A Digital Twin Based Condition Monitoring Method for Power Modules of Inverters

In this paper, a digital twin model-based method for condition monitoring of inverter power modules is proposed. The method utilizes the existing sensor data of the inverter to realize the health monitoring of the power module. The switching process of power modules is accurately reconstructed using an analytical model that considers circuit parasitic parameters and nonlinearities in the model parameters. A thermal network model and a fast loss calculation method are used to realize a rapid and accurate estimation of the junction temperature of power modules.