数控铣床铣刀磨破损监测技术研究

Research on CNC Milling Cutter Grinding Damage Monitoring Technology

针对目前缺少铣刀磨损量建模研究的现状,通过搭建铣刀磨破损监测信号采集实验系统,对铣刀的磨破损进行全寿命的切削加工实验,采集了铣刀在加工中的加速度、切削力和声发射信号并对采集的信号进行时域、频域和时频域特征指标的提取,分析和评估了各特征指标对铣刀磨破损分类的敏感程度,筛选出适用于铣刀磨破损建模的特征向量。利用支持向量机分类算法建立了铣刀磨破损的分类模型,同时引入了粒子群算法对模型的参数进行优化。

In view of the current situation of the lack of modeling research on wear loss of milling cutters,the experimental system of damage monitoring signal acquisition for milling cutter was built,and the whole life cutting test was carried out on the milling wear of the milling cutter.The acceleration,cutting force and acoustic emission signal of the milling cutter were collected,and the characteristic indexes such as the time domain,frequency domain and time-frequency domain of the collected signals were extracted.The sensitivity of each characteristic index to the classification of milling wear damage was analyzed and evaluated,thus a feature vector suitable for milling damage modeling was selected.The classification model of milling wear was established by the support vector machine(SVM)classification algorithm,and the particle swarm optimization(PSO)was introduced to optimize the parameters of the model.