Genetic algorithm based SHE-PWM for 1- ø and 3- ø voltage source inverters

Abstract

In medium voltage high power application of voltage source inverter (VSI), low switching frequency pulse width modulation (LSPWM) techniques offer better efficiency and better thermal management as the switching loss decreases drastically. The number of switching transitions are very limited, hence getting a good quality output waveform is a drawback in LSFPWM. To overcome this problem, selective harmonics elimination (SHE) PWM techniques are used to eliminate any desired harmonics from the output waveform by turning ON and OFF of semiconductor switches at predefined time. However, the main challenge in its implementation is to compute the exact instant of time to turn ON the switches. In this paper a Genetic Algorithm based Optimization technique is presented to compute the switching angles in bipolar waveform. The simulation and hardware results are given to verify the robustness of the algorithm.