Research And Development On State Monitoring And Intelligent Faults Diagnosis System For Gas Blowers

The gas blower is a very important equipment in chemical plant , in order to guarantee its security and dependability when operating, it is very significant to set up a intelligence system which can monitor the state of the gas blower in real time and predict, alarm, diagnose to the fault of the gas blower in time. Based on the fault characteristic of gas blower, A state monitoring and intelligent faults diagnosis system for gas blowers is developed by using object-oriented technique, database technique and intelligent diagnosis technique etc.. The paper introduces basic function of the system ,the general structure elements of the system and critical technology of the system. The key contents of the paper is the realization of intelligent diagnosis system for gas blowers.At present, Fuzzy Theory, Artificial Neural Networks (ANN) and Expert System (ES) have already become studied focuses in the Artificial Intelligence (AI) field and they are widely applied to the fault diagnosis field. Nevertheless, there are certain defects in each independently practical application. Through the analysis and comparison between their merits and demerits, it is educed that there are strong complementarities between them and that their deficiencies were made up by combining in this paper. Therefore, they have been combined together in this paper. A fault diagnosis system model has been proposed in which fuzzy neural network and expert system are integrated. Regarding the gas blower as the study target, a gas blower fault intelligence diagnosis system has been set up and realized by VB6.0 and Access2000 in this paper. The signals of vibration are deal with fuzzy method to reflect the randomicity and uncertainty of the fault freely. In this system, the BP neural network are selected to capture the Information. The modes of the neural network are used to express the knowledge, so the superiority of self-study of Neural network can be used sufficiently. The knowledge in traditional Expert system is expressed by a kind data structure. On contrast to the traditional Expert system, the new method can distribute the information of repository to the value of weight and threshold of the Neural network, the knowledge of expert can be expressed preferably. The imitation results show that the intelligent integrated fault diagnosis method is reasonable and feasible.To this day, This system can collect data and analyze signal, monitor running state and 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..