Research on power equipment diagnosis based on big data

Abstract

The continuous development of power system is an important guarantee for the rapid development of economy and science and technology in China. The emergence of smart grid has accelerated the development of power big data and artificial intelligence in power field. In smart grid, the state of power equipment determines the state of power system and is the decisive factor for the stable operation of power grid. spark distributed parallel processing system can meet the storage and calculation of big data, and provides a new research idea for realizing fault classification and diagnosis of power equipment under big data. The concept of deep learning provides a platform for the analysis and prediction of power equipment status under big data. This paper introduces the basic contents of power equipment handover test, analyzes the common fault types of power equipment in handover test, introduces the existing fault diagnosis methods, puts forward the scheme of power equipment state diagnosis-based on spark platform, and classifies the faults of power equipment by using the naive Bayesian network in spark, Taking the test data as sample input, the fault diagnosis of power equipment can be carried out. On this basis, through the research of deep learning network, the feasibility of power equipment fault diagnosis is analyzed. The test shows that the method used in this paper can diagnose and predict the state of power equipment effectively.