Artificial Intelligence-Based Open-Circuit Fault Diagnosis for Power Electronic Converters: Recent Advances and Future Prospects

Fault localization and identification for converters are related to the stable operation of power electronic-based systems and are receiving widespread attention. Advanced algorithms, represented by artificial intelligence (AI), are providing new solutions. This article explores the application of artificial intelligence in diagnosing open-circuit faults in power electronic converters, which are critical components of modern power systems. The fault diagnosis techniques are systematically categorized and summarized from an algorithmic perspective, covering a range of methods including traditional machine learning (e.g., SVM, ELM), neural networks (e.g., CNN, RNN), multimodal fusion, and hybrid models. Their performance is evaluated in terms of noise robustness, computational efficiency, and real-time capabilities. Finally, future prospects and recommended directions for AI-driven fault diagnosis are provided, advocating for innovations to enhance the reliability and cost-effectiveness of power converter systems. This review aims to provide a comprehensive reference, fostering the development of advanced diagnostic algorithms and techniques.