Research Of Vibration Signal Processing And Fault Diagnosis Method For Rotating Machinery

Rotating machinery vibration signal processing and fault diagnosis is the core part of rotating machinery fault detection and identification,in the field of engineering,the vibration signal of rotating machinery carries the important feature of the equipment running status information,the correct processing of these signals and accurate fault analysis to monitor the running state of the equipment to predict in advance equipment of sudden failure,reduce because of the risks of equipment failure,has the vital significance.This paper introduced stochastic resonance and its improved algorithm and bipolar fuzzy theory analysis method and other methods,through the in-depth study of these algorithms,and apply it to the vibration signal processing and fault diagnosis of rotating machinery,the main content is as follows:In that research of weak fault feature signal extraction under strong background noise,aiming at the defect that the parameter a,b,h of the stochastic resonance system and the penalty factor alpha and the number k of the modal decomposition of the variational modal decomposition are often poor in processing effect due to imperfect preset when the parameter a,b,h and the number k of the modal decomposition are preset in advance,method of combining signal processing is proposed based on chaos particle swarm optimization(pso)algorithm parameters optimization of stochastic resonance and variational mode decomposition.Using chaotic particle swarm optimization(pso)algorithm to optimize the parameters of the stochastic resonance and VMD respectively,then processing the signal,and through the simulation and experiment show that this method can effectively extract the fault signal characteristic under the strong background noise.The fuzzy c-means clustering fault classification method based on membership and hard c clustering is studied.FCM classification method is applied to fault classification in combination with stochastic resonance and variational mode decomposition.According to the characteristics of the sample entropy and mean square value of the signal processed by stochastic resonance and variational mode decomposition,the fault can be classified effectively by this method.The fault characteristic fuzzy theory analysis method is studied,and the complexity of fault degree diagnosis is deeply studied.the bipolar fuzzy theory analysis method is innovatively applied to the fault degree evaluation and compared with the traditional unipolar fuzzy theory analysis method,this method combines fault comprehensive evaluation method and the bipolar fuzzy theory,can be more in-depth analysis for fault conditions.To test in this paper,the fault feature extraction method and the effectiveness of the fault degree evaluation method,In this paper,by using DDS power transmission fault diagnosis comprehensive experimental platform buy from SpectraQuest Co.,LTD.for further research and further validation,through experimental platform to collect four signal of rolling bearing.At the same time compare using VMD,fuzzy c-means clustering method with using parameters based on chaos particle swarm optimization of stochastic resonance,VMD based on parameters of chaotic particle swarm optimization,the fuzzy c-means clustering method.Comparing with the traditional fuzzy theory analysis method with bipolar fuzzy theory analysis method.Through the experiment show that,the accuracy of the SRVMD combined with FCM method this article used is obviously better than the accuracy of the method VMD combined with FCM method,Based on the theory of bipolar fuzzy fault degree of evaluation method in this paper,is a theory evolved from the basis of bipolar fuzzy analysis method and comprehensive evaluation is more effective.