1.NARI Group Corporation (State Grid Electric Power Research Institute), Nanjing 211106, China;2.NARI Technology Co., Ltd., Nanjing 211106, China;3.State Key Laboratory of Smart Grid Protection and Control, Nanjing 211106, China

Inspired by the Internet user recommendation technology, an effective dispatch operation behavior pattern mining and recommendation technology is proposed based on the historical operation records of operators in the dispatch system, combined with data analysis and machine learning algorithms. Firstly, the proposed technology collects and records the daily operation data of dispatchers and performs the data preprocessing through page type division, session recognition, and routine identification. Then, the behavior pattern mining of dispatchers is completed by using pattern analysis and cluster analysis, and the pattern data is compressed to a frequent pattern tree (FP-tree) by association rules. Finally, the real-time recommendation system based on the pattern tree structure is used to provide the operation recommendation services for dispatchers. The results of the verification and application in the actual dispatch system show that the proposed technology can effectively identify the operation behavior pattern of dispatchers and provide more accurate operation recommendation for dispatchers, thereby improving the interaction efficiency between the dispatcher and the dispatch system, and reducing the workload of dispatchers.