Modelling of Machine Learning Controller Based Reduced Switch Multi Level Inverter

Abstract

This manuscript studies about a photovoltaic (PV) system. PV system consists of solar panels, boost converter, Reduced Switch Multi Level Inverter (RSMLI), proposed controller (Machine Learning Based Proportional Integral Derivative Controller (MLBPIDC)) and a two quadrant single phase rectifier DC motor drive. In this proposed controller an attempt is made to overcome challenges posed by classical Proportional Integral Derivative Controller (PIDC) by using modern machine learning techniques. All prospective machine learning algorithms including Levenberg Marquardt Algorithm (LMA), Bayesian Regularization Algorithm (BRA) and Scaled Conjugate Gradient Algorithm (SCGA) are compared and a best possible fit is chosen. The proposed PV system along with MLBPIDC has been checked for accuracy, robustness and stability through MATLAB/simulink.