工业工程

工业工程 ›› 2024, Vol. 27 ›› Issue (3): 64-77,86.doi: 10.3969/j.issn.1007-7375.240063

• 复杂系统建模与运筹优化 • 上一篇    下一篇

基于注意力机制堆叠LSTM的多传感器信息融合刀具磨损预测

成佳闻1, 赛希亚拉图1, 张超勇1,2, 罗敏2   

  1. 1. 华中科技大学 机械科学与工程学院，湖北 武汉 430074;
    2. 湖北汽车工业学院 电气与信息工程学院，湖北 十堰 442002
  • 收稿日期:2024-02-09 发布日期:2024-07-12
  • 通讯作者: 张超勇 (1972—)，男，江苏省人，教授，博士，主要研究方向为智能调度算法、网络化制造、绿色制造。Email: zcyhust@hust.edu.cn E-mail:zcyhust@hust.edu.cn
  • 作者简介:成佳闻 (1999—)， 男，湖北省人，硕士，主要研究方向为刀具磨损和数字孪生
  • 基金资助:
    中德重点研发资助项目 (2023ZY01089)；工信部高质量发展专项资助项目 (2023ZY01089)

Tool Wear Prediction Based on Multi-sensor Information Fusion Using Stacked LSTM with Attention Mechanism

CHENG Jiawen1, BAO Saixialt1, ZHANG Chaoyong1,2, LUO Min2   

  1. 1. School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China;
    2. School of Electrical and Information Engineering, Hubei University of Automotive Technology, Shiyan 442002, China
  • Received:2024-02-09 Published:2024-07-12

RichHTML

26

PDF

39

摘要： 刀具磨损是影响数控机床加工质量和加工效率的重要因素之一。针对现有铣刀磨损预测中信号单一和预测精度不足的问题，提出了一种基于注意力机制的堆叠LSTM (long short-term memory，长短期记忆网络) 的多传感器信息融合刀具磨损预测方法。对多传感器信号进行预处理，然后提取多域特征，利用核主成分分析法对其进行特征级信息融合，得到后续网络的输入。采用基于注意力机制的堆叠LSTM网络模型，使得网络能够自适应地学习数据的重要信息，在PHM2010的数据集上预测精度达到99.9%。通过与其他算法的对比试验和加入人工噪声的方法，验证了本文所提出的模型的高精度和鲁棒性。

关键词: 刀具磨损, 核主成分分析 (KPCA), 信息融合, 注意力机制, 鲁棒性

Abstract: Tool wear is one of the critical factors affecting the quality and efficiency of computer numerical control (CNC) machine tools. To address the current issues of signal singularity and insufficient prediction accuracy in milling cutter wear predictions, a novel method for tool wear prediction is proposed, which involves the fusion of multi-sensor information based on a stacked long short-term memory (LSTM) network with attention mechanism. Initially multiple sensor signals are preprocessed, subsequently multi-domain features are extracted. These features are fused at the feature level using Kernel Principal Component Analysis (KPCA), providing the input for the subsequent network. A stacked LSTM network model with attention mechanism is used to enable adaptive learning of crucial information, achieving a predictive accuracy of 99.9% on the PHM2010 dataset. Comparative experiments are conducted with other algorithms and incorporating artificial noise to verify the high accuracy and robustness of the proposed model.

Key words: tool wear, kernel principal component analysis (KPCA), information fusion, attention mechanism, robustness

中图分类号: