14 June 2024 in IEEE Sensors Journal
Modular Multilevel Converter (MMC) is widely used in high power motor drive, energy storage and flexible DC power transmission & distribution system, and the parameter identification for sub-module is essentially important for reliable operation of MMC. This paper proposes a digital twins (DT) based method for model construction and parameter identification of the MMC half-bridge sub-module (HBSM). Firstly, the DT model of the HBSM is built, which consists of a simplified MMC sub-module (SM) equivalent model and an arm current simulator based mission profile emulator (MPE). Secondly, a modified particle swarm optimization (MPSO) algorithm is proposed to achieve the online identification of internal parameters in the HBSM, further achieving the real-virtual interaction feedback for real-time sensing of the operation state of the HBSM. Finally, a non-intrusive DT MPE for the HBSM is constructed in MATLAB. The simulation and experimental result shows that the average errors of the actual and model-calculated capacitor voltage values are within 0.1%, and the parameter identification results combined error is less than 1%. The digital twin model established by the proposed method can effectively identify the parameters of MMC sub-modules under different operating conditions.