Optimum Harmonic Elimination Control Method for Multilevel Inverters Using Artificial Neural Network

摘要

In this paper, an artificial neural network is applied to estimate the optimum switching angles of cas-caded multilevel inverter which eliminates specified lower order harmonics while maintaining the re-quired fundamental voltage. This technique can be applied to multilevel inverters with any number of levels if the switching angles change continuously versus modulation index. As an example, in this pa-per a seven-level inverter is considered and neural network is trained with the optimum switching an-gles that achieved by iterative method. After that a large and memory-demanding look-up table re-placed with trained neural network to generate the switching angles for given modulation index and can be used to estimate continuously optimum switching angles which eliminate the fifth and seventh harmonics.