Identification of abnormal operation of electric energy metering equipment based on multi-dimensional and multi-layer association rule mining algorithm

Abstract In order to comprehensively and deeply explore the potential correlations and patterns in operational data, this paper proposes a method for identifying abnormal operation of electric energy metering equipment based on multi-dimensional and multi-layer association rule mining algorithms. A multi-level association rule mining model has been constructed to automatically discover potential correlations and abnormal patterns between different indicators, thereby accurately identifying abnormal operation of energy metering equipment. The raw running data is converted into a form suitable for association rule mining, and evaluation metrics such as support and confidence are calculated for each rule. Abnormal operation of energy metering equipment is identified based on these evaluation indicators. The experimental results show that it can effectively identify the abnormal operation of energy metering equipment, providing strong technical support for equipment maintenance and troubleshooting of power companies.