Multi-state Reliability Analysis Of Spindle System For NC Machine Tool Based On Probabilistic Model Checking

As the main part of industrial manufacturing,the reliability research of NC machine tool is paid more attention by users.Due to the reliability of CNC machine tool mechanical transmission structure and functional unit design and test process is not standard,causing frequent domestic machine tool numerical control system failure,so that its position in market competition at a disadvantage.Based on NC machine tool spindle system as the research object,combined with the damage degree of analysis,probabilistic model checking,markov model,multi-state system,methods of nonparametric kernel density estimation model to analyze reliability of NC machine tool,from the overall layer gradually positioning part of weak,for the reliability of NC machine tool research,analysis,to provide theoretical support.The main work of the thesis is as follows:(1)Research on the hazard degree of NC machine tool.Fault analysis for numerical control machine as a whole,through to each subsystem failure frequency statistics,and the fault categories,the frequency level,easy to check the degree of division evaluation,to carry out the empirical analysis of the harmfulness of the system.The weak subsystem of NC machine is constrained to the main shaft system of mechanical part.(2)A fault tree analysis method based on probabilistic model detection and makov process is studied.Probabilistic model detection can verify and analyze the random behaviors in the system,making fault tree analysis more accurate and efficient.At the same time,the problem of state explosion in the analysis of traditional fault tree can be avoided effectively.Compared with the research results of Probabilistic model detection and markov,it can be seen that this method can find the weak link of spindle system more precisely and provide theoretical basis for the further reliability analysis.(3)The reliability of spindle bearing is analyzed based on multi-state system theory.Based on the weak link bearing of spindle system,the reliability analysis of multi-state theory of bearing is carried out,and the coupling between multiple failure states based on multi-state system theory is analyzed.The variation trend of bearing reliability under multi-state theory is obtained.(4)A reliability analysis method based on nonparametric kernel density estimation is proposed.The non-parametric kernel density estimation of bearing failure time is obtained,and the kernel density prediction model is obtained.The prediction model is modified basedon the multi-state analysis of bearing parts and the related parameters based on 5M1E analysis.