线性回归模型诊断和在线预测刀具磨损量的方法研究  

A methodology of Tool wear diagnostic using Linear regression model and prognostic using double exponential smoothing

作　　者：吴越[1] 王玫[1] 陈勇[1] 

机构地区：[1]四川大学制造科学与工程学院,成都610065

出　　处：《机械设计与制造》2009年第6期236-238,共3页Machinery Design & Manufacture

摘　　要：目的是研究诊断端面铣刀磨损量和在线预测铣刀的剩余寿命的方法。采用线性回归模型估计测刀面的磨损量。线性回归模型的输入是从铣刀受力信号提取出的特征和切削条件,比如进给量、转速等。在诊断了刀具的磨损量后,采用双指数平滑方法跟随诊断结果预测铣刀的使用寿命。最后,通过实验验证了基于线性回归模型得到的刀具的磨损量和基于双指数平滑方法在线预测铣刀的剩余寿命的可行性。The primary objective of this research is to diagnostic the flank wear and prognostic the remaining useful life of the cutting tool in face milling. Linear regression model is trained to estimate the flank wear on cutter inserts. The input to the linear regression model are the features extracted from the force signal and other known cutting parameters such as feed rate ,and spindle speed. After the estimation of flank wear,the double exponential smoothing is then applied to follow the trend of estimated tool wear data and prognostic the remaining useful life. Finally,it is confirmed experimentally that the tool wear and remaining useful life can be well estimated by linear regression model and the double exponential smoothing.

关 键 词：刀具磨损 线性回归 双指数平滑 

分 类 号：TH16[机械工程—机械制造及自动化] TP391[自动化与计算机技术—计算机应用技术]