A Random Forest Regression Based Space Vector PWM Inverter Controller for the Induction Motor Drive

This paper presents a random forest (RF) regression based implementation of space vector pulse width modulation (SVPWM) for a two-level inverter to improve the performance of the three-phase induction motor (TIM) drive. The RF scheme offers the advantage of rapid implementation and improved prediction for the SVPWM algorithm to improve the performance of a conventional space vector modulation scheme. In order to show the superiority of the proposed RF technique to other techniques, an adaptive neuro fuzzy inference system (ANFIS) and artificial neural network (ANN) based SVPWM schemes are also used and compared. The proposed speed controller uses a backtracking search algorithm to search for the best values for the proportional-integral controller parameters. The robustness of the RF-based SVPWM is found superior to the ANFIS and ANN controllers in all tested cases in terms of damping capability, settling time, steady-state error, and transient response under different operating conditions. The prototype of the optimal RF-based SVPWM inverter controller of induction motor drive is fabricated and tested. Several experimental results show that there is a good agreement of the speed response and stator current with the simulation results which are verified and validated the performance of the proposed RF-based SVPWM inverter controller.