Weighted Differential Evolution Based PWM Optimization for Single Phase Voltage Source Inverter

Abstract

Programmed pulse width modulation (PPWM) techniques require the solution of a set of nonlinear transcendental equations in order to obtain the switching angles. The main problem of almost all of them is solving these complex equations. Normally iterative methods such as Conjugate Gradient Descent Method and Newton-Raphson Method are used to solve these equations. To obtain convergence with numerical techniques, the initial values of the variables must be selected considerably close to the exact solution. Also In most cases only a local minimum can be obtained after considerable computational time. In this paper, a variant of differential evolution algorithm (DEA) is used to find the switching angles of a PPWM controlled inverter in a way that selected harmonics are removed from its output and the magnitude of fundamental is set at any desired level. The output waveform and spectrum results from simulation and experimental Digital Signal Controller (dsPic) PPWM inverter are presented to confirm the effectiveness of the proposed Weighted Differential Evolution (W-DE) method.