Generalized Dynamic Model and Digital-Twin-Based Open-Switch Fault Classification of Multi-Phase Inverters

In this paper, a comprehensive digital twin system that virtually mimics a two-level multi-phase inverter system in both normal and faulty operation conditions is proposed. Conventional models of the multi-phase inverters only consider normal operation. It is not possible to develop a digital twin system using such models. This paper presents a generalized dynamic model of a multi-phase inverter that produces accurate outputs in open-switch fault conditions. The proposed model can be used to represent multi-phase two-level inverters with a high degree of fidelity. An experimental setup including a six-phase two-level inverter and a digital twin is developed to demonstrate the feasibility of using the proposed model to perform open-switch fault classification. The accuracy of the proposed model is calculated to be 3.15 % by comparing the outputs of the digital twin and hardware. The fault detection and identification were performed in 815.79 μs and 898.74 μs respectively.