Thermal Error Analysis And Experimental Study Of CNC Machine Tool Feeding System

Rolling bearings are widely used in machinery equipment and occupy an important position.Many faults in mechanical equipment are closely related to the running condition of rolling bearing.A large proportion of faults in rotating mechanical equipment is caused by the failure of bearing,so it is of great practical significance to carry out fault diagnosis of rolling bearing reasonably.In this paper,the identification of rolling bearing fault state are carried out by logistic regression and supported vector data description.Firstly,local mean decomposition and logistic regression model were used to identify the fault type of rolling bearing.The main problem was to determine the inner or outer fault.Then,identifing different damage process of a certain fault type from slight to severe state by local mean decomposition and support vector data description model mainly,which can identify the damage degree of rolling bearing.Local mean decomposition is an adaptive time-frequency analysis method for nonlinear and unstable faults vibration signals of rolling bearing.In this paper,the local mean decomposition method was applied to the vibration signal processing and analysis of the rolling bearing fault.Then,the eigenvalues were selected and the genetic algorithm was used to optimize the parameters of the logistic regression model.The logistic regression model was obtained and the fault type identification of the rolling bearing was performed.According to the vibration signal after local mean decomposition,the support vector data description model was selected and the support vector data description model was established and the damage degree of a certain rolling bearing fault type was identified.The verification of the rolling bearing fault data showed that the method can identify the fault type and damage degree of the rolling bearing.