Robust fault-tolerant sliding mode control and advanced fault diagnosis for doubly-fed induction generators

• Datum: 2025
• Hauptverfasser: Hamdi, N., Babaa, F., Touil, A., Merabet, N.
• Format: Artikel
• Sprache: English
• Veröffentlicht: National Technical University "Kharkiv Polytechnic Institute" and Аnatolii Pidhornyi Institute of Power Machines and Systems of NAS of Ukraine 2025
• Schlagworte: sliding mode control; doubly-fed induction generator; active fault tolerance control; renewable energy
• Online Zugang: http://eie.khpi.edu.ua/article/view/334672
• Назва журналу: Electrical Engineering & Electromechanics

Introduction. Doubly-fed induction generators (DFIGs) have become the preferred technology in modern wind energy systems due to their high efficiency and flexible variable-speed operation capabilities. Problem. Despite their advantages, DFIGs face significant challenges related to grid-connected power converters, which are susceptible to operational instability caused by voltage imbalances and electrical faults. Goal. This study aims to develop and validate a novel Active Fault-Tolerant Sliding Mode Control (AFT-SMC) strategy that integrates real-time fault diagnosis to enhance the reliability and stability of DFIG systems during grid disturbances. Unlike existing approaches, this work specifically addresses the reduction of false fault detections during transient events and improves fault characterization through spectral analysis. Methodology. The proposed control framework combines a robust sliding mode controller with a model-based fault detection and isolation system that employs adaptive thresholds and diagnostic residuals for accurate fault identification. The approach has been thoroughly tested through high-fidelity simulations under severe voltage unbalance scenarios. Results. Simulation outcomes demonstrate the superior performance of the proposed strategy in maintaining system stability under a 30 % voltage unbalance scenario. Specifically, the controller achieves a voltage recovery time of 0.28 s, compared to 0.42 s with conventional vector control, and reduces electromagnetic torque oscillations by approximately 45 %. Furthermore, the integrated spectral diagnosis method reaches a fault classification accuracy of 94.6 %, confirming its effectiveness in enabling early and reliable fault detection. These results validate the advantages of the proposed AFT-SMC framework in both dynamic response and fault resilience. Scientific novelty. The key innovation lies in the integration of a self-correcting «detect-and-adapt» mechanism that mitigates false triggers during transient grid conditions, alongside a novel spectral decomposition method for precise detection and characterization of voltage imbalances through negative-sequence component analysis. Practical value. This strategy significantly reduces operational costs at pilot wind farms and sets a new benchmark for intelligent fault management in renewable energy systems, with broad applicability to other power electronic interfaces in smart grids. References 35, figures 12.