Power user electric energy data acquire system is one of the indispensable key infrastructures in modern society, and its reliability and safety are very important for the normal operation of society. Effective monitoring and data transmission of power user electric energy data are key factors to ensure the stability of data acquire system. In this paper, an expert system based on message detection of power user electric energy data is proposed, which aims to identify errors and locate faults through full link messages, thus improving the reliability and security of power user electric energy data. First of all, we collect the data of power user electric energy data messages from various sources and preprocess the data. Then, based on Transformer algorithm, we developed data verification algorithm and anomaly detection technology, and introduced expert rules and logic check to identify unreasonable, wrong or abnormal data points, so as to locate faults. Experiments show that the expert system proposed in this paper can effectively identify errors in message information and locate faults.