Cascaded H-Bridge Multilevel Inverter Topology with Artificial Neural Network Technique

This paper deals with Cascaded H-bridge multilevel inverter topology with artificial neural network technique. This topology consists of lower blocking voltage on switches and it requires less number of dc voltage sources, power switches which results in decrease the complexity and total cost of the inverter. Moreover, a new algorithm is used to determine the magnitude of dc voltage sources for generation of all voltage levels. Artificial Neural Network (ANN) is trained by the back-propagation algorithm of the Mean Square Error (MSE) between the output and the desired value. The performance and functional accuracy of the proposed topology using ANN technique in generating all voltage levels for 31-level inverter are simulated using MATLAB simulink.