Failure Prediction Of NC Lathes Based On Fuzzy Reasoning

The improvement of the reliability level of CNC machine tools is beneficial to the development of manufacturing industry in China.In this paper,a series of CNC lathe is selected as the reliability analysis object,and the method to improve the overall reliability of CNC lathe is found out.Based on the fault data of a series of CNC lathes,this paper divides the fault levels of CNC lathes,fits the G-R curve of the fault levels,and uses B value and fuzzy theory to predict the faults.The main research work of this paper is as follows:(1)A large number of fault data of CNC lathe are collected.Based on the field fault data and considering the composition,function and working characteristics of CNC lathe,the subsystem of CNC lathe is divided.Through the overall analysis of CNC lathe by FMECA,the subsystem fault criticality of CNC lathe is obtained.Based on the subsystem hazard,the fault grade of CNC lathe is divided.(2)According to the fault data of numerically controlled lathes,combined with the fault level for sorting and statistical analysis,the G-R curve in seismology is used to build a numerically controlled lathe fault level model.The G-R curve is applied to the reliability research of CNC lathe.Some similar conclusions have been obtained in the research of CNC lathe.When the b value is close to 1.0,the CNC lathe works stably and the failure will not occur frequently.(3)Applying the parameter b value in G-R curve and combining with fuzzy theory,we try to predict the failure of CNC lathe.When the b value changes abnormally,the CNC lathe may fail.At this time,we should stop the machine for detection,in order to prevent the occurrence of large-scale failure and reduce unnecessary downtime,so as to improve the reliability of CNC lathe.