Failure Analysis And Reliability Growth Assessment Of CNC System

CNC system is the brain of machine tools.Its reliability level affects the performance of the whole machine,even the level of national manufacturing industry.Improving the reliability of CNC system is the key to the rapid development of machine tools,and an important way to comprehensive national strength and national status.This paper originated from the national science and technology major sub-project " brain switching engineering of domestic CNC system for manufacturing typical components of aeroengine ".In this paper,FMECA and FTA analysis of early faults are carried out for a domestic CNC system in the project,and corresponding measures are given according to the analysis results.The application of reliability growth model in reliability growth of CNC system is studied,which has certain theoretical significance and practical value for improving reliability of CNC system and the plan of developing reliability growth.The research content includes the following aspects.(1)Based on the hierarchical analysis of the CNC system and collected failure data,the FMECA method is used to determine the frequent occurrence of failures in the CNC system,the frequent failure modes,the frequent causes of failures and the hazards of each subsystem and its corresponding failure modes.(2)According to the characteristics of the bathtub curve and fault data,the early failure of the CNC system is analyzed.In order to determine the early failure period of the numerical control system,superposed weibull models,superposed PLP models and superposed LLP models are established respectively.Using the Minitab probability map,find the optimal life distribution of the CNC system,and determine the optimal model initially.By calculating the correlation coefficient of the model and analyzing the failure rate curve,the superposed LLP model is finally determined as the best model.The inflection point of the early failure period calculated by the model can be used to screen out the early failure from the collected data.For the early failures of the CNC system,the fault tree analysis is carried out.Based on fault tree analysis,this paper uses fuzzy analytic hierarchy Process and considers the weights of scoring experts to evaluate the risk of bottom event of fault tree,which is the root cause of fault,and to identify the weak points of CNC system in early failure period.(3)From the point of view of CNC system enterprises,the reliability assurance system of outsourced parts is established,the reliability control flow of outsourced parts is formulated,and the reliability assurance measures for each stage of the life cycle of outsourced parts are put forward.(4)Aiming at the reliability growth process of TAAF mode in continuous time(single stage),Duane model and AMSAA model based on failure time are used to evaluate the reliability growth.(5)For the multi-stage reliability growth process,combined with the Bayes method.According to the life distribution characteristics of the numerical control system,the dirchlet distribution is used as the prior distribution,the comprehensive historical information and the various growth stages.Fault data,given joint posterior distribution,Bayes inference and estimation by Gibbs sampling method.The reliability point estimation and interval estimation of CNC system in each stage are obtained.