Health State Assessment Of CNC Machine Tools Based On TOPSIS And Grey Clustering Theory

With the rapid development of manufacturing technologies and fierce competition in the international market,failure prediction and health management technologies for production equipment are becoming important means for manufacturing companies to reduce production costs and enhance their core competitiveness.This article takes the key equipment of the manufacturing system—CNC machine tools as the research object,and conducts research on the field of health assessment of CNC machine tools,which helps manufacturers to manage and maintain CNC machine tools,reduce unplanned machine downtime and reduce maintenance management costs,and has important research and practical significance.From the perspective of the user of the CNC machine tool,this paper gives the definition of the health state of the CNC machine tool so that it has a clear physical meaning.In order to quantify the health state level of CNC machine tools,combining the ideal state of the physical hypothesis,defines the health index of CNC machine tools as the degree of deviation of the CNC machine tools from its assumed ideal state.Based on the theory of TOPSIS(The Technique for Order Preference by Similarity to Ideal Solution),a numerical model and method for the health state of CNC machine tools are proposed,which takes the machine feature vector as input and outputs the corresponding health index.An improvement method based on the Mahalanobis distance quantization model is proposed to solve the problem that the performance characteristics of CNC machine tools do not completely satisfy the linear independence due to the mutual influence of some of the performances of CNC machine tools.In order to fit the expression habits of people in different fields for health,the health status levels of CNC machine tools are defined and divided.Considering the advantages of grey clustering in small sample and poor information environment,a hierarchical model of health status of CNC machine tools based on the grey clustering theory is proposed.The hierarchical model uses the health index calculated by the quantitative model as input and outputs the corresponding health state level.In order to solve the problem of determining the whiten weight function based on expert experience,a whiten weight function solving method based on sample data is proposed,which can reduce the dependence of the hierarchical model on expert experience.In order to verify and explain the above-mentioned health state quantification model and hierarchical model,designs a simulation experiment of the accelerated degradation of the main shaft.The experimental results show that and accuracy of the model for health status assessment can achieve 90%.Finally,based on the above methods and models,a prototype system for health assessment of CNC machine tools was designed and developed,then applied to manufacturing companies.The application results show that the feasibility and effectiveness of the above method were further verified.