Enhanced direct torque control and predictive torque control strategies of an open‐End winding induction motor drive to eliminate common‐mode voltage and weighting factors

Abstract

Direct torque control (DTC) and predictive torque control (PTC) strategies emerged as powerful tools for speed control of variable frequency drives (VFD). The limitations in classical DTC and PTC are: higher ripple in torque, flux, variable switching frequency, and higher common-mode voltage (CMV). High dv/dt and CMV results in shaft voltages, bearing currents, malfunction of power electronic devices and electromagnetic interference (EMI). To eliminate the CMVs and also to reduce the switching frequency, voltage vector selection-based DTC and PTC strategies are introduced to an open-end winding induction motor (OEWIM) drive. Another limitation of classical PTC is cumbersome tuning of weighting factors. To address this limitation, modified cost function-based PTC strategy has been developed to eliminate weighting factors. OEWIM drive operates with dual inverter configuration and the two inverters are operated with equal DC-link voltages; therefore, it delivers three-level output voltage. This article introduces various DTC and PTC strategies to an OEWIM drive to reduce torque, flux ripples, switching frequency, elimination of CMV, and weighting factors. The effectiveness of proposed DTC and PTC strategies is tested by dspace-1104 real-time interface controller and comparing the obtained results with classical DTC and PTC.

Speed-Sensorless Predictive Torque Control of the IM Based on MRAS

“…In this MRAS estimators, while the measured 𝑖 𝑠𝛼 and 𝑖 𝑠𝛽 are outputs of the reference model, the adaptive model outputs are obtained by Equations ( 6) and (7). 𝜑 𝑟𝛼 and 𝜑 𝑟𝛽 required for Equations ( 6) and ( 7) are provided by Equations ( 8) and (9), which are the current model. Here, Equations ( 6)-( 9) are obtained by discretizing Equations ( 1)-( 4) with forward Euler approximation.…”

“…In the literature, one of the most preferred MPC strategies for the control of IMs is the predictive torque control (PTC) [8]. In the PTC strategy, the electromagnetic torque and stator flux are controlled by the proper selection of the voltage vector, which determines the inverter switching states [9]. By determining the proper voltage vector, the reference flux and electromagnetic torque values are tracked by PTC [9].…”

“…In the PTC strategy, the electromagnetic torque and stator flux are controlled by the proper selection of the voltage vector, which determines the inverter switching states [9]. By determining the proper voltage vector, the reference flux and electromagnetic torque values are tracked by PTC [9]. By the minimization of the cost function, which is defined by the control objectives in PTC, the optimal voltage vector is selected.…”

“…As addressed by Rodriguez et al [4], by considering the computational optimization, the parameter sensitivity, the weight factor adjustment, and the switching frequency limitation, more efficient PTC strategies can be developed. Thus, there are some studies for the enhancement of PTC strategies by reducing switching frequency [9], minimizing power loss [10], optimizing or eliminating weight factor [11,12], and increasing robustness against the parameter variations [13].…”

Speed-sensorless predictive torque controlled induction motor drive with feed-forward control of load torque for electric vehicle applications

“…However, Rodriguez et al [10] state that research efforts in the parameter sensitivity of this strategy, adjustment of the weight factor in the cost function, limitation of the switching frequency, and computational optimization are required to develop more efficient FCS-PTC strategies. For this purpose, some efforts have been made on determining or eliminating the weighting factor [11][12][13], reducing or fixing the switching frequency [12][13][14][15], reducing torque and current harmonics [12,[14][15][16], overcurrent protection [11,[14][15][16], minimizing electrical power losses [17], robustness of parameter variation [17,18], dead time compensation [11,15,16,18], and optimizing computational complexity [12,14,16].…”