Powder Bed Fusion (PBF) is an additive manufacturing process in which laser heat liquefies blown powder particles on top of a powder bed, and cooling solidifies the melted powder particles. During this process, the laser beam heat interacts with the powder causing thermal emission and affecting the melt pool. This paper aims to predict heat emission in PBF by harnessing the strengths of recurrent neural networks. Long Short-Term Memory (LSTM) networks are developed to learn from sequential data (emission readings), while the learning is guided by process physics including laser power, laser speed, layer number, and scanning patterns. To reduce the computational efforts on model training, the LSTM models are integrated with a new approach for down-sampling the pyrometry raw data and extracting useful statistical features from raw data. The structure and hyperparameters of the LSTM model reflect several iterations of tuning based on the training on the pyrometer readings data. Results reveal useful knowledge on how raw pyrometer data should be processed to work the best with LSTM, how physics features are informative in predicting overheating, and the effectiveness of physics-guided LSTM in emission prediction.
{"title":"PHYSICS-GUIDED LONG SHORT-TERM MEMORY NETWORKS FOR EMISSION PREDICTION IN LASER POWDER BED FUSION","authors":"Rong Lei, Y.B. Guo, W. Guo","doi":"10.1115/1.4063270","DOIUrl":"https://doi.org/10.1115/1.4063270","url":null,"abstract":"\u0000 Powder Bed Fusion (PBF) is an additive manufacturing process in which laser heat liquefies blown powder particles on top of a powder bed, and cooling solidifies the melted powder particles. During this process, the laser beam heat interacts with the powder causing thermal emission and affecting the melt pool. This paper aims to predict heat emission in PBF by harnessing the strengths of recurrent neural networks. Long Short-Term Memory (LSTM) networks are developed to learn from sequential data (emission readings), while the learning is guided by process physics including laser power, laser speed, layer number, and scanning patterns. To reduce the computational efforts on model training, the LSTM models are integrated with a new approach for down-sampling the pyrometry raw data and extracting useful statistical features from raw data. The structure and hyperparameters of the LSTM model reflect several iterations of tuning based on the training on the pyrometer readings data. Results reveal useful knowledge on how raw pyrometer data should be processed to work the best with LSTM, how physics features are informative in predicting overheating, and the effectiveness of physics-guided LSTM in emission prediction.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48212078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the development and gradual maturity of additive manufacturing (AM) over the years, AM has reached a stage where implementation into a conventional production system becomes possible. With AM suitable for small volume of highly customized production, there are various ways of implementing AM in a conventional production line. The aim of this paper is to present a strategic design approach of implementing AM with conventional manufacturing in a complementary manner for parallel processing of production orders of large quantities in a make-to-stock environment. By assuming that a single machine in conventional manufacturing can be operated using AM, splitting of production orders is allowed. Therefore production can be conducted by both conventional and AM processes simultaneously, with the latter being able to produce various make-to-stock parts in a single build. A generic algorithm with a scheduling and rule-based heuristic for part allocation on build plate of AM process is used to solve a multi-objective implementation problem of AM with conventional manufacturing, with cost, scheduling and sustainability being the considered performance measures. By obtaining a knee-point solution using varying numbers of population size and generation number, an experiment involving an industry case study of implementing fused deposition modelling (FDM) process with injection moulding process shows the greatest impact, i.e., increase, in cost. Except for material efficiency, improvements are shown in scheduling and carbon footprint objectives.
{"title":"Strategic Production Process Design with Additive Manufacturing in a Make-to-stock Environment","authors":"P. C. Chua, S. K. Moon, Y. Ng, Manel Lopez","doi":"10.1115/1.4063285","DOIUrl":"https://doi.org/10.1115/1.4063285","url":null,"abstract":"\u0000 With the development and gradual maturity of additive manufacturing (AM) over the years, AM has reached a stage where implementation into a conventional production system becomes possible. With AM suitable for small volume of highly customized production, there are various ways of implementing AM in a conventional production line. The aim of this paper is to present a strategic design approach of implementing AM with conventional manufacturing in a complementary manner for parallel processing of production orders of large quantities in a make-to-stock environment. By assuming that a single machine in conventional manufacturing can be operated using AM, splitting of production orders is allowed. Therefore production can be conducted by both conventional and AM processes simultaneously, with the latter being able to produce various make-to-stock parts in a single build. A generic algorithm with a scheduling and rule-based heuristic for part allocation on build plate of AM process is used to solve a multi-objective implementation problem of AM with conventional manufacturing, with cost, scheduling and sustainability being the considered performance measures. By obtaining a knee-point solution using varying numbers of population size and generation number, an experiment involving an industry case study of implementing fused deposition modelling (FDM) process with injection moulding process shows the greatest impact, i.e., increase, in cost. Except for material efficiency, improvements are shown in scheduling and carbon footprint objectives.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42288570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The vigorous development of the human cyber-physical system (HCPS) and the next generation of artificial intelligence provide new ideas for smart manufacturing, where manufacturing quality prediction is an important issue in the manufacturing system. However, the small-scale data from humans in emerging HCPS-enabled manufacturing restricts the development of traditional quality prediction methods. To address this question, a data augmentation-based manufacturing quality prediction approach in human cyber-physical systems is proposed in this paper. Specifically, a Data Augmentation-Gradient Boosting Decision Tree (DA-GBDT) model is developed for quality prediction under the HCPS context. In addition, an adaptive selection algorithm of data augmentation rate is designed to balance the trade-off between the training time of the prediction model and the prediction accuracy. Finally, the experimental results of automobile covering products demonstrate that the proposed method can improve the average prediction error of the model compared with the prevailing quality prediction methods. Moreover, the predicted quality information can provide guidance for product optimization decisions in smart manufacturing systems.
{"title":"Data Augmentation-based Manufacturing Quality Prediction Approach in Human Cyber-Physical Systems","authors":"Tianyue Wang, Bingtao Hu, Yixiong Feng, Xiaoxie Gao, Chen Yang, Jianrong Tan","doi":"10.1115/1.4063269","DOIUrl":"https://doi.org/10.1115/1.4063269","url":null,"abstract":"\u0000 The vigorous development of the human cyber-physical system (HCPS) and the next generation of artificial intelligence provide new ideas for smart manufacturing, where manufacturing quality prediction is an important issue in the manufacturing system. However, the small-scale data from humans in emerging HCPS-enabled manufacturing restricts the development of traditional quality prediction methods. To address this question, a data augmentation-based manufacturing quality prediction approach in human cyber-physical systems is proposed in this paper. Specifically, a Data Augmentation-Gradient Boosting Decision Tree (DA-GBDT) model is developed for quality prediction under the HCPS context. In addition, an adaptive selection algorithm of data augmentation rate is designed to balance the trade-off between the training time of the prediction model and the prediction accuracy. Finally, the experimental results of automobile covering products demonstrate that the proposed method can improve the average prediction error of the model compared with the prevailing quality prediction methods. Moreover, the predicted quality information can provide guidance for product optimization decisions in smart manufacturing systems.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48401734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract The joining techniques between carbon fiber reinforced polymer (CFRP) and metal are of great importance in many areas of structural mechanics where the optimization of weight, rigidity, and strength is a necessity (such as aeronautics, vehicles, energy generation, and biomechanics). As a result, several types of metal–composite joints have been manufactured using different methods, with the 3D metal anchor solution attracting significant attention. This study evaluates different anchor geometries applied to single lap joints through preliminary finite element method (FEM) simulations and experimental validation on joints between CFRP and Inconel 625 produced via a laser beam powder bed fusion (LB-PBF) additive process. The models proposed increase in complexity. The homogenization process is employed to determine the equivalent properties of the joint region that is occupied by metal anchors and CFRP. The model also supports topology parametrization to assess the impact of anchor geometry on structural properties. The study provides experimental validation of joint strength under tensile load for various anchoring surface topologies.
{"title":"Modeling and Experimental Validation of CFRP-Metal Joints Utilizing 3D Additively Manufactured Anchors","authors":"Giorgio De Pasquale, Antonio Coluccia","doi":"10.1115/1.4063110","DOIUrl":"https://doi.org/10.1115/1.4063110","url":null,"abstract":"Abstract The joining techniques between carbon fiber reinforced polymer (CFRP) and metal are of great importance in many areas of structural mechanics where the optimization of weight, rigidity, and strength is a necessity (such as aeronautics, vehicles, energy generation, and biomechanics). As a result, several types of metal–composite joints have been manufactured using different methods, with the 3D metal anchor solution attracting significant attention. This study evaluates different anchor geometries applied to single lap joints through preliminary finite element method (FEM) simulations and experimental validation on joints between CFRP and Inconel 625 produced via a laser beam powder bed fusion (LB-PBF) additive process. The models proposed increase in complexity. The homogenization process is employed to determine the equivalent properties of the joint region that is occupied by metal anchors and CFRP. The model also supports topology parametrization to assess the impact of anchor geometry on structural properties. The study provides experimental validation of joint strength under tensile load for various anchoring surface topologies.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136244258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Wenjun Xu, Siqi Feng, Bitao Yao, Zhenrui Ji, Zhihao Liu
Human-robot collaboration (HRC) combines the repeatability and strength of robots and human's ability of cognition and planning to enable a flexible and efficient production mode. The ideal HRC process is that robots can smoothly assist workers in complex environments. This means that robots need to know the process's turn-taking earlier, adapt to the operating habits of different workers, and make reasonable plans in advance to improve the fluency of HRC. However, many of the current HRC systems ignore the fluent turn-taking between robots and humans, which results in unsatisfactory HRC and affects productivity. Moreover, there are uncertainties in humans as different humans have different operating proficiency, resulting in different operating speeds. This requires the robots to be able to make early predictions of turn-taking even when human is uncertain. Therefore, in this paper, an early turn-taking prediction method in HRC assembly tasks with Izhi neuron model-based spiking neuron network (SNN) is proposed. On this basis, dynamic motion primitives (DMP) are used to establish trajectory templates at different operating speeds. The length of the sequence sent to the SNN network is judged by the matching degree between the observed data and the template, so as to adjust to human uncertainty. The proposed method is verified by the gear assembly case. The results show that our method can shorten the human-robot turn-taking recognition time under human uncertainty.
{"title":"Turn-taking prediction for human-robot collaborative assembly considering human uncertainty","authors":"Wenjun Xu, Siqi Feng, Bitao Yao, Zhenrui Ji, Zhihao Liu","doi":"10.1115/1.4063231","DOIUrl":"https://doi.org/10.1115/1.4063231","url":null,"abstract":"\u0000 Human-robot collaboration (HRC) combines the repeatability and strength of robots and human's ability of cognition and planning to enable a flexible and efficient production mode. The ideal HRC process is that robots can smoothly assist workers in complex environments. This means that robots need to know the process's turn-taking earlier, adapt to the operating habits of different workers, and make reasonable plans in advance to improve the fluency of HRC. However, many of the current HRC systems ignore the fluent turn-taking between robots and humans, which results in unsatisfactory HRC and affects productivity. Moreover, there are uncertainties in humans as different humans have different operating proficiency, resulting in different operating speeds. This requires the robots to be able to make early predictions of turn-taking even when human is uncertain. Therefore, in this paper, an early turn-taking prediction method in HRC assembly tasks with Izhi neuron model-based spiking neuron network (SNN) is proposed. On this basis, dynamic motion primitives (DMP) are used to establish trajectory templates at different operating speeds. The length of the sequence sent to the SNN network is judged by the matching degree between the observed data and the template, so as to adjust to human uncertainty. The proposed method is verified by the gear assembly case. The results show that our method can shorten the human-robot turn-taking recognition time under human uncertainty.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48689787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The use of Human-Robot Collaboration (HRC) in assembly tasks has gained increasing attention in recent years as it allows for the combination of the flexibility and dexterity of human operators with the repeatability of robots, thus meeting the demands of the current market. However, the performance of these collaborative systems is known to be influenced by various factors, including the complexity perceived by operators. This study aimed to investigate the effects of perceived complexity on the performance measures of HRC assembly. An experimental campaign was conducted in which a sample of skilled operators was instructed to perform six different variants of electronic boards and express a complexity assessment based on a set of assembly complexity criteria. Performance measures such as assembly time, in-process defects, quality control times, offline defects, total defects, and human stress response were monitored. The results of the study showed that the perceived complexity had a significant effect on assembly time, in-process and total defects, and human stress response, while no significant effect was found for offline defects and quality control times. Specifically, product variants perceived as more complex resulted in lower performance measures compared to products perceived as less complex. These findings hold important implications for the design and implementation of HRC assembly systems and suggest that perceived complexity should be taken into consideration to increase HRC performance.
{"title":"Exploring the effects of perceived complexity criteria on performance measures of human-robot collaborative assembly","authors":"E. Verna, Stefano Puttero, G. Genta, M. Galetto","doi":"10.1115/1.4063232","DOIUrl":"https://doi.org/10.1115/1.4063232","url":null,"abstract":"\u0000 The use of Human-Robot Collaboration (HRC) in assembly tasks has gained increasing attention in recent years as it allows for the combination of the flexibility and dexterity of human operators with the repeatability of robots, thus meeting the demands of the current market. However, the performance of these collaborative systems is known to be influenced by various factors, including the complexity perceived by operators. This study aimed to investigate the effects of perceived complexity on the performance measures of HRC assembly. An experimental campaign was conducted in which a sample of skilled operators was instructed to perform six different variants of electronic boards and express a complexity assessment based on a set of assembly complexity criteria. Performance measures such as assembly time, in-process defects, quality control times, offline defects, total defects, and human stress response were monitored. The results of the study showed that the perceived complexity had a significant effect on assembly time, in-process and total defects, and human stress response, while no significant effect was found for offline defects and quality control times. Specifically, product variants perceived as more complex resulted in lower performance measures compared to products perceived as less complex. These findings hold important implications for the design and implementation of HRC assembly systems and suggest that perceived complexity should be taken into consideration to increase HRC performance.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47350204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Partha Protim Mondal, Placid Ferreira, S. Kapoor, Patrick N. Bless
As a popular applied artificial intelligence tool, Bayesian networks are increasingly being used to model multistage manufacturing processes for fault diagnosis purposes. However, the major issue limiting the practical adoption of Bayesian networks is the difficulty of learning the network structure for large multistage processes. Traditionally, Bayesian network structures are learned either with the help of domain experts or by utilizing data-driven structure learning algorithms through trial and error. Both approaches have their limitations. On one hand, expert-driven approach is costly, time-consuming, cumbersome for large networks, susceptible to errors in assessing probabilities and on the other hand, data-driven approaches suffer from noise, biases, inadequacy of training data and often fail to capture the physical causal structure of the data. Therefore, in this paper, we propose a Bayesian network structure learning approach where popular manufacturing knowledge sources like the Failure Mode and Effect Analysis (FMEA) and hierarchical variable ordering are used as structural priors to guide the data-driven structure learning process. In addition, to introduce modularity and flexibility into the learning process, we present a sequential modeling approach for structure learning so that large multistage networks can be learned stage by stage progressively. Furthermore, through simulation studies, we compare and analyze the performance of the knowledge source based structurally-biased networks in the context of multistage process fault diagnosis.
{"title":"SEQUENTIAL MODELING AND KNOWLEDGE SOURCE INTEGRATION FOR IDENTIFYING THE STRUCTURE OF A BAYESIAN NETWORK FOR MULTISTAGE PROCESS MONITORING AND DIAGNOSIS","authors":"Partha Protim Mondal, Placid Ferreira, S. Kapoor, Patrick N. Bless","doi":"10.1115/1.4063235","DOIUrl":"https://doi.org/10.1115/1.4063235","url":null,"abstract":"\u0000 As a popular applied artificial intelligence tool, Bayesian networks are increasingly being used to model multistage manufacturing processes for fault diagnosis purposes. However, the major issue limiting the practical adoption of Bayesian networks is the difficulty of learning the network structure for large multistage processes. Traditionally, Bayesian network structures are learned either with the help of domain experts or by utilizing data-driven structure learning algorithms through trial and error. Both approaches have their limitations. On one hand, expert-driven approach is costly, time-consuming, cumbersome for large networks, susceptible to errors in assessing probabilities and on the other hand, data-driven approaches suffer from noise, biases, inadequacy of training data and often fail to capture the physical causal structure of the data. Therefore, in this paper, we propose a Bayesian network structure learning approach where popular manufacturing knowledge sources like the Failure Mode and Effect Analysis (FMEA) and hierarchical variable ordering are used as structural priors to guide the data-driven structure learning process. In addition, to introduce modularity and flexibility into the learning process, we present a sequential modeling approach for structure learning so that large multistage networks can be learned stage by stage progressively. Furthermore, through simulation studies, we compare and analyze the performance of the knowledge source based structurally-biased networks in the context of multistage process fault diagnosis.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41819234","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article introduces a new control method for web tension control on a complex roll-to-roll winding machine used in battery production. Traditional web tension control method cannot perform well enough under high winding speed: the parameter tuning process is time-consuming, and the disturbance rejection performance is not satisfying, and the control performance is not stable. A hybrid control method is proposed, and it is easy to be implemented on common programming platform for commercial winding machines with an easy tuning process, while providing superior control performance to traditional control method. The system modeling used in the control method is much simpler than the modeling in most of tension control research, providing better feasibility for industrial application.
{"title":"Li-ion Battery Electrode Manufacturing Control System in Winding Process: Tension Control in Industrial Complex Roll-to-roll Winding Machine via SMC-FLC Hybrid Control Method","authors":"Haozhen Chen, J. Ni","doi":"10.1115/1.4063233","DOIUrl":"https://doi.org/10.1115/1.4063233","url":null,"abstract":"\u0000 This article introduces a new control method for web tension control on a complex roll-to-roll winding machine used in battery production. Traditional web tension control method cannot perform well enough under high winding speed: the parameter tuning process is time-consuming, and the disturbance rejection performance is not satisfying, and the control performance is not stable. A hybrid control method is proposed, and it is easy to be implemented on common programming platform for commercial winding machines with an easy tuning process, while providing superior control performance to traditional control method. The system modeling used in the control method is much simpler than the modeling in most of tension control research, providing better feasibility for industrial application.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45519793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Metamaterials are designed with intrinsic geometries to deliver unique properties, and recent years have witnessed an upsurge in leveraging additive manufacturing (AM) to produce metamaterials. However, the frequent occurrence of geometric defects in AM poses a critical obstacle to realizing the desired properties of fabricated metamaterials. Advances in three-dimensional (3D) scanning technologies enable the capture of fine-grained 3D geometric patterns, thereby providing a great opportunity for detecting geometric defects in fabricated metamaterials for property-oriented quality assurance. Realizing the full potential of 3D scanning-based quality control hinges largely on devising effective approaches to process scanned point clouds and extract geometric-pertinent information. In this study, a novel framework is developed to integrate recurrence network-based 3D geometry profiling with deep one-class learning for geometric defect detection in AM of metamaterials. First, we extend existing recurrence network models that focus on image data to representing 3D point clouds, by designing a new mechanism that characterizes points' geometric pattern affinities and spatial proximities. Then, a one-class graph neural network (GNN) approach is tailored to uncover topological variations of the recurrence network and detect anomalies that associated with geometric defects in the fabricated metamaterial. The developed methodology is evaluated through comprehensive simulated and real-world case studies. Experimental results have highlighted the efficacy of the developed methodology in identifying both global and local geometric defects in AM-fabricated metamaterials.
{"title":"Recurrence Network based 3D Geometry Representation Learning for Quality Control in Additive Manufacturing of Metamaterials","authors":"Yujing Yang, Chen Kan","doi":"10.1115/1.4063236","DOIUrl":"https://doi.org/10.1115/1.4063236","url":null,"abstract":"\u0000 Metamaterials are designed with intrinsic geometries to deliver unique properties, and recent years have witnessed an upsurge in leveraging additive manufacturing (AM) to produce metamaterials. However, the frequent occurrence of geometric defects in AM poses a critical obstacle to realizing the desired properties of fabricated metamaterials. Advances in three-dimensional (3D) scanning technologies enable the capture of fine-grained 3D geometric patterns, thereby providing a great opportunity for detecting geometric defects in fabricated metamaterials for property-oriented quality assurance. Realizing the full potential of 3D scanning-based quality control hinges largely on devising effective approaches to process scanned point clouds and extract geometric-pertinent information. In this study, a novel framework is developed to integrate recurrence network-based 3D geometry profiling with deep one-class learning for geometric defect detection in AM of metamaterials. First, we extend existing recurrence network models that focus on image data to representing 3D point clouds, by designing a new mechanism that characterizes points' geometric pattern affinities and spatial proximities. Then, a one-class graph neural network (GNN) approach is tailored to uncover topological variations of the recurrence network and detect anomalies that associated with geometric defects in the fabricated metamaterial. The developed methodology is evaluated through comprehensive simulated and real-world case studies. Experimental results have highlighted the efficacy of the developed methodology in identifying both global and local geometric defects in AM-fabricated metamaterials.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43993417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The new wave of Industry 4.0 is transforming manufacturing factories into data-rich environments. This provides an unprecedented opportunity to feed large amounts of sensing data collected from the physical factory into the construction of digital twin (DT) in cyberspace. However, little has been done to fully utilize the DT technology to improve the smartness and autonomous levels of small and medium-sized manufacturing factories. Indeed, only a small fraction of small and medium-sized manufacturers (SMMs) has considered implementing DT technology. There is an urgent need to exploit the full potential of data analytics and simulation-enabled DTs for advanced manufacturing. Hence, this paper presents the design and development of DT models for simulation optimization of manufacturing process flows. First, we develop a multi-agent simulation model that describes nonlinear and stochastic dynamics among a network of interactive manufacturing things, including customers, machines, automated guided vehicles (AGVs), queues, and jobs. Second, we propose a statistical metamodeling approach to design sequential computer experiments to optimize the utilization of AGV under uncertainty. Third, we construct two new graph models - job flow graph and AGV traveling graph - to track and monitor the real-time performance of manufacturing jobshops. The proposed simulation-enabled DT approach is evaluated and validated with experimental studies for the representation of a real-world manufacturing factory. Experimental results show that the proposed methodology effectively transforms a manufacturing jobshop into a new generation of DT-enabled smart factories.
{"title":"Digital Twinning and Optimization of Manufacturing Process Flows","authors":"Hankang Lee, Hui Yang","doi":"10.1115/1.4063234","DOIUrl":"https://doi.org/10.1115/1.4063234","url":null,"abstract":"\u0000 The new wave of Industry 4.0 is transforming manufacturing factories into data-rich environments. This provides an unprecedented opportunity to feed large amounts of sensing data collected from the physical factory into the construction of digital twin (DT) in cyberspace. However, little has been done to fully utilize the DT technology to improve the smartness and autonomous levels of small and medium-sized manufacturing factories. Indeed, only a small fraction of small and medium-sized manufacturers (SMMs) has considered implementing DT technology. There is an urgent need to exploit the full potential of data analytics and simulation-enabled DTs for advanced manufacturing. Hence, this paper presents the design and development of DT models for simulation optimization of manufacturing process flows. First, we develop a multi-agent simulation model that describes nonlinear and stochastic dynamics among a network of interactive manufacturing things, including customers, machines, automated guided vehicles (AGVs), queues, and jobs. Second, we propose a statistical metamodeling approach to design sequential computer experiments to optimize the utilization of AGV under uncertainty. Third, we construct two new graph models - job flow graph and AGV traveling graph - to track and monitor the real-time performance of manufacturing jobshops. The proposed simulation-enabled DT approach is evaluated and validated with experimental studies for the representation of a real-world manufacturing factory. Experimental results show that the proposed methodology effectively transforms a manufacturing jobshop into a new generation of DT-enabled smart factories.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43520609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：