Taekwang Ha, Torgeir Welo, Geir Ringen, Jyhwen Wang
Abstract Springback is one of the factors that causes decreased product quality in metal forming. Advanced 2D and 3D stretch bending process can be used to manufacture a complex geometry a profile with springback reduction. For a non-linear springback problem, an artificial neural network (ANN) is an attractive data-driven approach to achieving springback prediction and control. The main objective of the present work is to control springback and improve geometrical quality with an ANN in 2D and 3D stretch bending. In general, an ANN is trained with collected data sets from a large number of experiments, causing expensive costs and time-consuming work. In the present work, the training data sets for the proposed ANN are obtained from both experiments and an analytical springback model. As the analytical model can adopt different bending angles, material properties, and geometries, supplementary data by the analytical model significantly reduced the number of experiments needed for ANN training. Contrary to the typical springback predictions, the proposed ANN synthesizes the machine settings based on the desired dimensions as the inputs. It is shown that springback can be controlled by specifying the bend angles provided by the ANN prediction. The proposed ANN method was validated in 2D and 3D stretch bending, and its prediction and control performance is favorably compared to an ANN trained with only experimental data sets.
{"title":"Smart Control of Springback in Stretch Bending of a Rectangular Tube by an Artificial Neural Network","authors":"Taekwang Ha, Torgeir Welo, Geir Ringen, Jyhwen Wang","doi":"10.1115/1.4063737","DOIUrl":"https://doi.org/10.1115/1.4063737","url":null,"abstract":"Abstract Springback is one of the factors that causes decreased product quality in metal forming. Advanced 2D and 3D stretch bending process can be used to manufacture a complex geometry a profile with springback reduction. For a non-linear springback problem, an artificial neural network (ANN) is an attractive data-driven approach to achieving springback prediction and control. The main objective of the present work is to control springback and improve geometrical quality with an ANN in 2D and 3D stretch bending. In general, an ANN is trained with collected data sets from a large number of experiments, causing expensive costs and time-consuming work. In the present work, the training data sets for the proposed ANN are obtained from both experiments and an analytical springback model. As the analytical model can adopt different bending angles, material properties, and geometries, supplementary data by the analytical model significantly reduced the number of experiments needed for ANN training. Contrary to the typical springback predictions, the proposed ANN synthesizes the machine settings based on the desired dimensions as the inputs. It is shown that springback can be controlled by specifying the bend angles provided by the ANN prediction. The proposed ANN method was validated in 2D and 3D stretch bending, and its prediction and control performance is favorably compared to an ANN trained with only experimental data sets.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136210186","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}
Anna Komodromos, Joshua Grodotzki, Felix Kolpak, A. Erman Tekkaya
Abstract By Directed Energy Deposition (DED) a flexible design of cooling channels in forming tools, e.g. hot stamping, with a variety of sizes and a high positioning flexibility compared to machining processes is possible. The subsequent ball burnishing of the tool surfaces in combination with a variation of the DED process parameters enables a control of the tool surface properties and the friction behavior. Parameters such as the ball burnishing pressure or the path overlapping in the DED process are investigated to quantify their effects on roughness, hardness, friction, residual stresses and heat transfer coefficient of generic tool surfaces. The friction coefficient at elevated temperatures depends strongly on the surface roughness of the tool steel surfaces generated by DED and ball burnishing. The latter process improves the surface integrity: the roughness peaks are leveled by up to 75 %, the hardness and the residual stresses are enhanced by up to 20 % and 70 %, respectively. However, the roughness of the tool surfaces is determined mainly by the path overlapping of the welded beads in the DED process. Despite the higher surface roughness, the heat transfer coefficient is in the range of conventionally manufactured tool surfaces of up to 2,700 W/m2K for contact pressures up to 40 MPa. First hot stamping experiments demonstrate that the tools manufactured by the novel process combination are able to manufacture 22MnB5 hat profiles with an increased and more homogenous hardness as well as more homogeneous thickness distribution compared to conventionally manufactured tools.
{"title":"Characterization of Tool Surface Properties Generated by Directed Energy Deposition and Subsequent Ball Burnishing","authors":"Anna Komodromos, Joshua Grodotzki, Felix Kolpak, A. Erman Tekkaya","doi":"10.1115/1.4063736","DOIUrl":"https://doi.org/10.1115/1.4063736","url":null,"abstract":"Abstract By Directed Energy Deposition (DED) a flexible design of cooling channels in forming tools, e.g. hot stamping, with a variety of sizes and a high positioning flexibility compared to machining processes is possible. The subsequent ball burnishing of the tool surfaces in combination with a variation of the DED process parameters enables a control of the tool surface properties and the friction behavior. Parameters such as the ball burnishing pressure or the path overlapping in the DED process are investigated to quantify their effects on roughness, hardness, friction, residual stresses and heat transfer coefficient of generic tool surfaces. The friction coefficient at elevated temperatures depends strongly on the surface roughness of the tool steel surfaces generated by DED and ball burnishing. The latter process improves the surface integrity: the roughness peaks are leveled by up to 75 %, the hardness and the residual stresses are enhanced by up to 20 % and 70 %, respectively. However, the roughness of the tool surfaces is determined mainly by the path overlapping of the welded beads in the DED process. Despite the higher surface roughness, the heat transfer coefficient is in the range of conventionally manufactured tool surfaces of up to 2,700 W/m2K for contact pressures up to 40 MPa. First hot stamping experiments demonstrate that the tools manufactured by the novel process combination are able to manufacture 22MnB5 hat profiles with an increased and more homogenous hardness as well as more homogeneous thickness distribution compared to conventionally manufactured tools.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"195 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136209942","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 Selecting suitable cutting conditions is crucial in maintaining chatter stability and achieving acceptable surface quality. However, the selection of a constant set of cutting parameters is not feasible due to the time-varying dynamics of highly flexible thin-walled blades. This paper presents an optimal selection of tool orientation and spindle speed along the tool path as the metal is removed during the ball end milling of blades. The effects of tool orientation and speed on the mechanics and dynamics of the ball-end milling process are formulated. Test case simulations are used to demonstrate the impact of tool orientation and speed on chatter stability and forced vibrations. The proposed algorithm identifies the optimal spindle speed and tool orientation by continuously updating the workpiece dynamics as a function of time and tool position to achieve improved stability and surface quality. Stability simulations are conducted to assess the optimization approach's performance, and the results are compared with experiments by machining a series of thin-walled twisted fan blades.
{"title":"Chatter Avoidance by Spindle Speed and Orientation Planning in Five-Axis Ball-End Milling of Thin-Walled Blades","authors":"Behnam Karimi, Yusuf Altintas","doi":"10.1115/1.4063654","DOIUrl":"https://doi.org/10.1115/1.4063654","url":null,"abstract":"Abstract Selecting suitable cutting conditions is crucial in maintaining chatter stability and achieving acceptable surface quality. However, the selection of a constant set of cutting parameters is not feasible due to the time-varying dynamics of highly flexible thin-walled blades. This paper presents an optimal selection of tool orientation and spindle speed along the tool path as the metal is removed during the ball end milling of blades. The effects of tool orientation and speed on the mechanics and dynamics of the ball-end milling process are formulated. Test case simulations are used to demonstrate the impact of tool orientation and speed on chatter stability and forced vibrations. The proposed algorithm identifies the optimal spindle speed and tool orientation by continuously updating the workpiece dynamics as a function of time and tool position to achieve improved stability and surface quality. Stability simulations are conducted to assess the optimization approach's performance, and the results are compared with experiments by machining a series of thin-walled twisted fan blades.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135590871","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}
Michael Biehler, Reinaldo Mock, Shriyanshu Kode, Maham Mehmood, Palin Bhardwaj, Jianjun Shi
Abstract Additive manufacturing (AM) has revolutionized the way we design, prototype, and produce complex parts with unprecedented geometries. However, the lack of understanding of the functional properties of 3D printed parts has hindered their adoption in critical applications where reliability and durability are paramount. This paper proposes a novel approach to the functional qualification of 3D-printed parts via physical and digital twins. Physical twins are parts that are printed under the same process conditions as the functional parts and undergo a wide range of (destructive) tests to determine their mechanical, thermal, and chemical properties. Digital twins are virtual replicas of the physical twins that are generated using finite element analysis (FEA) simulations based on the 3D shape of the part of interest. We propose a novel approach to transfer learning, specifically designed for the fusion of diverse, unstructured 3D shape data and process inputs from multiple sources. The proposed approach has demonstrated remarkable results in predicting the functional properties of 3D-printed lattice structures. From an engineering standpoint, this paper introduces a comprehensive and innovative methodology for the functional qualification of 3D-printed parts. By combining the strengths of physical and digital twins with transfer learning, our approach opens up possibilities for the widespread adoption of 3D printing in safety-critical applications. Methodologically, this work presents a significant advancement in transfer learning techniques, specifically addressing the challenges of multi-source (e.g., digital and physical twins) and multi-input (e.g., 3D shapes and process variables) transfer learning.
{"title":"AUDIT: Function<u>a</u>l Q<u>u</u>alification in A<u>d</u>ditive Manufacturing via Physical and Dig<u>i</u>tal <u>T</u>wins","authors":"Michael Biehler, Reinaldo Mock, Shriyanshu Kode, Maham Mehmood, Palin Bhardwaj, Jianjun Shi","doi":"10.1115/1.4063655","DOIUrl":"https://doi.org/10.1115/1.4063655","url":null,"abstract":"Abstract Additive manufacturing (AM) has revolutionized the way we design, prototype, and produce complex parts with unprecedented geometries. However, the lack of understanding of the functional properties of 3D printed parts has hindered their adoption in critical applications where reliability and durability are paramount. This paper proposes a novel approach to the functional qualification of 3D-printed parts via physical and digital twins. Physical twins are parts that are printed under the same process conditions as the functional parts and undergo a wide range of (destructive) tests to determine their mechanical, thermal, and chemical properties. Digital twins are virtual replicas of the physical twins that are generated using finite element analysis (FEA) simulations based on the 3D shape of the part of interest. We propose a novel approach to transfer learning, specifically designed for the fusion of diverse, unstructured 3D shape data and process inputs from multiple sources. The proposed approach has demonstrated remarkable results in predicting the functional properties of 3D-printed lattice structures. From an engineering standpoint, this paper introduces a comprehensive and innovative methodology for the functional qualification of 3D-printed parts. By combining the strengths of physical and digital twins with transfer learning, our approach opens up possibilities for the widespread adoption of 3D printing in safety-critical applications. Methodologically, this work presents a significant advancement in transfer learning techniques, specifically addressing the challenges of multi-source (e.g., digital and physical twins) and multi-input (e.g., 3D shapes and process variables) transfer learning.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135591937","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}
Hemant Agiwal, Hwasung Yeom, Kumar Sridharan, Shiva Rudraraju, Frank E. Pfefferkorn
Abstract The ‘radius of contact’ or the ‘real-rotational contact plane’, has been increasingly mentioned terminology in friction surfacing. However, the fundamental understanding of the flow dynamics behind this phenomenon is still very limited. The goal of this study was to understand the influence of spindle speed and consumable rod diameter on the flow dynamics and radius of contact during friction surfacing of 304L stainless steel over a substrate of the same material. Friction surfacing was performed using consumable rods with diameters of 4.76 mm, 9.52 mm, and 12.7 mm while using spindle speeds from 1,500 RPM to 20,000 RPM. The impact of spindle speed on deposition morphology, including the radius of contact, was studied. The radius of contact was calculated empirically and was found to be inversely proportional to the tangential velocity of the rod. The coupling between flow stresses and localized forces is hypothesized to be the key factor behind the variation of the radius of contact with processing conditions.
{"title":"Radius of Contact During Friction Surfacing of Stainless Steel 304L: Effect of Spindle Speed and Rod Diameter","authors":"Hemant Agiwal, Hwasung Yeom, Kumar Sridharan, Shiva Rudraraju, Frank E. Pfefferkorn","doi":"10.1115/1.4063653","DOIUrl":"https://doi.org/10.1115/1.4063653","url":null,"abstract":"Abstract The ‘radius of contact’ or the ‘real-rotational contact plane’, has been increasingly mentioned terminology in friction surfacing. However, the fundamental understanding of the flow dynamics behind this phenomenon is still very limited. The goal of this study was to understand the influence of spindle speed and consumable rod diameter on the flow dynamics and radius of contact during friction surfacing of 304L stainless steel over a substrate of the same material. Friction surfacing was performed using consumable rods with diameters of 4.76 mm, 9.52 mm, and 12.7 mm while using spindle speeds from 1,500 RPM to 20,000 RPM. The impact of spindle speed on deposition morphology, including the radius of contact, was studied. The radius of contact was calculated empirically and was found to be inversely proportional to the tangential velocity of the rod. The coupling between flow stresses and localized forces is hypothesized to be the key factor behind the variation of the radius of contact with processing conditions.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135549270","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 Residual stresses have been characterized in four Wire Arc Additive Manufacturing specimens with neutron diffraction technique. Firstly, two methods are investigated for obtaining the reference diffracted angle θ0 that is required for the computation of micro-strains and, thus, the stresses. θ0 was obtained using two approaches. The first one required a strain-free specimen in order to get directly the reference diffracted angles θ0 in three directions. The second one is based on the plane stress assumption to get θ0 indirectly by imposing that the normal stress was equal to zero. Both methods led to similar residual stress profiles for the 1-layer specimen what validated this approach for all specimens that did not have a strain-free specimen available. The second part of this work focused on the effect of addition of a new layer on residual stresses. The measurements showed that the longitudinal stress was tensile in the Heat Affected Zone (HAZ) and Fusion Zone (FZ) with a maximum value located at the parent material - layers interface where the thermal loadings were applied. A decrease of this maximum value from 257 MPa to 199 MPa appeared after deposition of a new layer which is due to some stress relaxation effect. Inside the parent material, a large zone presents compressive longitudinal stress up to -170 MPa. The bottom part of the parent material is under tensile stress likely due to its upward bending following the thermal contraction of the deposited layers during cooling to ambient temperature.
{"title":"Effect of layer addition on residual stresses of wire arc additive manufactured stainless steel specimens","authors":"Sebastien Rouquette, Camille Cambon, Issam Bendaoud, Sandra Cabeza, Fabien Soulié","doi":"10.1115/1.4063446","DOIUrl":"https://doi.org/10.1115/1.4063446","url":null,"abstract":"Abstract Residual stresses have been characterized in four Wire Arc Additive Manufacturing specimens with neutron diffraction technique. Firstly, two methods are investigated for obtaining the reference diffracted angle θ0 that is required for the computation of micro-strains and, thus, the stresses. θ0 was obtained using two approaches. The first one required a strain-free specimen in order to get directly the reference diffracted angles θ0 in three directions. The second one is based on the plane stress assumption to get θ0 indirectly by imposing that the normal stress was equal to zero. Both methods led to similar residual stress profiles for the 1-layer specimen what validated this approach for all specimens that did not have a strain-free specimen available. The second part of this work focused on the effect of addition of a new layer on residual stresses. The measurements showed that the longitudinal stress was tensile in the Heat Affected Zone (HAZ) and Fusion Zone (FZ) with a maximum value located at the parent material - layers interface where the thermal loadings were applied. A decrease of this maximum value from 257 MPa to 199 MPa appeared after deposition of a new layer which is due to some stress relaxation effect. Inside the parent material, a large zone presents compressive longitudinal stress up to -170 MPa. The bottom part of the parent material is under tensile stress likely due to its upward bending following the thermal contraction of the deposited layers during cooling to ambient temperature.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135436438","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}
Saravana Sundar A, Krishna Kishore Mugada, Adepu Kumar
The present study explores the application of static shoulder friction stir welding (SSFSW) to address the challenges of poor mechanical properties in conventional Al-Ti dissimilar friction stir joints, which arise due to significant material mixing and the formation of thick intermetallic layers. The results show that SSFSW inhibited material mixing and the mutual diffusion of Al and Ti were suppressed due to lower heat input. Mutual interdiffusion of Al and Ti was directed by an exothermic chemical reaction, forming an Al5Ti2 – Al3Ti sequence due to sluggish diffusion of Al in Ti at a temperature of 512°C achieved in this study. The microstructure at stir zone (SZ) comprised equiaxed grains with Ti particles acting as dispersoids for nucleation, whereas the presence of large Ti blocks at SZ of Conventional FSW (CFSW) resisted plastic deformation, resulting in non-homogeneous concentration of dislocations near its interface. A significant decrease in grain size at all the critical zones of weldment was due to rearrangement of dislocations through slip-and-climb mechanism, as evidenced by the occurrence of dynamic recrystallization. Emergence of γ-fiber and basal fiber texture increased the tensile strength of SSFSW to 289 MPa, which is about 11.2% higher than CFSW, with joint efficiency of about 88%. The study also analysed the contribution of various strengthening mechanisms to the yield strength improvement of SSFSW weldments in detail of SSFSW weldments in detail, and the results showed that grain boundary strengthening contributed the most to strength improvement in SSFSW.
{"title":"Enhancing Microstructural, Textural, and Mechanical Properties of Al-Ti Dissimilar Joints via Static Shoulder Friction Stir Welding","authors":"Saravana Sundar A, Krishna Kishore Mugada, Adepu Kumar","doi":"10.1115/1.4063358","DOIUrl":"https://doi.org/10.1115/1.4063358","url":null,"abstract":"\u0000 The present study explores the application of static shoulder friction stir welding (SSFSW) to address the challenges of poor mechanical properties in conventional Al-Ti dissimilar friction stir joints, which arise due to significant material mixing and the formation of thick intermetallic layers. The results show that SSFSW inhibited material mixing and the mutual diffusion of Al and Ti were suppressed due to lower heat input. Mutual interdiffusion of Al and Ti was directed by an exothermic chemical reaction, forming an Al5Ti2 – Al3Ti sequence due to sluggish diffusion of Al in Ti at a temperature of 512°C achieved in this study. The microstructure at stir zone (SZ) comprised equiaxed grains with Ti particles acting as dispersoids for nucleation, whereas the presence of large Ti blocks at SZ of Conventional FSW (CFSW) resisted plastic deformation, resulting in non-homogeneous concentration of dislocations near its interface. A significant decrease in grain size at all the critical zones of weldment was due to rearrangement of dislocations through slip-and-climb mechanism, as evidenced by the occurrence of dynamic recrystallization. Emergence of γ-fiber and basal fiber texture increased the tensile strength of SSFSW to 289 MPa, which is about 11.2% higher than CFSW, with joint efficiency of about 88%. The study also analysed the contribution of various strengthening mechanisms to the yield strength improvement of SSFSW weldments in detail of SSFSW weldments in detail, and the results showed that grain boundary strengthening contributed the most to strength improvement in SSFSW.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42067640","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}
Pengjie Gao, Junliang Wang, Min Xia, Zijin Qin, Jie Zhang
As an important application of human-robot collaboration, intelligent detection of surface defects is crucial for production quality control, which also helps in relieving the workload of technical staff in human-centric smart manufacturing. To accurately detect defects with limited samples in industrial practice, a dual-metric neural network with attention guided is proposed. First, an attention-guided recognition network with channel attention and position attention module is designed to efficiently learn representative defect features with limited samples. Second, aiming to detect defects with confusing surface images, a dual-metric function is presented to learn the classification boundary by controlling the distance of samples in feature space from intra-class and inter-class. The experiment results on the fabric defect dataset demonstrate that the proposed approach outperforms state-of-the-art methods in accuracy, recall, precision, F1-score, and few-shot accuracy. Further comparative experiments reveal that the dual-metric function is superior in improving the few-shot detection accuracy for the defect patterns of fabric.
{"title":"Dual-Metric Neural Network with Attention Guidance for Surface Defect Few-Shot Detection in Smart Manufacturing","authors":"Pengjie Gao, Junliang Wang, Min Xia, Zijin Qin, Jie Zhang","doi":"10.1115/1.4063356","DOIUrl":"https://doi.org/10.1115/1.4063356","url":null,"abstract":"\u0000 As an important application of human-robot collaboration, intelligent detection of surface defects is crucial for production quality control, which also helps in relieving the workload of technical staff in human-centric smart manufacturing. To accurately detect defects with limited samples in industrial practice, a dual-metric neural network with attention guided is proposed. First, an attention-guided recognition network with channel attention and position attention module is designed to efficiently learn representative defect features with limited samples. Second, aiming to detect defects with confusing surface images, a dual-metric function is presented to learn the classification boundary by controlling the distance of samples in feature space from intra-class and inter-class. The experiment results on the fabric defect dataset demonstrate that the proposed approach outperforms state-of-the-art methods in accuracy, recall, precision, F1-score, and few-shot accuracy. Further comparative experiments reveal that the dual-metric function is superior in improving the few-shot detection accuracy for the defect patterns of fabric.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46460502","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 five-axis ball-end milling dynamics of thin-walled blades is presented. The cutting forces are predicted from the ball end mill–blade geometry engagement maps along the tool path. The Frequency Response Function (FRF) of the thin-walled blade is predicted using Finite Element shell elements, and it is updated along the toolpath as the metal is removed. The predicted cutting forces are applied on both the workpiece and tool FRFs to predict the forced vibrations and chatter stability at each tool location. A simplified method to update the cutter–workpiece engagement (CWE) is used to obtain the three-dimensional stability lobe diagram at each desired point on the blade. The integrated model is used to simulate the 5-axis machining of thin-walled blades in the digital environment. The proposed digital model is experimentally validated by machining a series of thin-walled rectangular plates and a twisted fan blade.
{"title":"Virtual Dynamics Model for 5-axis Machining of Thin-Walled Blades","authors":"B. Karimi, Y. Altintas","doi":"10.1115/1.4063286","DOIUrl":"https://doi.org/10.1115/1.4063286","url":null,"abstract":"\u0000 The five-axis ball-end milling dynamics of thin-walled blades is presented. The cutting forces are predicted from the ball end mill–blade geometry engagement maps along the tool path. The Frequency Response Function (FRF) of the thin-walled blade is predicted using Finite Element shell elements, and it is updated along the toolpath as the metal is removed. The predicted cutting forces are applied on both the workpiece and tool FRFs to predict the forced vibrations and chatter stability at each tool location. A simplified method to update the cutter–workpiece engagement (CWE) is used to obtain the three-dimensional stability lobe diagram at each desired point on the blade. The integrated model is used to simulate the 5-axis machining of thin-walled blades in the digital environment. The proposed digital model is experimentally validated by machining a series of thin-walled rectangular plates and a twisted fan blade.","PeriodicalId":16299,"journal":{"name":"Journal of Manufacturing Science and Engineering-transactions of The Asme","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2023-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41382106","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 manufacturing industry is currently facing an increasing demand for customized products, leading to a shift from mass production to mass customization. As a result, operators are required to produce multiple product variants with varying complexity levels while maintaining high-quality standards. Further, in line with the human-centered paradigm of Industry 5.0, ensuring the well-being of workers is equally important as production quality. This paper proposes a novel tool, the “Human-Robot Collaboration Quality and Well-Being Assessment Tool” (HRC-QWAT), which combines the analysis of overall defects generated during product variant manufacturing with the evaluation of human well-being in terms of stress response. The HRC-QWAT enables the evaluation and monitoring of human-robot collaboration systems during product variant production from a broader standpoint. A case study of collaborative human-robot assembly is used to demonstrate the applicability of the proposed approach. The results suggest that the HRC-QWAT can evaluate both production quality and human well-being, providing a useful tool for companies to monitor and improve their manufacturing processes. Overall, this paper contributes to developing a human-centric approach to quality monitoring in the context of human-robot collaborative manufacturing.
{"title":"A Novel Diagnostic Tool for Human-Centric Quality Monitoring in Human-Robot Collaboration Manufacturing","authors":"E. Verna, Stefano Puttero, G. Genta, M. Galetto","doi":"10.1115/1.4063284","DOIUrl":"https://doi.org/10.1115/1.4063284","url":null,"abstract":"\u0000 The manufacturing industry is currently facing an increasing demand for customized products, leading to a shift from mass production to mass customization. As a result, operators are required to produce multiple product variants with varying complexity levels while maintaining high-quality standards. Further, in line with the human-centered paradigm of Industry 5.0, ensuring the well-being of workers is equally important as production quality. This paper proposes a novel tool, the “Human-Robot Collaboration Quality and Well-Being Assessment Tool” (HRC-QWAT), which combines the analysis of overall defects generated during product variant manufacturing with the evaluation of human well-being in terms of stress response. The HRC-QWAT enables the evaluation and monitoring of human-robot collaboration systems during product variant production from a broader standpoint. A case study of collaborative human-robot assembly is used to demonstrate the applicability of the proposed approach. The results suggest that the HRC-QWAT can evaluate both production quality and human well-being, providing a useful tool for companies to monitor and improve their manufacturing processes. Overall, this paper contributes to developing a human-centric approach to quality monitoring in the context of human-robot collaborative manufacturing.","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":"49147762","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：