The state of the tool is an important factor that affects the processing quality and production efficiency.The development of signal processing technology,such as sensors,feature extraction and machine learning,provides new ideas for tool condition monitoring.However,the establishment of high precision and good robustness of the multi-sensor system is still the difficulty of tool condition monitoring.Based on the experiment of milling 30CrMnMoRE,the physical parameters such as cutting force and vibration under different tool conditions are analyzed,and a more reliable method is proposed by comparing various feature extraction and pattern recognition methods.Cutter state monitoring system is used to achieve high-strength steel cutting tool wear state recognition.%刀具磨损状态是影响加工质量和生产效率的重要因素之一,传感器、特征提取、信息融合和机器学习等技术的发展为刀具状态监测提供了新思路,然而,建立精度高、鲁棒性好的多传感器系统仍是刀具状态检测的难点.以涂层硬质合金刀具对难加工材料30CrMnMoRE的铣削试验为基础,通过对不同刀具磨损状态下的切削力和切削振动进行采集分析,并比较多种特征提取和模式识别方法,建立了一种更为可靠的铣刀状态监测系统,实现了对高强度钢切削时的刀具磨损状态识别.
展开▼