Fault Diagnosis Expert System Of Equipment Based On Neural Network

The function of modern equipment is more and more diverse and intelligent,and its structure becomes more and more complex which leads to equipment maintenance and management difficulties and workload and complexity improved greatly,especially the equipment fault diagnosis problem.Because diverse factors caused fault,the maintenance and repairing personnel fault diagnosis accuracy and the work efficiency is reduced.The fault diagnosis technology research is the key technology to protect the equipment safety,high effective and reliable operation.This thesis is supported by Shanxi Province Education Department of special research program "based on embedded system of weapon equipment fault diagnosis method research".Taking fault diagnosis expert system for equipment as the research object,the thesis puts forward fault diagnosis expert system of equipment based on Neural Network combined with its characteristics of equipment and the problems encountered by fault diagnosis.Through the analysis of the advantages and disadvantages of various intelligent fault diagnosis technologies,intelligent diagnosis technology suitable for the subject is selected.Multi-mode and real-time fault diagnosis expert system based on neural network is constructed by comparing and analyzing the differences and connections between the neural network and expert system as well as in-depth studying the combination way of neural network and expert system.The fault diagnosis expert system is also transplanted to mobile handheld terminal providing a more convenient and flexible way to fault diagnosis.The real-time monitoring of the diagnosis equipment about the working voltage provides a more directly and effective basis for fault diagnosis,and reduces or avoids to manual participate in the process of fault diagnosis.The testing results indicates that the Elman neural network is better than BP neural network in the aspects of convergence speed and accuracy,which is based on the analysis of the structure and function of the Elman neural network model and the detection of the practical effect of the fault diagnosis process of a equipment.In addition,the dynamic characteristics of the Elman neural network makes it more suitable for dynamic working environment and it is more helpful for fault diagnosis expert system to have real-time fault diagnosis ability.The multi-model and real-time fault diagnosis expert system constructed has been tested.It can effectively solve the problems of low accuracy and low efficiency in the processing of fault diagnosis,and reduce the artificial participation,which provides convenient and flexible fault diagnosis methods.