| Tool is a key component of CNC machine tools,and its wear directly affects the surface quality of the workpiece and the accuracy of machining dimensions.The traditional tool changing method based on experience or quantitative method not only increases the labor intensity of the operator,but also reduces the service life of the tool and the processing efficiency of the workpiece,which seriously restricts the development of CNC machine tools in the field of intelligent manufacturing.Therefore,it is of great significance to study the performance degradation risk assessment of tools without sudden failure for the formulation of preventive tool replacement strategies.Aiming at the performance degradation process of the tool without sudden failure,this paper calculates the performance health index and models the degradation probability based on the collected vibration signal and the vibration signal considering the external load impact,combined with the severity of the tool state and the degradation consequences,the performance degradation risk of the tool with and without external impact is evaluated,and the degradation probability,degradation state and degradation consequences are used as variables to analyze its sensitivity to the performance degradation risk,which provides evidence for tool reliability management.The main research contents are as follows:（1）Based on the collected tool vibration signals,model the tool degradation probability with and without external impact.The vibration signal of the precision milling die of the machining center is collected,and it is fused with the normal random impact signal to obtain the tool vibration signal considering the external load impact,and the wavelet semi-soft threshold function is used to reduce the vibration signal.Combine correlation,monotonicity and robustness to extract time domain features,use empirical mode decomposition and singular value decomposition to extract time-frequency domain features,use time-frequency feature fusion to construct Mahalanobis distance,use CUMSUM（Cumulative Sum）method to calculate performance health indicators,through fitting test and the least squares parameter estimation,the tool degradation probability model is constructed.（2）Evaluation of the consequences and severity of tool performance degradation.The distribution fitting model of the workpiece surface roughness and the effective value of the tool vibration signal is constructed,and the performance degradation consequence is represented by the ratio of the workpiece surface roughness and its accuracy at different times.According to the failure threshold and performance health index,the tool performance degradation rate is calculated,and the support vector machine is used to classify the frequency domain energy.The performance degradation rate,significant frequency domain energy and its weight are used to construct the degradation state index;the degradation state and performance degradation consequences are combined to evaluate the performance degradation severity with and without external impact.（3）Combine the probability of tool degradation and the severity of performance degradation,and evaluate the risk of tool performance degradation based on the risk principle;taking degradation state and performance degradation consequences as variables,analyze its sensitivity to the severity of performance degradation;the degradation state,degradation consequence and degradation probability corresponding to Monte Carlo simulation are comprehensively analyzed,and their sensitivity to performance degradation risk is analyzed respectively,which provides a basis for tool reliability growth. |
