Torque ripple minimization of PMSM using an adaptive Elman neural network-controlled feedback linearization-based direct torque control strategy

Torque ripple and harmonics in permanent magnet synchronous motor (PMSM) cause negative impacts on its speed control and overall efficiency. In this article, an Elman neural network (ENN)-controlled feedback linearization-based direct torque control strategy has been proposed to mitigate torque ripples and undesirable harmonics of PMSM. In this approach, the stator flux, stator flux position, and developed torque are used to generate a switching pulse to the impedance source inverter employing ENN controller. By changing the switching pattern of the impedance source inverter through modifying the pulse width modulation using the proposed control strategy. It is possible to control the armature current of the PMSM at various orders and corresponding time harmonics under different operating conditions. The outstanding aspect of the proposed control strategy is that it boosts the inverter operation and reduces the requirement of the DC-DC converter in boosting. For evaluating the performance of the developed control strategy, an extensive simulation study has been conducted in the MATLAB/Simulink environment, and experimental analysis is carried out using the Xilinx-FPGA kit.