In order to strengthen the real-time monitor of the field metering device, collecting device and the distribution network operation, and to enhance the accuracy of anomaly analysis for the user"s power use behavior, a system embedded in the power user electric energy data acquisition system has been developed, which based on online monitoring and intelligent diagnosis technology, using the data mining methods,by combination of abnormal event with the abnormity of external system and fault treatment process,and implements the status analysis of metering device and acquisition equipment, the judgment of breach of a contract for electricity using and electricity stealing, and the intelligent diagnosis of abnormal metering. Adopting the SOA technique structure, the system consists of three parts, the expert database of abnormal index ,the real-time monitoring and intelligent diagnosis,and the web application . The web application integrates the GPS positioning of abnormity with the GIS show, distribution and the dispatching process of abnormity. Trial operation in Anhui electric power company shows that the system which based on SOA structure is flexible deployment, can easily interact with external systems or data center information,and have the strong uniform quality of openness and security. The system strengthen the power consumption and the metering department on the analysis,positioning and dispatching efficiency of exception on the function. This can greatly improved the real-time monitoring, comprehensive analysis and rapid reaction ability of power company for the abnormality which may cause the accident of electric power and metering device fault. It also plays an important role in ensuring the security and stability operation of power grids, and causing public sentiment of anti-electricity stealing and reduing the metering deviation.