Neural modified super-twisting filed-oriented approach of asynchronous machine drives

Abstract

This paper proposes a novel field-oriented command (FOC) strategy with a neural-modified super-twisting algorithm (NMSTA) controller for a category of nonlinear systems. This designed control is used to adjust the speed and torque of the asynchronous machine (AM) to enhance the dynamic response, tracking error and robustness. Also, to reduce the torque, stream, and flux undulations of the AM. The FOC-NMSTA technique of the AM has been simulated in MATLAB and compared with the FOC technique. The comparative results that were obtained under different operating conditions prove the superior performance and robustness of this proposed new intelligent nonlinear technique (FOC-NMSTA). In all four tests, the current harmonic distortion value was improved by 85.96% and 77.91%. Also, the FOC-NMSTA provided a better ripple value of flux than the FOC, where the reduction rates in all tests were 87.76% and 86%. Torque overshoot in all tests was also reduced compared to the FOC with percentages estimated at 41.02% and 44%. These relatively high percentages indicate the extent of FOC-NMSTA's competence and ability to get better the features of the control system.