Fault Diagnosis Of Rolling Bearing Based On Morphological Filtering And MCKD

Rolling bearing is one of the most commonly used and easily damaged components in all kinds of rotating machinery.Therefore,it is a topic of great concern in the field of engineering and technology at home and abroad.Aiming at a series of problems in the running process of bearings,this paper takes the bearing vibration signal as the goal,and studies the problem of fault frequency extractionfrom fault signals.The main contents are as follows:First,based on the theoretical foundation and practical application,the research status of rolling bearing fault diagnosis is analyzed,and the development direction of rolling bearing fault diagnosis in recent years is also introduced.On this basis,the important significance of fault diagnosis to the stability and safety of high-speed locomotive operation is expounded.In view of the problems and difficulties that exist in the process of fault diagnosis,this paper lists the contents and technical solutions for the problems to be studied in this paper.The common fault types and reasons of bearing are analyzed through bearing material,and the fault frequency is derived by formula.Secondly,in view of the blindness in the selection of structural elements in the process of morphological filtering,the paper compares and analyzes the filtering effect of different shape elements,and finds out the shape of the structural elements that have the best filtering effect.The length of the structural elements is critical for the impact of the extraction of signal characteristics,in order to avoid unreasonable structure element selection cause the failure of the impact of the feature extraction,this paper designed the kurtosis and impact the composition ratio of the composite evaluation index to guide the selection of structure element length.The experiment shows that a good filtering effect can be obtained by using the optimal structural elements to filter the signal by morphological filtering.Then,the improved MCKD algorithm is used to process the signal after morphological filtering.Morphological filtering can only achieve the effect of noise reduction.After analyzing the filtered signal,we can not extract the impact characteristics accurately,so we need to extract the impact feature through the MCKD algorithm.MCKD algorithm L is selected according to the experience of value selection,this paper designs the fault characteristic frequency energy ratio of K to make a quantitative evaluation of the selection of L,in order to find the best extraction effect of L,and the Teager Operator Demodulation of the signal after processing,can greatly increase the impact characteristics of components,and then find the fault impact characteristics.Finally,by setting parameters in particle swarm optimization,the parameters in MCKD are adaptively optimized,and the fault frequency is extracted quickly and accurately.The extracted fault frequency curve is clear.The MCKD algorithm of particle swarm optimization(PSO)is used to deal with the signal,which can achieve high accuracy of fault diagnosis and high precision of the amplitude of the extracted fault frequency.