Research status and future directions of artificial intelligence in the fault diagnosis of electrical equipment

In recent years, artificial intelligence (AI) technology has been widely used in the field of electrical equipment fault diagnosis, and has shown excellent fault detection ability. The traditional fault detection methods are limited by the inefficiency and accuracy under complex working conditions, while AI technology realizes the efficient monitoring and intelligent prediction of the operating state of electrical equipment through neural network, deep learning, unsupervised learning and other methods. This paper systematically analyzes the research trend, key technologies and future development direction of this field in the past decade. It is found that the deep learning technologies such as generative countermeasure network (GAN), convolutional neural network (CNN), long and short-term memory network (LSTM) have shown significant advantages in complex data processing and real-time fault detection. At the same time, the literature analysis reveals that the current research mainly focuses on the diagnosis and optimization of single equipment, and lacks the systematic research under the environment of multiple equipment and multiple working conditions.