Analysis of Different Meta Heuristics Method in Intelligent Fault Detection of Multilevel Inverter with Photovoltaic Power Generation Source

Abstract

Meta Heuristic methods have made a deep impact in the area of optimization in different streams of engineering. The performance of these algorithms is of importance because the hardware implementation of these algorithms is to be carried out for different engineering applications. As an important application in High Voltage DC (HVDC) transmission and Industrial Drives the multilevel inverter fault diagnosis is carried out using the different meta-heuristic methods with Neural Network as the decision making algorithm. The optimization of the weight and the bias values in the neural network diagnosis system is carried out in order to analyze the performance by means of comparing the Mean Square Error (MSE) while the Neural Network is getting trained for different fault conditions in the multilevel inverter. Matlab based implementation is carried out and the results are tabulated and inferred for a Multilevel Inverter fed from the Photovoltaic power generation system. In order to increase the robustness of the fault detection, with renewable energy based power generation system as the source for the Multilevel Inverter, the feature extracted from the multilevel inverter are positive, negative and zero sequence voltage along with the THD of the output voltage. The optimization algorithm used is Particle Swarm Optimization (PSO), Cuckoo Search Algorithm(CSA), Genetic Algorithm(GA) and Tabu Search Algorithm (TSA).