The Design And Implementation The Fault Diagnosis System Of Thermal Power Plant Based On The Information Fusion

Thermal power plant is the main place of producing electric energy in our country. All sorts of equipment have a very complex coupling in power plant, and the equipment have very complicated features. In addition, this equipment in the plant also has the extremely complex working environment. These factors determine the thermal power plant is a place where a prone to accidents and these failures can produce very big damage and serious consequences. With the development of market economy, the electric power production department pays more and more attention to the safety and reliability of electric power operation, because these are very important factors of power companies bidding on the internet. The original fault diagnosis method is no longer meeting the requirements of current electric power production. The current electricity market requires implementing preventive detection, so that it can do some early detection to reduce the economic loss and the risk of electricity production. Therefore it needs to establish a fault diagnosis system in thermal power plant to service for power production.In fault diagnosis technology, it is generally according to single test information to judge whether the fault is reliability. In the actual fault diagnosis, only get more information for the fault to be able better judge. Multi-sensor information fusion technology is based on the relationships between data and information, and used for testing equipment, and then based on D-S evidence theory to detect the data fusion, to get more effective data. This topic introduced the information fusion technology into the fault diagnosis system, so as to improve the accuracy of fault diagnosis.In this article, through reading a large number of literatures and searching for related information, we studied the basic fault theory of the main production equipment in thermal power plants. This paper introduced the features of power plant main production equipment, and the development of related technology. The design of the system is based on VB object-oriented languages as the framework, and the Matlab program is as development tool for all kinds of intelligent algorithm, and the SQL database is for the system’s database service tools. The system combined these technologies and multi-sensor information fusion to build a fault diagnosis system of thermal power plants. This paper describes the overall framework of the fault diagnosis system and the design process of each function module. The implementation effect of each function module were also introduced, and then the condenser equipment failure data is loaded into the fault diagnosis system, using BP neural network algorithm and D- S evidence theory to verify the accuracy of the system fault diagnosis. In the end, the paper introduced the system overall function test and performance test. The results showed that the fault diagnosis system can be accurate and stable operation. This system also has a good human interface and simple operation, which is easy improved and upgraded in the future.