Research On On-line Monitoring And Fault Diagnosis System Of Fan Based On Fuzzy Neural Network

Fan is the important ventilation equipment in coal mine enterprises, onc failureof fan can cause gas accumulation even gas explosion, endanger undergroundworkers’ safety. Therefore, the on-line monitoring and fault diagnosis of fan isparticularly important. Intelligent fault diagnosis as a diagnostic technology belongsto an emerging discipline, which is a hotspot of current fault diagnosis technology.Combining the advantages of fuzzy neural network and expert systemrespectively,we designed a fault diagnosis model of mine fan combination. Based onfuzzy theory, fuzzy applied to reasoning neural network and solveing the problem ofsystem uncertainty model. By analyzing the time and frequency of fan signal,Usingwavelet analysis method to extract the fan vibration signal eigenvalues,Fuzzyprocessing for neural network training samples,so as to lay the foundation for faultdiagnosis of fan.The signal collection system is designed based on virtual instrument technology.Combined with the advanced of Labview, Matlab, SQL Server2008and VB languageand modular, the establishment of fuzzy neural network fault diagnostic expert system,it improves the timeliness and accuracy of mine fan fault diagnosis.This paper proposes a fault diagnosis system which is divided into front-end ofon-line monitoring and fault diagnosis of simple end of two parts. The use of Labview,SQL Server2008to establish a simple online monitoring and fault diagnosis of fanfront, data acquisition and monitoring interface, including friendly visual operationstate of each part of the display, storage, query, running state trend prediction and faultalarm etc. The use of VB, SQL Server2008and Matlab back-end intelligent faultdiagnosis of fan fault diagnosis, according to the reasoning mechanism of fuzzyneural system to establish the knowledge base of expert system, the neural networkand expert reasoning for fault diagnosis.In the end,to application the system,the results show that the system user-friendly,convenient and intuitive monitoring, with strong reliability and practicability.