Comprehensive Intelligent Diagnosis for Mechanical and Insulation Faults of Power Equipment in the Power Internet of Things Context

Abstract

During the construction of the power Internet of Things, the data of the entire process of equipment operation will be monitored and retained. Therefore, the representative and comprehensive problem of the fault sample is solved. In this way, artificial intelligence technology can be used to carry out in-depth mining in order to digitally and intelligently diagnose power equipment failures. To this end, this paper proposes an efficient lightweight convolutional neural network for comprehensive intelligent diagnosis of mechanical and insulation faults in power equipment. This paper first introduces the process of comprehensive intelligent fault diagnosis under the power Internet of Things. Then a lightweight convolutional neural network (LCNN) for comprehensive intelligent fault diagnosis was constructed. Next, this paper validates the method on the GIS partial discharge data set and the mechanical fault data set. Compared with the traditional method, the accuracy of the method proposed in this paper is 99.91% on the mechanical dataset, and 94.52% on the insulation dataset, which has a significant improvement. Moreover, the model is one-tenth of the traditional model in terms of parameter quantity and storage space, which is conducive to real-time and fast processing of signals under the power Internet of Things.