Danny Hoang , Hamza Errahmouni , Hanning Chen , Sriniket Rachuri , Nasir Mannan , Ruby ElKharboutly , Mohsen Imani , Ruimin Chen , Farhad Imani
{"title":"分层表示和可解释学习用于加速加工过程质量监控","authors":"Danny Hoang , Hamza Errahmouni , Hanning Chen , Sriniket Rachuri , Nasir Mannan , Ruby ElKharboutly , Mohsen Imani , Ruimin Chen , Farhad Imani","doi":"10.1016/j.cirpj.2024.02.010","DOIUrl":null,"url":null,"abstract":"<div><p>While modern 5-axis computer numerical control (CNC) systems offer enhanced design flexibility and reduced production time, the dimensional accuracy of the workpiece is significantly compromised by geometric errors, thermal deformations, cutting forces, tool wear, and fixture-related factors. In-situ sensing, in conjunction with machine learning (ML), has recently been implemented on edge devices to synchronously acquire and agilely analyze high-frequency and multifaceted data for the prediction of workpiece quality. However, limited edge computational resources and lack of interpretability in ML models obscure the understanding of key quality-influencing signals. This research introduces <span><math><mi>InterpHD</mi></math></span>, a novel graph-based hyperdimensional computing framework that not only assesses workpiece quality in 5-axis CNC on edge, but also characterizes key signals vital for evaluating the quality from in-situ multichannel data. Specifically, a hierarchical graph structure is designed to represent the relationship between channels (e.g., spindle rotation, three linear axes movements, and the rotary A and C axes), parameters (e.g., torque, current, power, and tool speed), and the workpiece dimensional accuracy. Additionally, memory refinement, separability, and parameter significance are proposed to assess the interpretability of the framework. Experimental results on a hybrid 5-axis LASERTEC 65 DED CNC machine indicate that <span><math><mi>InterpHD</mi></math></span> not only achieves a 90.7% F1-Score in characterizing a 25.4 mm counterbore feature deviation but also surpasses other ML models with an F1-Score margin of up to 73.0%. The interpretability of the framework reveals that load and torque have 12 times greater impact than power and velocity feed forward for the characterization of geometrical dimensions. <span><math><mi>InterpHD</mi></math></span> offers the potential to facilitate causal discovery and provide insights into the relationships between process parameters and part quality in manufacturing.</p></div>","PeriodicalId":56011,"journal":{"name":"CIRP Journal of Manufacturing Science and Technology","volume":"50 ","pages":"Pages 198-212"},"PeriodicalIF":4.6000,"publicationDate":"2024-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Hierarchical representation and interpretable learning for accelerated quality monitoring in machining process\",\"authors\":\"Danny Hoang , Hamza Errahmouni , Hanning Chen , Sriniket Rachuri , Nasir Mannan , Ruby ElKharboutly , Mohsen Imani , Ruimin Chen , Farhad Imani\",\"doi\":\"10.1016/j.cirpj.2024.02.010\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>While modern 5-axis computer numerical control (CNC) systems offer enhanced design flexibility and reduced production time, the dimensional accuracy of the workpiece is significantly compromised by geometric errors, thermal deformations, cutting forces, tool wear, and fixture-related factors. In-situ sensing, in conjunction with machine learning (ML), has recently been implemented on edge devices to synchronously acquire and agilely analyze high-frequency and multifaceted data for the prediction of workpiece quality. However, limited edge computational resources and lack of interpretability in ML models obscure the understanding of key quality-influencing signals. This research introduces <span><math><mi>InterpHD</mi></math></span>, a novel graph-based hyperdimensional computing framework that not only assesses workpiece quality in 5-axis CNC on edge, but also characterizes key signals vital for evaluating the quality from in-situ multichannel data. Specifically, a hierarchical graph structure is designed to represent the relationship between channels (e.g., spindle rotation, three linear axes movements, and the rotary A and C axes), parameters (e.g., torque, current, power, and tool speed), and the workpiece dimensional accuracy. Additionally, memory refinement, separability, and parameter significance are proposed to assess the interpretability of the framework. Experimental results on a hybrid 5-axis LASERTEC 65 DED CNC machine indicate that <span><math><mi>InterpHD</mi></math></span> not only achieves a 90.7% F1-Score in characterizing a 25.4 mm counterbore feature deviation but also surpasses other ML models with an F1-Score margin of up to 73.0%. The interpretability of the framework reveals that load and torque have 12 times greater impact than power and velocity feed forward for the characterization of geometrical dimensions. <span><math><mi>InterpHD</mi></math></span> offers the potential to facilitate causal discovery and provide insights into the relationships between process parameters and part quality in manufacturing.</p></div>\",\"PeriodicalId\":56011,\"journal\":{\"name\":\"CIRP Journal of Manufacturing Science and Technology\",\"volume\":\"50 \",\"pages\":\"Pages 198-212\"},\"PeriodicalIF\":4.6000,\"publicationDate\":\"2024-03-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"CIRP Journal of Manufacturing Science and Technology\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1755581724000300\",\"RegionNum\":2,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, MANUFACTURING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"CIRP Journal of Manufacturing Science and Technology","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1755581724000300","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, MANUFACTURING","Score":null,"Total":0}
引用次数: 0
摘要
虽然现代五轴计算机数控(CNC)系统提高了设计灵活性并缩短了生产时间,但工件的尺寸精度却因几何误差、热变形、切削力、刀具磨损和夹具相关因素而大打折扣。最近,人们在边缘设备上实现了原位传感与机器学习(ML)相结合,以同步获取和灵活分析高频率、多方面的数据,从而预测工件质量。然而,有限的边缘计算资源和缺乏可解释性的 ML 模型阻碍了对关键质量影响信号的理解。本研究介绍了 InterpHD,这是一种基于图的新型超维计算框架,它不仅能评估边缘五轴数控系统中的工件质量,还能描述对评估现场多通道数据质量至关重要的关键信号。具体来说,设计了一种分层图结构来表示通道(如主轴旋转、三个线性轴运动、旋转 A 轴和 C 轴)、参数(如扭矩、电流、功率和刀具速度)和工件尺寸精度之间的关系。此外,还提出了记忆细化、可分离性和参数重要性,以评估该框架的可解释性。在混合五轴 LASERTEC 65 DED 数控机床上进行的实验结果表明,InterpHD 不仅在表征 25.4 毫米对孔特征偏差方面取得了 90.7% 的 F1 分数,而且还以高达 73.0% 的 F1 分数裕度超越了其他 ML 模型。该框架的可解释性表明,在表征几何尺寸方面,载荷和扭矩的影响是功率和速度前馈的 12 倍。InterpHD 具有促进因果关系发现的潜力,并能深入了解制造过程中工艺参数与零件质量之间的关系。
Hierarchical representation and interpretable learning for accelerated quality monitoring in machining process
While modern 5-axis computer numerical control (CNC) systems offer enhanced design flexibility and reduced production time, the dimensional accuracy of the workpiece is significantly compromised by geometric errors, thermal deformations, cutting forces, tool wear, and fixture-related factors. In-situ sensing, in conjunction with machine learning (ML), has recently been implemented on edge devices to synchronously acquire and agilely analyze high-frequency and multifaceted data for the prediction of workpiece quality. However, limited edge computational resources and lack of interpretability in ML models obscure the understanding of key quality-influencing signals. This research introduces , a novel graph-based hyperdimensional computing framework that not only assesses workpiece quality in 5-axis CNC on edge, but also characterizes key signals vital for evaluating the quality from in-situ multichannel data. Specifically, a hierarchical graph structure is designed to represent the relationship between channels (e.g., spindle rotation, three linear axes movements, and the rotary A and C axes), parameters (e.g., torque, current, power, and tool speed), and the workpiece dimensional accuracy. Additionally, memory refinement, separability, and parameter significance are proposed to assess the interpretability of the framework. Experimental results on a hybrid 5-axis LASERTEC 65 DED CNC machine indicate that not only achieves a 90.7% F1-Score in characterizing a 25.4 mm counterbore feature deviation but also surpasses other ML models with an F1-Score margin of up to 73.0%. The interpretability of the framework reveals that load and torque have 12 times greater impact than power and velocity feed forward for the characterization of geometrical dimensions. offers the potential to facilitate causal discovery and provide insights into the relationships between process parameters and part quality in manufacturing.
期刊介绍:
The CIRP Journal of Manufacturing Science and Technology (CIRP-JMST) publishes fundamental papers on manufacturing processes, production equipment and automation, product design, manufacturing systems and production organisations up to the level of the production networks, including all the related technical, human and economic factors. Preference is given to contributions describing research results whose feasibility has been demonstrated either in a laboratory or in the industrial praxis. Case studies and review papers on specific issues in manufacturing science and technology are equally encouraged.