Research on information aggregation methods of electric equipment state monitoring based on data driver

In view of the unsatisfactory results and the low utilization rates of electric equipment application, this paper proposed a hierarchical information aggregation structure and model in data analysis and application, taking the autocorrelation analysis based on data probability distribution as the foundation. The model aimed at predicting equipment state trends and assessing equipment service lifetime, and driven by dada mining, analyzed and processed the state monitoring data of large scale electric equipment. A threshold setting method grounded on the probability distribution of historical data was put forward in this paper, and solved the equipment state boundary division for online monitoring. In addition, an equipment abnormal state analysis and identification approach, taking the correlation feature changes of continuous and multidimensional data as the core, was suggested in this paper. The concepts and methods provided in this paper was used to dig for and analyze the online real data of an 110kV transformer in Shanxi Province, North China, and verified the feasibility of correlation analysis based on data driver in state monitoring application practice.