Electrolytic capacitors in Modular Multilevel Converter-based Unified Power Flow Controllers (MMC-UPFC) are prone to parameter degradation, significantly affecting system reliability. Accurate identification of their degradation parameters—capacitance (C) and equivalent series resistance (ESR)—is essential for equipment health management. This study proposes a non-intrusive degradation parameter identification method for electrolytic capacitors in MMC-UPFC full-bridge submodules based on digital twin technology and an improved intelligent optimization algorithm. First, the degradation mechanism of electrolytic capacitors under long-term operational conditions is systematically analyzed. Second, a high-fidelity digital twin model integrating the main circuit, sampling circuit, and modulation circuit of the full-bridge submodule is established, enabling real-time synchronization between the physical and virtual systems. Third, based on the degradation characteristics of aluminum electrolytic capacitors, a condition monitoring objective function is formulated. An improved Particle Swarm Optimization (PSO) algorithm is proposed, which dynamically adjusts acceleration coefficients to iteratively optimize the monitored capacitance and ESR values, demonstrating superior convergence performance compared to conventional PSO. Verification results validate the effectiveness of the proposed method under various operating conditions. The computational efficiency of the method makes it suitable for quasi-online applications, and the monitoring cycle can be adjusted according to accuracy requirements. This study provides a practical solution for non-intrusive health monitoring of MMC-UPFC systems, contributing to improved equipment health management, extended system lifespan, and enhanced operational reliability in modern Flexible AC Transmission Systems (FACTS).
Yu et al. (Thu,) studied this question.