Open Circuit Switch Fault Detection in Multilevel Inverter Topology using Machine Learning Techniques

Abstract

Multilevel inverter is mostly used in high voltage and high power applications in industry. The possibility of faults raises with an increment in the number of switches in multilevel inverter. In power industries, the reliability of multilevel inverters is one of the main concerns. Hence methods for detecting switch faults are required to improve in the reliability. This paper is mainly focused on open circuit switch fault detection for multilevel inverter. The proposed scheme identifies failed switches by monitoring capacitor current and switches current data. The diagnosis techniques are Artificial Neural Network (ANN), k-Nearest Neighbors (KNN), Support Vector Machines (SVM) and Decision Tree (DT). These methods are only capable for diagnosing failed switches. On identification of the faulty switch, switching sequence has to be reconfigured such that the output voltage is restored to its healthy operating conditions.