Digital Twin Approach for Degradation Parameters Identification of a Single-Phase DC-AC Inverter

The digital twin concept has attracted a lot of attention in recent years since it can achieve cost-effective, noninvasive and reliable condition monitoring for the physical power converters. In this paper, a digital twin approach for degradation parameters identification of the single-phase dc/ac inverter is proposed. First of all, the digital twin model of the inverter circuits is established under the steady-state, including the output side subsystem, the main circuit, PI controller and SPWM modulation. Then, an optimized objective function is constructed based on the sampling data from the physical circuit and incoming data from the digital twin, and the particle swarm optimization (PSO) algorithm is applied to minimize the difference between both of them. It is noteworthy that the values of output side filter capacitance $C$ and Load resistance $R$ are determined through the digital twin subsystem of the inverter, then the digital twin inverter system determines other quantities except $C$ and $R$ . The key degradation parameters, such as capacitance, switch's on-state resistance and inductance can be identified and monitored online through this approach. Finally, a single-phase dc/ac inverter was built, and some simulation and experiment results verified the effectiveness of the proposed approach.