Research On Diagnosis Method Of Bearing Fault Of Coal Mine Belt Conveyor

The belt conveyor is responsible for the transportation of materials during the coal production process,and its normal working state is an important guarantee for the safe production of coal mines.As the core component of the coal mine belt conveyor,the running state of the rolling bearing directly affects the stability and safety of the belt conveyor.Therefore,this paper takes the rolling bearing of belt conveyor as the research object,and conducts the following research around how to optimize the extraction of fault features and improve the accuracy of fault diagnosis:(1)The working environment of the coal mine belt conveyor is complex,and it is difficult to extract the fault characteristics of the bearing signal.In response to this problem,the wavelet packet noise reduction method is used to reduce the noise interference in the vibration signal;the advantages and disadvantages of empirical mode decomposition(EMD)and ensemble empirical mode decomposition(EEMD)are compared and analyzed,and the signal is decomposed by the better performance EEMD method Inherent mode function(IMF),and then extract the characteristic vector of the fault by calculating the energy contained in different IMF components.(2)In the process of fault classification,it is proposed to use artificial bee colony algorithm(ABC)to optimize the problem that support vector machine(SVM)penalty factor and kernel function parameters are not easy to determine.Aiming at the disadvantages of the traditional ABC algorithm that the global search ability is weak and easy to fall into the local optimum,the crossover operation and the global optimum solution are introduced to improve the colony search method,forming the cross global artificial bee colony algorithm(CGABC).After simulation analysis,it is found that the convergence speed and accuracy of the CGABC algorithm are significantly better than that of the ABC algorithm.Therefore,the CGABC-SVM model is constructed to diagnose the data samples.The results show that the classification accuracy and classification efficiency of the CGABC-SVM fault diagnosis model are higher,and the fault type can be effectively diagnosed.(3)Aiming at the problem that the degree of bearing failure is small and difficult to distinguish,the multi-class correlation vector machine(MRVM)is used as a fault diagnosis model to make full use of the advantages of MRVM,such as fewer adjustment parameters and direct realization of multiple classifications,to effectively diagnose the degree of bearing failure.Aiming at the premature and local optimization problems of particle swarm optimization(PSO),the inertia weight is improved to make it adaptively change,forming the adaptive weighted particle swarm optimization(APSO).On this basis,the APSO-MRVM fault diagnosis model is constructed and the bearing fault degree is diagnosed.The simulation found that the accuracy of the APSO-MRVM model for fault diagnosis is as high as 96%,and it has the advantages of low computational complexity and strong diagnostic ability.The paper has 39 pictures,10 tables,and 81 references.