Wear Condition Monitoring of Milling Cutter Based on Random Forest Method
Aiming at the problem of tool condition monitoring is difficult in milling machine tool processing and incomplete infor-mation in single sensor monitoring method,this paper proposes a tool wear condition monitoring method which uses Random For-ests to complete multi-sensor information fusion.Under different cutting parameters,acoustic emission sensor and vibration sen-sor are used as signal acquisition elements to collect multi-directional signals,and the signals are analyzed in time domain,fre-quency domain and wavelet packet.A total of 23 characteristic quantities that are sensitive to the cutting tool state are extracted,and Random Forests is constructed to monitor the milling cutter state.The experimental results show that the accuracy of the method is90.0%,which is feasible.
万方数据
维普
