Design And Construction Of Bearing Intelligent Fault Diagnosis System

As a general component of mechanical equipment,the normality of rolling bearing’s state directly affects the safe operation of the equipment.Therefore,research on rolling bearing fault diagnosis technology has important theoretical significance and practical value.Aiming at the problems of the fault diagnosis instrument,such as less detection function,poor diagnostic ability and inconvenient carrying,a bearing intelligent fault diagnosis system was researched and designed.The system is based on a data acquisition instrument and the Android platform.Combined with adaptive resonance demodulation technology and support vector machine method,it achieved good diagnostic results.The main work of the study is as follows:(1)Research on adaptive resonance demodulation technology.For the selection problem of optimal band-pass filter,the composite index and particle swarm optimization algorithm are introduced to determine the optimal band-pass filter.Firstly,the possible range of the optimal filter parameters is determined by kurtosis.Then the fault pulse energy factor is used as the fitness function.The optimal band-pass filter is found by PSO.The verification of fault feature frequency extraction was performed in the digital simulation and actual signals.The results show that the method can effectively extract early weak faults.(2)Research on rolling bearing faults using support vector machine.The classification principle of SVM was studied to optimize the parameters of SVM by the improved PSO algorithm.For the SVM input vector determination problem,the adaptive resonance demodulation technology is used to denoise the signal.Then the sensitivity function is introduced to screen the fault characteristics of the rolling bearing.Different types of fault signals were collected for fault diagnosis experiments and the results verify the effectiveness of the method for fault location of rolling bearings.(3)Design and build a bearing intelligent fault diagnosis system.The entire system consists of two parts: a data acquisition instrument and a fault diagnosis software.The SVM algorithm and the adaptive resonance demodulation method are transplanted into the diagnostic software to improve the classification accuracy.The cloud database is added to implement remote fault diagnosis.Finally,the diagnostic system was tested on the rolling bearing failure test platform.The results show that the system has better diagnostic accuracy for multiple faults of bearings.