Human-robot collaboration (HRC) has been regarded as one of the most promising paradigms for human-centric smart manufacturing in the context of Industry 5.0. To improve human well-being and robotic flexibility in HRC, a plethora of works around human body perception have emerged over the years, but most of them only considered a specific facade of human recognition while lacking a holistic perspective of the human operator. To this end, this study proposes an exemplary vision-based Human Digital Twin (HDT) model for highly dynamic HRC applications. The model mainly consists of a convolutional neural network that can simultaneously model the hierarchical human status including 3D human posture, action intention, and ergonomic risk. Then, on the basis of the constructed HDT, a robotic motion planning strategy is further introduced with the aim of adaptively optimizing the robotic motion trajectory. Further experiments and case studies are conducted in an HRC scenario to demonstrate the effectiveness of our approach.
{"title":"A Vision-based Human Digital Twin Modelling Approach for Adaptive Human-Robot Collaboration","authors":"Junming Fan, Pai Zheng, Carman K. M. Lee","doi":"10.1115/1.4062430","DOIUrl":"https://doi.org/10.1115/1.4062430","url":null,"abstract":"\u0000 Human-robot collaboration (HRC) has been regarded as one of the most promising paradigms for human-centric smart manufacturing in the context of Industry 5.0. To improve human well-being and robotic flexibility in HRC, a plethora of works around human body perception have emerged over the years, but most of them only considered a specific facade of human recognition while lacking a holistic perspective of the human operator. To this end, this study proposes an exemplary vision-based Human Digital Twin (HDT) model for highly dynamic HRC applications. The model mainly consists of a convolutional neural network that can simultaneously model the hierarchical human status including 3D human posture, action intention, and ergonomic risk. Then, on the basis of the constructed HDT, a robotic motion planning strategy is further introduced with the aim of adaptively optimizing the robotic motion trajectory. Further experiments and case studies are conducted in an HRC scenario to demonstrate the effectiveness of our approach.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45745663","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}
Pee-Yew Lee, C. Weng, H. Huang, Li-Yan Wu, Guo-Hao Lu, Chao-Feng Liu, Cheng-You Chen, Ting-Yu Li, Yung-Sheng Lin
Micro/nano-textured Si wafers manufactured using metal-assisted chemical etching (MACE) have been the focus of several studies, but the mechanism of bubble generation during the MACE process affecting textured surfaces has rarely been reported. This study investigated the bubble effect due to the different placement patterns of the Si wafer (face-up, stirred face-down, and face-down). The results indicated that the placement pattern of the Si wafer notably influences the uniformity of outward appearance. However, no significant differences were noted in the scanning electron microscopy images of Si nanowires (SiNWs) at 0.5 h of etching. At 2 h of etching, the outward appearance uniformity of face-up etching was more homogeneous than that of stirred face-down and face-down patterns, and the SiNWs processed through face-up etching were longer (41 μm) than those subjected to stirred face-down etching (36 μm) and face-down etching (32 μm). Therefore, the placement pattern of Si wafer can affect the uniformity and properties of SiNWs because of bubbles trapped inside cavities or between SiNWs.
{"title":"Bubble Effects on Manufacturing of Silicon Nanowires by Metal-Assisted Chemical Etching","authors":"Pee-Yew Lee, C. Weng, H. Huang, Li-Yan Wu, Guo-Hao Lu, Chao-Feng Liu, Cheng-You Chen, Ting-Yu Li, Yung-Sheng Lin","doi":"10.1115/1.4062392","DOIUrl":"https://doi.org/10.1115/1.4062392","url":null,"abstract":"\u0000 Micro/nano-textured Si wafers manufactured using metal-assisted chemical etching (MACE) have been the focus of several studies, but the mechanism of bubble generation during the MACE process affecting textured surfaces has rarely been reported. This study investigated the bubble effect due to the different placement patterns of the Si wafer (face-up, stirred face-down, and face-down). The results indicated that the placement pattern of the Si wafer notably influences the uniformity of outward appearance. However, no significant differences were noted in the scanning electron microscopy images of Si nanowires (SiNWs) at 0.5 h of etching. At 2 h of etching, the outward appearance uniformity of face-up etching was more homogeneous than that of stirred face-down and face-down patterns, and the SiNWs processed through face-up etching were longer (41 μm) than those subjected to stirred face-down etching (36 μm) and face-down etching (32 μm). Therefore, the placement pattern of Si wafer can affect the uniformity and properties of SiNWs because of bubbles trapped inside cavities or between SiNWs.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41477361","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}
Semih Akin, Puyuan Wu, Chandra Nath, Jun Chen, M. Jun
Supersonic cold spraying of liquid droplets containing functional nanomaterials is of particular interest in advanced thin-film coating, that enabling high-adhesion strength particle deposition. In cold spraying, the optimum design of the supersonic nozzle is essential for accelerating particles to desired velocities. However, research on the supersonic nozzle design for liquid droplets is limited. Thus, we thoroughly investigate the influence of nozzle geometrical parameters (i.e., throat diameter, exit diameter, divergent length) on droplets acceleration by numerical modeling followed by experimental validation, and a case study on surface coating application. Two-phase flow modeling was used to predict droplets' behavior in continuous gas flow for different nozzle configurations. The results show that the nozzle expansion ratio - a function of throat and exit diameters - has a significant influence on droplet velocity, followed by divergent length. In particular, to correctly accelerate “low-inertia liquid droplets”, optimum nozzle expansion ratio for an axisymmetric convergent-divergent nozzle is found to be in a range of 1.5-2.5 for different sets of parameters, which is different than the recommended expansion ratio (i.e., 5-9) for cold spraying of conventional “metal” particles. Based on the simulation results, an optimal design of supersonic nozzle is selected and prototyped for the experimental studies. Numerical modeling results are validated by particle image velocimetry (PIV) measurements. Moreover, coating experiments confirm the adaptability of the optimized nozzle for supersonic cold spraying of droplets containing nanoparticles, which thereby has the potential for rapid production of advanced thin films.
{"title":"A study on converging-diverging nozzle design for supersonic spraying of liquid droplets toward nanocoating applications","authors":"Semih Akin, Puyuan Wu, Chandra Nath, Jun Chen, M. Jun","doi":"10.1115/1.4062351","DOIUrl":"https://doi.org/10.1115/1.4062351","url":null,"abstract":"\u0000 Supersonic cold spraying of liquid droplets containing functional nanomaterials is of particular interest in advanced thin-film coating, that enabling high-adhesion strength particle deposition. In cold spraying, the optimum design of the supersonic nozzle is essential for accelerating particles to desired velocities. However, research on the supersonic nozzle design for liquid droplets is limited. Thus, we thoroughly investigate the influence of nozzle geometrical parameters (i.e., throat diameter, exit diameter, divergent length) on droplets acceleration by numerical modeling followed by experimental validation, and a case study on surface coating application. Two-phase flow modeling was used to predict droplets' behavior in continuous gas flow for different nozzle configurations. The results show that the nozzle expansion ratio - a function of throat and exit diameters - has a significant influence on droplet velocity, followed by divergent length. In particular, to correctly accelerate “low-inertia liquid droplets”, optimum nozzle expansion ratio for an axisymmetric convergent-divergent nozzle is found to be in a range of 1.5-2.5 for different sets of parameters, which is different than the recommended expansion ratio (i.e., 5-9) for cold spraying of conventional “metal” particles. Based on the simulation results, an optimal design of supersonic nozzle is selected and prototyped for the experimental studies. Numerical modeling results are validated by particle image velocimetry (PIV) measurements. Moreover, coating experiments confirm the adaptability of the optimized nozzle for supersonic cold spraying of droplets containing nanoparticles, which thereby has the potential for rapid production of advanced thin films.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45353497","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 development of feasible methods for the design of power-skiving tools without cutting interference is essential in ensuring the accuracy of involute internal machined gears. One of the most crucial points in obtaining interference-free and re-sharpenable power-skiving tools is that of determining the cutting edge and clearance surface. The present study introduces a tilt angle during the power-skiving process to design a simple cylindrical interference-free tool shape, in which the shape of the cutting edge remains unchanged after re-sharpening. The relative position between the new tool center point and gear during machining is similarly unchanged after the re-sharpening process. In addition, the clearance angle between the tool and the gear can be easily adjusted simply by changing the tilt angle of the tool during power-skiving. The validity of the proposed design method is demonstrated through a simple numerical example. The simulation results confirm the feasibility of the proposed method.
{"title":"Algebraic modeling of cylindrical interference-free power-skiving tool for involute internal gear cutting with tilt angle","authors":"","doi":"10.1115/1.4062312","DOIUrl":"https://doi.org/10.1115/1.4062312","url":null,"abstract":"\u0000 The development of feasible methods for the design of power-skiving tools without cutting interference is essential in ensuring the accuracy of involute internal machined gears. One of the most crucial points in obtaining interference-free and re-sharpenable power-skiving tools is that of determining the cutting edge and clearance surface. The present study introduces a tilt angle during the power-skiving process to design a simple cylindrical interference-free tool shape, in which the shape of the cutting edge remains unchanged after re-sharpening. The relative position between the new tool center point and gear during machining is similarly unchanged after the re-sharpening process. In addition, the clearance angle between the tool and the gear can be easily adjusted simply by changing the tilt angle of the tool during power-skiving. The validity of the proposed design method is demonstrated through a simple numerical example. The simulation results confirm the feasibility of the proposed method.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47793554","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 Fiber-reinforced polymer (FRP) additive manufacturing has transformed fused filament fabrication (FFF) by manufacturing products with excellent mechanical characteristics. However, the surface finish and dimensional characteristics of printed FRP parts are typically poor due to protruding fibers and the stair-stepping effect. This parametric study examined an in-process combined mechanical plus chemical finishing technique to improve the surface finish of FRPs manufactured through FFF. This process is particularly useful for internal or complex features that cannot be otherwise finished after printing. In this work, a custom-built three-axis machine with printing, machining, and chemical finishing capabilities was used for the experiments. The effect of mechanical finishing on surface characteristics was first quantified using chip load and spindle speed as independent parameters. Following that, chemical treatment was performed on the already machined surface at two pressing depths (PD), which control the normal contact force acting on the surface. The best surface characteristics were observed at a low chip load of 0.007 mm and a moderately high spindle speed of 20,000 rpm. After chemical treatment using a lower PD, a surface roughness reduction was observed (from 8.041 to 4.988 µm). Increased PD led to even lower Ra values (from 4.988 to 3.538 µm) due to the enhanced fiber encapsulation phenomenon. Finally, the dimensional analysis revealed that the final combined finished samples had less than 1%-dimensional error (0.05 mm), which is an order of magnitude less than the typical error in FFF-printed parts (0.5 mm). This study provides means to conduct finishing in an additive manufacturing environment to reduce the time, labor, and cost associated with post-processing.
{"title":"Surface Characterization of Three-Dimensional Printed Fiber-Reinforced Polymer Following an In-Process Mechanical–Chemical Finishing Method","authors":"Aman Nigam, Bruce L. Tai","doi":"10.1115/1.4062146","DOIUrl":"https://doi.org/10.1115/1.4062146","url":null,"abstract":"Abstract Fiber-reinforced polymer (FRP) additive manufacturing has transformed fused filament fabrication (FFF) by manufacturing products with excellent mechanical characteristics. However, the surface finish and dimensional characteristics of printed FRP parts are typically poor due to protruding fibers and the stair-stepping effect. This parametric study examined an in-process combined mechanical plus chemical finishing technique to improve the surface finish of FRPs manufactured through FFF. This process is particularly useful for internal or complex features that cannot be otherwise finished after printing. In this work, a custom-built three-axis machine with printing, machining, and chemical finishing capabilities was used for the experiments. The effect of mechanical finishing on surface characteristics was first quantified using chip load and spindle speed as independent parameters. Following that, chemical treatment was performed on the already machined surface at two pressing depths (PD), which control the normal contact force acting on the surface. The best surface characteristics were observed at a low chip load of 0.007 mm and a moderately high spindle speed of 20,000 rpm. After chemical treatment using a lower PD, a surface roughness reduction was observed (from 8.041 to 4.988 µm). Increased PD led to even lower Ra values (from 4.988 to 3.538 µm) due to the enhanced fiber encapsulation phenomenon. Finally, the dimensional analysis revealed that the final combined finished samples had less than 1%-dimensional error (0.05 mm), which is an order of magnitude less than the typical error in FFF-printed parts (0.5 mm). This study provides means to conduct finishing in an additive manufacturing environment to reduce the time, labor, and cost associated with post-processing.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134955097","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}
M. Ansari, Hemant Agiwal, D. Franke, M. Zinn, F. Pfefferkorn, S. Rudraraju
This study employs a high-fidelity numerical framework to determine the plastic material flow patterns and temperature distributions that lead to void formation during friction stir welding (FSW), and to relate the void morphologies to the underlying alloy material properties and process conditions. Three aluminum alloys, viz., 6061-T6, 7075-T6, and 5053-H18 were investigated under varying traverse speeds. The choice of aluminum alloys enables investigation of a wide range of thermal and mechanical properties. The numerical simulations were validated using experimental observations of void morphologies in these three alloys. Temperatures, plastic strain rates, and material flow patterns are considered. The key results from this study are: (1) The predicted stir zone and void morphology are in good agreement with the experimental observations, (2) The temperature and plastic strain-rate maps in the steady-state process conditions show a strong dependency on the alloy type and traverse speeds, (3) The material velocity contours provide a good insight into the material flow in the stir zone for the FSW process conditions that result in voids as well as those that do not result in voids. The numerical model and the ensuing parametric studies presented in this work provide a framework for understanding material flow under different process conditions in aluminum alloys, and potentially in other alloys. Furthermore, the utility of the numerical model for making quantitative predictions and investigating different process parameters to reduce void formation is demonstrated.
{"title":"Numerical investigation into the influence of alloy type and thermo-mechanics on void formation in Friction Stir Welding of Aluminium alloys","authors":"M. Ansari, Hemant Agiwal, D. Franke, M. Zinn, F. Pfefferkorn, S. Rudraraju","doi":"10.1115/1.4062270","DOIUrl":"https://doi.org/10.1115/1.4062270","url":null,"abstract":"\u0000 This study employs a high-fidelity numerical framework to determine the plastic material flow patterns and temperature distributions that lead to void formation during friction stir welding (FSW), and to relate the void morphologies to the underlying alloy material properties and process conditions. Three aluminum alloys, viz., 6061-T6, 7075-T6, and 5053-H18 were investigated under varying traverse speeds. The choice of aluminum alloys enables investigation of a wide range of thermal and mechanical properties. The numerical simulations were validated using experimental observations of void morphologies in these three alloys. Temperatures, plastic strain rates, and material flow patterns are considered. The key results from this study are: (1) The predicted stir zone and void morphology are in good agreement with the experimental observations, (2) The temperature and plastic strain-rate maps in the steady-state process conditions show a strong dependency on the alloy type and traverse speeds, (3) The material velocity contours provide a good insight into the material flow in the stir zone for the FSW process conditions that result in voids as well as those that do not result in voids. The numerical model and the ensuing parametric studies presented in this work provide a framework for understanding material flow under different process conditions in aluminum alloys, and potentially in other alloys. Furthermore, the utility of the numerical model for making quantitative predictions and investigating different process parameters to reduce void formation is demonstrated.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"68 11","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41277575","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}
In order to establish a high-fidelity mechanism model for investigating the molten pool behaviors during directed energy deposition (DED) process, a molten pool dynamics model combined with the discrete element method is developed in present study. The proposed model contains newly added particle sources to further intuitively reproduce the interaction between the discrete powder particles and the molten pool. Meanwhile, the effects of the nozzle structure, carrier gas and shielding gas on the feedstock feeding process are simulated in detail using the gas-powder flow model based on the multi-phase flow theory. The gas-powder flow model is used to provide the reasonable outlet velocities, focal distance and radius of the focal point for the particle sources in the molten pool dynamics model, which solves the difficulty that the motion state of the powder streams obtained by the molten pool dynamics simulation are hard to reproduce the actual situation. Besides, relevant experiments are conducted to verify the accuracy of the developed models. The predicted parameters of the powder streams are consistent with the experiment, and the deviations of the predicted molten pool dimensions are less than 10%. The heat and mass transfer phenomena inside the molten pool are also revealed. Furthermore, the maximum size of the spherical pore defects is predicted to be 18.6 µm, which is underestimated by 7% compared to the microscopic observation. Taken together, the developed numerical model could augment and improve the training samples for the machine learning modelling of DED process.
{"title":"Multi-physics investigations on the gas-powder flow and the molten pool dynamics during directed energy deposition process","authors":"Chenghong Duan, X. Cao, Xiangpeng Luo, Dazhi Shang, X. Hao","doi":"10.1115/1.4062259","DOIUrl":"https://doi.org/10.1115/1.4062259","url":null,"abstract":"\u0000 In order to establish a high-fidelity mechanism model for investigating the molten pool behaviors during directed energy deposition (DED) process, a molten pool dynamics model combined with the discrete element method is developed in present study. The proposed model contains newly added particle sources to further intuitively reproduce the interaction between the discrete powder particles and the molten pool. Meanwhile, the effects of the nozzle structure, carrier gas and shielding gas on the feedstock feeding process are simulated in detail using the gas-powder flow model based on the multi-phase flow theory. The gas-powder flow model is used to provide the reasonable outlet velocities, focal distance and radius of the focal point for the particle sources in the molten pool dynamics model, which solves the difficulty that the motion state of the powder streams obtained by the molten pool dynamics simulation are hard to reproduce the actual situation. Besides, relevant experiments are conducted to verify the accuracy of the developed models. The predicted parameters of the powder streams are consistent with the experiment, and the deviations of the predicted molten pool dimensions are less than 10%. The heat and mass transfer phenomena inside the molten pool are also revealed. Furthermore, the maximum size of the spherical pore defects is predicted to be 18.6 µm, which is underestimated by 7% compared to the microscopic observation. Taken together, the developed numerical model could augment and improve the training samples for the machine learning modelling of DED process.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41738532","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}
Suyog Ghungrad, Meysam Faegh, B. Gould, S. Wolff, Azadeh Haghighi
Physics-informed deep learning (PIDL) is one of the emerging topics in additive manufacturing (AM). However, the success of previous PIDL approaches is generally significantly dependent on the existence of massive datasets. As the data collection in AM is usually challenging, a novel Architecture-driven PIDL structure named APIDL based on the deep unfolding approach for limited data scenarios has been proposed in the current study for predicting thermal history in the laser powder bed fusion process. The connections in this machine learning architecture are inspired by iterative thermal model equations. In other words, each iteration of the thermal model is mapped to a layer of the neural network. The hyper-parameters of the APIDL model are tuned, and its performance is analyzed. The APIDL for 1000 points with 80:20 split ratio achieves a testing mean absolute percentage error (MAPE) of 2.8% and R2 value of 0.936. The APIDL is compared with the artificial neural network, extra trees regressor (ETR), support vector regressor, and long short-term memory algorithms. It was shown that the proposed APIDL model outperforms the others. The MAPE and R2 of APIDL are 55.7% lower and 15.6% higher than the ETR, which had the best performance among other pure ML models.
{"title":"Architecture-driven Physics-informed Deep Learning for Temperature Prediction in Laser Powder Bed Fusion Additive Manufacturing with Limited Data","authors":"Suyog Ghungrad, Meysam Faegh, B. Gould, S. Wolff, Azadeh Haghighi","doi":"10.1115/1.4062237","DOIUrl":"https://doi.org/10.1115/1.4062237","url":null,"abstract":"\u0000 Physics-informed deep learning (PIDL) is one of the emerging topics in additive manufacturing (AM). However, the success of previous PIDL approaches is generally significantly dependent on the existence of massive datasets. As the data collection in AM is usually challenging, a novel Architecture-driven PIDL structure named APIDL based on the deep unfolding approach for limited data scenarios has been proposed in the current study for predicting thermal history in the laser powder bed fusion process. The connections in this machine learning architecture are inspired by iterative thermal model equations. In other words, each iteration of the thermal model is mapped to a layer of the neural network. The hyper-parameters of the APIDL model are tuned, and its performance is analyzed. The APIDL for 1000 points with 80:20 split ratio achieves a testing mean absolute percentage error (MAPE) of 2.8% and R2 value of 0.936. The APIDL is compared with the artificial neural network, extra trees regressor (ETR), support vector regressor, and long short-term memory algorithms. It was shown that the proposed APIDL model outperforms the others. The MAPE and R2 of APIDL are 55.7% lower and 15.6% higher than the ETR, which had the best performance among other pure ML models.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42822968","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}
Seng Xiang, Xingtao Liu, Li-cong An, Haozheng J. Qu, G. Cheng
Modulating the heating and cooling during plastic deformation has been critical to control the microstructure and phase change in metals. During laser shock peening under optimal elevated temperatures, high-density dislocations and nanoprecipitates can be generated to greatly enhance material strength and fatigue life in metals. In this paper, we propose a general methodology to modulate the heating and cooling during laser shock processing via temporal pulse shaping, namely dual pulse laser shock peening (DP-LSP), which combines both ultrafast-heating and laser shock peening in one operation to generate desired microstructure and mechanical property. Single pulse LSP was able to remelt large second phase precipitates due to fast cooling, resulting in smaller grains (500nm), while using DP-LSP with appropriate pulse durations, dynamic precipitation effects can generate nanosized (30nm) intermetallic phase Al3Ti with high density. By generation of grain size refinement, high density nanoscale precipitates, and dislocations, the yield strength increase by 18% and 102% compared with single pulse processing, and original sample respectively. A phase-field model (PFM) and multiscale dislocation dynamics (MDD) were applied to study dislocation dynamics and nanoprecipitation generation during DP-LSP, and their interaction. The work provides a basis for controlling microstructure by DP-LSP to enhance mechanical properties in metals.
{"title":"Nanoengineered Laser Shock Processing via Pulse Shaping for Nanostructuring in Metals: Multiscale simulations and experiments","authors":"Seng Xiang, Xingtao Liu, Li-cong An, Haozheng J. Qu, G. Cheng","doi":"10.1115/1.4062234","DOIUrl":"https://doi.org/10.1115/1.4062234","url":null,"abstract":"\u0000 Modulating the heating and cooling during plastic deformation has been critical to control the microstructure and phase change in metals. During laser shock peening under optimal elevated temperatures, high-density dislocations and nanoprecipitates can be generated to greatly enhance material strength and fatigue life in metals. In this paper, we propose a general methodology to modulate the heating and cooling during laser shock processing via temporal pulse shaping, namely dual pulse laser shock peening (DP-LSP), which combines both ultrafast-heating and laser shock peening in one operation to generate desired microstructure and mechanical property. Single pulse LSP was able to remelt large second phase precipitates due to fast cooling, resulting in smaller grains (500nm), while using DP-LSP with appropriate pulse durations, dynamic precipitation effects can generate nanosized (30nm) intermetallic phase Al3Ti with high density. By generation of grain size refinement, high density nanoscale precipitates, and dislocations, the yield strength increase by 18% and 102% compared with single pulse processing, and original sample respectively. A phase-field model (PFM) and multiscale dislocation dynamics (MDD) were applied to study dislocation dynamics and nanoprecipitation generation during DP-LSP, and their interaction. The work provides a basis for controlling microstructure by DP-LSP to enhance mechanical properties in metals.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48496380","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 wide application of new electric vehicle (EV) battery in various industrial fields, it is important to establish a systematic intelligent battery recycling system that can be used to find out the resource wastes and environmental impacts for the retired EV battery. By combining the disassembly and echelon utilization of EV battery recycling in the re-manufacturing fields, human-robot collaboration (HRC) disassembly method can be used to solve many huge challenges about the efficiency and safety of retired EV battery recycling. In order to find out the common problems in the human-robot collaboration disassembly process of EV battery recycling, a dynamic disassembly process optimization method based on Multi-Agent Reinforcement Learning (MARL) algorithm is proposed. Furthermore, it is necessary to disassemble the EV battery disassembly task trajectory based on human-robot collaboration disassembly task in 2D planar, which can be used to acquire the optimal disassembly paths in the same disassembly planar combining the Q-learning algorithm. The disassembly task sequence can be completed through standard trajectory matching. Finally the feasibility of the method is verified by disassembly operations for a specific battery module case.
{"title":"Multi-agent Reinforcement Learning Method for Disassembly Sequential Task Optimization Based on Human-Robot Collaborative Disassembly in Electric Vehicle Battery Recycling","authors":"Jinhua Xiao, Jiaxu Gao, N. Anwer, B. Eynard","doi":"10.1115/1.4062235","DOIUrl":"https://doi.org/10.1115/1.4062235","url":null,"abstract":"\u0000 With the wide application of new electric vehicle (EV) battery in various industrial fields, it is important to establish a systematic intelligent battery recycling system that can be used to find out the resource wastes and environmental impacts for the retired EV battery. By combining the disassembly and echelon utilization of EV battery recycling in the re-manufacturing fields, human-robot collaboration (HRC) disassembly method can be used to solve many huge challenges about the efficiency and safety of retired EV battery recycling. In order to find out the common problems in the human-robot collaboration disassembly process of EV battery recycling, a dynamic disassembly process optimization method based on Multi-Agent Reinforcement Learning (MARL) algorithm is proposed. Furthermore, it is necessary to disassemble the EV battery disassembly task trajectory based on human-robot collaboration disassembly task in 2D planar, which can be used to acquire the optimal disassembly paths in the same disassembly planar combining the Q-learning algorithm. The disassembly task sequence can be completed through standard trajectory matching. Finally the feasibility of the method is verified by disassembly operations for a specific battery module case.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44396908","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：