Inter national J our nal of A pplied P o wer Engineering (IJ APE) V ol. 14, No. 3, September 2025, pp. 513 ∼ 521 ISSN: 2252-8792, DOI: 10.11591/ijape.v14.i3.pp513-521 ❒ 513 Ev aluation of sensorless VF-MRAS and FOC-MRAS of IM electrical dri v e system Moustapha Diop, Abdoulay e K ebe, Ibrahima Guey e Department STI, Ecole Normale Sup ´ erieure d’Enseignement T echnique et Professionnel, Uni v ersite Cheikh Anta Diop, Dakar , Sene g al Laboratoire L3EPI, Ecole Sup ´ erieure Polytechnique, Uni v ersit ´ e Cheikh Anta Diop, Dakar , S ´ en ´ eg al Article Inf o Article history: Recei v ed Jul 20, 2024 Re vised Dec 26, 2024 Accepted Jan 19, 2025 K eyw ords: Field-oriented control Induction motor MRAS Scalar V/f Sensorless control ABSTRA CT This paper e v aluates the performance of sensorless v ector and scalar control methods, namely eld-oriented control-based model reference adapti v e system (FOC-MRAS) and v oltage frequenc y-based model reference adapti v e system (VF-MRAS), applied to an induction motor (IM) dri v en by a space v ector mod- ulation in v erter . In motorized systems, con v entional control m ethods use me- chanical sensors, which can be cumbersome and costly . T o o v ercome these limi- tations, sensorless control techniques based on speed estimation ha v e been intro- duced. In this paper , MRAS-based sensorless speed control for IM dri v es using rotor ux is used. This adapti v e sys tem uses a reference model based on rotor ux and implements closed-loop control. The estimated speed deri v ed from the current and v oltage models is compared to the desired speed and adjusted by the proportional-inte gral (PI) controllers. The performances of the approaches are e v aluated in terms of speed re gulation and minimization of electromagnetic torque and rotor ux ripples, through a comparati v e analysis of sensor and sen- sorless controls under v arious operating conditions, including v ariable loads and speed re v ersal. The simulation results obtained, using consistent criteria for both methods, conrm the ef fecti v eness of sensorless control. This is an open access article under the CC BY -SA license . Corresponding A uthor: Moustapha Diop Department STI, Ecole Normale Sup ´ erieure d’Enseignement T echnique et Professionnel Uni v ersite Cheikh Anta Diop Dakar , Sene g al Email: moustapha17.diop@ucad.edu.sn 1. INTR ODUCTION Induction motors are the most widely used v ariable speed dri v es because of their rob ustness, rel ia- bility , simplicity and straightforw ard control process [1]-[5]. Ho we v er , to achie v e high performance with the induction motors (IMs), it is necessary to select an ef fecti v e control strate gy tailored to specic applications. F or induction motor (IM) control, se v eral techniques ha v e been proposed, such as traditional direct torque control (DTC), scalar control (V/f), and eld-oriented control (FOC) [6]. The DTC is characterized by good dynamic torque response and easy to implem ent. Ho we v er , both torque and electromagnetic ux e xhibit ripples [7]. F or the FOC, it is dis tinguished by its good dynamic response [8]. It maintains ef cienc y o v er a wide speed range and tak es into account the load torque v ariations. As for scalar control, it is simple, less costly , allo ws slo w speed v ariation and is easy to implement, b ut it e xhibits poor dynamic performance and is typically used in lo w-cost, and lo w-performance system dri v es. J ournal homepage: http://ijape .iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
514 ❒ ISSN: 2252-8792 The abo v e-mentioned controls ha v e adv antages and limitations depending on the application. As the ability to operate at v ariable speed is a x ed objecti v e in this w ork, the FOC and v oltage frequenc y (VF) con- trols are used with the aim of impro ving their performance by e xplori ng alternati v e implementation strate gies. Ho we v er , ef cient e x ecution of these closed-loop strate gies requires speed information. T o this end, the use of mechanical sensors has often been necessary . Ho we v er , their use presents certain limitations, such as increased de vice cost, reduced reliability and lo w noise immunity . F or this reason, recent research has lar gely focused on sensorless dri v es. The sensorless control enables speed to be reconstructed from the IM model. Se v eral techniques are used, including open-loop technique, sliding mode observ er (SMO), e xtended Kalman lter observ er , and model refer ence adapti v e syst ems (MRAS) observ er [9]-[13]. Open-loop estimators ha v e al w ays at tracted at- tention due to their simplicity , b ut the y e xhibit lo w rob ustness [14]. The Kalman method is widely used for ux and speed estimation [15]-[17]. Although the estimated v ariables are well ltered, the method is often impractical at lo w speeds, sensiti v e to parameter v ariations, and dif cult to implement. As for t he SMO, it is rob ust and simple to implement. Ho we v er , its use is limited by oscillations that can lead to instability , and com- putational comple xity is signicantly increased. The Luenber ger technique yields good results b ut is sensiti v e to disturbances and parameter v ariations, and is also challenging to synthesize [18]. F or the MRAS technique, widely used for estimating motor speed or motor resistance, it is based on comparing a reference model that is independent of speed to an adjustable adapti v e model that depends on speed [19]. The error is corrected by an adaptation mechanism using a proportional-inte gral (PI) controller that determines the estimated motor speed [20]. This method pro vides good results, is straightforw ard to implement has good accurac y and requires less computational ef fort compared to the mentioned estimator techniques [21]. In t his w ork, the ux-based MRAS technique is adopted for speed estimation applied to FOC and VF dri v es. In this respect, this paper proposes an alternati v e methodology for estimating speed via MRAS of a three-phase IM dri v en by a space v ector modulation (SVM) v oltage source in v erter (VSI). The SVM strate gy is emplo yed to achie v e reduced harmonic distortion, and minimized switching losses [22]-[24]. The proposed control methods are e v aluated in terms of speed re gulation, as we ll as reductions in torque and rotor ux ripple. A comparati v e study of sensor and sensorless control strate gies w as conducted under dif ferent operating conditions. This research is di vided into four sections. F ollo wi ng an introduction in section 1, section 2 presents the mathematical models of the dri v e system components and the proposed control strate gies. Section 3 presents and discusses the simulation results. Finally , the paper concludes in section 4. 2. MA TERIALS AND METHOD In this paper , the performances of sensorless FOC and VF controls, based on the closed-loop MRAS technique are studied. The dri v e system consists of a three-phase IM dri v en by a SVM VSI. The IM models in (d-q) and (alpha-beta) reference frames are de v eloped initially . This model will be used to establish equations for estimating speed using the MRAS method. The parameters of the controllers used are calculated to achie v e f ast adapti v e loop that is independent of load torque v ariation. The models and control strate gies are then implemented and simulated using the MA TLAB-Simulink softw are tool under v arious load conditions. 2.1. Induction motor modeling In this w ork, the dq and α , β models are used. The dq model described by (1)-(4) al lo ws for estab- lishing con v entional controls, while the α , β model is essential for sensorless control. The dq model can be subsequently transformed into stationary coordinates ( α , β ) using the in v erse P ark transformation. v ds = R s i ds + L s 1 − L 2 m L r L s i ds dt + L m L r ϕ dr dt − L s 1 − L 2 m L r L s ω s i q s − ω s L m L r ϕ q r (1) v q s = R s i q s + L s 1 − L 2 m L r L s i q s dt + L m L r ϕ q r dt − L s 1 − L 2 m L r L s ω s i q s + ω s L m L r ϕ dr (2) ϕ dr = − T r ϕ dr dt + T r ω r Φ q r + L m i ds ϕ q r = − T r ϕ q r dt − T r ω r Φ dr + L m i q s (3) J d Ω dt = T em − T L − f r Ω T em = p L m L r ( ϕ dr i q s − i ds ϕ q r ) (4) Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 513–521 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 ❒ 515 2.2. SVM in v erter modeling The diagram of the VSI comprises three branches and si x switches whose switching depends is de- termined by the SVM scheme depicted in Figure 1. W ith the SVM, the VSI require a v olt age v ector space, i.e., 8 poss ible switching states that are then transformed into v oltage v ectors in α , β frame corresponding to well-dened sequences [25]. Re g arding the VSI model, only six of the eight v oltage v ectors (001), (101), (100), (110), (010), and (011) are acti v e as control elements. The v ectors (000) and (111) are null. The acti v e switching states, the line v oltages, and the tw o-phase v oltage le v els v α and v β are summarized in T able 1. Figure 1. General diagram of SVM in v erter T able 1. Switching states and v oltages table States v a v b v c v α v β 000 0 0 0 0 0 100 2 V dc / 3 − V dc / 3 − V dc / 3 2 V dc / 3 0 110 V dc / 3 V dc / 3 − 2 V dc / 3 V dc / 3 √ 3 V dc / 3 010 - V dc / 3 2 V dc / 3 − V dc / 3 − V dc / 3 √ 3 V dc / 3 011 - 2 V dc / 3 V dc / 3 V dc / 3 - 2 V dc / 3 0 001 - V dc / 3 - V dc / 3 2 V dc / 3 - V dc / 3 - √ 3 V dc / 3 101 V dc / 3 -2 V dc / 3 V dc / 3 V dc / 3 - √ 3 V dc / 3 111 0 0 0 0 0 2.3. Speed obser v er based on MRAS The principle for controls relies on comparing reference model (RM) and adapti v e model (AM), as depicted in Figure 2 [19]. The RM corresponds to the rotor ux calculated from current and v oltage feedback, while the AM corresponds to the rot or ux deri v ed from rotor equations. The error between these models is used to dri v e an appropriate adaptation m echanism using a PI controller that generates the estimated speed [20]. Rotor ux es ( Φ ∗ α , Φ ∗ β ) of the RM and rotor ux es of adapti v e model ( ˆ ϕ α , ˆ ϕ β ) are gi v en by: dϕ ∗ αr dt = L r L m v αs − R s i αs − L s 1 − L 2 m L r L s di αs dt (5) dϕ ∗ β r dt = L r L m v β s − R s i β s − L s 1 − L 2 m L r L s di β s dt (6) d ˆ ϕ αr dt = − 1 T r ˆ ϕ αr + L m T r i sα − ω ˆ Φ β r (7) d ˆ ϕ β r dt = − 1 T r ˆ ϕ β r + L m T r i sβ + ω ˆ Φ αr (8) Evaluation of sensorless VF-MRAS and FOC-MRAS of IM electrical drive system (Moustapha Diop) Evaluation Warning : The document was created with Spire.PDF for Python.
516 ❒ ISSN: 2252-8792 v αs and v β s denote the stator v oltages, while i αs and i β s represent the stator currents, all e xpressed in the α β frame, which are used as feedback inputs to the motor control system as repreented in Figure 2. According to the error signals used as input to the PI controller , the estimated speed is determined. It can be e xpressed as (9). ω = ϵ k 1 + k 2 s = k 1 + k 2 s ˆ ϕ α ϕ ∗ β − ˆ ϕ β ϕ ∗ α (9) The PI controller g ains k 1 and k 2 are tuned based on the MRAS closed-loop transfer function (CL TF), ob- tained by linearizing the adaptation model. Ass uming perfect ux orientation with constant ux and current magnitudes ( ϕ ∗ α = Φ r and ϕ ∗ β = 0 ), the (10) gi v es the linearized e xpression of the error . δ ϵ = Φ ∗ β δ ˆ Φ α + ˆ Φ α δ Φ ∗ β − ˆ Φ β δ Φ ∗ α − Φ ∗ α δ ˆ Φ β = − Φ ∗ α δ ˆ Φ β (10) The linearization of the adapti v e model e xpressions gi v es as (11). s δ ˆ Φ α = − 1 T r δ ˆ Φ α + ω δ ˆ Φ β s δ ˆ Φ β = − 1 T r δ ˆ Φ β − ω δ ˆ Φ α + ˆ Φ α δ ω (11) The substitution of the pre vious equations gi v es as (12). δ ˆ Φ β = − Φ r ( T r s + 1) ( T r s + 1) 2 + ( ω T r ) 2 δ ω (12) Since δ ϵ = − ϕ ∗ α δ ˆ ϕ β , the open-loop transfer function is as (13). G 0 ( s ) = δ ϵ δ ω = − Φ 2 r ( T r s + 1) ( T r s + 1) 2 + ( ω T r ) 2 (13) Using the PI controller , the transfer function G f ( s ) is e xpressed. Pol e-placement-tuned PI controllers g ains are calculated where ζ and ω n est are the damping coef cient and the natural frequenc y . G f ( s ) = ( k 1 s + k 2 ) ϕ 2 r s 2 + ( 1 T r + Φ 2 r k 1 ) s + k 2 ϕ 2 r (14) k 1 = 2 ζ T r ω n est − 1 T r Φ 2 r k 2 = ω 2 n est Φ 2 r (15) Reference Model Adaptati v e Model v αs v β s x i αs i β s x - + k 1 + k 2 s ϕ ∗ α ϕ ∗ β ˆ ϕ β ˆ ϕ α ω ω ϵ Figure 2. MRAS structural diagram 2.4. VF-MRAS Scalar control aims to k eep the magnetic ux constant at its maximum v alue by k eeping the v olt- age/frequenc y rat io constant and boosting it to mini mize the v ol tage drop across the st ator resistance at lo w speed. F or scalar control, torque is controlled by slip v ariation and its e xpression is gi v en by (16). The control system of the closed-loop V/f-MRAS depicted in Figure 3 includes an outer -loop estimator to determine the Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 513–521 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 ❒ 517 motor speed ( ω ). Acc ording to the scheme, the estimated speed is compared with a reference speed ( ω ∗ ). The resulting error is fed i nto the PI speed controller , which calculates a slip angular speed that i s then added to the estimated motor speed to deri v e the frequenc y . Figure 4 sho ws the synthesis of speed corrector . The controller g ains can be calculated using (17). T em − max = 3 p 2 R r L 2 m L 2 s V 2 n ω 2 n × ω sl (16) k p = 2R r 3p L 2 s L 2 m ω 2 n V 2 n (2Jki ζ ω c − f r ) k i = 2R r 3p L 2 s L 2 m ω 2 n V 2 n J ω 2 c (17) 2.5. FOC-MRAS F or the FOC technique, the rotor ux and motor torque are controlled respecti v ely by the direct and the quadrature current components. F or determining the rotor position, the sensorless MRAS technique is used to estimate rotor position information in real time. Figure 5 illustrates the FOC-MRAS control scheme. Three phase in v erter SVPWM References IM v αs v β s V ∗ - + ω s ω ∗ PI ω est MRAS abc to α β I V v αs v β s i αs i β s + + Three phase in v erter Figure 3. Synoptic diagram of the proposed V/f-MRAS strate gy 1 Js + f r 3p 2R r L 2 m L 2 s V 2 n ω 2 n k p + k i s T em ω sl T L - + Ω ∗ - + Ω Figure 4. Synthesis of the speed corrector for scalar control Three phase in v erter SVPWM dq to α β Decoupling T orque controller Flux controller Speed controller Field weak ening IM v αs v β s v q s v ds - + ω ∗ - + - + abc to α β abc to dq I V MRAS v αs v β s i αs i β s Estimation ω s and θ s θ s ω s ϕ r i q s i ds ω est Figure 5. Synoptic diagram of the proposed FOC-MRAS strate gy Evaluation of sensorless VF-MRAS and FOC-MRAS of IM electrical drive system (Moustapha Diop) Evaluation Warning : The document was created with Spire.PDF for Python.
518 ❒ ISSN: 2252-8792 The control structure consists of an outer speed controller in a closed-loop and a series of inner current controllers. On the basis of a cascade control, the current controllers dynamics is suf ciently f ast with respect to speed loop. The diagrams is represented in Figure 6. The controller g ains can be calculated using (18). k i Ω = J ω 2 c Ω k p Ω = 2 J ζ ω c Ω − f r (18) ζ is damping coef cient, and ω c Ω the o wn pulsation. 1 Js + f r k pΩ + k iΩ s T em T L - + Ω ∗ - + Ω Figure 6. Synthesis of the speed corrector for FOC control 3. RESUL TS AND DISCUSSION The FOC-MRAS and V/f-MRAS controls are e v aluated through simulation using M A TLAB/Simulink and a 2.2 kW IM in this section. The performances are assessed by comparing the results with those obtained from sensor -based controls under v arious operating conditions. The rst test w as conducted to v erify the performance in terms of reference speed tracking. The results are presented in Figures 7 to 11. F or each case, the reference speed w as set to 314 rad/s, with a load torque of 3.7 Nm (half load) applied at t = 3 s and the rated load of 7.37 Nm at t = 5 s. A second test w as performed to assess the strate gies under re v erse operation mode. According to results represented respecti v ely in Figures 8 and 11, the torque follo ws the v ariation of the load with a slight of fset due to the mechanical parameters of the motor , and e xhibits noise associated to uctuations in the estimated speed. Additionally , the magnetic ux is properly oriented to w ards the direct axis. 0 1 2 3 4 5 6 7 8 Times (s) 0 100 200 300 Speeds   (rd/s) ref est 3 4 5 6 310 315 Figure 7. Reference, real, and estimated rotor speeds VF-MRAS 0 1 2 3 4 5 6 7 8 Times (s) 0 5 10 Torques (N.m) T e m T L Figure 8. Load and motor torques VF-MRAS Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 513–521 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 ❒ 519 0 1 2 3 4 5 6 7 8 Times (s) -400 -200 0 200 400 Speeds   (rd/s) ref est 4 6 295 300 305 310 315 Figure 9. Reference, real, and estimated rotor speeds FOC-MRAS 0 1 2 3 4 5 6 7 8 Times (s) 0 0.5 1 Fluxes  dq  (Wb) d q Figure 10. Rotor ux es FOC-MRAS 0 1 2 3 4 5 6 7 8 Times (s) 0 5 10 Torques (N.m) T e m T L Figure 11. Load and motor torques FOC-MRAS T o test the rob ustness of the sensorless strate gies ag ainst disturbances, the system is simulated using a v ariable load prole. The m easured, estimated, and reference speeds are sho wn in the Figures 7 and 9. The results sho w that the proposed MRAS speed estimator enables accurate tracking of the actual motor speed, with a f ast response time. The zooms around the instants at 3 s and 5 s, when loads are applied, demonstrate accurate tracking that reects the ef fecti v eness and rob ustness of the control in rejecting disturbances, despite proportional v ariations in the applied load. Moreo v er , the results sho w the speed estimation error . It indi- cates the instantaneous di f ference between the actual speeds and t he estimated speeds, which is almost zero. T o further e v aluate the strate gy , at t = 6.5 s, a re v ersal of rotation direction follo wed by a speed reference of +314 rad/s and -314 rad/s is performed to test the performance in these re gimes. Figure 9 sho ws similar be- ha vior , with a slight delay observ ed in the tr ansient phases. Thus, the MRAS technique enables accurate speed estimation in these re gimes, ensuring good agreement with the reference and near -perfect approximation with the estimated speed. In summary , the gures presented abo v e illustrate a series of simulation results aimed at e v aluating the performance of the FOC-MRAS and VF-MRAS controls. T ests include load v ariations, re v ersals of rotation direction, and detailed analyses of motor ux and t orque. The results sho w that the estimated speed using MRAS control enables accurate tracking of the motor’ s actual speed, with f ast response and good rob ustness to disturbances. Ov erall, the control strate gies demonstrate an ef fecti v e ability to maintain motor speed close to reference despite v arying load and operating condi tions. The strate gies e xhibit e xcellent reference tracking, and the speed transient re gime is characterized by f ast response times without signicant o v ershoot. 4. CONCLUSION The conclusion of this paper summarizes the v alidation and benets of a sensorless v ector and scal ar IM dri v e based on the MRAS method with speed estimation. The proposed control scheme is simulated under dif ferent operation conditions, including load v ariation, and rotation in v ersion. Ov erall, the simulation results highlights the rob ustness and impro v ed performance of the proposed control strate gies. The results emphasize enhanced speed tracking, instantaneous speed estimation, and reduced ux and torque ripples. The de v eloped control schemes sho w promising potential for practical applications, as sensorless induction motor control inte grated with speed esti mation mitig ates the disadv antages associated with speed sensors, such as reduced reliability , noise sensiti vity , increased cost, weight, and system comple xity of the dri v e system. FUNDING INFORMA TION The authors declare that no funding w as recei v ed for this research. A UTHOR CONTRIB UTIONS ST A TEMENT This journal uses the Contrib utor Roles T axonomy (CRediT) to recognize indi vidual author contrib u- tions, reduce authorship disputes, and f acilitate collaboration. Evaluation of sensorless VF-MRAS and FOC-MRAS of IM electrical drive system (Moustapha Diop) Evaluation Warning : The document was created with Spire.PDF for Python.
520 ❒ ISSN: 2252-8792 Name of A uthor C M So V a F o I R D O E V i Su P Fu Moustapha Diop ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Abdoulaye K ebe ✓ ✓ ✓ ✓ ✓ ✓ ✓ Ibrahima Gue ye ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ 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 Administration 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 The authors declare that the y ha v e no kno wn competing nancial interests or personal rela tionships that could ha v e appeared to inuence the w ork reported in this paper . D A T A A V AILABILITY Data a v ailability is not applicable to this paper as no ne w data were created or analyzed in this study . REFERENCES [1] A. Sahu, K. B. Mohanty , R. N. Mishra, “De v elopment and e xperimental realization of an adapti v e neural-based discrete model predicti v e direct torque and ux controller for induction motor dri v e, ” Applied Soft Computing , v ol. 108, Sep. 2021, doi: 10.1016/j.asoc.2021.107418. [2] A . Gundogdu, H. Kizmaz, R. Celik el, and M. Y ilmaz, “Speed sensorless adapti v e po wer control for photo v oltaic-fed w ater pump using e xtended Kalman Buc y lter , ” Ener gy Reports , v ol. 10, pp. 1785-1795, 2023, doi: 10.1016/j.e gyr .2023.08.036. [3] K. Saad, K. Abdellah, H. Ahmed, and A. Iqbal, “In v estig ation on SVM-backsteppi ng sensorless control of v e-phase open-end winding induction motor based on model reference adapti v e system and parameter estimation, ” Engineering Science and T echnol- ogy , an International Journal , v ol. 22, no. 4, pp. 1013-1026, 2019, doi: 10.1016/j.jestch.2019.02.008. [4] S. A. Bernarz and M. Dybk o wski, “Estimation of the induction motor stator and rotor resistance usi ng acti v e and reacti v e po wer based model reference adapti v e system estimator , ” Applied Sciences , v ol. 9, no. 23, 2019, doi: 10.3390/app9235145. [5] H . Hadla and F . Santos, “Perform ance comparison of eld-oriented control, direct torque control, and model-predi cti v e control for SynRMs, ” Chinese Journal of Electrical Engineering , v ol. 8, no. 1, pp. 24-37, Mar . 2022, doi: 10.23919/CJEE.2022.00000. [6] N. El Ouanjli et al. , “Impro v ed twelv e sectors DTC strate gy of induction motor dri v e using backstepping speed controller and P-MRAS stator resistance identication-design and v alidation, ” Ale xandria Engineering Journal , v ol. 80, pp. 358-371, 2023, doi: 10.1016/j.aej.2023.08.077. [7] Y . Ren, R. W ang, S. J. Rind, P . Zeng, and L. Jiang, “Speed sensorless nonlinear adapti v e control of induction motor using combined speed and perturbation observ er , ” Control Engineering Practice , v ol. 123, 105166, 2022, doi: 10.1016/j.conengprac.2022.105166. [8] Y . Zahraoui, A. Bennassar , M. Akherraz, and A. Essa lmi, “Indirect v ector control of induction motor using an e xtended Kalman observ er a nd fuzzy l ogic controllers, ” 2015 3rd International Rene w able and Sustainable Ener gy Conference (IRSEC) , Morocco, Dec. 2015, doi: 10.1109/IRSEC.2015.7455046. [9] F . Z. Latrech, A. B. Rhouma, and A. Khedher , “FPGA implementation of a rob ust DTC-SVM based sliding mode ux ob- serv er for a double star induction motor: Hardw are in the loop v alidation, ” Microelectronics Reliability , v ol. 150, 2023, doi: 10.1016/j.microrel.2023.115118. [10] J. Gamazo-Real, V . Mart ´ ınez-Mart ´ ınez, and J. Gomez-Gil, “ ANN-based position and speed sensorless estimation for BLDC motors, ” Measurement , v ol. 188, 2022, doi: 10.1016/j.measurement.2021.110602. [11] C. U. Reddy , K. K. Prabhakar , A. K. Singh, and P . K umar , “Speed estimation technique using modied stator current error -based MRAS for direct torque controlled induction motor dri v es, ” in IEEE Journal of Emer ging and Selected T opics in Po wer Electronics , v ol. 8, no. 2, pp. 1223-1235, Jun. 2020, doi: 10.1109/JESTPE.2019.2901523. [12] A. Bo yar , E. Kabalci, and Y . Kabalci, “Sensorless speed controller of an induction motor with MRAS-based model predicti v e control, ” Computers and Electrical Engineering , v ol. 118, 2024, doi: 10.1016/j.compeleceng.2024.109350. [13] M. T ouam, M. Chenaf a, S. Chekroun, and R. Salim, “Sensorless nonlinear sliding mode control of the induction machine at v ery lo w speed using FM-MRAS observ er , ” International Journal of Po wer Electronics and Dri v e Systems , v ol. 12, no. 4, pp. 1987-1998, doi: 10.11591/ijpeds.v12.i4.pp1987-1998. [14] F . Lima, W . Kais er , I. N. da Silv a, and A. A. A. de Oli v eira, “Open-loop neuro-fuzzy speed estimator applied to v ector and scalar induction motor dri v es, ” Applied Soft Computing , v ol. 21, pp. 469-480, 2014, doi: 10.1016/j.asoc.2014.03.044. [15] J. Li, “Model predicti v e control for e xtended Kalman lter based speed sensorless induction motor dri v es, ” 2016 IEEE Applied Po wer Electronics Conference and Exposition (APEC) , USA, Mar . 2016, pp. 2770-2775, doi: 10.1109/APEC.2016.7468256. [16] S. Gue ye, L. Thia w , M. F . Ndiaye, I. Ngom, M. Diop, and E. H. M. Ndiaye, “ A sensorless speed control of DC motor based on an adapti v e reference model, ” 4th Biennial International Conference on Nascent T echnologies in Engineering (ICNTE) , India, 2021, pp. 1-5, doi: 10.1109/ICNTE51185.2021.9487787. Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 513–521 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 ❒ 521 [17] K. Horv a th, “Comparison of e xtended and unscented Kalman lters with and without using mechanical model for speed sensorless control of induction machines, ” 2023 18th Conference on Electrical Machines, Dri v es and Po wer Systems (ELMA) , Bulg aria, 2023, pp. 1-4, doi: 10.1109/ELMA58392.2023.10202302. [18] M. V idlak, L. Gorel, and P . Mak ys, “Performance e v aluation, analysis, and comparison of the back-EMF-based sensorless FOC and stable V/f control for PMSM, ” 2022 International Symposium on Po wer Electronics, Electrical Dri v es, Automation and Motion (SPEED AM) , Italy , 2022, pp. 318-323, doi: 10.1109/SPEED AM53979.2022.9842027. [19] J. Mohanalakshmi, and H. N. Suresh, “Sensorless speed estimation and v ector control of an induction motor dri v e using model reference adapti v e control, ” 2015 International Conference on Po wer and Adv anced Control Engineering (ICP A CE) , India, Aug. 2015, pp. 377-382, doi: 10.1109/ICP A CE.2015.7274976. [20] S. Dhundhara, P . K umar , and Y . P . V erma, “Sensor less speed control of PMSM using space v ector pulse width modulation based on MRAS method, ” 015 2nd International Conference on Recent Adv ances in Engineering & Computational Sciences (RAECS) , India, Dec. 2015, doi: 10.1109/RAECS.2015.7453408. [21] Y . B. Zbede, S. M. Gadoue, and D. J. Atkinson, “Model predicti v e MRAS estimator for sensorless induction motor dri v es, ” IEEE T ransactions on Industrial Electronics , v ol. 63, no. 6, pp. 3511-3521, 2016, doi: 10.1109/TIE.2016.2521721. [22] P . Ramasamy , and V . Krishnasamy , “SVPWM control strate gy for a three phase v e le v el dual in v erter fed open-end winding induction motor , ” ISA T ransactions , v ol. 102, pp. 105-116, 2020, doi: 10.1016/j.isatra.2020.02.034. [23] T . Arao ye, E. C. Ashigwuik e, A. C. Ade yemi, S. V . Egoigwe, N. G. Ajah, and E. Eronu, “Reduction and control of harmonic on three-phase s quirrel cage induction motors with v oltage source in v erter (VSI) using ANN-grasshopper optimization shunt acti v e lters (ANN-GOSAF), ” Scientic African , v ol. 21, 2023, doi: 10.1016/j.sciaf.2023.e01785. [24] Q. T an, L. Mao, Y . Cai, B. Zhang, and Z. Ruan, “Comparati v e e v aluation and analysis of GaN-based VSIs and CSIs, ” Ener gy Reports , v ol. 9, no. 1, pp 568-576, 2023, doi: 10.1016/j.e gyr .2022.11.082. [25] F . Shira v ani, P . Alk orta, J . A. Cortajarena, and O. Barambones, “ An impro v ed predicti v e current control for IM dri v es, ” Ain Shams Engineering Journal , v ol. 14, no. 8, 2023, doi: 10.1016/j.asej.2022.102037. BIOGRAPHIES OF A UTHORS Moustapha Diop holds a Ph.D. in Electrical Systems and Rene w able Ener gies. He is cur - rently Associate Professor in the Department of Science and Industrial T echnology at the Higher Normal School of T echnical and V ocational Education (ENSETP) at Cheikh Anta Diop Uni v er - sity in Dakar , Sene g al, and a permanent researcher in the Ener gy , W ater , En vironment and Indus- trial Processes Laboratory (L 3EPI). His current research focuses on po wer con v erters, control and modeling systems, and rene w able ener gies. He is the author or co-author of se v eral papers, pub- lished in international scientic journals and conference proceedings. He can be contacted at email: moustapha17.diop@ucad.edu.sn. Abdoulay e K ebe is Professor in the Department of Science and Industrial T echnology at the Higher Normal School of T echnical and V ocational Education (ENSETP) at Cheikh Anta Diop Uni v ersity in Dakar , Sene g al. He holds a Ph.D. in Ph ysics from the Uni v ersity of P aris S ud in 2013, and a master’ s de gree in Analysis, Design and Research in the Field of Engineering T echnologies in Education (A CREDITE) at the Uni v ersity of Cer gy Pontoise in 2016. His research is mainly oriented to w ards rene w able ener gies. He is author of se v eral publications in the eld of ener gy con v ersion. He is currently Director of ENSETP . He can be contacted at email: abdoulaye.k ebe@ucad.edu.sn. Ibrahima Guey e holds a Ph.D. in Automation, Production, Signal and Image, and Cog- niti v e Engineering from the Uni v ersity of Bordeaux. Currently , He is an Associate Professor at the Higher National School of T echnical and V ocational Education. In addition to his pedagogical acti vities, he is af liated with the Ener gy , W ater , En vironment, and Industrial Processes Laboratory (L3EPI) at the Polytechnic Superior School of Dakar , where he collaborate with a team of researchers on the design of electronic systems in v olv ed in the ener gy chain of photo v oltaic solar systems. His aim is to optimize the transfer of electrical ener gy produced and promote the local manuf acturing of electronic systems. He can be contacted at email: ibrahima64.gue ye@ucad.edu.sn. Evaluation of sensorless VF-MRAS and FOC-MRAS of IM electrical drive system (Moustapha Diop) Evaluation Warning : The document was created with Spire.PDF for Python.