Neural classification method in fault detection and diagnosis for voltage source inverter in variable speed drive with induction motor

These days, electrical drives generally associate inverter and induction machine. Thus, these two elements must be taken into account in order to provide a relevant diagnosis of these electrical systems. The aim of this paper is to study the feasibility of fault detection and diagnosis in a three-phase inverter feeding an induction motor. The proposed approach is a neural network classification applied to the fault diagnosis of a field oriented drive of induction motor. Multilayer perception (MLP) networks are used to identify the type and location of occurring fault using the stator Concordia mean current vector. In the case of a single fault occurrence, a localization domain made with seven patterns is built. With the possibility of occurrence of two faults simultaneously, there are twenty-two different patterns. Simulated experimental results on 1.5-kW induction motor drives show the effectiveness of the proposed approach with a classification performance over than 95%.