An Improved Artificial Bee Colony Algorithm based Harmonic control for Multilevel Inverter

Abstract

In this paper, an Improved Artificial Bee Colony (IABC) Algorithm is proposed for eliminating voltage harmonics in the Multilevel Inverter (MLI). It is achieved by reducing the THD present in the MLI output voltage. In a MLI, the harmonics can be eliminated by an optimal selection of switching angles. The proposed IABC technique utilizes an Artificial Bee Colony (ABC) Algorithm and Recurrent Neural Networks (RNN) for the optimal selection of switching angles. An IABC algorithm is used to evaluate the optimum switching angles that are obtained from the iteration as well as the RNN. Here, the voltage variation of the MLI is determined from the actual load voltage and the reference voltage. The voltage variation at different time interval has been applied to the IABC algorithm. According to the voltage variations, the switching angles can be generated from the RNN. These switching angles can make the MLI output voltage with reduced THD. The proposed algorithm is tested with a seven-level inverter and the resultant fundamental and harmonics voltages are analyzed. The experimental result shows the efficiency of the proposed algorithm in eliminating the harmonics that are generated by the inverter. The proposed algorithm is implemented in the MATLAB/Simulink working platform. The effectiveness of the proposed algorithm is evaluated by the MLI output voltage with some traditional techniques like genetic algorithm (GA), ABC.