| Fault diagnosis is a new muti-subject-crossed technique, it has been rapidly in last twenty years, and it has brought huge benefit. Information fusion is a subject formed recently; it has been researched and applied in many fields. But, it is in starting stage in fault diagnosis. By reason of the combination of self-learning characteristic of neural network and local characteristic of wavelet analysis, wavelet neural network has self-adapting and favorable fault tolerant ability, so it has been applied in fault diagnosis field far and wide. The fault diagnosis model based on the integrated neural network is put forward in this paper, firstly, information fusion in machine equipment and the basic methods of prediction the fault characteristic information from the vibrating signal for the equipment are presented. Then the tight wavelet neural network was constituted taking the nonlinear Morlet wavelet radices as the stimulant function, the idiographic algorithm was presented. And contrast with BP network by experimental analysis, it proved that wavelet neural network is adaptive in complex pattern classifying, so wavelet neural network is used in integrated fault diagnosis system, the realizable policy of it and the established principle of the sub-wavelet neural networks were given. Lastly, the integrated- wavelet neural network fault-diagnosis system were set up, it based on both information fusion technology and actual fault diagnosis takes the sub-wavelet neural network as primary diagnosis from different sides, and then gains the conclusions through decision-making fusion, It proved that the integrated- wavelet neural network fault-diagnosis system takes more advantage of diversified characteristic information than the single wavelet neural network fault-diagnosis system, and it can solve the difficulty and problem what could not solved by single neural network, moreover, the result is more all-sided and accurate than the single network.
|
PDF Full Text Request