João Sousa, R. Darabi, A. Reis, Marco Parente, L. Reis, J. C. de Sá
During the last decades, metal additive manufacturing (AM) technology has transitioned from rapid prototyping application to industrial adoption owing to its flexibility in product design, tooling, and process planning. Thus, understanding the behavior, interaction, and influence of the involved processing parameters on the overall AM production system in order to obtain high-quality parts and stabilized manufacturing process is crucial. Despite many advantages of the AM technologies, difficulties arise due to modelling the complex nature of the process-structure-property relations, which prevents its wide utilization in various industrial sectors. It is known that many of the most important defects in direct energy deposition (DED) are associated with the volume and timescales of the evolving melt pool. Thus, the development of methodologies for monitoring, and controlling the melt pool is critical. In this study, an adaptive numerical transient solution is developed, which is fed from the set of experiments for single-track scanning of super-alloy Inconel 625 on the hot-tempered steel type 42CrMo4. An established exponential formula based on the response surface methodology (RSM) that quantifies the influence of process parameters and geometries of deposited layers from experiments are considered to activate the volume fraction of passive elements in the finite element discretization. By resorting to the FORTRAN language framework capabilities, commercial finite element method software ABAQUS has been steered in order to control unfavorable defects induced by localized rapid heating and cooling, and unstable volume of the melt pool. A thermodynamic consistent phase-field model is coupled with a transient thermal simulation to track the material history. A Lagrangian description for the spatial and time discretization is used. The goal is to present a closed-loop approach to track the melt pool morphology and temperature to a reference deposition volume profile which is established based on deep reinforcement learning (RL) architecture aiming to avoid instabilities, defects and anomalies by controlling the laser power density adaptability. Despite the small number of iterations during RL model training, the agent was able to learn the desired behaviour and two different reward functions were evaluated. This approach allows us to show the possibility of using RL with openAI Gym for process control and its interconnection with ABAQUS framework to train a model first in a simulation environment, and thus take advantage of RL capabilities without creating waste or machine time in real-world.
{"title":"An Adaptive Thermal Finite Element Simulation of Direct Energy Deposition With Reinforcement Learning: A Conceptual Framework","authors":"João Sousa, R. Darabi, A. Reis, Marco Parente, L. Reis, J. C. de Sá","doi":"10.1115/imece2022-95055","DOIUrl":"https://doi.org/10.1115/imece2022-95055","url":null,"abstract":"\u0000 During the last decades, metal additive manufacturing (AM) technology has transitioned from rapid prototyping application to industrial adoption owing to its flexibility in product design, tooling, and process planning. Thus, understanding the behavior, interaction, and influence of the involved processing parameters on the overall AM production system in order to obtain high-quality parts and stabilized manufacturing process is crucial. Despite many advantages of the AM technologies, difficulties arise due to modelling the complex nature of the process-structure-property relations, which prevents its wide utilization in various industrial sectors. It is known that many of the most important defects in direct energy deposition (DED) are associated with the volume and timescales of the evolving melt pool. Thus, the development of methodologies for monitoring, and controlling the melt pool is critical. In this study, an adaptive numerical transient solution is developed, which is fed from the set of experiments for single-track scanning of super-alloy Inconel 625 on the hot-tempered steel type 42CrMo4. An established exponential formula based on the response surface methodology (RSM) that quantifies the influence of process parameters and geometries of deposited layers from experiments are considered to activate the volume fraction of passive elements in the finite element discretization. By resorting to the FORTRAN language framework capabilities, commercial finite element method software ABAQUS has been steered in order to control unfavorable defects induced by localized rapid heating and cooling, and unstable volume of the melt pool. A thermodynamic consistent phase-field model is coupled with a transient thermal simulation to track the material history. A Lagrangian description for the spatial and time discretization is used. The goal is to present a closed-loop approach to track the melt pool morphology and temperature to a reference deposition volume profile which is established based on deep reinforcement learning (RL) architecture aiming to avoid instabilities, defects and anomalies by controlling the laser power density adaptability. Despite the small number of iterations during RL model training, the agent was able to learn the desired behaviour and two different reward functions were evaluated. This approach allows us to show the possibility of using RL with openAI Gym for process control and its interconnection with ABAQUS framework to train a model first in a simulation environment, and thus take advantage of RL capabilities without creating waste or machine time in real-world.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"231 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125699779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Roham Sadeghi Tabar, B. Lindau, L. Lindkvist, Kristina Wärmefjord, R. Söderberg
Geometric variation is one of the causes of aesthetic and functional issues in mechanical assemblies. To predict the geometric variation in assemblies of rigid and non-rigid parts, statistical variation simulation is introduced. For non-rigid parts, bending and deformation occur during the assembly process. In non-rigid variation simulation, contact modeling is utilized to avoid the virtual penetration of the components in the adjacent areas. Contact modeling imposes non-linear behavior to the MIC approach for variation simulation, and thereby the problem complexity and simulation time increase. Traditionally, iterative node search is used to identify and define the computational contact nodes. However, iterative search is time-demanding, specifically in large-scale models, as the search space increases by the number of nodes included in the assembly. To allow for faster contact search, a data structuring method using Kd-trees and nearest neighbor search (NN) is implemented and integrated into a computer aided tolerancing tool, enhancing the search functionality and reducing the search time compared to iterative one-by-one node search. The method is applied to three reference assemblies of different size, and the identified contact nodes and the time needed to perform the search is compared to an iterative node search. The results show that the K-tree structure and nearest neighbor search perform considerably, 96%, faster than the iterative node search. The method increases the search performance, while the identified contact points are similar to the ones identified by an iterative search. The approach efficiently enables the contact search of large models and reduces the modeling time required for non-rigid variation simulation.
{"title":"Contact Search Using a Kd-Tree for Non-Rigid Variation Simulation","authors":"Roham Sadeghi Tabar, B. Lindau, L. Lindkvist, Kristina Wärmefjord, R. Söderberg","doi":"10.1115/imece2022-94989","DOIUrl":"https://doi.org/10.1115/imece2022-94989","url":null,"abstract":"\u0000 Geometric variation is one of the causes of aesthetic and functional issues in mechanical assemblies. To predict the geometric variation in assemblies of rigid and non-rigid parts, statistical variation simulation is introduced. For non-rigid parts, bending and deformation occur during the assembly process. In non-rigid variation simulation, contact modeling is utilized to avoid the virtual penetration of the components in the adjacent areas. Contact modeling imposes non-linear behavior to the MIC approach for variation simulation, and thereby the problem complexity and simulation time increase. Traditionally, iterative node search is used to identify and define the computational contact nodes. However, iterative search is time-demanding, specifically in large-scale models, as the search space increases by the number of nodes included in the assembly. To allow for faster contact search, a data structuring method using Kd-trees and nearest neighbor search (NN) is implemented and integrated into a computer aided tolerancing tool, enhancing the search functionality and reducing the search time compared to iterative one-by-one node search. The method is applied to three reference assemblies of different size, and the identified contact nodes and the time needed to perform the search is compared to an iterative node search. The results show that the K-tree structure and nearest neighbor search perform considerably, 96%, faster than the iterative node search. The method increases the search performance, while the identified contact points are similar to the ones identified by an iterative search. The approach efficiently enables the contact search of large models and reduces the modeling time required for non-rigid variation simulation.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130822937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To address the difficulties faced by the existing 3-axis incremental sheet forming (ISF) processes, toolpaths of 5-axis incremental sheet forming were developed to form the components with features having forming angles greater than 90°. This paper compares the STL and point cloud-based methodologies for the toolpaths generated for 5 axes incremental sheet forming. The STL models are used in the existing 5-axis toolpath techniques to determine the point normal at a contact point and hence the tool posture angles. However, there are several drawbacks associated with this approach of using STL models. Thus, a point cloud-based posture calculation strategy is proposed in this work. It was observed that the proposed approach performs better than the STL-based strategy in terms of computing efficiency. Numerical simulations were performed to validate the approach utilizing the toolpaths generated by both strategies. The simulation of the ISF process was performed to determine the feasibility of the 5-axis incremental sheet forming toolpath. Finite element simulations were performed using the shell elements for the geometry with the curvature of the wall greater than 90°. Results based on geometrical accuracy are compared to understand the differences between the two strategies.
{"title":"A Numerical Investigation to Compare Point Cloud and STL-Based Toolpath Strategies for 5-Axis Incremental Sheet Forming","authors":"Ayushi Gupta, A. Nagargoje, A. Dubey, P. Tandon","doi":"10.1115/imece2022-94589","DOIUrl":"https://doi.org/10.1115/imece2022-94589","url":null,"abstract":"\u0000 To address the difficulties faced by the existing 3-axis incremental sheet forming (ISF) processes, toolpaths of 5-axis incremental sheet forming were developed to form the components with features having forming angles greater than 90°. This paper compares the STL and point cloud-based methodologies for the toolpaths generated for 5 axes incremental sheet forming. The STL models are used in the existing 5-axis toolpath techniques to determine the point normal at a contact point and hence the tool posture angles. However, there are several drawbacks associated with this approach of using STL models. Thus, a point cloud-based posture calculation strategy is proposed in this work. It was observed that the proposed approach performs better than the STL-based strategy in terms of computing efficiency. Numerical simulations were performed to validate the approach utilizing the toolpaths generated by both strategies. The simulation of the ISF process was performed to determine the feasibility of the 5-axis incremental sheet forming toolpath. Finite element simulations were performed using the shell elements for the geometry with the curvature of the wall greater than 90°. Results based on geometrical accuracy are compared to understand the differences between the two strategies.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132059753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anwar Q. Al-Gamal, Muhammad Ali Ablat, Lakshmi Ramineni, Majed Ali, Abdalmageed Almotari, A. Alafaghani, Jian-Qiao Sun, A. Qattawi
The sustainability of sheet metal parts often has multiple facets depending on the phase under consideration. The work presented in this paper focuses on cradle-to-gate Life Cycle Analysis (LCA) of the Origami-based Sheet Metal (OSM) folding process. OSM is an emerging fabrication technique that utilizes the principle of folding sheet metal parts by creating Material Discontinuities (MD) along the bend line. MD enables sheet metal folding (i.e., bending) with minimal force requirements and machinery. The anticipated reduction in force and machinery will result in a reduction in the required manufacturing energy. In addition, the OSM has less dependency on dies and shape-dedicated equipment. Hence, the cost associated with sheet metal parts development is reduced. This study attempts to establish the environmental impacts of the OSM for sheet metal parts by utilizing cradle-to-gate life cycle analysis. Environmental impacts of OSM are highlighted by comparing the OSM with the conventional stamping process. In the LCA, consumed energy and emissions are considered environmental impact indicators. Energy and emissions data are collected from published literature, machinery manuals, and available empirical models for energy consumption. A case study of a vehicle floor panel is presented as an example. Finite element analysis (FEA) is employed to achieve a more accurate energy estimation since the LCA inventory data displays a significant discrepancy. The findings of this study reveal that OSM requires less energy and produces fewer emissions than the stamping process.
{"title":"Cradle-to-Gate Life Cycle Analysis of Origami-Based Sheet Metal for Automobile Parts","authors":"Anwar Q. Al-Gamal, Muhammad Ali Ablat, Lakshmi Ramineni, Majed Ali, Abdalmageed Almotari, A. Alafaghani, Jian-Qiao Sun, A. Qattawi","doi":"10.1115/imece2022-96922","DOIUrl":"https://doi.org/10.1115/imece2022-96922","url":null,"abstract":"\u0000 The sustainability of sheet metal parts often has multiple facets depending on the phase under consideration. The work presented in this paper focuses on cradle-to-gate Life Cycle Analysis (LCA) of the Origami-based Sheet Metal (OSM) folding process. OSM is an emerging fabrication technique that utilizes the principle of folding sheet metal parts by creating Material Discontinuities (MD) along the bend line. MD enables sheet metal folding (i.e., bending) with minimal force requirements and machinery. The anticipated reduction in force and machinery will result in a reduction in the required manufacturing energy. In addition, the OSM has less dependency on dies and shape-dedicated equipment. Hence, the cost associated with sheet metal parts development is reduced.\u0000 This study attempts to establish the environmental impacts of the OSM for sheet metal parts by utilizing cradle-to-gate life cycle analysis. Environmental impacts of OSM are highlighted by comparing the OSM with the conventional stamping process. In the LCA, consumed energy and emissions are considered environmental impact indicators. Energy and emissions data are collected from published literature, machinery manuals, and available empirical models for energy consumption. A case study of a vehicle floor panel is presented as an example. Finite element analysis (FEA) is employed to achieve a more accurate energy estimation since the LCA inventory data displays a significant discrepancy. The findings of this study reveal that OSM requires less energy and produces fewer emissions than the stamping process.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116224507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Pokkalla, A. Hassen, J. Heineman, Thomas Snape, J. Arimond, V. Kunc, Seokpum Kim
Autoclave processing is a commonly used state-of-the-art fiber-reinforced composite manufacturing technology, albeit with high capital cost, long cycle times and high energy consumption. Alternatively, out-of-autoclave processing reduces the initial and operating costs while producing composite structures with similar quality as that of autoclave parts. Additive Manufacturing (AM) the scaled-up molds for out-of-autoclave process using carbon fiber (CF) reinforced composite offers design flexibility, enhanced mechanical, and thermal properties in addition to reduction in weight and cost. However, heating of these molds using an oven is still expensive and necessitates an energy-efficient heating process. In this study, resistive heating through heating elements embedded within fiber reinforced composite molds is used as an efficient heating mechanism. The goal is to design wire embeddings and determine the optimal heat flux density to achieve a target uniform temperature of 80°C across the mold surface. To this end, numerical analyses were performed to evaluate the temperature distribution across the composite mold surface for a given wire placement and mold configuration. Constant thermal properties of the 20 wt.% short CF reinforced acrylonitrile butadiene styrene (ABS) were used in the thermal analysis. Time taken to reach the steady state temperature was also estimated. Design guidelines for wire embeddings were included to enable efficient manufacturing of fiber-reinforced composites through out-of-autoclave molds.
{"title":"Thermal Analysis and Design of Self-Heating Molds Using Large-Scale Additive Manufacturing for Out-of-Autoclave Applications","authors":"D. Pokkalla, A. Hassen, J. Heineman, Thomas Snape, J. Arimond, V. Kunc, Seokpum Kim","doi":"10.1115/imece2022-95790","DOIUrl":"https://doi.org/10.1115/imece2022-95790","url":null,"abstract":"\u0000 Autoclave processing is a commonly used state-of-the-art fiber-reinforced composite manufacturing technology, albeit with high capital cost, long cycle times and high energy consumption. Alternatively, out-of-autoclave processing reduces the initial and operating costs while producing composite structures with similar quality as that of autoclave parts. Additive Manufacturing (AM) the scaled-up molds for out-of-autoclave process using carbon fiber (CF) reinforced composite offers design flexibility, enhanced mechanical, and thermal properties in addition to reduction in weight and cost. However, heating of these molds using an oven is still expensive and necessitates an energy-efficient heating process. In this study, resistive heating through heating elements embedded within fiber reinforced composite molds is used as an efficient heating mechanism. The goal is to design wire embeddings and determine the optimal heat flux density to achieve a target uniform temperature of 80°C across the mold surface. To this end, numerical analyses were performed to evaluate the temperature distribution across the composite mold surface for a given wire placement and mold configuration. Constant thermal properties of the 20 wt.% short CF reinforced acrylonitrile butadiene styrene (ABS) were used in the thermal analysis. Time taken to reach the steady state temperature was also estimated. Design guidelines for wire embeddings were included to enable efficient manufacturing of fiber-reinforced composites through out-of-autoclave molds.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"199 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114619922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rashid Ali Laghari, S. Mekid, S. S. Akhtar, A. Laghari, Muhammad Jamil
In this paper, the face milling experiments were performed to investigate the cutting process of SiCp/Al (SiCp 65%) volume percentage and their effect on tool life and tool wear mechanism. The study was performed based on different cutting parameters (cutting speed vc, feed per tooth fz, and axial depth of cut ap, 1mm, and width of cut ae 8mm,) and cutting environments adopted as Dry MQL and CO2 Snow to analyze the effect of lubri-cooling machining process using polycrystalline diamond (PCD) cutting tools. A total of 18 experiments were performed during milling of SiCp/Al (65%) with each experimental run involved the 321 mm3 of the volume of material removal. The study found that lubrication and cooling can effectively reduce the tool wear and improve the tool life up to 29% for SiCp65% on moderate cutting parameters. The major wear mechanisms of PCD cutting tools are perceived as abrasive wear and adhesive wear mechanisms, which develop the flank wear and build-up-edge formations. Seemingly, a sub-zero cooling environment and dry cutting are found convenient to produce a build-up edge on the rake face of the cutting tool while MQL-assisted machining provides the benefit to prevent the cutting tool from material adhesion.
{"title":"Investigating the Tribological Aspects of Tool Wear Mechanism and Tool Life in Sustainable Lubri-Cooling Face Milling Process of Particle Reinforced SiCp/Al Metal Matrix Composites","authors":"Rashid Ali Laghari, S. Mekid, S. S. Akhtar, A. Laghari, Muhammad Jamil","doi":"10.1115/imece2022-95183","DOIUrl":"https://doi.org/10.1115/imece2022-95183","url":null,"abstract":"\u0000 In this paper, the face milling experiments were performed to investigate the cutting process of SiCp/Al (SiCp 65%) volume percentage and their effect on tool life and tool wear mechanism. The study was performed based on different cutting parameters (cutting speed vc, feed per tooth fz, and axial depth of cut ap, 1mm, and width of cut ae 8mm,) and cutting environments adopted as Dry MQL and CO2 Snow to analyze the effect of lubri-cooling machining process using polycrystalline diamond (PCD) cutting tools. A total of 18 experiments were performed during milling of SiCp/Al (65%) with each experimental run involved the 321 mm3 of the volume of material removal. The study found that lubrication and cooling can effectively reduce the tool wear and improve the tool life up to 29% for SiCp65% on moderate cutting parameters. The major wear mechanisms of PCD cutting tools are perceived as abrasive wear and adhesive wear mechanisms, which develop the flank wear and build-up-edge formations. Seemingly, a sub-zero cooling environment and dry cutting are found convenient to produce a build-up edge on the rake face of the cutting tool while MQL-assisted machining provides the benefit to prevent the cutting tool from material adhesion.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133153318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Micro-, and milli-scale robots have been of great R&D interest, due to their ability to accomplish difficult tasks such as minimally invasive diagnosis and treatment for human bodies, and underground or deep-sea tests for environment monitoring. A good solution to this design need is a multi-unit deployable tensegrity microrobot. The microrobot can be folded to only 15% of its deployed length, so as to easily enter a desired working area with a small entrance. When deployed, the tensegrity body of the robot displays lightweight and high stiffness to sustain loads and prevent damage when burrowing through tightly packed tissues or high-pressure environments. In this work, topology, initial configuration and locomotion of a deployable tensegrity microrobot are determined optimally. Based on the design, a centimeter-scale prototype is manufactured by using a fused deposition modelling advanced additive manufacturing or 3-D printing system for proof of concept. As shown in experimental results, the deployable tensegrity microrobot prototype designed and manufactured can achieve an extremely high folding ratio, while be lightweight and rigid. The locomotion design, that mimics a crawling motion of an earthworm, is proved to be efficient by the prototype equipped with stepper motors, actuation cables, control boards and a braking system.
{"title":"Prototype Design and Manufacture of a Deployable Tensegrity Microrobot","authors":"Christian Kazoleas, Kaushik Mehta, S. Yuan","doi":"10.1115/imece2022-93929","DOIUrl":"https://doi.org/10.1115/imece2022-93929","url":null,"abstract":"\u0000 Micro-, and milli-scale robots have been of great R&D interest, due to their ability to accomplish difficult tasks such as minimally invasive diagnosis and treatment for human bodies, and underground or deep-sea tests for environment monitoring. A good solution to this design need is a multi-unit deployable tensegrity microrobot. The microrobot can be folded to only 15% of its deployed length, so as to easily enter a desired working area with a small entrance. When deployed, the tensegrity body of the robot displays lightweight and high stiffness to sustain loads and prevent damage when burrowing through tightly packed tissues or high-pressure environments. In this work, topology, initial configuration and locomotion of a deployable tensegrity microrobot are determined optimally. Based on the design, a centimeter-scale prototype is manufactured by using a fused deposition modelling advanced additive manufacturing or 3-D printing system for proof of concept. As shown in experimental results, the deployable tensegrity microrobot prototype designed and manufactured can achieve an extremely high folding ratio, while be lightweight and rigid. The locomotion design, that mimics a crawling motion of an earthworm, is proved to be efficient by the prototype equipped with stepper motors, actuation cables, control boards and a braking system.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129570123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Variation management is a responsible task for product developers, which have to balance the ever-increasing quality demands and cost pressures, while considering the product design as well as the manufacturing and assembly process. These aspects have a direct impact on its subsequent success in the market. Therefore, a large number of different activities of variation management are necessary. In this area, a wide variety of mostly document-centered methods support product developers, which have individual interfaces. Thus, it is currently not possible to map the entire variation management process in a single model. Especially with regard to the increasing availability of large amounts of data, the potential of an integrated variation management cannot be exploited efficiently. For this reason, this paper presents a novel model-based approach for the development of a combined system and tolerancing model. This model contains the processes and activities of integrated variation management and links them with further system models and the corresponding data. The presented approach is a superordinate model for variation management as well as its processes and provides the modeling of individual views of different stakeholders. In addition, process- and program-specific solutions can be integrated into the model, which enables a cross-linking of the data beyond their interfaces. In this paper, the approach is realized using the systems modeling language (SysML).
{"title":"A Model-Based Approach for Integrated Variation Management","authors":"D. Horber, Stefan Götz, B. Schleich, S. Wartzack","doi":"10.1115/imece2022-90956","DOIUrl":"https://doi.org/10.1115/imece2022-90956","url":null,"abstract":"\u0000 Variation management is a responsible task for product developers, which have to balance the ever-increasing quality demands and cost pressures, while considering the product design as well as the manufacturing and assembly process. These aspects have a direct impact on its subsequent success in the market. Therefore, a large number of different activities of variation management are necessary. In this area, a wide variety of mostly document-centered methods support product developers, which have individual interfaces. Thus, it is currently not possible to map the entire variation management process in a single model. Especially with regard to the increasing availability of large amounts of data, the potential of an integrated variation management cannot be exploited efficiently. For this reason, this paper presents a novel model-based approach for the development of a combined system and tolerancing model. This model contains the processes and activities of integrated variation management and links them with further system models and the corresponding data. The presented approach is a superordinate model for variation management as well as its processes and provides the modeling of individual views of different stakeholders. In addition, process- and program-specific solutions can be integrated into the model, which enables a cross-linking of the data beyond their interfaces. In this paper, the approach is realized using the systems modeling language (SysML).","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127627471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The field of mechanical engineering is evolving with latest technologies such as artificial intelligence. the blend of AI technologies such as deep convolutional neural network (DCNN), convolutional neural network (CNN), artificial neural network (ANN) which contributes more to control the process parameters, process planning, machining, quality control and optimization for a better product or system. The implementation of AI in mechanical engineering applications results in minimizing the rejection of machine components which helps the whole process to be economical with better quality outputs. Considering the stiff competition among the manufacturers in the market, increasing the production rate while maintaining stringent quality control is a big challenge. In this perspective, artificial intelligence is gaining popularity in production lines to maintain a high quality for the products. A CNN is a deep learning algorithm, that is analogous to that the connectivity pattern of neurons in the human brain, has become popular and effective to image classification problems recently. It takes in the image of the object and assigns importance to various aspects/objects in the image so as to differentiate one from the other. In fruit-sorting process, manual classification is time-consuming, expensive, and requires experienced experts whose availability is often limited. To address these issues, various machine learning algorithms have been proposed to support the automated classification of fruits. In this paper, to classify “regular apples” and “damaged apples”, deep learning algorithm is applied. The pre-trained, deep learning models namely, VGG 16, ResNet50, Inceptionv3, Mobilenet_v2 along with a basic sequential convolutional model are applied to differentiate the damaged apples from regular ones and their performance variation is also analyzed. For this work, the data set containing damaged and regular apples was garnered from various local stores and farms. The data set consisted of 400 color images of both regular and damaged apples. Though the number of samples is smaller, the above-mentioned deep learning models demonstrated to overcome this deficit. For the training of model, 80% of the total sample (280) images were utilized while 20% and 10% of the sample (80 & 40) were applied for the validation and testing the model. The results show more than 90% accuracy for all the models except ResNet 50. The performance of these models can be improved even further by increasing the size of data set by adding more fruit images through better training of the models. Our experimental study demonstrates the application of artificial intelligence through four different transfer learning techniques works well for deep neural network-based fruit classification. It minimizes the labor and human errors involved in the fruit-sorting process which results in saving money and time.
{"title":"Damaged Apple Detection Using Artificial Intelligence","authors":"S. Gurupatham, Caleb Bailey","doi":"10.1115/imece2022-96162","DOIUrl":"https://doi.org/10.1115/imece2022-96162","url":null,"abstract":"\u0000 The field of mechanical engineering is evolving with latest technologies such as artificial intelligence. the blend of AI technologies such as deep convolutional neural network (DCNN), convolutional neural network (CNN), artificial neural network (ANN) which contributes more to control the process parameters, process planning, machining, quality control and optimization for a better product or system. The implementation of AI in mechanical engineering applications results in minimizing the rejection of machine components which helps the whole process to be economical with better quality outputs. Considering the stiff competition among the manufacturers in the market, increasing the production rate while maintaining stringent quality control is a big challenge. In this perspective, artificial intelligence is gaining popularity in production lines to maintain a high quality for the products. A CNN is a deep learning algorithm, that is analogous to that the connectivity pattern of neurons in the human brain, has become popular and effective to image classification problems recently. It takes in the image of the object and assigns importance to various aspects/objects in the image so as to differentiate one from the other. In fruit-sorting process, manual classification is time-consuming, expensive, and requires experienced experts whose availability is often limited. To address these issues, various machine learning algorithms have been proposed to support the automated classification of fruits. In this paper, to classify “regular apples” and “damaged apples”, deep learning algorithm is applied. The pre-trained, deep learning models namely, VGG 16, ResNet50, Inceptionv3, Mobilenet_v2 along with a basic sequential convolutional model are applied to differentiate the damaged apples from regular ones and their performance variation is also analyzed. For this work, the data set containing damaged and regular apples was garnered from various local stores and farms. The data set consisted of 400 color images of both regular and damaged apples. Though the number of samples is smaller, the above-mentioned deep learning models demonstrated to overcome this deficit. For the training of model, 80% of the total sample (280) images were utilized while 20% and 10% of the sample (80 & 40) were applied for the validation and testing the model. The results show more than 90% accuracy for all the models except ResNet 50. The performance of these models can be improved even further by increasing the size of data set by adding more fruit images through better training of the models. Our experimental study demonstrates the application of artificial intelligence through four different transfer learning techniques works well for deep neural network-based fruit classification. It minimizes the labor and human errors involved in the fruit-sorting process which results in saving money and time.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123337793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fused Filament Fabrication (FFF) is an extrusion-based additive manufacturing process that utilizes a filament material melted through a hot end extruder to generate a component. Despite the great potential of the process to drastically reduce time-to-produce, cost and material waste for the creation of geometrically complex components, the presence of diverse defects deteriorate the quality of the final build. Defects in FFF (e.g., voids, stringing, and varying track width) are primarily linked to improper calibration of parameters, including feed speed, extrusion speed, extruder temperature, and build plate temperature. Trial and error is the most common practice implemented to manually offset baseline parameters using an array of components generated with varying process parameters. However, fabrication with manual adjustment not only is time consuming, but also leads to a suboptimal solution that jeopardizes the strength and integrity of the generated components. We propose a novel Bayesian Optimization (BO) methodology in conjunction with heterogeneous sensing to determine optimal process parameters with a minimum number of experiments. BO consists of two steps: First, a Gaussian Process as a surrogate model that maps the relationship between controllable parameters (e.g., feed rate/flow rate ratio, extrusion temperature, and layer height) and build quality (i.e., the objective function that is derived from sensing data). Second, an acquisition function is defined from this surrogate to decide where to sample. We design build quality characterization model that formulated as an objective-scoring algorithm that returns the proportion of the effective specimen sensor measurements divided by the desired values. The experimental results on real-world case study shows that the proposed BO is capable of determining the values for parameters in just 7 steps with quality improvement of 0.036 from the best trial quality.
{"title":"Heterogeneous Sensing and Bayesian Optimization for Smart Calibration in Additive Manufacturing Process","authors":"Sean Rescsanski, Mahdi Imani, Farhad Imani","doi":"10.1115/imece2022-96010","DOIUrl":"https://doi.org/10.1115/imece2022-96010","url":null,"abstract":"\u0000 Fused Filament Fabrication (FFF) is an extrusion-based additive manufacturing process that utilizes a filament material melted through a hot end extruder to generate a component. Despite the great potential of the process to drastically reduce time-to-produce, cost and material waste for the creation of geometrically complex components, the presence of diverse defects deteriorate the quality of the final build. Defects in FFF (e.g., voids, stringing, and varying track width) are primarily linked to improper calibration of parameters, including feed speed, extrusion speed, extruder temperature, and build plate temperature. Trial and error is the most common practice implemented to manually offset baseline parameters using an array of components generated with varying process parameters. However, fabrication with manual adjustment not only is time consuming, but also leads to a suboptimal solution that jeopardizes the strength and integrity of the generated components. We propose a novel Bayesian Optimization (BO) methodology in conjunction with heterogeneous sensing to determine optimal process parameters with a minimum number of experiments. BO consists of two steps: First, a Gaussian Process as a surrogate model that maps the relationship between controllable parameters (e.g., feed rate/flow rate ratio, extrusion temperature, and layer height) and build quality (i.e., the objective function that is derived from sensing data). Second, an acquisition function is defined from this surrogate to decide where to sample. We design build quality characterization model that formulated as an objective-scoring algorithm that returns the proportion of the effective specimen sensor measurements divided by the desired values. The experimental results on real-world case study shows that the proposed BO is capable of determining the values for parameters in just 7 steps with quality improvement of 0.036 from the best trial quality.","PeriodicalId":113474,"journal":{"name":"Volume 2B: Advanced Manufacturing","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115483272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：