tims ar 0.12s and 0.32s at high spd. Aftr ncountring road umps, th adaptiv fuzzy PID control can quickly ract and rturn to normal spd, whras th convntional PID control is vidntly affctd y th intrfrnc. Kywords: fuzzy Control; PID control; simulation; lctric vhicl (Sumittd on August 31, 2017; Rvisd on Octor 5, 2017; Accptd on Octor 23, 2017) 2017 Totm Pulishr, Inc. All rights rsrvd. 1. Introduction Nw nrgy vhicls, particularly hyrid and pur lctric vhicls, ar currntly dvloping at a rapid rat. Rsarch on th prcis control of automotiv driv motors is coming incrasingly important. A growing numr of xisting motor driv vhicls ar using prmannt magnt rushlss DC motors inconsidration of motor torqu, powr dnsity, and cost factors [1]. Howvr, such motors xhiit strong coupling and nonlinar charactristics, whras traditional studis on motor control us th proportional intgral drivativ (PID) control [2], which has a simpl structur that can satisfy crtain roustnss rquirmnts for common applications in th industrial fild [3]. Howvr, as a powr sourc for vhicls, th prcision of PID control cannot satisfy th dmands of drivrs for control and comfort [4]. Th typical rsults of currnt local and intrnational rsarch on motor control xhiit disadvantags, including th low prcision of th traditional PID control [5]. Svral scholars adoptd th fuzzy control algorithm to solv th aformntiond prolms. Howvr, fuzzy control must uilt asd on a crtain amount of historical data and an mpirical attriut is important in slcting a fuzzy mmrship function [6]. Similar to th us of th fuzzy control mthod, som nginrs introducd a nural ntwork into rushlss DC motor rotation control and dsignd a control systm with a digital signal procssor as th control unit [7,10]. In summary, although rsarch rsults on motor control ar aundant, studis on th applications of such motors to vhicls and on th prcis control of vhicls asd on th rquirmnts of drivrs for driving and riding comfort rmain lacking [8]. In th currnt work, a rushlss DC driving motor is slctd as th ojct of study. Thn, a motor control mthod asd on fuzzy control and adaptiv PID control is proposd to optimiz motor control accuracy [9]. 2. Mathmatical Modl of Motor A prmannt magnt rushlss DC motor has a trapzoidal ack lctromotiv forc (EMF) and a rctangular currnt wav form. Th mutual inductanc producd y th rotor and th stator is nonlinar [10]. It is frquntly simplifid in rsarch y assuming th following: In asnc of th alvolar ffct, th windings ar vnly distriutd insid th stator, armatur * Corrsponding author. addrss: @qq.com.

2 Spd Control Simulation of th Elctric Vhicl Driving Motor 1141 raction is disrgardd, th winding symmtry has thr phass, and losss du to hystrsis and ddy currnts ar not considrd. Th voltag alanc quation for th motor winding is simplifid as follows: ua r 0 0 ia L M M ia a u 0 r 0 i M L M p i u 0 0 r M M L c i c i c c whr th winding voltag of th stator is dnotd as i, i, i a c ; th winding EMF of th stator is dnotd as ua, u, uc a,, c (1) ; th winding lctric currnt of th stator is dnotd as ; and th inductanc of ach phas winding, th mutual inductanc twn th aritrary two-phas windings, and th diffrntial oprators ar rspctivly dnotd as L, M, and p. Th rsulting phass otain a trapzoidal ack EMF and ar rctangular currnt wavform, as shown in Figur 1. E,I 120 a i a 2 wt Figur 1. Phas currnt and ack EMF Whn th rushlss DC motor is considrd, no cntral lin is osrvd among th thr windings, i.., ia i ic 0 (2) Mia Mi Mic 0 (3) Thrfor, th voltag alanc quation for th nw motor winding is otaind as follows: ua r 0 0 ia L 0 0 ia a u 0 r 0 i 0 L 0 p i u 0 0 r 0 0 L c i c i c c (4) Th motor circuit rprsntd y th uppr typ may quivalnt to Figur 2. i u a a R L-M a u i R L-M c u c i c R L-M Figur 2. Equivalnt motor circuit Furthrmor, th torqu and motion quations for th simplifid rushlss DC motor ar as follows: T i i i w a a c c (5)

3 1142 Wanmin Li, Mnglu Gu, and Lulu Wi dw T TL J Bw (6) dt Whr th motor lctromagntic torqu is dnotd as T, th rotor angular vlocity is dnotd as w, th load torqu is dnotd as T l, and th motor slf-rotating inrtia is dnotd as J. 3. Doul closd-loop control systm asd on fuzzy PID mthod 3.1. Systm structur Th motor adopts th doul closd-loop control stratgy for rotation spd and phas currnt. Th main and auxiliary rings control rotation spd and phas currnt, rspctivly. Th idal motor control procss is th vhicl raching th maximum at th starting stag to gnrat maximum acclration. Th driv currnt drops rapidly whn vhicl spd acclrats to tal stag, which alancs th forc and load of th driving motor. Th phas currnt cannot rapidly rducd caus of th prsnc of inductanc in th motor, and a crtain amount of tim is rquird to start th vhicl rapidly within accptal limits aftr th currnt rachs its maximum valu [11]. Thrfor, th doul loop control stratgy for rotation spd and phas currnt is introducd into th motor control. Th motor control procss is illustratd in Figur Dsign of adaptiv fuzzy PID control Figur 3. Flowchart of motor control procdur Th fuzzy PID control consists of two parts: fuzzy control and PID control. To rciv rapid and corrct rsponss aftr systm inputs, modl rror and rror chang rat c ar continually dtctd whil th systm is oprating. Thn, in accordanc with th fuzzy control rul, th PID control paramtrs k, k, k ar corrspondingly rgulatd asd on thir p i d rlations to nal modl rror and rror chang rat c to satisfy all systm rquirmnts. Th structur of th fuzzy PID controllr is shown in Figur 4. In th figur, modl rror and rror chang rat c ar otaind aftr comparing systm inputs and outputs. To fulfill th rquirmnts of fuzzy filds, a quantization factor is adoptd to magnify or diminish th signal to an appropriat xtnt, which is thn rgardd as th final input of th fuzzy control systm. Th fuzzy controllr calculats th corrction of th PID adjustal paramtr asd on th scaling paramtr. Through this procss, nw control paramtrs, namly, k, k, k, ar otaind y th PID controllr, which ar thn applid to th controlld targt. p i d Figur 4. Flowchart of fuzzy PID control Thr fuzzy controllrs ar adoptd in this study. Modl rror and rror chang rat c ar usd as th inputs of ach controllr. Thr adjustal paramtrs ar usd as th output sin PID control; that is, a doul-input singl-output mthod is implmntd. In accordanc with actual dmands, this study considrs Mandani and dsigns a fuzzy infrnc systm (FIS)

4 Spd Control Simulation of th Elctric Vhicl Driving Motor 1143 control in using MATLAB. Th fuzzy domain is st to [ 6, 6]. Th fuzzy susts of ach controllr ar all st to {NB, NM, NS, ZO, PS, PM, and PB}.Th susts rprsnt ngativ ig, ngativ middl, ngativ small, zro, plus small, plus middl, and plus ig. Th triangl mmrship function is slctd with high snsitivity and rsolution to acquir good roustnss and practicaility of th fuzzy controllr. Th mmrship functions of th fuzzy controllr ar dfind in Figur 5. 1 NB NM NS ZO PS PM PB Figur 5. Mmrship functions of fuzzy controllr Th stalishmnt of fuzzy control ruls, which is ralizd mostly through th xprinc of xprts or th statistical induction of historical data, is an important aspct of fuzzy control. Th corrction factors of PID paramtrs ar otaind y considring tuning, as shown low: Tal 1. k p Fuzzy ruls c NB NM NS ZO PS PM PB NB PB PB PM PM PS ZO ZO NM PB PB PM PS PS ZO NS NS PM PM PM PS ZO NS NS ZO PM PM PS ZO NS NM NM PS PS PS ZO NS NS NM NM PM PS ZO NS NM NM NM NB PB ZO ZO NM NM NM NB NB Tal 2. k i Fuzzy ruls c NB NM NS ZO PS PM PB NB NB NB NM NM NS ZO ZO NM NB NB NM NS NS ZO ZO NS NM NM NS NS ZO PS PS ZO NM NM NS ZO PS PM PS PS NM NS ZO PS PS PM PB PM ZO ZO PS PS PM PB PB PB ZO ZO PS PM PM PB PB Tal 3. k d Fuzzy ruls c NB NM NS ZO PS PM PB NB PS NS NB NB NB NM PS NM PS NS NB NM NM NS ZO NS ZO NS NM NM NS NS ZO ZO ZO NS NS NS NS NS ZO PS ZO ZO ZO ZO ZO ZO ZO PM PB PS PS PS PS PS PB PB PB PM PM PM PS PS PB Th outputs of adopting fuzzy control ruls ar svral fuzzy sts. Howvr, prcis inputs ar rquird to control th motor, and thus, th fuzzy control consqunc should clar. Existing mthods for achiving clarnss includ th avrag ara mthod, th maximum mmrship dgr mthod, and th cntr-of-gravity mthod. In this study, th simpl cntr-of-gravity mthod is usd to conduct clarnss procssing. Th modifid control paramtrs ar drivd y sustituting th control varials {, c}p otaind from th fuzzy control into th aov quations and comining th quantization factor of th fuzzy control. Th final fuzzy PID control output can acquird aftr th modifid paramtrs ar addd to th initial control paramtrs. k = k k {, } p (7) p p0 u1 c

5 1144 Wanmin Li, Mnglu Gu, and Lulu Wi k = k k {, } i (8) i i0 u 2 c k = k k {, } d (9) d d 0 u3 c k, k, k In th last quation, k p0, ki0, k rprsnt th initial control paramtrs, whras ar th fuzzy controllr d 0 quantization factors. Th initial control paramtrs ar dtrmind y trial and rror, and a convntional PID control modl for a rushlss DC motor is stalishd forhand. Th paramtrs ar constantly adjustd, and th influnc of such adjustmnt is analyzd. Thn, th PID paramtrs ar slctd, which provids a stady and accptal ovrshoot rsult as th initial input valu of th fuzzy PID control. Th initial PID control paramtrs that ar adoptd aftr th first tuning ar 3, 0.15, and Estalishmnt of fuzzy PID control modl u1 u 2 u3 Th workload for indpndntly dvloping and stalishing a motor control modl is havy. A modul lirary is intgratd into th Simulink platform of MATLAB to usd for th motor control modl. This procdur provids convninc for motor modling and control, such as th insulatd gat ipolar transistor modul, which is ncssary for stalishing an invrtr. In this study, th modl for th fuzzy PID control is uilt using th Sim Powr Systm toolox, as shown in Figur Modl simulation and analysis Figur 6. Block diagram of fuzzy PID control Th stalishd modl is tstd in Simulink nvironmnt, and th sampling tim T of th simulation systm is st to s. A common typ of vhicl-driving rushlss DC motor is usd as th motor prototyp, and its asic paramtrs ar as follows: winding inductanc L is 0.02H, mutual inductanc Mis 0.067H, damping cofficint B is , total rsistanc R is 1Ω, rotary inrtia of th motor rotor J is 0.005kg m2, pol log P is 1, ack EMF cofficint k is 0.382, and motor voltag rating U is standard at 220V. On th asis of th aformntiond sttings, th PID control initial paramtrs, including th quantification factor and th proportion factors k, k u1, k u2, and k u3, ar st to 0.01, 0.38, 0.01, and 0.01, rspctivly. Th vhicl paramtrs mainly considr th componnts at th lowr nd of th motor output and th asic paramtrs of th vhicl undr running condition. It includs th minimum running spd; atypical uran road is slctd as an xampl. Th minimum spd is 30km/h, th maximum spd is 80km/h, th rolling radius of th vhicl driving whl is 0.367m, and th transmission ratio is 1. In accordanc with th rlationship twn vhicl spd and motor spd rn ua (10) ii g o Th low and high motor spd ar approximatly 220r/min and 580r/min, rspctivly. Th aformntiond initial stat is inputtd to otain th simulation rsults at low and high spds, as shown in Figurs 7 and 8. From th simulation rsults, th targt spd for low spds is 220r/min. Th tim rquird for systm tuning is approximatly 0.2s for th convntional non-adaptiv fuzzy PID control systm. Although th raction tim is longr, no ovrshoot phnomnon occurs. Th rgulation tim of th modl proposd in this study is only 0.08s. Rspons coms fastr at th sam spd and no ovrshoot occurs. Th targt spd for high spds is 580r/min. Furthrmor, th rquird systm tuning tim is approximatly 0.32s for th convntional fuzzy PID control systm. Corrspondingly, th rgulation tim is only 0.12s whn th slf-adaptiv fuzzy PID control systm proposd in this study is adoptd. Th diffrncs twn th two systms ar apparnt in trms of raction tim. On th asis of th raction tim rquird y th systm, tim is xtrmly short and asically fulfills th rquirmnts of systm rspons tim.

6 Spd Control Simulation of th Elctric Vhicl Driving Motor 1145 Figur 7. Simulation rsults at low spd Figur 8. Simulation rsults at high spd In th normal driving procss, particularly in a low-spd cas, th vhicl is sujctd to umps on th road surfac. Thrfor, whn rushlss DC motor spd control xprimnts ar prformd, th conditions of th road whr th vhicl is travling must considrd. In th simulation procss, a crtain intrfrnc load is applid to th motor, and th load is addd at s. Thn, th diffrnt ffcts of th convntional PID control systm and th slf-adaptiv fuzzy PID control systm ar analyzd. Th rsults ar prsntd in Figur 9. As shown in th figur, motor spd is rducd significantly aftr th load is addd. Th ffct of th convntional PID control is not apparnt and motor spd is in a disturd stat. In contrast, th ffct of th adaptiv fuzzy PID control is vidnt. Th adaptiv fuzzy PID control can ract quickly and rturn to normal spd, whras th convntional PID control is affctd considraly y intrfrnc. Th adaptiv fuzzy PID control xhiits strong adaptaility. 5. Conclusions Figur 9. Simulation rsults undr load application Th matur PID control stratgy is adoptd and comind with an xisting optimizd fuzzy control that is commonly usd in th control fild to raliz prcis control of th spd of a driving motor. An adaptiv fuzzy PID control stratgy for

7 1146 Wanmin Li, Mnglu Gu, and Lulu Wi rushlss DC motors is stalishd and th motor spd control modl is uilt in MATLAB/Simulink nvironmnt. Simulation tst ar prformd and th following conclusions ar drawn. In low-spd cas, th adaptiv fuzzy PID control adjustmnt tim is 0.08s, whras th convntional PID control adjustmnt tim is 0.22s. In a high-spd cas, th adaptiv fuzzy PID control adjustmnt tim is 0.12s, whras th convntional PID control adjustmnt tim is 0.32s. Accordingly, th adaptiv fuzzy PID control is mor adaptal than th convntional PID control. An intrfrnc load is addd to th simulation to considr th umps causd y road roughnss whil travling. Th rsults show that th adaptiv fuzzy PID control can ract rapidly and rturn to normal spd, whras th convntional PID control is vidntly affctd y such intrfrnc. Acknowldgmnts This work was partly financially supportd through grants from Scintific Rsarch Projct of Univrsitis in Gansu Provinc (2017A-110) and Young Scinc and Tchnology Innovation Projct of Lanzhou Institut of Tchnology(16K- 004). Rfrncs 1. L. Briman, Random Forsts, Machin Larning, vol. 45, no. 1, pp. 5-32, A. Choudhury, P. Pillay, DC-Link Voltag Balancing for a Thr-Lvl Elctric Vhicl Traction Invrtr Using an Innovativ Switching Squnc Control Schm, IEEE Journal of Emrging and Slctd Topics in Powr Elctronics, vol. 2, no. 2, pp , B. Hamd, M. Almoaid, Fuzzy PID Controllrs Using FPGA Tchniqu for Ral Tim DC Motor Spd Control, Intllignt Control and Automation, vol. 2, no. 3, pp , J. J. Justo, F. Mwasilu, Fuzzy Modl Prdictiv Dirct Torqu Control of IPMSMs for Elctric Vhicl Applications, IEEE/ASME Transactions on Mchatronics, vol. 22, no. 4, pp , M. Jalali, A. Khajpour, Intgratd staility and traction control for lctric vhicls using modl prdictiv control, Control Enginring Practic, vol. 54, no. 9, pp , M. Muruganandam, M. Madhswaran, Exprimntal vrification of choppr fd DC sris motor with ANN controllr, Frontirs of Elctrical and Elctronic Enginring, vol. 7, no. 4, pp , S. D. Pinto, P. Camocardi, Torqu-Fill Control and Enrgy Managmnt for a Four-Whl-Driv Elctric Vhicl Layout with Two-Spd Transmissions, IEEE Transactions on Industry Applications, vol. 53, no. 1, pp , R. Shi, S. Smsar, Constant Currnt Fast Charging of Elctric Vhicls via a DC Grid Using a Dual-Invrtr Driv, IEEE Transactions on Industrial Elctronics, vol. 64, no. 9, pp , X. Shi, M. Krishnamurthy, Survival Opration of Induction Machin Drivs with Smooth Transition Stratgy for EV Applications, IEEE Journal of Emrging and Slctd Topics in Powr Elctronics, vol. 2, no. 3, pp , P. Walkr, B. Zhu, Powrtrain dynamics and control of a two spd dual clutch transmission for lctric vhicls, Mchanical Systms and Signal Procssing, vol. 85, no. 2, pp. 1-15, Z. Yang, F. Shang, Comparativ Study of Intrior Prmannt Magnt, Induction, and Switchd Rluctanc Motor Drivs for EV and HEV Applications, IEEE Transactions on Transportation Elctrification, vol. 1, no. 3, pp , 2015