Journal of Manufacturing Systems最新文献_第6页

Integrated assembly, measurement, and adjustment method of reconfigurable flexible fixture for aircraft panels based on augmented reality and human-computer interaction

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-16 DOI: 10.1016/j.jmsy.2025.01.003

Xiangrong Zhang , Shuang Meng , Binbin Wang , Lianyu Zheng , Rui Zhang , Xufei Li

Owing to the characteristics of reconfigurable flexible fixtures (RFFs) for aircraft panels, the automation of their assembly is limited by technology and cost. As a result, manual assembly remains the predominant method. During the manual assembly, workers are affected by several challenges, such as difficulty in understanding the process documents and drawings, distinguishing the assembly positions of the similar components, inefficient data transmission, and determining the adjustment direction of contour board locators (CBLs). These issues arise because of inadequate digital assistance to workers. This paper proposes an integrated assembly, measurement, and adjustment (AMA) method for RFFs based on augmented reality (AR) and human–computer interaction (HCI) to assist workers. First, an information model based on core process elements for the assembly process is constructed. This model clarifies the correlations among multi-source data. Then, measured data is transformed into six-dimensional (6D) parameters to assist in the adjustment of CBLs. Based on the model and 6D parameters, the multiple visual assembly guidance is established by AR virtual-reality fusion technology. Subsequently, the HCI technology is introduced, to adaptively provide guidance via the hand-free head pointer. Workers use the AR head-mounted devices (HMDs) as a medium to interact with the laser tracker, which enables to quickly and accurately obtain the measurement data. Finally, AR and HCI technology are combined to establish an integrated process of AMA of RFFs. This method significantly improves the collaboration between workers and information during the assembly process of RFF. Field-assembly validation demonstrates that, compared with conventional methods, the proposed method achieves a positioning accuracy of ± 0.12 mm and enhances assembly efficiency by 32.87 %.

{"title":"Integrated assembly, measurement, and adjustment method of reconfigurable flexible fixture for aircraft panels based on augmented reality and human-computer interaction","authors":"Xiangrong Zhang , Shuang Meng , Binbin Wang , Lianyu Zheng , Rui Zhang , Xufei Li","doi":"10.1016/j.jmsy.2025.01.003","DOIUrl":"10.1016/j.jmsy.2025.01.003","url":null,"abstract":"<div><div>Owing to the characteristics of reconfigurable flexible fixtures (RFFs) for aircraft panels, the automation of their assembly is limited by technology and cost. As a result, manual assembly remains the predominant method. During the manual assembly, workers are affected by several challenges, such as difficulty in understanding the process documents and drawings, distinguishing the assembly positions of the similar components, inefficient data transmission, and determining the adjustment direction of contour board locators (CBLs). These issues arise because of inadequate digital assistance to workers. This paper proposes an integrated assembly, measurement, and adjustment (AMA) method for RFFs based on augmented reality (AR) and human–computer interaction (HCI) to assist workers. First, an information model based on core process elements for the assembly process is constructed. This model clarifies the correlations among multi-source data. Then, measured data is transformed into six-dimensional (6D) parameters to assist in the adjustment of CBLs. Based on the model and 6D parameters, the multiple visual assembly guidance is established by AR virtual-reality fusion technology. Subsequently, the HCI technology is introduced, to adaptively provide guidance via the hand-free head pointer. Workers use the AR head-mounted devices (HMDs) as a medium to interact with the laser tracker, which enables to quickly and accurately obtain the measurement data. Finally, AR and HCI technology are combined to establish an integrated process of AMA of RFFs. This method significantly improves the collaboration between workers and information during the assembly process of RFF. Field-assembly validation demonstrates that, compared with conventional methods, the proposed method achieves a positioning accuracy of ± 0.12 mm and enhances assembly efficiency by 32.87 %.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 117-133"},"PeriodicalIF":12.2,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143169567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Systematic AR-based assembly guidance for small-scale, high-density industrial components

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-15 DOI: 10.1016/j.jmsy.2025.01.002

Junhao Geng , Yuntao Wang , Yu Cheng , Xin Zhang , Yingpeng Xu , Wenjie Lv

Small-scale high-density industrial components (SHIC) are widely used in high-end electromechanical equipment with higher requirements for functional integration. SHIC assembly operation is a typical manual fine operation, which is labor-intensive, time-consuming, and error-prone and seriously affects assembly efficiency and quality consistency. The current augmented reality (AR) technology cannot realize the complete, accurate, and stable cognition and guidance for SHIC because of the single use of visual cognition. Meanwhile, manual preparation of prior knowledge dramatically reduces AR applications’ automation level and practicability. Addressing this research gap, this paper develops a technical framework of AR-based assembly guidance for SHIC, which integrates several novel intelligent methods and deeply combines prior knowledge reasoning and computer vision cognition in a simple and interpretable way. First, computer vision and deep learning are used to automatically generate prior knowledge from the three-dimensional (3D) model of SHIC. Then, 3D tracking, visual recognition, virtual-real matching, and rule reasoning are combined to complete, sequence, and locate the assembly targets in the case of insufficient visual information. Finally, based on the adaptive threshold, two kinds of guidance modes, i.e., position-based precise guidance and region-based heuristic guidance, are automatically switched to realize the AR guidance of the whole assembly process. The case study of complex electrical connectors shows that this framework with novel intelligent methods can meet the industrial requirements of performance, availability, and effectiveness and improve the robustness and efficiency of SHIC assembly operation by 92.95% and 87.06%, respectively. This study realizes AR guidance technology’s systematic and practical application in SHIC assembly operation. Also, it provides a practicable technology architecture for intelligent AR assistance in the fine assembly field.

{"title":"Systematic AR-based assembly guidance for small-scale, high-density industrial components","authors":"Junhao Geng , Yuntao Wang , Yu Cheng , Xin Zhang , Yingpeng Xu , Wenjie Lv","doi":"10.1016/j.jmsy.2025.01.002","DOIUrl":"10.1016/j.jmsy.2025.01.002","url":null,"abstract":"<div><div>Small-scale high-density industrial components (SHIC) are widely used in high-end electromechanical equipment with higher requirements for functional integration. SHIC assembly operation is a typical manual fine operation, which is labor-intensive, time-consuming, and error-prone and seriously affects assembly efficiency and quality consistency. The current augmented reality (AR) technology cannot realize the complete, accurate, and stable cognition and guidance for SHIC because of the single use of visual cognition. Meanwhile, manual preparation of prior knowledge dramatically reduces AR applications’ automation level and practicability. Addressing this research gap, this paper develops a technical framework of AR-based assembly guidance for SHIC, which integrates several novel intelligent methods and deeply combines prior knowledge reasoning and computer vision cognition in a simple and interpretable way. First, computer vision and deep learning are used to automatically generate prior knowledge from the three-dimensional (3D) model of SHIC. Then, 3D tracking, visual recognition, virtual-real matching, and rule reasoning are combined to complete, sequence, and locate the assembly targets in the case of insufficient visual information. Finally, based on the adaptive threshold, two kinds of guidance modes, i.e., position-based precise guidance and region-based heuristic guidance, are automatically switched to realize the AR guidance of the whole assembly process. The case study of complex electrical connectors shows that this framework with novel intelligent methods can meet the industrial requirements of performance, availability, and effectiveness and improve the robustness and efficiency of SHIC assembly operation by 92.95% and 87.06%, respectively. This study realizes AR guidance technology’s systematic and practical application in SHIC assembly operation. Also, it provides a practicable technology architecture for intelligent AR assistance in the fine assembly field.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 86-100"},"PeriodicalIF":12.2,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143169565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Towards perceptive assembly: Real-time full-field deformation twin perception method for large components under parametric load

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-15 DOI: 10.1016/j.jmsy.2024.12.013

Jiacheng Cui, Yang Zhang, Yongkang Lu, Pengbo Yin, Qihang Chen, Lei Han, Wei Liu

Enhancing information perception capabilities during the manufacturing and assembly of large-scale components is pivotal for advancing intelligent systems, particularly in the aerospace industry. This paper presents a perceptive assembly approach utilizing a full-field deformation twin perception method based on parametric loads, enabling real-time and accurate reconstruction of deformation fields in large components through a binocular vision system. This method centers around parametric load definitions, proposing the PPOD (Parametric Proper Orthogonal Decomposition) technique for deformation reconstruction, followed by an in-depth analysis of the factors contributing to perception errors. To meet the demands of online deployment, a comprehensive framework is established, deeply integrating measurement instruments, measurement data, and physical models to enhance measurement efficiency and perception robustness. Extensive simulations and experimental results demonstrate that this approach reduces perception errors by over 70% compared to traditional methods, achieving real-time, high-precision, and robust monitoring of deformation in large components. This perceptive assembly framework holds significant promise as a foundational infrastructure for real-time state perception in the intelligent manufacturing of large-scale components.

{"title":"Towards perceptive assembly: Real-time full-field deformation twin perception method for large components under parametric load","authors":"Jiacheng Cui, Yang Zhang, Yongkang Lu, Pengbo Yin, Qihang Chen, Lei Han, Wei Liu","doi":"10.1016/j.jmsy.2024.12.013","DOIUrl":"10.1016/j.jmsy.2024.12.013","url":null,"abstract":"<div><div>Enhancing information perception capabilities during the manufacturing and assembly of large-scale components is pivotal for advancing intelligent systems, particularly in the aerospace industry. This paper presents a perceptive assembly approach utilizing a full-field deformation twin perception method based on parametric loads, enabling real-time and accurate reconstruction of deformation fields in large components through a binocular vision system. This method centers around parametric load definitions, proposing the PPOD (Parametric Proper Orthogonal Decomposition) technique for deformation reconstruction, followed by an in-depth analysis of the factors contributing to perception errors. To meet the demands of online deployment, a comprehensive framework is established, deeply integrating measurement instruments, measurement data, and physical models to enhance measurement efficiency and perception robustness. Extensive simulations and experimental results demonstrate that this approach reduces perception errors by over 70% compared to traditional methods, achieving real-time, high-precision, and robust monitoring of deformation in large components. This perceptive assembly framework holds significant promise as a foundational infrastructure for real-time state perception in the intelligent manufacturing of large-scale components.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 101-116"},"PeriodicalIF":12.2,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143169566","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Intelligent path planning algorithm system for printed display manufacturing using graph convolutional neural network and reinforcement learning

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-14 DOI: 10.1016/j.jmsy.2024.12.016

Jiacong Xiong , Jiankui Chen , Wei Chen , Xiao Yue , Ziwei Zhao , Zhouping Yin

Inkjet printing technology is considered one of the core components of next-generation display technologies for manufacturing organic light-emitting diode (OLED). However, the patterning process for novel display inkjet printing involves diverse characteristics across different dimensions, such as varying printing scales and resolutions. Existing patterning modules using a single planning algorithm for all inkjet printing scenarios often result in long planning times and unstable planning quality. Therefore, a more comprehensive algorithm system is needed to evaluate inkjet planning problems and select the most suitable planning algorithm. This paper proposes a multi-algorithm integrated online patterning intelligence planning system, which includes three patterning algorithms specific to the inkjet display field and an algorithm selection network based on Proximal Policy Optimization (PPO). We first identify the core metrics of the inkjet planning problem as planning time and solution quality, analyzing how different characteristics of the planning problem affect these metrics. We then propose three algorithms suited to different performance needs: an integer programming method based on graph convolutional neural networks, a binary greedy algorithm, and a maximum contiguous interval search algorithm, each corresponding to high overall performance, high solution quality, and short solution time, respectively, to address complex inkjet planning scenarios. Additionally, the PPO-based algorithm selection network refines the features of the inkjet planning problem to achieve intelligent algorithm selection. Finally, we validate the multi-algorithm integrated online patterning intelligence planning system using the self-developed NEJ-PRG4.5 inkjet equipment.

{"title":"Intelligent path planning algorithm system for printed display manufacturing using graph convolutional neural network and reinforcement learning","authors":"Jiacong Xiong , Jiankui Chen , Wei Chen , Xiao Yue , Ziwei Zhao , Zhouping Yin","doi":"10.1016/j.jmsy.2024.12.016","DOIUrl":"10.1016/j.jmsy.2024.12.016","url":null,"abstract":"<div><div>Inkjet printing technology is considered one of the core components of next-generation display technologies for manufacturing organic light-emitting diode (OLED). However, the patterning process for novel display inkjet printing involves diverse characteristics across different dimensions, such as varying printing scales and resolutions. Existing patterning modules using a single planning algorithm for all inkjet printing scenarios often result in long planning times and unstable planning quality. Therefore, a more comprehensive algorithm system is needed to evaluate inkjet planning problems and select the most suitable planning algorithm. This paper proposes a multi-algorithm integrated online patterning intelligence planning system, which includes three patterning algorithms specific to the inkjet display field and an algorithm selection network based on Proximal Policy Optimization (PPO). We first identify the core metrics of the inkjet planning problem as planning time and solution quality, analyzing how different characteristics of the planning problem affect these metrics. We then propose three algorithms suited to different performance needs: an integer programming method based on graph convolutional neural networks, a binary greedy algorithm, and a maximum contiguous interval search algorithm, each corresponding to high overall performance, high solution quality, and short solution time, respectively, to address complex inkjet planning scenarios. Additionally, the PPO-based algorithm selection network refines the features of the inkjet planning problem to achieve intelligent algorithm selection. Finally, we validate the multi-algorithm integrated online patterning intelligence planning system using the self-developed NEJ-PRG4.5 inkjet equipment.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 73-85"},"PeriodicalIF":12.2,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143170024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Application and trends of point cloud in intelligent welding: State of the art review

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-09 DOI: 10.1016/j.jmsy.2025.01.001

Hui Wang , Youmin Rong , Jiajun Xu , Yu Huang , Guojun Zhang

Point cloud is an important data format for expressing 3D scenes. With the development of digitalization and intelligence in welding, the application of point cloud in welding is gradually increasing. This paper provides a comprehensive overview of the application of point clouds in intelligent welding, divided into pre-welding stage, in-welding stage, and post-welding stage according to the welding sequence. A detailed analysis was conducted on the pre-welding stage of intelligent welding operations, including the point cloud construction of the workpiece, weld seam detection in the point cloud, path and posture planning, automatic programming, and weld seam positioning. The in-welding stage was divided into welding pool monitoring, weld seam tracking and adjustment, and a comparative analysis was conducted on point cloud processing algorithms and control methods. The post-welding stage is divided into weld seam identification, defect detection, and quality detection, which are analyzed and discussed separately. This paper introduces the application of point clouds in the previous stages, and compares in detail the selection of sensors, application scenarios, point cloud processing algorithms, and processing effects. Furthermore, in-depth analysis was conducted on the effects and limitations achieved in various application scenarios. Finally, this paper provides an outlook on future work, summarizes the relevant applications of machine learning in welding point cloud processing, and analyzes future development trends.

{"title":"Application and trends of point cloud in intelligent welding: State of the art review","authors":"Hui Wang , Youmin Rong , Jiajun Xu , Yu Huang , Guojun Zhang","doi":"10.1016/j.jmsy.2025.01.001","DOIUrl":"10.1016/j.jmsy.2025.01.001","url":null,"abstract":"<div><div>Point cloud is an important data format for expressing 3D scenes. With the development of digitalization and intelligence in welding, the application of point cloud in welding is gradually increasing. This paper provides a comprehensive overview of the application of point clouds in intelligent welding, divided into pre-welding stage, in-welding stage, and post-welding stage according to the welding sequence. A detailed analysis was conducted on the pre-welding stage of intelligent welding operations, including the point cloud construction of the workpiece, weld seam detection in the point cloud, path and posture planning, automatic programming, and weld seam positioning. The in-welding stage was divided into welding pool monitoring, weld seam tracking and adjustment, and a comparative analysis was conducted on point cloud processing algorithms and control methods. The post-welding stage is divided into weld seam identification, defect detection, and quality detection, which are analyzed and discussed separately. This paper introduces the application of point clouds in the previous stages, and compares in detail the selection of sensors, application scenarios, point cloud processing algorithms, and processing effects. Furthermore, in-depth analysis was conducted on the effects and limitations achieved in various application scenarios. Finally, this paper provides an outlook on future work, summarizes the relevant applications of machine learning in welding point cloud processing, and analyzes future development trends.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 48-72"},"PeriodicalIF":12.2,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143169036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Intelligent monitoring system for production lines in smart factories: A hybrid method integrating Transformer and Kalman filter

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-06 DOI: 10.1016/j.jmsy.2024.12.014

Xiaohui Fang , Qinghua Song , Zhenyang Li , Xiaojuan Wang , Haifeng Ma , Zhanqiang Liu

Intelligent monitoring systems for production lines in smart factories are used to ensure production efficiency, quality control and fault warning, promoting the optimization of production processes and resource allocation. Tool wear monitoring (TWM) bridges the gap between the perception of machining state information and accurate health management of tools. However, signal features undergo significant and abnormal changes during the late-stage of tool wear, posing substantial challenges to the development of an accurate tool wear intelligent monitoring model. In this paper, a hybrid TWM model integrating the Transformer and Kalman filter is proposed, with a state-space model of tool wear constructed and dynamically updated to address the gap in late-stage prediction accuracy of existing TWM methods. Specifically, the Transformer model is designed to describe the observation model of early tool wear states based on monitoring data. A system model is developed based on the actual tool wear mechanism to describe the relationship between the tool wear rate and tool-workpiece contact load over time. The Kalman filter is used to estimate the parameters of the mechanism model and track the evolution of wear. Within the Bayesian inference framework, measurement noise in the monitoring data is accounted for to optimize and update state estimation deviations and mechanism model parameters, enabling late-stage wear prediction through posterior estimation. The effectiveness and generalization of the proposed method are validated through milling experiments on both thin-walled and rectangular block parts. The experimental results indicate that the average RMSE error of tool wear prediction for thin-walled parts is 6.02, while for rectangular block parts, it is 4.70. The average RMSE errors of the proposed method are reduced by 16.34 % and 11.31 %, respectively, with respect to the single model. More importantly, the proposed method for TWM demonstrates strong predictive performance in the late-stage of tool wear while quantifying the uncertainty of wear prediction.

{"title":"Intelligent monitoring system for production lines in smart factories: A hybrid method integrating Transformer and Kalman filter","authors":"Xiaohui Fang , Qinghua Song , Zhenyang Li , Xiaojuan Wang , Haifeng Ma , Zhanqiang Liu","doi":"10.1016/j.jmsy.2024.12.014","DOIUrl":"10.1016/j.jmsy.2024.12.014","url":null,"abstract":"<div><div>Intelligent monitoring systems for production lines in smart factories are used to ensure production efficiency, quality control and fault warning, promoting the optimization of production processes and resource allocation. Tool wear monitoring (TWM) bridges the gap between the perception of machining state information and accurate health management of tools. However, signal features undergo significant and abnormal changes during the late-stage of tool wear, posing substantial challenges to the development of an accurate tool wear intelligent monitoring model. In this paper, a hybrid TWM model integrating the Transformer and Kalman filter is proposed, with a state-space model of tool wear constructed and dynamically updated to address the gap in late-stage prediction accuracy of existing TWM methods. Specifically, the Transformer model is designed to describe the observation model of early tool wear states based on monitoring data. A system model is developed based on the actual tool wear mechanism to describe the relationship between the tool wear rate and tool-workpiece contact load over time. The Kalman filter is used to estimate the parameters of the mechanism model and track the evolution of wear. Within the Bayesian inference framework, measurement noise in the monitoring data is accounted for to optimize and update state estimation deviations and mechanism model parameters, enabling late-stage wear prediction through posterior estimation. The effectiveness and generalization of the proposed method are validated through milling experiments on both thin-walled and rectangular block parts. The experimental results indicate that the average <em>RMSE</em> error of tool wear prediction for thin-walled parts is 6.02, while for rectangular block parts, it is 4.70. The average <em>RMSE</em> errors of the proposed method are reduced by 16.34 % and 11.31 %, respectively, with respect to the single model. More importantly, the proposed method for TWM demonstrates strong predictive performance in the late-stage of tool wear while quantifying the uncertainty of wear prediction.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 27-47"},"PeriodicalIF":12.2,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143169564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Residual stress field inference method using structured latent Gaussian process with structured-covariances

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-03 DOI: 10.1016/j.jmsy.2024.12.011

Zhiwei Zhao , Yingguang Li , Changqing Liu , Yifan Zhang

Residual stress field distributed within materials is one of the important properties of the material, which influences the manufacturing precision and fatigue life of part, especially for component with large size and complex residual stress field. Accurately obtaining the overall residual stress field within parts serves as a vital reference for optimizing manufacturing and assembly processes. Although the residual stress fields could be estimate by related observable physical quantity, the characteristics of residual stress fields, including a large number of unknowns and multiple fields, present significant challenge for the solving process. Introducing prior knowledge as the constrains for the possible solution is an effective method for achieving accurate inference of residual stress fields. In this paper, a method for inferring residual stress field which employed Structured Latent Gaussian Process with structured covariance is proposed. The unobservable residual stress field is modeled as a latent Gaussian process with structured covariance, formed by Kronecker product considering the correlations across different directions of the residual stress and the prior knowledge of similarities between partial fields. By introducing structured prior information, the large covariance matrix estimation is transformed into the estimation of several smaller matrices, which significantly reduces the number of unknowns to be estimated, and it makes the inference of residual stress fields with numerous variables using limited observable data feasible. Simulation and experimental validation results demonstrate that the incorporation of such prior information can enhance the accuracy and reliability of the inferred residual stress fields.

{"title":"Residual stress field inference method using structured latent Gaussian process with structured-covariances","authors":"Zhiwei Zhao , Yingguang Li , Changqing Liu , Yifan Zhang","doi":"10.1016/j.jmsy.2024.12.011","DOIUrl":"10.1016/j.jmsy.2024.12.011","url":null,"abstract":"<div><div>Residual stress field distributed within materials is one of the important properties of the material, which influences the manufacturing precision and fatigue life of part, especially for component with large size and complex residual stress field. Accurately obtaining the overall residual stress field within parts serves as a vital reference for optimizing manufacturing and assembly processes. Although the residual stress fields could be estimate by related observable physical quantity, the characteristics of residual stress fields, including a large number of unknowns and multiple fields, present significant challenge for the solving process. Introducing prior knowledge as the constrains for the possible solution is an effective method for achieving accurate inference of residual stress fields. In this paper, a method for inferring residual stress field which employed Structured Latent Gaussian Process with structured covariance is proposed. The unobservable residual stress field is modeled as a latent Gaussian process with structured covariance, formed by Kronecker product considering the correlations across different directions of the residual stress and the prior knowledge of similarities between partial fields. By introducing structured prior information, the large covariance matrix estimation is transformed into the estimation of several smaller matrices, which significantly reduces the number of unknowns to be estimated, and it makes the inference of residual stress fields with numerous variables using limited observable data feasible. Simulation and experimental validation results demonstrate that the incorporation of such prior information can enhance the accuracy and reliability of the inferred residual stress fields.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 14-26"},"PeriodicalIF":12.2,"publicationDate":"2025-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143169563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Towards automatic anomaly detection and diagnosis in positional arc-directed energy deposition based on deep learning

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2025-01-02 DOI: 10.1016/j.jmsy.2024.12.015

Yuhua Cai, Chaonan Li, Hui Chen, Jun Xiong

Positional arc-directed energy deposition (DED) is a highly anticipated technique with high degrees of freedom for fabricating overhang structures without support structures and positioners. Nevertheless, the immaturity of monitoring and control techniques for ensuring a stable deposition process in positional arc-DED is the main challenge in achieving reliable and automatic manufacturing of metal components. This study aims to identify the molten pool state and diagnose the occurrence of hump and drop defects in positional arc-DED based on deep learning. Five convolutional neural network (CNN) models are used to perform the classification task of molten pool states to determine the optimal architecture of the defect detection model based on their classification performance. A target recognition framework, called YOLOv5s, is used to construct a hump defect diagnosis model and a drop defect diagnosis model to diagnose the occurrence of hump and drop defects, respectively. Compared to other CNN classification models, ResNet18 can effectively balance the performance and the computational resource requirement, obtaining an excellent classification accuracy of 0.996. The maximum detection error of the hump defect diagnosis model for extracting the molten pool height and length is less than 0.68 mm. The molten pool dimensional ratio is used to evaluate the probability of the hump defect occurrence. The molten pool dimensional ratio increases gradually during one formation cycle of the hump defect. The drop defect diagnosis model successfully avoids the interference of the burning arc and obtains an accurate detection result, with a maximum detection error of less than 2.09 mm². The molten pool area is introduced to evaluate the probability of the drop defect occurrence in positional arc-DED, which decreases first and then increases as the drop falls from the molten pool. This study lays a solid foundation for controlling the stability of the deposition process in positional arc-DED.

{"title":"Towards automatic anomaly detection and diagnosis in positional arc-directed energy deposition based on deep learning","authors":"Yuhua Cai, Chaonan Li, Hui Chen, Jun Xiong","doi":"10.1016/j.jmsy.2024.12.015","DOIUrl":"10.1016/j.jmsy.2024.12.015","url":null,"abstract":"<div><div>Positional arc-directed energy deposition (DED) is a highly anticipated technique with high degrees of freedom for fabricating overhang structures without support structures and positioners. Nevertheless, the immaturity of monitoring and control techniques for ensuring a stable deposition process in positional arc-DED is the main challenge in achieving reliable and automatic manufacturing of metal components. This study aims to identify the molten pool state and diagnose the occurrence of hump and drop defects in positional arc-DED based on deep learning. Five convolutional neural network (CNN) models are used to perform the classification task of molten pool states to determine the optimal architecture of the defect detection model based on their classification performance. A target recognition framework, called YOLOv5s, is used to construct a hump defect diagnosis model and a drop defect diagnosis model to diagnose the occurrence of hump and drop defects, respectively. Compared to other CNN classification models, ResNet18 can effectively balance the performance and the computational resource requirement, obtaining an excellent classification accuracy of 0.996. The maximum detection error of the hump defect diagnosis model for extracting the molten pool height and length is less than 0.68 mm. The molten pool dimensional ratio is used to evaluate the probability of the hump defect occurrence. The molten pool dimensional ratio increases gradually during one formation cycle of the hump defect. The drop defect diagnosis model successfully avoids the interference of the burning arc and obtains an accurate detection result, with a maximum detection error of less than 2.09 mm<sup>2</sup>. The molten pool area is introduced to evaluate the probability of the drop defect occurrence in positional arc-DED, which decreases first and then increases as the drop falls from the molten pool. This study lays a solid foundation for controlling the stability of the deposition process in positional arc-DED.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"79 ","pages":"Pages 1-13"},"PeriodicalIF":12.2,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143169562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Investigation of assistance systems in assembly in the context of digitalization: A systematic literature review 数字化背景下装配辅助系统的研究：系统的文献综述

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-11-29 DOI: 10.1016/j.jmsy.2024.11.015

Mathias König , Herwig Winkler

Assistance systems play a crucial role in enhancing working conditions and efficiency in industrial assembly. In the context of Industry 4.0, it is important to determine the types of assistance systems that contribute to assembly goals as well as their economic benefits. First, the significance of the topic will be introduced, and the research questions will be presented. Second, the basic technical terms will be defined, and third, the research methodology of a structured literature review (SLR) will be delineated. The fourth section presents an overview of the ergonomic and information assistance systems used in operational practice and academic test set-ups. It further explains the reasons for using assistance systems in assembly and their economic benefits, particularly in terms of reducing assembly times and errors. In the fifth section, the research perspectives of the respective publications are evaluated and summarized in both a qualitative and quantitative way. The present mixed-methods study is not generalizable due to its limitations such as a small sample size, the geographical scope of the study, type of databanks, time of publication and language of the reviewed articles, and methods of data collection. It does, however, identify potential areas for future research and provide recommendations for further investigation.

辅助系统在提高工业装配的工作条件和效率方面发挥着至关重要的作用。在工业4.0的背景下，确定有助于实现装配目标及其经济效益的辅助系统类型非常重要。首先，介绍课题的意义，提出研究问题。第二，基本的技术术语将被定义，第三，结构化文献综述（SLR）的研究方法将被描绘。第四部分概述了在操作实践和学术测试设置中使用的人体工程学和信息辅助系统。它进一步解释了在装配中使用辅助系统的原因及其经济效益，特别是在减少装配时间和错误方面。第五部分从定性和定量两方面对各自出版物的研究视角进行了评价和总结。目前的混合方法研究由于样本量小、研究的地理范围、数据库类型、发表时间和审查文章的语言以及数据收集方法等限制而不能推广。然而，它确实确定了未来研究的潜在领域，并为进一步调查提供了建议。

{"title":"Investigation of assistance systems in assembly in the context of digitalization: A systematic literature review","authors":"Mathias König , Herwig Winkler","doi":"10.1016/j.jmsy.2024.11.015","DOIUrl":"10.1016/j.jmsy.2024.11.015","url":null,"abstract":"<div><div>Assistance systems play a crucial role in enhancing working conditions and efficiency in industrial assembly. In the context of Industry 4.0, it is important to determine the types of assistance systems that contribute to assembly goals as well as their economic benefits. First, the significance of the topic will be introduced, and the research questions will be presented. Second, the basic technical terms will be defined, and third, the research methodology of a structured literature review (SLR) will be delineated. The fourth section presents an overview of the ergonomic and information assistance systems used in operational practice and academic test set-ups. It further explains the reasons for using assistance systems in assembly and their economic benefits, particularly in terms of reducing assembly times and errors. In the fifth section, the research perspectives of the respective publications are evaluated and summarized in both a qualitative and quantitative way. The present mixed-methods study is not generalizable due to its limitations such as a small sample size, the geographical scope of the study, type of databanks, time of publication and language of the reviewed articles, and methods of data collection. It does, however, identify potential areas for future research and provide recommendations for further investigation.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"78 ","pages":"Pages 187-199"},"PeriodicalIF":12.2,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Material removal rate optimization with bayesian optimized differential evolution based on deep learning in robotic polishing 基于深度学习的贝叶斯优化微分进化机器人抛光材料去除率优化

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-11-29 DOI: 10.1016/j.jmsy.2024.11.014

Ruoxin Wang , Chi Fai Cheung , Yikai Zang , Chunjin Wang , Changlin Liu

Large aperture aspheric optical surfaces (LAAOS) have been applied in many industries, but their high requirements for precision and efficiency make manufacturing more challenging. Robotic polishing is a representative computer-controlled optical surfacing technique to manufacture LAAOS with low-cost and high-efficiency. However, how to achieve the highest material removal rate (MRR) involves many process parameters. It is difficult to determine the optimal parameter settings since the complex relationships among them. In this paper, a novel Bayesian optimized differential evolution based on deep learning method is proposed to optimize the MRR, in which the designed deep neural network is responsible for MRR modeling and Bayesian optimized differential evolution is used for MRR optimization. Bayesian optimization is used to find the best hyperparameter of differential evolution method so as to improve optimization performance. To evaluate the proposed method, a series of robotic polishing experiments are conducted to build the MRR model. The optimization performance comparison experiments show the superiority of our proposed method, which increases MRR by an average of 0.16.

大口径非球面光学表面（LAAOS）在许多行业中得到了应用，但其对精度和效率的高要求使其制造更具挑战性。机器人抛光是一种具有代表性的低成本、高效率制造LAAOS的计算机控制光学抛光技术。然而，如何达到最高的材料去除率（MRR）涉及到许多工艺参数。由于各参数之间的关系复杂，确定最佳参数设置比较困难。本文提出了一种新的基于深度学习的贝叶斯优化微分进化方法来优化MRR，其中设计的深度神经网络负责MRR建模，并使用贝叶斯优化微分进化进行MRR优化。采用贝叶斯优化方法寻找微分进化方法的最佳超参数，以提高优化性能。为了验证所提出的方法，进行了一系列机器人抛光实验来建立MRR模型。优化后的性能对比实验表明了本文方法的优越性，MRR平均提高了0.16。

{"title":"Material removal rate optimization with bayesian optimized differential evolution based on deep learning in robotic polishing","authors":"Ruoxin Wang , Chi Fai Cheung , Yikai Zang , Chunjin Wang , Changlin Liu","doi":"10.1016/j.jmsy.2024.11.014","DOIUrl":"10.1016/j.jmsy.2024.11.014","url":null,"abstract":"<div><div>Large aperture aspheric optical surfaces (LAAOS) have been applied in many industries, but their high requirements for precision and efficiency make manufacturing more challenging. Robotic polishing is a representative computer-controlled optical surfacing technique to manufacture LAAOS with low-cost and high-efficiency. However, how to achieve the highest material removal rate (MRR) involves many process parameters. It is difficult to determine the optimal parameter settings since the complex relationships among them. In this paper, a novel Bayesian optimized differential evolution based on deep learning method is proposed to optimize the MRR, in which the designed deep neural network is responsible for MRR modeling and Bayesian optimized differential evolution is used for MRR optimization. Bayesian optimization is used to find the best hyperparameter of differential evolution method so as to improve optimization performance. To evaluate the proposed method, a series of robotic polishing experiments are conducted to build the MRR model. The optimization performance comparison experiments show the superiority of our proposed method, which increases MRR by an average of 0.16.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"78 ","pages":"Pages 178-186"},"PeriodicalIF":12.2,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0