Multi-information Fusion Technology In The Research And Application Of Fault Diagnosis Of Fan

In the process of modern industrial production, in order to make good use of the equipment, we always hope to predict the equipment fault, to nip in the bud, ensure that the equipment runs under the condition of safety, stabilization, long-period and full-load, so the discipline of equipment fault diagnosis comes into being and attracts larger numbers of scientific researcher to research on it, so as to keep it on being developed and perfected.Modern equipment is moving toward high-speed, high-power, high-reliability, large-scale, so that equipment faults become more complex and diverse, the traditional fault diagnosis methods have failed to accomplish the complexity of modern equipment fault diagnosis. For the purpose of this, this thesis introduces the idea of artificial intelligence to the field of equipment fault diagnosis. Thus solve the problem of complex equipment fault diagnosis.First, we carry out a detailed discussion on the type of fan fault in this thesis, analyzing the composition of fan, cases of fan fault, fault type, feasibility analysis for application of information fusion technology in fault diagnosis.Secondly, using the principle of wavelet packet to discuss the methods for extracting feature parameters of fan fault signal, to construct the feature vector of fault use the method of wavelet packet coefficients energy, as a source for the objective evidence of fault diagnosis. Then, discuss the fundamental of information fusion technology in detail. Describe the level of information fusion and its method, and the relationship between information fusion technology and equipment fault diagnosis, and its fault diagnosis model.Once more, information fusion fault diagnosis method based on D-S evidence theory to be discussed, Fault diagnosis model based of D-S evidence theory has been bult.Multi-fault feature information fusion diagnosis frame has been built, and provide a new method to construct belief function.Finally, explain the implement steps and methods of D-S evidence theory in detail through project cases of fan fault diagnosis, it has been proved that it can effectively improve the diagnosis reliability after multi-faults feature have been fused. Using the US excellent software of D-S Engine that achieved a higher accuracy has been proved by an experiment of fusion simulation.Application results show that using the non-stationary signal of complex equipment system to establish its artificial intelligent fault diagnosis system can obtain more physical meaning, to achieve accurate diagnosis.