Study On Faults Diagnosis System For Gas Blower Based On Fuzzy Clustering Optimized By Particle Swarm Optimization

The gas blower is a very important equipment in chemical plant, if it is shut down the gas supply of all companies will not be well balanced, at the same time, coke oven gas dischargeing to air, so it is very important to monitor status real-timely,give a fairly accurate forecast of which fault may happen,set up an intelligent fault diagnosis system. Based on the property of gas blower and the former On-line monitoring system, this paper put forward a diagnosis algorithm based on fuzzy clustering optimized by Particle Swarm Optimization, and use it in practice.This paper introduces the basic principle and realization issues of the algorithm, the composition of system, and the each function module.At present, The clustering analysis has become an active area in the research on artificial intelligence pattern recognition machine learning. The clustering analysis based on objective function have been applied widely in fault diagnosis field,but the algorithm has some drawback in actual practice,it is so sensitive to initial values that it easy to get into the local best and can't get the optimal solution, it also has bade timeliness processing mass data and multiple dimensions data.Particle swarm optimization (PSO) is an evolutionary computation technique which is used to solve optimization problems through simulating social behavior of bird flocking. Based on the research the Fuzzy C-Means, this paper put forward a fault diagnose algorithm which is based on fuzzy clustering optimized by Particle Swarm Optimization and use the performance of global optimization get out of the semilocal convergence and find optimal solutions. Simulation results show that fuzzy clustering optimized by Particle Swarm Optimization has an ability of fast convergence speed and clustering performance. Based on the former program,this paper describe the detail algorithm realization through the hybrid programming method of MATLAB and VC++,Visual C++ and SQL SERVER 2000 as software development tools are adopted to set up database system based on Particle swarm optimization, a fault diagnosis system based clustering optimized by Particle Swarm Optimization which has well application performance.To this day,This system can collect data and analyze signal,monitor running stateand forecast the trend,give an alarm and diagnosis for abnormal detected parameters and unwonted state.All kinds of functions are achieved the expected design object and meted the request of the user and obtained the appraisement of the user.