This paper proposes a deep learning-based method for power system fault diagnosis and prediction. By leveraging real-time monitoring data and deep neural networks (DNN) and convolutional neural networks (CNN), the model extracts features and accurately classifies fault patterns. Additionally, Long Short-Term Memory (LSTM) networks are used to predict potential faults by capturing long-term dependencies in sequential data. Experimental results show that the proposed method outperforms traditional techniques in accuracy, real-time performance, and robustness. Simulations and real-case studies demonstrate its effectiveness in enhancing the reliability and safety of power systems.
Le‐Hang Guo (Sun,) studied this question.