Performance Enhancement of Evolutionary Algorithm Based Multilevel Inverter: A Comparative Study

This paper aims to extend the knowledge about the performance enhancement of cascaded multilevel inverter by using heuristic optimization methods. For improving efficiency of multilevel inverter, harmonics has to be eliminated. To eliminate the harmonics, Selective Harmonics Elimination Pulse Width Modulation technique is used for a better power quality output. This technique eliminates the pre-selected lower order harmonics which are more dominant into the nature while satisfying the fundamental component of output voltage, but the major issue is to solve the set of nonlinear harmonic equations derived from Fourier analysis of the output waveform of multilevel inverter. Using Conventional approach to solve those equations does not guaranteed the optimal solution and may stuck in the local minima. Hence, Evolutionary optimization algorithms are used for solving those equation. Evolutionary algorithm consist a subset of evolutionary computation and the results are obtained. Among of all a comparison is done between simulated annealing, Genetic Algorithm and multi objective genetic algorithm. Whereas, Evolutionary optimization algorithm are inspired by the biological process of evolution. Using those algorithm the main concern is to find the notches angles for giving the gate pulse to the individual switches of multilevel inverter. Using all of the three algorithms data sets are collected for different modulation indexes and output are verified by results. Results also shows the comparison between genetic algorithm and multiobjective genetic algorithm for getting the optimum results.