Tool Condition Monitoring And Remaining Life Prediction Based On NHPP-WPHM Model

As a complex equipment integrating electromechanical and hydraulic,CNC lathe plays an crucial part in the operation of industrial growth.Therefore,it is very key to check its healthy operation.The CNC lathe is a complex system composed of multiple subsystems,only the reliable operation of each part can ensure the dependable working of the whole machine.As an important part of CNC machine tools,tools can be repaired during use.The remaining life evaluation of tools after different repair times can not only make full use of the service life of the tools,but also make reasonable spare parts while satisfying the quality of the workpieces.Change strategy.In this paper,a certain type of CNC lathe tool is used as an example to design an experiment,and the tool degradation data and the values of environmental influence factors in the machining process are collected.On the basis of obtaining the inherent reliability model of the tool,the Weibull proportional hazards model of the tool is obtained by processing,analyzing and modeling the collected data,that is,the comprehensive reliability model.On this basis,the proportional hazards model of tool update after different repair times is obtained by inhomogeneous Poisson process.In this paper,a tool availability model is established and the maximum availability is the monitoring target.On this basis,the monitoring threshold is analyzed,and the collected vibration signal is used as the input variable to monitor the real-time status of the tool.The remaining life prediction of the tool after different repair times is carried out.The main contents of this paper are as follows:(1)In order to analyze the overall(inherent)reliability level of the tool,on the basis of collecting the tool degradation data of the NC lathe,the tool failure time is obtained according to the tool failure threshold specified by the machinery industry.The optimal distribution of tool failure time is obtained by fitting and optimizing the distribution of failure time data.Different parameter estimation methods are used to estimate the parameters of the best distribution,and the parameters with the best fitting effect are obtained by taking the root mean square error as the fitting measurement standard.On this basis,the tool reliability model based on failure time is established.(2)Considering that the individual reliability of NC lathe tool is related to the machining process,and the vibration in the machining process has an obvious impact on the tool,the vibration information of the tool is selected as the external covariate factor affecting the tool reliability evaluation.So as to reduce the influence of noise on the real signal,the collected vibration signal is denoised by wavelet packet denoising algorithm and the signal characteristic parameters are extracted.In order to overcome the limitation of small sample data,the artificial fish swarm algorithm is used to optimize and improve the traditional grey prediction model,in order to find the law in the chaotic original signal characteristic parameters,so as to obtain the predicted value of signal characteristic parameters.On this basis,combined with the inherent reliability model of the tool,the Weibull proportional risk model,namely the tool comprehensive reliability model,is established.(3)Because the tool performance after repair is better than that before repair,but it can not be restored to the original performance,combined with the tool comprehensive reliability model,the reliability distribution form of the tool after the kth repair is obtained through inhomogeneous Poisson process.Taking the maximum availability as the goal,the monitoring threshold of reliability is set,and the real-time monitoring of tool wear state is realized through the collected covariate data.The residual life of NC lathe tools after the k-th repair is predicted and the validity of the model is verified by comparing with the actual remaining life and the remaining life prediction model based on the Weibull model.The application appears that the tack offered in this paper not only considers the overall reliability level of the tool,but also considers the influence of external factors on the reliability of the tool during operation.The method can correct the errors caused by the reliability model established only based on the tool degradation life data,and can better reflect the wear level of the tool,so as to accurately evaluate the remaining life of the tool.