数据机理双模型融合的刀具磨损监测方法

doi:10.3901/JME.260104

摘要/Abstract

摘要： 针对刀具磨损监测中数据驱动方法缺乏物理可解释性、机理模型实时性不足的问题，提出一种数据与机理双模型融合的预测算法。首先，构建振动信号多尺度时空特征提取网络，通过融合多尺度卷积神经网络与Transformer，捕捉刀具退化的动态特性；建立铣削过程的有限元仿真模型，基于Usui磨损率方程模拟刀具几何演化；设计强跟踪卡尔曼滤波算法，利用机理仿真结果作为状态转移先验、数据预测值作为观测更新量，实现刀具磨损量的动态最优估计。试验结果表明，与单一模型驱动的预测方法相比，所提的数据机理双模型融合预测算法能显著提高刀具磨损监测的精度与鲁棒性。

关键词: 刀具磨损监测, 数据和机理融合模型, 强跟踪卡尔曼滤波算法, 多尺度时空特征提取网络, 有限元

Abstract: To address the limitations of data-driven methods lacking physical interpretability and mechanistic models' insufficient real-time capability in tool wear monitoring, a novel predictive algorithm is proposed by fusing data-driven and mechanism-based models. First, a multi-scale spatiotemporal feature extraction network for vibration signals is constructed, capturing the dynamic characteristics of tool degradation by integrating multi-scale convolutional neural network and Transformer model. A finite element simulation model for the milling process is established, and the Usui wear rate equation is used to simulate the tool’s geometric evolution pattern. A strong tracking Kalman filter algorithm is designed, where the mechanism simulation results serve as the prior for state transition, and data prediction values are used for observation updates, enabling dynamic optimal estimation of tool wear. Experimental results demonstrate that, compared to single-model-driven prediction methods, the proposed data-mechanism dual-model fusion prediction algorithm significantly improves the accuracy and robustness of tool wear monitoring.

Key words: tool wear monitoring, data and mechanism fusion model, strong tracking Kalman filter, multi-scale spatiotemporal feature extraction network, finite element

中图分类号:

TH165

陈安航, 张海霞, 袁东风, 韩乔剑, 李娜, 曹凤. 数据机理双模型融合的刀具磨损监测方法[J]. 机械工程学报, 2026, 62(4): 37-51.

CHEN Anhang, ZHANG Haixia, YUAN Dongfeng, HAN Qiaojian, LI Na, CAO Feng. Data-mechanism Dual-model Fusion Method for Tool Wear Monitoring[J]. Journal of Mechanical Engineering, 2026, 62(4): 37-51.

参考文献

[1] LIU S，BAO J，ZHENG P. A review of digital twin-driven machining：From digitization to intellectualization[J]. Journal of Manufacturing Systems，2023，67：361-378.
[2] REHORN A G，JIANG J，ORBAN P E. State-of-the-art methods and results in tool condition monitoring：A review[J]. The International Journal of Advanced Manufacturing Technology，2005，26：693-710.
[3] 王宇鹏. TC4铣削过程球头铣刀磨损仿真研究[D]. 哈尔滨：哈尔滨理工大学，2020. WANG Yupeng. Wear simulation of ball-end milling cutter during milling of TC4[D]. Harbin：Harbin University of Science and Technology，2021.
[4] YEN Y C，SÖHNER J，LILLY B，et al. Estimation of tool wear in orthogonal cutting using the finite element analysis[J]. Journal of Materials Processing Technology，2004，146(1)：82-91.
[5] 孙玉晶. 钛合金铣削加工过程参量建模及刀具磨损状态预测[D]. 济南：山东大学，2014. SUN Yujing. Parametric modeling of milling titanium alloy and prediction of tool wear state[D]. Jinan：Shandong University，2014.
[6] YADAV R K，KUMAR A，MAHAPATRA S S. A simulation approach for estimating flank wear and material removal rate in turning of Inconel 718[J]. Simulation Modelling Practice and Theory，2015(52)：1-14.
[7] 于晓琳，吕诚凯，黄树涛，等. 铣削高强度钢刀具磨损有限元仿真研究[J]. 工具技术，2022，56(1)：9-14. YU Xiaolin，LÜ Chengkai，HUANG Shutao，et al. Finite element simulation study on tool wear of milling high- strength steel[J]. Tool Technology，2022，56(1)：9-14.
[8] PAUL P S，VARADARAJAN A S. A multi-sensor fusion model based on artificial neural network to predict tool wear during hard turning[J]. Proceedings of the Institution of Mechanical Engineers，Part B：Journal of Engineering Manufacture，2012，226(5)：853-860.
[9] WU D，JENNINGS C，TERPENNY J，et al. A comparative study on machine learning algorithms for smart manufacturing：Tool wear prediction using random forests[J]. Journal of Manufacturing Science and Engineering，2017，139(7)：071018.
[10] 文成林，吕菲亚. 基于深度学习的故障诊断方法综述[J]. 电子与信息学报，2020，42(1)：234-248. WEN Chenglin，LÜ Feiya. Review on deep learning based fault diagnosis[J]. Journal of Electronics & Information Technology，2020，42(1)：234-248.
[11] 何彦，凌俊杰，王禹林，等. 基于长短时记忆卷积神经网络的刀具磨损在线监测模型[J].中国机械工程，2020，31(16)：1959-1967. HE Yan，LING Junjie，WANG Yulin，et al. Inprocess tool wear monitoring model based on LSTM-CNN[J]. China Mechanical Engineering，2020，31(16)：1959-1967.
[12] 葛慧，韩林池，麻俊方，等. 采用门控循环单元神经网络和多特征融合的铣削刀具磨损监测[J]. 机械科学与技术，2024，43(4)：667-673. GE Hui，HAN Linchi，MA Junfang，et al. Milling tool wear monitoring by using gated recurrent unit neural network and multi-feature fusion[J]. Mechanical Science and Technology for Aerospace Engineering，2024，43(4)：667-673.
[13] KUREK J，ŚWIDERSKA E，SZYMANOWSKI K. Tool wear classification in chipboard milling processes using 1-D CNN and LSTM based on sequential features[J]. Applied Sciences，2024，14(11)：4730.
[14] 李东阳，袁东风，张海霞，等. 云边端协同的机床刀具故障智能诊断系统研究[J]. 中国机械工程，2023，34(5)：584-594. LI Dongyang，YUAN Dongfeng，ZHANG Haixia，et al. Research on intelligent tool fault diagnosis system of machine tools with cloud-edge-device collaboration[J]. China Mechanical Engineering，2023，34(5)：584-594.
[15] ZHOU C，ZHANG K，XU J，et al. Tool wear monitoring for robotic milling based on multi-dimensional stacked sparse autoencoders and bidirectional LSTM networks with singularity features[J]. The International Journal of Advanced Manufacturing Technology，2025，136：3534-3557.
[16] ZHU M，ZHANG J，BU L，et al. Methodology and experimental verification for predicting the remaining useful life of milling cutters based on hybrid CNN-LSTM-Attention-PSA[J]. Machines，2024，12(11)：752.
[17] HAO C，MAO X，MA T，et al. A novel deep learning method with partly explainable：Intelligent milling tool wear prediction model based on transformer informed physics[J]. Advanced Engineering Informatics，2023，57：102106.
[18] LIU H，LIU Z，JIA W，et al. Tool wear estimation using a CNN-transformer model with semi-supervised learning[J]. Measurement Science and Technology，2021，32(12)：125010.
[19] 赵凤，耿苗苗，刘汉强，等. 卷积神经网络与视觉Transformer联合驱动的跨层多尺度融合网络高光谱图像分类方法[J]. 电子与信息学报，2024，46(5)：2237-2248. ZHAO Feng，GENG Miaomiao，LIU Hanqiang，et al. A cross-layer multiscale fusion network hyperspectral image classification method jointly driven by convolutional neural network and visual transformer[J]. Journal of Electronics and Information，2024，46(5)：2237-2248.
[20] LIU X，JIANG D，TAO B，et al. A systematic review of digital twin about physical entities，virtual models，twin data，and applications[J]. Advanced Engineering Informatics，2023，55：101876.
[21] LU Y，LIU C，KEVIN I，et al. Digital twin-driven smart manufacturing：Connotation，reference model，applications and research issues[J]. Robotics and Computer-integrated Manufacturing，2020，61：101837.
[22] LUO W，HU T，YE Y，et al. A hybrid predictive maintenance approach for CNC machine tool driven by digital twin[J]. Robotics and Computer-Integrated Manufacturing，2020，65：101974.
[23] 张春霖，周婷婷，胡天亮，等. 刀具切削变工况数字孪生模型构建方法[J]. 计算机集成制造系统，2023，29(6)：1852-1866. ZHANG Chunlin，ZHOU Tingting，HU Tianliang，et al. Construction method of digital twin model for cutting tools under variable working conditions[J]. Computer Integrated Manufacturing System，2023，29(6)：1852-1866.
[24] DAI J，TIAN L，HAN T，et al. Digital twin for wear degradation of sliding bearing based on PFENN[J]. Advanced Engineering Informatics，2024，61：102512.
[25] 信苗苗，曹凤，江铭炎，等. SDU-QIT立铣刀磨损试验数据集[J]. 机械工程学报，2022，58(9)：166-171. XIN Miaomiao，CAO Feng，JIANG Mingyan，et al. SDU-QIT end milling cutter accelerated life test datasets[J]. Journal of Mechanical Engineering，2022，58(9)：166-171.
[26] TAO F，XIAO B，QI Q，et al. Digital twin modeling[J]. Journal of Manufacturing Systems，2022，64：372-389.
[27] KYRATSIS P，TZOTZIS A，DAVIM J P. A comparative study between 2D and 3D finite element methods in machining[M]//3D FEA Simulations in Machining. Cham：Springer International Publishing，2023.
[28] 蒋志涛，刘泓滨，王飞，等. ABAQUS的高速铣削二维仿真分析[J]. 现代制造工程，2008(8)：45-47. JIANG Zhitao，LIU Hongbin，WANG Fei，et al. Finite element analysis of two-dimensional high speed milling process based on ABAQUS[J]. Modern Manufacturing Engineering，2008(8)：45-47.
[29] 董辉跃. 航空整体结构件加工过程的数值仿真[D]. 杭州：浙江大学，2004. DONG Yuehui. Machining process simulation of aerospace monolithic component[D]. Hangzhou：Zhejiang University，2004.
[30] LI B. A review of tool wear estimation using theoretical analysis and numerical simulation technologies[J]. International Journal of Refractory Metals and Hard Materials，2012，35：143-151.
[31] 朱志武，宁建国. Johnson-Cook冲击动态本构模型研究[C]//中国力学学会，北京. 塑性力学新进展——2011年全国塑性力学会议论文集，2011：78-83. ZHU Zhiwu，NING Jianguo. Research on Johnson-Cook impact dynamic constitutive model[C]//Proceedings of National Conference on Plastic Mechanics. Beijing：Chinese Society of Theoretical and Applied Mechanics，2011：78-83.
[32] 姜增辉，宋亚洲，贾民飞. 基于Usui模型的硬质合金刀具切削高强度钢磨损仿真研究[J]. 制造技术与机床，2019(5)：112-115，121. JIANG Zenghui，SONG Yazhou，JIA Minfei. Research on simulation of carbide tool wear in cutting high-strength steel based on the Usui model[J]. Manufacturing Technology & Machine Tool，2019(5)：112-115，121.