Inter national J our nal of P o wer Electr onics and Dri v e System (IJPEDS) V ol. 17, No. 2, June 2026, pp. 885 ∼ 893 ISSN: 2088-8694, DOI: 10.11591/ijpeds.v17.i2.pp885-893 ❒ 885 T or que ripple r eduction in PMSM f or FCEVs using ANFIS contr oller Shilpa Rao Hosabettu, Pushpa Rajesh V iswanathan Department of Electrical Engineering, Jain (Deemed to be Uni v ersity), Beng aluru, India Article Inf o Article history: Recei v ed Oct 21, 2025 Re vised Jan 29, 2026 Accepted Feb 21, 2026 K eyw ords: ANFIS controller FCEVs Field oriented control PI controller PMSM ABSTRA CT Globally , there is a gro wing emphasis on switching to green ener gy , particularly in the transportation sector , due to the ef fects of global w arming, as seen by rising carbon footprints. Fuel cell electric v ehicles (FCEVs) are one such technology that has attracted a lot of interest because of their a v ailability , ease of use, high ef cienc y , and silent operation. Fuel cells are emplo yed along with batteries to dri v e the v ehicle much f arther . Motors lik e permanent magnet synchronous motor (PMSM) pro vide the dri ving force for the v ehicle, o wing to their high torque at v ariable speeds and compactness. In such systems, it is necessary to ha v e intelligent controllers that can align with the load requirement by means of a consis tent and optimized po wer distrib ution. The torque ripple phenomenon, which has an impact on dynamic performance and operational stability , is one of the main limitations in the operation of PMSMs. In this w ork, smart control techniques, which are a combination of adapti v e neuro fuzzy inference systems (ANFIS) and proportional-inte gral (PI) control, are emplo yed to demonstrate the application of PMSM in conjunction with eld-oriented control (FOC). Simul ation results indicate that the proposed ANFIS-based FOC reduces torque ripple as compared to con v entional PI control under v arying load conditions. This is an open access article under the CC BY -SA license . Corresponding A uthor: Shilpa Rao Hosabettu Department of Electrical Engineering, Jain (Deemed to be Uni v ersity) Beng aluru, India Email: shilpaach16@gmail.com 1. INTR ODUCTION Battery electric v ehicles (BEVs) are limited by range and can be b ulk y for long-distance tra v el. Furthermore, the bigger the battery , the longer it tak es to get char ged. Batteries also ha v e a limited lifespan; their performance depletes with time, and we need to consider the mitig ation of the disposal of aged-out batteries, which pose an en vironment al hazard [1]. In comparison with a battery , fuel cells generate electrical ener gy instead of storing it and do so as long as the fuel supply is maintained. Fuel cell electric v ehicles (FCEVs) of fer v arious benets, including a rapid refuelling time that mak es them suitable for long-distance journe ys. Fuel cells cannot react to sudden transient speed and torque v ariations due to the time required to change t he fuel supply rate and fuel reaction rate. This supply time g ap should be mitig ated intelligently by the use of battery packs. Although direct current (DC) machines are well kno wn for their ease of control, alternating current (A C) motors ha v e adv antages such as lo wer maintenance and higher ef cienc y [2]. In the case of electric v ehicles (EVs), the choice for motors includes induction motors, brushless DC motors, and permanent magnet J ournal homepage: http://ijpeds.iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
886 ❒ ISSN: 2088-8694 synchronous motors (PMSMs). As loads are non-linear , the control system for fuel injection should be dynamic and f ast-responsi v e [3], which leads to reduced ef cienc y . PMSMs are generally preferred in EVs due to lo wer torque ripple, higher torque density , good lo w-speed performance, higher ef cienc y and lo wer acoustic noise compared to brus hless DC (BLDC) motors [4]. BLDC motors are well kno wn for simpler control strate gies, such as step control, which limit performance optimization. Ha and V an Hai [5] present an adapti v e neuro-fuzzy inference system (ANFIS)–based torque controller for an in-wheel single-sided axial ux permanent magnet synchronous motor (AFPMSM) used in electric v ehicles. The proposed ANFIS controller is designed within a eld-oriented control frame w ork and compared with con v entional proportional inte gral (PI) and fuzzy logic controllers. MA TLAB/Simulink results demonstrate that ANFIS pro vides superior torque tracking, reduced torque ripple, impro v ed ef cienc y , and better rob ustness to parameter v ariations. The study highlights the ef fecti v eness of intelligent h ybrid control strate gies for enhancing torque performance and stability in in-wheel AFPMSM-based elec tric v ehicle traction systems. Chaudhary et al. [6] present a topology featuring a modied boost DC-DC con v erter connected to the PMSM via eld-oriented control (FOC). The study highlights t he use of the FOC strate gy for fuel-cell-based EVs equipped with PMSMs. Re generati v e braking ar tef acts by means of a bi-directional con v erter ha v e not been considered. Basappa and V isw anathan [7] propose FOC on ANFIS for PMSM-based EVs handling nonlinearities. ANFIS-based approach gi v es smoother performance, b ut does not e v aluate torque ripple, which is a k e y parameter for use in EV’ s. T able 1 presents a comparison of commonl y used control techniques, including FLC [8], particle sw arm optimization (PSO), and FOC. The FLC-based v alues e xhibit a maximum torque ripple of 13.5% as compared to multiobjecti v e PSO of 7.3% and 77.6% with h ysteresis band current controller [9]. The ndings from the table abo v e highlight the signicant ripple in traditional methods, such as FLC, PI control-based FOC, and traditional FOC. By le v eraging smart control techniques such as PSO or ANFIS-based control, EVs can operate with much higher precision and pro vide a smoother dri v e e xperience in EVs. T able 1. Comparison of control techniques for PMSM dri v es Controller T orque ripple Inference FLC 0.135 Good Multiobjecti v e PSO 0.073 F ast FOC 0.776 Poor 2. METHOD The inherent issue with FCEVs is the torque ripple, which leads to jerks while dri ving the v ehicle. It also increases ambient noise and mechanical stress, and reduces dri v e-train ef cienc y . The primary reasons for torque ripple are fuel injection, control strate gy sl o wness, in v erter harmonics, and battery internal resistance. This paper shall address countering fuel injection, control strate gy slo wness, and torque ripple reduction by using an ANFIS controller . A closed-loop FOC for fuel cell electric v ehicles emplo ying PMSM motors. The system parameters, v oltage, current, and speed, are sensed and fed to the controller , as sho wn in Figure 1(a). Figure 1(b) sho ws the control operation with the ANFIS controller . The initial research be g an with choosing an of f-the-shelf PMSM motor simulation in MA TLAB. It is crucial to design a smart controller , such as ANFIS, to produce a f ast and more concise output response to v ariations in the load. Fuel cells operate at a nominal DC v oltage and are slo w to react to sudden changes in load, leading to transients. Thus, an additional battery is emplo yed in parallel, which not only supplies transient po wer b ut can also store ener gy during braking [4]. There are dif ferent types of controllers s u gges ted in [5], such as the PI controller , ANFIS controller , fuzzy controller , and neural netw ork [6], sliding mode controller [7]. DC-A C con v erter/in v erter is needed to feed the DC v oltage to the PMSM motor . The control strate gy is based on techniques such as direct torque control (DTC) [10] and FOC. The proposed topology includes 2 ener gy sources, viz.: fuel cell and battery packs. Fuel cells cannot be rechar ged by supplying current and can only dissipate po wer through the use of fuel, and thus can be called a unidirectional po wer source [8]. Thus, on light loads, the residual po wer from the fuel ce ll’ s operation is used to rechar ge the battery . The output v oltage of the fuel cell drops under v arious scenarios, such as concentration drop, acti v ation drop, and an increase in the output current. Thus, there is a need for v oltage stabilization at the output of the fuel cell [9]. Batteries, on the other hand, can either supply po wer to the load under v arious light-load conditions or can be rechar ged as well with residual po wer and re generati v e braking. Int J Po w Elec & Dri Syst, V ol. 17, No. 2, June 2026: 885–893 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Po w Elec & Dri Syst ISSN: 2088-8694 ❒ 887 (a) (b) Figure 1. Proposed system block diagram: (a) proposed system of fuel cell with battery EV system and (b) FOC block diagram with ANFIS 2.1. Bidir ectional con v erter A bidirectional DC-DC con v erter is emplo yed to bridge the connection between a fuel cell and a battery , as sho wn in Figure 2, due to the dif ference in rating and dynamic load. This con v erter shall play a crucial role in re gulating and managing the desired output v oltage, current, and po wer . The con v erter achie v es b uck-boost functionality by transferring ener gy between tw o inductors and capacitors through controlled switching, which controls the output v oltage [11]. The control can be adjusted by v arying the duty c ycle of the switch, the switching frequenc y , and other aspects based on the load and input conditions. The o wchart, as sho wn in Figure 3 e xplains the 3 modes of operation of bidirectional con v erter , viz.: i) fuel cell supply mode (boost operation) e xplains that the fuel cell supply mode indicates that all the load po wer is supplied by the fuel cell, ii) battery assist mode indicates that po wer to the load is supplied by both the fuel cell and battery combined, and iii) re generati v e braking mode which supplies po wer back to the battery during braking operation. This means that the transients of the load are not directly seen by the fuel cell and are borne by the battery output. The output sho wn in Figure 4(a) depicts the re gulation of 0.1% with the steady state con v erter v oltage of 58 V . Figure 4(b) sho ws the equi v alent MA TLAB simulation model. Figure 2. Bidirectional con v erter schematic Figure 3. Bidirectional con v erter T or que ripple r eduction in PMSM for FCEVs using ANFIS contr oller (Shilpa Rao Hosabettu) Evaluation Warning : The document was created with Spire.PDF for Python.
888 ❒ ISSN: 2088-8694 (a) (b) Figure 4. Bidirectional con v erter: (a) DC output and (b) MA TLAB simulation 2.2. P ermanent magnet synchr onous motor PMSM motors are the preferred choice for FCEVs due to their superior characteristics, impro v ed ef cienc y , compact size, reduced noise le v els, and rotor inertia [12]. The motor rating needs to be carefully chosen to match the maximum torque required and current rating, and this can be achie v ed by selecting intelligent controllers. The intelligent controller controls the DC-DC con v erter as well as the DC-A C in v erter . The main purpose of FOC is to maintain the stator and rotor elds perpendicular to each other to produce maximum torque [13]. The control signal is aligned with the magnetic eld of the rotor . The three-phase stator currents I a , I b , I c are con v er ted into 2-phase stationary frame (I, Q) kno wn as Clark e’ s T ransform. By using the angle θ r and P ark’ s transform, stationary currents are con v erted into a rotating reference frame kno wn as the d-q frame ( I d , I q ). The error observ ed between the reference current, deri v ed from the computation of stator ux, and the actual current is fed to the controller for current control, which con v erts the dif ference into v oltage terms used for PWM generation [14]. T w o control modes are used in FOC, viz.: current control and speed control. As the loads are nonlinear , the motor torque changes rapidly , which af fects the current requirement; maintaining v oltage re gulation during load swit ching [15]. The fundamental equation of torque ( T e ) in a PMSM is as in (1). T e = 3 p 2 ( ψ m i q + ( L d − L q ) i d i q ) (1) Where p number of poles, ψ m rotor ux linkage. i d , i q , L d , L q D and Q axis stator current and inductance. Int J Po w Elec & Dri Syst, V ol. 17, No. 2, June 2026: 885–893 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Po w Elec & Dri Syst ISSN: 2088-8694 ❒ 889 The torque comprises tw o parts: magnetic torque, the more dominant, produced by the inte raction between the permanent magnet ux and the quadrature current i q , and reluctance torque, which is produced by the dif ferences in inductance between d and q ax es. The problem with operating the PMSM at rated ux is that the maximum speed is limited by stator v oltage, rated current, and back electromagnetic force (EMF) [16]. In such scenarios, the option is to go with eld weak ening control of the motor , which controls the D-axis current by inducing a ne g ati v e v alue. By doing so, the rotor ux linkage reduces and thus higher speeds abo v e base speed are possible. PMSM motor has a sinusoidal type of back EMF that interacts with the stator currents to produce the motor torque [17]. Distortions caused in the back EMF shall further increase the torque ripple. Owing to its sinusoidal back EMF , the motor shall ha v e smooth torque production and lo wer harmonics as compared to other motors. Stator current analysis is essential for understanding it s control characteristics, performance e xpectations, and achie v able ef cienc y . The y are 120-de gree phase-shifted and balanced in ideal conditions [18]. Under loaded conditions, the st ator current increases proportionally to maintain the torque, thereby introducing harmonics due to saturation and potentially non-linearities [19]. These can be balanced out by means of a closed-loop control by emplo ying an additional PI controller . The P ark transform in v olv es con v erting 3-phase stator currents, which are sensed using a current sensor , into a non-rotating DQ-axis frame [20]–[22]. The purpose of doing so is to simplify the operation in a 3-phase rotating frame. The de gree of freedom in a PMSM includes current, v oltage, torque, and speed [23]. The MA TLAB model used in this w ork is sho wn in Figure 5. Figure 5. MA TLAB simulation model 2.3. PI contr oller and ANFIS contr oller A PI controller is a v ery commonly used feedback controller to minimize the error as compared to t he reference, which combines proportional and inte gral action to past error [24]. The system transfer function can thus be dened as (2). u ( t ) = K p e ( t ) + K i Z e ( t ) dt (2) Where K p is the proportional g ain K i is the inte gral g ain, and e ( t ) i s the error . PI controllers in this w ork ha v e been used in a couple of stages as: i) control of I d and I q with respect to reference; ii) ring angle control of bi-directional DC-DC con v erter . ANFIS controller is an articial intelligence-based control technique that incorporates a neural netw ork with a rule-based reasoning of fuzzy logic. It represents an adv anced control strate gy for PMSM that combines the strengths of neural netw orks and fuzzy logic systems. ANFIS controllers are particularly v aluable in electric v ehicle dri v e sys tems requiring precise torque control, high precision positioning systems and robotics [25], [26]. Due to this h ybrid approach of using both PI and ANFIS, we can ef fecti v ely mitig ate the ef fects of non-linear motor and load dynamics. Non-linear control, as pro vided by the ANFIS controller , learns and models PMSMs’ characteri stics ef fecti v ely [27]. It can dynamically adjust its control rules in real time as conditions (load, speed, temperature) change. Thereby , we can say that ANFIS minimizes ener gy losses and impro v es dri v e ef cienc y [28] and achie v es good speed and torque tracking. Furthermore, it optimiz es control parameters without requiring manual tuning. Therefore, we can say that this controller w orks in tandem with the PI controller to ensure ripple-reduced torque and intelligently balance fuel cell and battery po wer . T or que ripple r eduction in PMSM for FCEVs using ANFIS contr oller (Shilpa Rao Hosabettu) Evaluation Warning : The document was created with Spire.PDF for Python.
890 ❒ ISSN: 2088-8694 The ANFIS controller design for fuel cell EVs comprises the follo wing stages [29], and it is as sho wn in Figure 6. Fuzzication of the input to generate the membership grade of the input. Rule-based optimization by means of remo ving rarely used rules and by the use of the genetic algorithm (GA), to select optimal v alues. In this w ork, h ybrid controllers are emplo yed, which are a combination of PI and ANFIS, to stabilize performance. The fuzzy logic controller emplo ys a proportional-inte gral (PI) conguration, where a fuzzy inference system (FIS) utilizes speed error and its deri v ati v e v alues to generate the required q-axis current v alues, thereby maintaini ng the desired motor speed. No changes were made to the ANFIS structure from MA TLAB; the rules belo w were optimized to impro v e the performance with FCEVs. – If the error is ne g ati v e and the rate is also ne g ati v e, the output is -1. – If the error is positi v e and the rate is also positi v e, the output is 1. – All other cases (error and rate dif fering in sign), output is 0. The model congurations include the follo wing parameters for the ANFIS tuning, viz.: Kp, Ki, controller scaling f actors ( C 0 , C e , C d ) are deri v ed from the con v entional PI controller g ains. T able 2 gi v es the ANFIS tuning parameters , and T able 3 gi v es the PI Controller tuning parameters for current control and ring angle control. These v alues are chosen on a trial-and-error basis. Figure 6. ANFIS control block diagram T able 2. ANFIS tuning parameters P arameter V alue Error scaling f actor (Ce) 1 Change in error scaling f actor (Cd) 0.15149 Output scaling f actor (C0) 15.177 T able 3. PI tuning parameters P arameter Current controller (1) Firing angle control (2) Kp 5 0.3005 Ki 1 0.2291 3. RESUL TS AND DISCUSSION Reference speed w as chosen to depict changing load conditions at time interv als of 1 second. As sho wn in Figure 7(a), at startup, there is a slight o v ershoot in rotor speed before it settles into its steady-state. The speed feedback closely follo ws the changes in t he reference load speed with rapid settling (0.25 s), indicating a well-tuned loop. At each speed step, brief oscillations appear in speed and torque. Figure 7(b) sho ws the v olta ge transients w a v eform, which also indicates the transient se ttling with o v ershoot of approximately 5%. Figure 8(a) sho ws the motor v oltage closely follo wing the change in load demand. The controller e xhibits good dynamic performance with good transient settling time (0.25 s), accurate tracking, and stable bidirectional speed control, thus making it suitable for applications such as PMSM dri v es in FCEVs. ANFIS controllers clearly demonstrate superior motor control with swift acceleration. Further , the lo w o v ershoot from the desired speed demonstrates crisp and precise speed control. The transients seen at the time of phase switching are a typical characteristic of PMSM. Although the sampling time of the simulation w as chosen as 100 us, i n c reasing the sampling time does not bring forw ard an y further impro v ement in ripple reduction. Only changes with re spect to the control technique to use, machine learning methods such as reinforcement learning, can further impro v e the transient settling time. Int J Po w Elec & Dri Syst, V ol. 17, No. 2, June 2026: 885–893 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Po w Elec & Dri Syst ISSN: 2088-8694 ❒ 891 T able 4 gi v es the motor parameters chosen for simulation, and T able 5 gi v es the simulation output. Figure 8(b) sho ws the motor torque. Since the Iq directly controls torque as from the PMSM fundamental torque equation, it can be inferred directly that the torque control is e xactly as e xpected. (a) (b) Figure 7. Speed and current comparison with reference: (a) speed comparison with reference and (b) v oltage transients (a) (b) Figure 8. Motor v oltage and torque comparison with reference: (a) motor v oltage and (b) motor torque T able 4. Motor parameters in simulation P arameter V alue P arameter V alue Number of pole pairs 7 Stator inductance per phase 87.678 uH Rated current 7.26 A Nominal base speed 3476 rpm Rated torque 0.3471 Nm Rated po wer 200 W Stator resistance per phase 0.293 Ω T able 5. Simulation output Simulation output V alue V oltage o v ershoot 5% T orque ripple 2.35% Settling time 0.25 s Sampling time 100 us 4. CONCLUSION Fuel cell-based electric v ehicles ha v e been modeled by means of a fuel cell battery system, which serv es as an optimum control system to re gulate fuel us age as well as char ge the battery under light load conditions. The simulations ha v e been done considering v arying load conditions in practical scenarios. Smart controllers can signicantly reduce torque ripple and re gulate speed. In FCEVs, torque and speed must be controlled ef ciently to achie v e the best results. The obtained speed responses conrm that the implemented control strate gy achie v es accurate and stable speed re gulation o v er a wide operating range. The motor speed consistently tracks the reference commands with minimal steady-state error , indicating ef fecti v e control action. Although small transient o v ershoots and oscillations are observ ed during sudden speed changes and direction re v ersals, the y are well damped and settle quickly , demonstrating good dynamic stability . The controller maintains smooth transitions between motoring, zero-speed, and re generati v e (re v erse) operation, which is critical for applications such as PMSM-based FCEVs. Ov erall, the res ults v alidate that the control scheme pro vides rob ust bidirectional speed control, f ast transient response, and reliable performance under v arying speed commands, making it suitable for practical traction dri v e applications. The future room for impro v ement is the reduction of harmonics of A C v oltage, and further reduction in torque-speed ripple. It is proposed that this can be further reduced by the use of ML techniques. T or que ripple r eduction in PMSM for FCEVs using ANFIS contr oller (Shilpa Rao Hosabettu) Evaluation Warning : The document was created with Spire.PDF for Python.
892 ❒ ISSN: 2088-8694 FUNDING INFORMA TION Authors state no funding in v olv ed. A UTHOR CONTRIB UTIONS ST A TEMENT This journal uses the Contrib utor Roles T axonomy (CRediT) to recognize indi vidual author contrib utions, reduce authorship disputes, and f acilitate collaboration. Name of A uthor C M So V a F o I R D O E V i Su P Fu Shilpa Rao Hosabettu ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Pushpa Rajesh V isw anathan ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ C : C onceptualization I : I n v estig ation V i : V i sualization M : M ethodology R : R esources Su : Su pervision So : So ftw are D : D ata Curation P : P roject Administrati on V a : V a lidation O : Writing - O riginal Draft Fu : Fu nding Acquisition F o : F o rmal Analysis E : Writing - Re vie w & E diting CONFLICT OF INTEREST ST A TEMENT Authors state no conict of interest. D A T A A V AILABILITY The data that support the ndings of this study are a v ailable from the corresponding author , [SRH], upon reasonable request. REFERENCES [1] Z. J. Baum, R. E. Bird, X. Y u, and J. Ma, “Lithium-ion battery rec ycling–o v ervie w of techniques and trends, ” A CS Ener gy Letter s , v ol. 7, no. 2, pp. 712–719, Feb . 2022, doi: 10.1021/acsener gylett.1c02602. [2] M . ˙ Inci, M. B ¨ uy ¨ uk, M. H. Demir , and G. ˙ Ilbe y , “ A re vie w and research on fuel cell ele ctric v ehicles: T opologies, po wer elect ronic con v erters, ener gy management methods, technical challenges, mark eting and future aspects, ” Rene wable and Sustainable Ener gy Re vie ws , v ol. 137, p. 110648, Mar . 2021, doi: 10.1016/j.rser .2020.110648. [3] P . Purnima and S. Jayanti, “Fuel processor -battery-fuel cell h ybrid dri v etrai n for e xtended range operation of passenger v ehicles, ” International J ournal of Hydr o g en Ener gy , v ol. 44, no. 29, pp. 15494–15510, Jun. 2019, doi: 10.1016/j.ijh ydene.2019.04.081. [4] M . Subbarao, K. Dasari, S. S. Duvvuri, K. R. K. V . Prasad, B. K. Narendra, and V . B. Murali Krishna, “Design, control and performance comparison of PI and ANFIS controllers for BLDC motor dri v en electric v ehicles, ” Measur ement: Sensor s , v ol. 31, 2024, doi: 10.1016/j.measen.2023.101001. [5] V . T . Ha and N. V an Hai, “ Adv anced control strate gy for electric v ehicle AFPMSMs using intelligent h ybrid controllers. ” 2025, Preprints. doi: 10.20944/preprints202503.2210.v1. [6] K. Chaudhary , P . Singh, and A. J. Singh, “Fuel cell input based PMSM motor dri v e for electric v ehicle applications, ” in 2023 IEEE 3r d International Confer ence on Smart T ec hnolo gies for P ower , Ener gy and Contr ol (STPEC) , Dec. 2023, pp. 1–6. doi: 10.1109/STPEC59253.2023.10431161. [7] M. H. Basappa and P . V isw anathan, “V arious control methods of permanent magnet synchronous motor dri v es in electric v ehicle: a technical re vie w , ” TELK OMNIKA (T elecommunication Computing Electr onics and Contr ol) , v ol. 20, no. 6, p. 1225, Dec. 2022, doi: 10.12928/telk omnika.v20i6.24236. [8] P . B, R. Kaur , and A. B. K umar Mukkapati, “Fuzzy logic based eld oriented control of PMSM for electric v ehicle, ” in 2024 International Confer ence on Signal Pr ocessing , Computation, Electr onics, P ower and T elecommunication (IConSCEPT) , Jul. 2024, pp. 1–6. doi: 10.1109/IConSCEPT61884.2024.10627897. [9] J. Huang, Y . Sui, Z. Y in, G. Liu, P . Zheng, and Y . Li, “Multiobjecti v e particle sw arm optimization design of permanent magnet machine for torque density impro v ement and torque ripple suppression, ” 2022 International P ower Electr onics Confer ence , IPEC-Himeji 2022-ECCE Asia , pp. 601–606, 2022, doi: 10.23919/IPEC-Himeji2022-ECCE53331.2022.9807208. [10] A. Nallathambi, S. Nallathambi, and B. Gopal Samy , “ Autonomous electric v ehicles and edge computing, ” Articial Intellig ence Applications in Battery Mana g ement Systems and Routing Pr oblems in Electric V ehicles , pp. 233–248, 2023, doi: 10.4018/978-1-6684-6631-5.ch011. [11] J. S. V . S. K umar and P . Mallikarjuna Rao, “Performance analysis of PID controller and sliding mode control for electric v ehicle applications in interlea v ed double boost con v erter , ” Lectur e Notes in Electrical Engineering , v ol. 702, pp. 47–59, 2021, doi: 10.1007/978-981-15-8439-8 5. [12] F . Labchir , A. El Aa, K. Benkirane, and M. Khaf allah, “Impro v ed direct torque control of dual three-phase permanent magnet synchronous motor , ” Advances in Science , T ec hnolo gy and Inno vation , pp. 3–10, 2024, doi: 10.1007/978-3-031-51796-9 1. Int J Po w Elec & Dri Syst, V ol. 17, No. 2, June 2026: 885–893 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Po w Elec & Dri Syst ISSN: 2088-8694 ❒ 893 [13] T . Jarin, S. Akkara, S. S. Sreeja Mole, A. Mani v annan, and A. Immanuel Selv akumar , “Fuel v ehicle impro v ement using high v oltage g ain in DC-DC boost con v erter , ” Rene wable Ener gy F ocus , v ol. 43, pp. 228–238, Dec. 2022, doi: 10.1016/j.ref.2022.09.008. [14] S. F arhani, E. M. Barhoumi, and F . Bacha, “Design and hardw are in v estig ation of a ne w conguration of an isolated DC-DC con v erter for fuel cell v ehicle, ” Ain Shams Engineering J ournal , v ol. 12, no. 1, pp. 591–598, Mar . 2021, doi: 10.1016/j.asej.2020.07.014. [15] J. C. Nustes, D. Piet ro P au, and G. Gruosso, “Modelling the eld oriented control applied to a 3-phase permanent magnet synchronous motor , ” Softwar e Impacts , v ol. 15, p. 100479, Mar . 2023, doi: 10.1016/j.simpa.2023.100479. [16] W . W u, F . Xie, G. Li, K. Liang, C. Qiu, and H. Jiang, “Research on direct torque control based on RZVSVPWM of PMSM, ” in 2019 14th IEEE Confer ence on Industrial Electr onics and Appl ications (ICIEA) , Jun. 2019, pp. 2507–2511. doi: 10.1109/ICIEA.2019.8833809. [17] T .-L. Le, “ A rob ust control strate gy for ef fecti v e eld-oriented control of PMSMs, ” Engineering , T ec hnolo gy & Applied Science Resear c h , v ol. 14, no. 6, pp. 18469–18475, Dec. 2024, doi: 10.48084/etasr .8893. [18] P . Ramesh, M. Uma v athi, C. Bharatiraja, G. Ramanathan, and S. Athikkal, “De v elopment of a PMSM motor eld-oriented control algorithm for electrical v ehicles, ” Materials T oday: Pr oceedings , v ol. 65, pp. 176–187, 2022, doi: 10.1016/j.matpr .2022.06.080. [19] E. K. Beser , “Electrical equi v alent circuit for modelling permanent magnet synchronous motors, ” J ournal of Electrical Engineering , v ol. 72, no. 3, pp. 176–183, Jun. 2021, doi: 10.2478/jee-2021-0024. [20] D. Mohanraj, J. Gopalakrishnan, B. Chokkaling am, and L. Mihet-Popa, “Critical aspects of electric motor dri v e controllers and mitig ation of torque ripple—re vie w , ” IEEE Access , v ol. 10, pp. 73635–73674, 2022, doi: 10.1109/A CCESS.2022.3187515. [21] P . K. P athak, A. K. Y ada v , S. P admanaban, P . A. Alvi, and I. Kamw a, “Fuel cell-based topologies and multi-input DC–DC po wer con v erters for h ybrid electric v ehicles: A comprehensi v e re vie w , ” IET Gener ation, T r ansmi ssion & Distrib ution , v ol. 16, no. 11, pp. 2111–2139, Jun. 2022, doi: 10.1049/gtd2.12439. [22] S. Da v e, M. V asa v ada, S. S. N, and B. K. A, “Speed control of PMSM for wind turbine applicati on: a comparati v e e v aluation of FOC and DTC strate gies, ” The IUP J ournal of Electrical & Electr onics Engineering , v ol. 15, no. 4, p. 70, 2022. [23] M. P ala Prasad Reddy , S. Si v a Prasad, T . V enu Gopal, G. Madhusudhana Rao, and N. Sreeramula Reddy , “Optimisation of torque ripples in permanent magnet synchronous motor using h ysteresis current controller , ” E3S W eb of Confer ences , v ol. 472, p. 01001, Jan. 2024, doi: 10.1051/e3sconf/202447201001. [24] K. K. Pedapenki, S. P . Gupta, and M. K. P athak, “Shunt acti v e po wer lter with articial intelligent controllers, ” in 2015 International Confer ence on Contr ol, Instrumentation, Communication and Computational T ec hnolo gies (ICCICCT) , Dec. 2015, pp. 74–77. doi: 10.1109/ICCICCT .2015.7475252. [25] K. M. N. C. kumar Reddy and D. N. Kanag asabai, “Performance analysis of ANFIS-PID controller based speed re gulation and harmonic reduction in BLDC motor application, ” International J ournal of Electrical and Electr onics Resear c h , v ol. 12, no. 1, pp. 187–194, Mar . 2024, doi: 10.37391/IJEER.120127. [26] N. W alia, H. Singh, and A. Sharma, “ ANFIS: adapti v e neuro-fuzzy inference system- a surv e y , ” International J ournal of Computer Applications , v ol. 123, no. 13, pp. 32–38, Aug. 2015, doi: 10.5120/ijca2015905635. [27] K. K umar , M. Das, and A. K. Karn, “ ANFIS rob ust control application and analysis for load frequenc y control with nonlinearity , ” J ournal of Electrical Systems and Information T ec hnolo gy , v ol. 11, no. 1, 2024, doi: 10.1186/s43067-024-00175-9. [28] A. Intidam et al. , “De v elopment and e xperimental implementation of optimized PI-ANFIS controller for speed control of a brushless DC motor in fuel cell electric v ehicles, ” Ener gies , v ol. 16, no. 11, p. 4395, May 2023, doi: 10.3390/en16114395. [29] V . T . Ha and N. V an Hai, “ Adapti v e neuro-fuzzy control of a single-sided AFPMSM motor for electric v ehicle applications, ” SSRG International J ournal of Electrical and Electr onics Engineering , v ol. 11, no. 4, pp. 118–129, 2024, doi: 10.14445/23488379/IJEEE-V11I4P113. BIOGRAPHIES OF A UTHORS Shilpa Rao Hosabettu is an assistant professor in the Electrical Engineering Department at the AMC Colle ge of Engineering, Beng aluru, India, since 2017. She recei v ed her B.E., M.T ech. de gree in Ele ctrical Engineering from V isv esw araiah T echnological Uni v ersity , Karnataka, in 2006 and 2011, respecti v ely . Currently pursuing a Ph.D. since 2022 in the eld of fuel cell electric v ehicles. Her research interests include the eld of EVs, po wer electronics, motor dri v es, rene w able ener gy , and intelligent controllers. She can be contacted at email: shilpaach16@gmail.com. Pushpa Rajesh V iswanathan has been serving as a professor in the Electrical Engineering Department and Placement Of cer at Jain (Deemed to be Uni v ersity), Beng aluru, India, since 2017. He has completed a Ph.D. from Anna Uni v ersity , in Electrical Engineering, specializing in Po wer Electronics and Special Electrical Dri v es. He has recei v ed the International Best Research A w ard for the year 2018-2019, instituted by SDF International, London, UK. He also recei v ed national a w ards lik e best aca demic researcher , outstanding f a culty a w ard, best f aculty a w ard, and best placement coordinator from v arious research or g anizations. He has published man y papers in Internati onal and National Journals with a high impact f actor . He can be contacted at email: v .pushparajesh@jainuni v ersity .ac.in. T or que ripple r eduction in PMSM for FCEVs using ANFIS contr oller (Shilpa Rao Hosabettu) Evaluation Warning : The document was created with Spire.PDF for Python.