MLMVNN for parameter fault detection in PWM DC-DC converters and its applications for buck DC-DC converter

This paper presents an effective approach to the fault diagnosis of PWM DC-DC power converters. It is based on a MultiLayer Multi-Valued Neuron Neural Network (MLMVNN) with a complex QR decomposition. The network is used to identify the converter parameter values running out of tolerance. This technique is applied in the fault detection of parameters of a Buck DC-DC PWM converter and is based on steady state measurements of voltages and currents.