Artificial neural network-based fault diagnosis in the AC-DC converter of the power supply of series hybrid electric vehicle

AC-DC converter switches of the drive train of series hybrid electric vehicles (SHEVs) are generally exposed to the possibility of outbreak open-phase faults because of troubles with the switching devices. In this framework, the present study proposes an artificial neural network (ANN)-based method for fault diagnosis after extraction of a new pattern. The new pattern under AC-DC converter failure in view of SHEV application has been used for train-proposed ANN. To achieve this goal, four different levels of switches fault are considered on the basis of both simulation and experimental results. Ensuring the accuracy and generalisation of the introduced pattern, several parameters have been considered, namely: capacitor size changes, load, and speed variations. The experimental results validate the simulation results thoroughly.