Mode identification of hybrid system in the presence of fault using residual bond graph sinks

Mode identification is the premise to monitor the state of hybrid system, especially when fault occurs, it's essential to identify the current mode of the system timely and accuracy in order to accomplish fault diagnosis. In this paper, a quantitative and qualitative mode identification method is proposed and applied to a three-phase inverter in the presence of fault. Considering the structural faults of switching devices in three-phase inverter, residual bond graph sinks are used for quantitative fault detection by coupling a model of non-fault system to a model of the system that is subject to disturbances caused by faults. A qualitative approach is used to identify the fault mode of the system, which is based on signal amplitude behavior analysis when a fault occurs in the system. Finally, simulation result suggests the feasibility and accuracy of the presented mode identification algorithm.