基于数字孪生的铣刀磨损状态识别方法研究  

Research on Recognition Method of Milling Cutter Wear State Based on Digital Twin

作　　者：水星 容芷君[1,2] 但斌斌[1,2] 何强鉴 杨鑫 SHUI Xing;RONG Zhijun;DAN Binbin;HE Qiangjian;YANG Xin(School of Mechanical Automation,Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081;Key Laboratory of Metallurgical Equipment and its Control,Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081)

机构地区：[1]武汉科技大学机械自动化学院,武汉430081 [2]武汉科技大学冶金装备及其控制教育部重点实验室,武汉430081

出　　处：《组合机床与自动化加工技术》2024年第9期20-24,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(51475340);湖北省重点研发项目(2022BAA059)。

摘　　要：实时精准地监测铣刀磨损状态对于提高加工质量与加工效率具有重要意义,提出一种基于数字孪生的铣刀磨损状态识别方法,该方法通过结合VMD-MPE特征提取方法和GA-SVM状态识别模型构建数字孪生体对铣刀磨损状态进行实时监测。首先,利用变分模态分解算法(VMD)分解铣刀振动信号得到包含磨损状态信息的模态分量;其次,引入多尺度排列熵(MPE)从包含磨损状态信息的模态分量中提取铣刀的非线性动力学特征,并取各有效模态分量的多尺度排列熵平均值作为特征矩阵;最后,通过遗传算法(GA)优化支持向量机(SVM)构建铣刀磨损状态识别模型。实验结果表明,所构建的数字孪生体具有良好识别效果,其识别精度可达97.33%。Real-time and accurate monitoring of the wear state of milling cutter is of great significance to improve the machining quality and efficiency.In this paper,a milling cutter wear state recognition method based on digital twin is proposed.This method combines VMD-MPE feature extraction method and GA-SVM state recognition model to construct digital twins to monitor the wear state of milling cutter in real time.Firstly,the variational mode decomposition(VMD)algorithm is used to decompose the vibration signal of the milling cutter to obtain the modal component containing the wear state information.Secondly,the multi-scale permutation entropy(MPE)is introduced to extract the nonlinear dynamic characteristics of the milling cutter from the modal components containing the wear state information,and the average value of the multi-scale permutation entropy of each effective modal component is taken as the characteristic matrix.Finally,the genetic algorithm(GA)is used to optimize the support vector machine(SVM)to construct the milling cutter wear state recognition model.The experimental results show that the digital twin constructed in this paper has a good recognition effect,and its recognition accuracy can reach 97.33%.

关 键 词：数字孪生 刀具磨损 状态识别 变分模态分解 多尺度排列熵 支持向量机 

分 类 号：TH164[机械工程—机械制造及自动化] TG502[金属学及工艺—金属切削加工及机床]