Digital Twin Approach for Three-phase Interleaved Parallel Buck Converters Based on Improved PSO

Establishing a digital twin (DT) for power electronic converters is beneficial for accurately understanding the operational status of physical devices. This article proposes a DT approach for a three-phase interleaved parallel buck converter with peak current control. After introducing the fundamentals of the converter, a three-layer DT framework is proposed. In the analysis layer of DT, an adaptive inertia weight particle swarm optimization (PSO) algorithm is used and added with simulated annealing algorithm to give particles a greater chance of exploring the whole design space in the early stages of iteration. The difference between the output waveforms of DT and physical devices is used as an evaluation indicator of fitting. Simulation results show that the built DT can generate an excellent fitting and the proposed improved PSO algorithm is more suitable for the construction of complex DT.