基于机理-数据融合的刀具磨损预测方法

doi:10.3901/JME.260419

摘要/Abstract

摘要： 在航空发动机叶片、机匣等关键零部件的铣削加工过程中，钛合金和镍基高温合金等难切削材料导致刀具快速磨损，显著影响工件的几何精度和表面质量。针对传统端到端的黑箱模型可解释性差、跨工况泛化能力差等问题，提出了一种基于机理-数据融合加工刀具磨损预测模型。首先，通过建立考虑刀具磨损的铣削力模型和基于功率的铣削力预测模型，构建了具有物理意义的刀具磨损预测机理模型；其次，针对机床加工过程中的多维信息，采用时频域处理及互信息筛选方法，提取与刀具磨损高相关的特征信息，构建数据特征向量集；然后基于门控循环单元(Gated recurrent unit neural network model, GRU)设计了机理数据融合的GRU神经网络模型(Mechanism-data fusion gated recurrent unit neural network model, MDF-GRU)，并通过构建具有物理意义约束的损失函数，实现模型在物理一致性空间内的优化训练。并通过加工试验进行验证，结果表明，该模型相比传统方法的均方根误差和平均绝对误差分别降低了71.5%和68.7%，较好地提高了预测精度。研究成果为难切削材料加工中的刀具磨损预测提供了新的技术方案。

关键词: 刀具磨损, 机理-数据融合, 传感信号特征, 损失函数, 神经网络

Abstract: In the milling process of critical aero-engine components such as blades and casings, the rapid tool wear caused by difficult-to-cut materials like titanium alloys and nickel-based superalloys significantly affects the geometric accuracy and surface quality of workpieces. To address the limitations of traditional end-to-end black-box models, including poor interpretability and weak cross-working condition generalization capability, a mechanism-data fusion tool wear prediction model is proposed. Firstly, a physically meaningful tool wear prediction mechanism model is established by developing a cutting force model considering tool wear and a power-based cutting force prediction model. Secondly, for the multi-dimensional information in machine tool processing, time-frequency domain processing and mutual information screening methods are employed to extract features highly correlated with tool wear and construct data feature vectors. Then, a mechanism-data fusion gated recurrent unit neural network model (MDF-GRU) is designed based on GRU model, and through the construction of a physically constrained loss function, the model optimization is achieved within the physical consistency space. Machining experiments demonstrate that the proposed model reduces the root mean square error and mean absolute error by 71.5% and 68.7% respectively compared to traditional methods, significantly improving prediction accuracy. This research provides a novel technical solution for tool wear prediction in difficult-to-cut material machining.

Key words: tool wear, mechanism-data fusion, sensing signal characteristics, loss function, neural network

中图分类号:

TG714

朱昀, 李方春, 李俊龙, 张莹, 吴宝海. 基于机理-数据融合的刀具磨损预测方法[J]. 机械工程学报, 2026, 62(9): 238-253.

ZHU Yun, LI Fangchun, LI Junlong, ZHANG Ying, WU Baohai. Tool Wear Prediction Method Based on Mechanism-data Fusion[J]. Journal of Mechanical Engineering, 2026, 62(9): 238-253.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260419

http://www.cjmenet.com.cn/CN/Y2026/V62/I9/238

参考文献

[1] MOHAMMADPOUR P，YUAN H，PHILLION A B. Microstructure evolution of Inconel 625 alloy during single-track laser powder bed fusion[J]. Additive Manufacturing，2022，55：102824.
[2] 赖旭伟，丁昆，张楷，等. 基于可解释物理引导空间注意力改进的跨工艺参数立铣刀磨损辨识[J]. 机械工程学报，2024，60(12)：147-157. LAI Xuwei，DING Kun，ZHANG Kai，et al. Tool wear identification of end milling across machining parameters based on interpretable physics-guided spatial attention improvement[J]. Journal of Mechanical Engineering，2024，60(12)：147-157.
[3] CHENG Y H，LI Y G，ZHUANG Q Y，et al. Mechanism-informed friction-dynamics coupling GRU neural network for real-time cutting force prediction[J]. Mechanical Systems and Signal Processing，2024，221：111749.
[4] 王洋. 高速切削Ti6Al4V硬质合金刀具磨损预测研究[D]. 马鞍山：安徽工业大学，2016. WANG Yang. The wear prediction and study of cemented carbide tool for cutting Ti6Al4V in high speed [D]. Maanshan：Anhui University of Technology，2016.
[5] RECH J，GIOVENCO A，COURBON C，et al. Toward a new tribological approach to predict cutting tool wear[J]. CIRP Annals - Manufacturing Technology，2018，67(1)：65-68.
[6] 张佳弈. PCBN刀具磨损对切削过程影响的有限元仿真研究[J]. 工具技术，2017，51(1)：40-43. ZHANG Jiayi. Finite element simulation study on effect of PCBN tool wear on cutting process[J]. Tool Engineering，2017，51(1)：40-43.
[7] AN D，KIM N H，CHOI J H. Practical options for selecting data-driven or physics-based prognostics algorithms with reviews[J]. Reliability Engineering and System Safety，2015，133：223-236.
[8] PALMAI Z. Proposal for a new theoretical model of the cutting tool’s flank wear[J]. Wear，2013，303(1-2)：437-445.
[9] CHINCHANIKAR S，CHOUDHURY S K. Predictive modeling for flank wear progression of coated carbide tool in turning hardened steel under practical machining conditions[J]. The International Journal of Advanced Manufacturing Technology，2015，76(5-8)：1185-1201.
[10] ZHU K，ZHANG Y. A generic tool wear model and its application to force modeling and wear monitoring in high speed milling[J]. Mechanical Systems and Signal Processing，2019，115：147-161.
[11] 刘会永. 基于振动和功率信号融合的刀具磨损状态监测方法[D]. 济南：山东大学，2022. LIU Huiyong. Tool wear state monitoring method based on vibration and power signal fusion[D]. Jinan：Shandong University，2022.
[12] 黄豪. 基于振动和切削力信号的刀具磨损状态监测方法研究[D]. 重庆：重庆邮电大学，2022. HUANG Hao. Research on monitoring method of tool wear state based on vibration and cutting force signal[D]. Chongqing：Chongqing University of Posts and Telecommunications，2022.
[13] MONKA P，MONKOVA K，UBAN M，et al. Vibrodiagnostics as the tool of a tap wear monitoring[J]. Procedia Structural Integrity，2018，13(7)：959-964.
[14] 刘思辰，杨飞然，杨军. 基于多传感器融合的刀具剩余寿命预测[J]. 振动与冲击，2021，40(17)：47-54. LIU Sichen，YANG Feiran，YANG Jun. Tool residual life prediction based on multi-sensor fusion[J]. Vibration and Shock，2021，40(17)：47-54.
[15] XU W，MIAO H，ZHAO Z，et al. Multi-scale convolutional gated recurrent unit networks for tool wear prediction in smart manufacturing[J]. Chinese Journal of Mechanical Engineering，2021，34(1)：53.
[16] 吴雪峰，刘亚辉，毕淞泽. 基于卷积神经网络刀具磨损类型的智能识别[J]. 计算机集成制造系统，2020，26(10)：2762-2771. WU Xuefeng，LIU Yahui，BI Songze. Intelligent recognition of tool wear type based on convolutional neural networks[J]. Computer integrated manufacturing system. 2020，26(10)：2762-2771.
[17] 史丽晨，史炜椿，王海涛，等. 基于DRSN-BiLSTM模型的刀具磨损预测方法研究[J]. 机械工程学报，2024，60(24)：66-74. SHI Lichen，SHI Weichun，WANG Haitao，et al. Research on tool wear prediction method based on DRSN-BiLSTM model[J]. Journal of Mechanical Engineering，2024，60(24)：66-74.
[18] 刘大同，郭凯，王本宽. 数字孪生技术综述与展望[J]. 仪器仪表学报，2018，39(11)：1-10. LIU Datong，GUO Kai，WANG Benkuan，et al. Summary and perspective survey on digital twin technology[J]. Journal of Scientific Instrument，2018，39(11)：1-10.
[19] WANG J，WANG P，GAO R. Enhanced particle filter for tool wear prediction[J]. Journal of Manufacturing Systems，2015，36(6)：35-45.
[20] HANACHI H，YU W，KIM Y，et al. Hybrid data-driven physics-based model fusion framework for tool wear prediction[J]. The International Journal of Advanced Manufacturing Technology，2019，101：2861-2872.
[21] 王军. 基于数字孪生的刀具磨损监测和预测方法研究[D]. 济南：齐鲁工业大学，2021. WANG Jun. Research on tool wear monitoring and prediction method based on digital twin[D]. Jinan：Qilu University of Technology，2021.
[22] AN H，WANG G，DONG Y，et al. Tool life predicted based on Gauss importance resampling particle filter[J]. The International Journal of Advanced Manufacturing Technology，2019，103(9-12)：4627-4634.
[23] 万鹏，李迎光，等. 基于元学习和PINN的变工况刀具磨损精确预测方法[J]. 南京航空航天大学学报，2022，54(3)：387-396. WAN Peng，LI Yingguang，et al. Accurate prediction method of tool wear under varying cutting conditions based on meta learning and PINN[J]. Journal of Nanjing University of Aeronautics and Astronautics，2022，54(3)：387-396.
[24] KARANDIKAR J，SCHMITZ T，SMITH S. Physics-guided logistic classification for tool life modeling and process parameter optimization in machining[J]. Journal of Manufacturing Systems，2021，59：522-534.
[25] 徐健. 基于铣削功率的刀具磨损预测及参数优化方法[D]. 西安：西北工业大学，2021. XU Jian，Tool wear prediction and parameter optimization based on milling power[D]. Xi'an：Northwestern Polytechnical University，2021.
[26] TEITENBERG T M，BAYOUMI A E，YUSCESAN G. Tool wear modeling through an analytic mechanistic model of milling processes[J]. Wear，1992，154(2)：287-304.
[27] 王奇. 考虑刀具磨损的钻孔过程能耗模型及其应用研究[D]. 西安：西北工业大学，2021. WANG Qi. Energy consumption model of drilling process considering tool wear and its application[D]. Xi'an：Northwestern Polytechnical University，2021.
[28] NAIR V，HINTON G E. Rectified linear units improve restricted Boltzmann machines[C]// Proceedings of the 27th International Conference on Machine Learning (ICML-10)，2010：807-814.