The Application Of Particle Swarm Optimization RBF Neural Network In Condenser Vacuum Fault Diagnosis

From the point of view of the abstract，the Coal-fired power plants is a Rankine’scirculation process of action. The condenser as the cold end of Rankine’s cycle, itspros and cons of running state to the whole power plant has a great influence on theEconomy and safety. In the operation of the unit, the condenser will continue to leakinto the air, results in the decrease of the vacuum, thermal efficiency decline, reducingefficiency. So in the actual operation of the power plant, the staff want to be able totimely treatment failure as soon as possible，so as to ensure the unit’s safe and reliablework. Therefore, the vacuum monitoring and fault diagnosis model of condensersystem research has theory meaning and practical value in improving the economicoperation level of the unit.In this paper, the main work is divided into the followingsections:(1)According to the condenser heat transfer theory，we analyzes the main factorsaffecting the vacuum, Using the theory of heat transfer completed the condenservacuum on-line monitoring.(2)Through consulting a large number of relevant literature and sum upof experience from the power plant staff, I set up the condenser fault diagnosisknowledge base. Depend on the particle swarm radial nerve network, I design andestablish the fault diagnosis model of condenser vacuum system theoretically, and theMatlab program is written in accordance with the design to verify the diagnosiseffect.(3)Using C++Builder development tools combining SQL Sever databasesoftware, I developed the condenser vacuum monitoring and fault diagnosis systemsoftware. this software has a simple system interface to operate and can get diagnosisresults intuitively, has the very good guidance function to the operation of the unit.In this paper, according to the theory of data and field investigation, the vacuumof online monitoring developed. Combined with the optimized intelligent algorithm,condenser vacuum fault diagnosis theory analysis and software development achieved, which provides the appropriate operation instruction for power plant operatingpersonnel. And in the vacuum failure occurs, fault prediction is given to make quicklyfind fault under the complex industrial site so that a quick fix to become possible.