Electrical Equipment State Assessment System Based On Business Intelligence And Realization

The operation reliability of the electric power equipment is closely related to the safety and stability of electric power system. For a long time, the test, check and maintenance to electric power equipments are carried out according to "Regulation on the preventive test of electric power equipment" in time-based manner during power cut. This kind of maintenance mechanism has played a very important role in guaranteeing safety of electric power system operation. With the increasing development of economic and expanding scale of electric power system, the implementation of condition-based maintenance of electric power equipment is expected strongly. In order to reach the goal, the key problem to be solved is how to evaluate the current health conditions of the equipment. The system of economic power equipment condition assessment based on business intelligence carries on the quantification evaluation according to the equipment condition information and provides the scientific basis for the condition-based maintenance.In this article, an electric power corporation as the main subject, through in-depth field research and rigorous research and development, and a large number of theoretical learning, including KPI theory, business intelligence theory of data warehouse, data mining and OLAP technology, the author gives the new idea of condition assessment analyzing model which based on business intelligence. This paper introduces the thinking of KPI management, and concludes calculating equipment health condition to find KPI of condition assessment analyzing model. Method: First, find out all the possible attributions that influence the condition assessment. Then build the condition assessment analyzing subject data warehouse. And then abstract event gene from the data warehouse, including integration index and support index. For the integration index, its corresponding value is calculated by constructing BP neural net. The condition assessment of equipment is finally calculated by Constructed BP neural net again, making the event gene as input and the condition health assessment as output, and training the BP neural net using training data sets until it is converged .Until now, the condition assessment analyzing model is successful. At the end, the author validated and evaluated the model using the really data sets. The result proves the idea of entire intelligence solution is feasible. Meanwhile, use OLAP technology, analyzing the causes of ill health equipment. These results will be used to help the condition-based maintenance.