An Early Warning Analysis Technique of Electric Equipment Temperature Based on Association Rule Mining

Ensuring the safe operation of power equipment is the basis for the stability of power system and the temperature is one of the most concerned parameters. Most power equipment failures are accompanied by the abnormal fever. In order to analyze current status and running trend of the power equipment under the power communication networks, a temperature pre-warning analysis technique based on association rules mining is proposed in this paper. By analyzing multiple influencing factors and equipment temperature data, we establish an equipment temperature characteristics analysis model. Then, an intelligent analysis and early warning algorithm based on association rules is proposed. By a top-down method, the connection relations of influencing factors and the power equipment temperature are found out. Based on these association rules, the trend monitoring of power equipment temperature status and the active warning based on the perception information can be realized, providing theoretical support for safe operation of the power grid. The simulation results show that the algorithm can accurately predict the temperature of the power device in the acceptable prediction error rate and realize accurate temperature warning in time.