Methods And Applications Of Reliability Estimation Of CNC Equipment Based On Performance Parameters

Computerized Numerical Control Machine is a machine tool,whose health state is directly related to the quality of the products and the safety of the enterprise.Therefore,the enterprise needs to grasp and evaluates the health state of Computerized Numerical Control Machine tools.Reliability evaluation techniques can access machine tools' health state by analysising the field inspection data,which can provide scientific basis for equipment maintenance.There are many researches about the reliability evaluation technology of CNC machine.The traditional reliability evaluation method requires a large number of samples,and it is difficult to evaluate the products with high reliability and long life.Meanwhile it cannot access the reliability of a single machine.Therefore,it is difficult to truly reflect the changes of the reliability of CNC machine tools.Meanwhile,the reliability evaluation method based on operational data didn't consider the different importances of the performance index and the influence degree of different subsystems on the system's reliability.In this paper,the reliability evaluation model of machine tool components based on improved hidden markov model and the reliability evaluation model of machine tool system based on Bayesian network are proposed.The specific work is as follows:Firstly,according to the monitoring data of CNC machine,multiple indexes are used instead of single index,meanwhile the importance of the index is also considered.Based on the existing implicit markov model,an improved model scheme is proposed.This model proposed an analysis method of Spearman weight,and the multi performance observation sequence and the Spearman rank correlation analysis on the health status was studied to get the weight of each performance index,which reflects the contribution of performance index in equipments' y health assessment.The multi observation sequence was transformed into a single observation sequence by using vector quantization and weighted analysis.The equipment state transition probability was obtained through the improved hidden Markov model,so as to realize the reliability assessment of machine parts level.Secondly,the fault tree analysis of the whole machine was done to obtain the Bayesian network nodes,and the fault tree and correlation relationship Pearson analysis of components were also done to obtain the correlation degree.The state probability of components which was obtained from improved hidden Markov model was put into the Bayesian network model in CNC machine tool to evaluate the reliability of CNC machine tool system;Finally,the advanced computer network technology was used based on the above analysis model.A CNC machine tool reliability evaluation system was developed in Microsoft Visual Studio whose programming language is CSharp,which realizes the state evaluation function of CNC machine tool performance data characteristics,to provide the scientific basis for maintenance decision-making,and then the summary and prospect were made.