A new adaptive selective harmonic elimination method for cascaded multilevel inverters using evolutionary methods

In this paper, a new approach for modulation of an 11-level cascaded multilevel inverter using selective harmonic elimination (SHE) is presented. The dc sources feeding the inverter are considered to be varying in time. In this approach the switching angles are obtained offline for different dc source values. Then an artificial neural network (ANN) is trained to determine the switching angles that correspond to the real-time values of the dc sources in each phase. In fact, each one of the dc sources can have different values at any time, but the output fundamental voltage will stay constant and the harmonic content will still meet the desired specifications. Mathematical methods for harmonic elimination are presented in some of the literatures but solving a non-linear transcendental equation set describing the SHE problem using these methods are not suitable for high level inverters. In this paper, the genetic algorithm (GA) and the particle swarm optimization (PSO) are applied to obtain the switching angles. These techniques can be applied to cascaded multilevel inverters with any number of levels. This paper gives details on the both evolutionary methods. Finally, the results obtained using GA and PSO are compared together and an ANN is trained by the best answer between GA and PSO.