Robot assistance can improve the outcome of microsurgery by scaling down the surgeon's hand motions. However, the high cost of surgical robots has prevented their use in small hospitals or medical facilities in several developing countries. As a novel alternative, a fully mechanical motion-scaling instrument, which can be operated without computers and motors, was proposed based on the pantograph mechanism. However, it had several problems owing to the cumbersome and heavy structures during the prototype test. This study aims to solve the problems found in the first design and prove the advantages of the improvement, based on the design and performance criteria. The pantograph structure was simplified, and the gravity compensation method was modified to reduce inertia by using a constant force spring instead of a counter-mass. The improvement was computationally predicted using a mathematical model, and the results were verified through trajectory measurements in a micro-positioning task. Finally, the evaluation of dynamic performance is quantitatively presented through iterative positioning tasks.
{"title":"Fully mechanical motion-scaling instrument for microsurgery assistance: design improvement for enhancing the dynamic performance","authors":"Tae-Hoon Lee, Dongeun Choi, Chunwoo Kim","doi":"10.1093/jcde/qwad034","DOIUrl":"https://doi.org/10.1093/jcde/qwad034","url":null,"abstract":"\u0000 Robot assistance can improve the outcome of microsurgery by scaling down the surgeon's hand motions. However, the high cost of surgical robots has prevented their use in small hospitals or medical facilities in several developing countries. As a novel alternative, a fully mechanical motion-scaling instrument, which can be operated without computers and motors, was proposed based on the pantograph mechanism. However, it had several problems owing to the cumbersome and heavy structures during the prototype test. This study aims to solve the problems found in the first design and prove the advantages of the improvement, based on the design and performance criteria. The pantograph structure was simplified, and the gravity compensation method was modified to reduce inertia by using a constant force spring instead of a counter-mass. The improvement was computationally predicted using a mathematical model, and the results were verified through trajectory measurements in a micro-positioning task. Finally, the evaluation of dynamic performance is quantitatively presented through iterative positioning tasks.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"22 1","pages":"1010-1025"},"PeriodicalIF":4.9,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85192621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Suling Duan, Shan Jiang, H. Dai, Luping Wang, Zhenan He
Combinatorial optimization problems have very important applications in information technology, transportation, economics, management, network communication, and other fields. Since the problem size in real-scenario application is in large-scale, the demand for real-time and efficient solving approaches increases rapidly. The traditional exact methods guarantee the optimality of the final solution, but these methods can hardly solve the problem in acceptable time due to extremely high computational costs. Heuristic approaches can find feasible solutions in a limited time, while these approaches cannot meet the demand of solution quality. In recent years, hybrid algorithms based on exact methods and heuristic algorithms show outstanding performance in solving large-scale combinatorial optimization problems. The hybridization not only overcomes the shortcomings from single algorithm but also fully utilizes the search ability for population-based approaches as well as the interpretability in exact methods, which promotes the application of combinatorial optimization in real-world problems. This paper reviews existing studies on hybrid algorithms combining exact method and evolutionary computation, summarizes the characteristics of the existing algorithms, and directs the future research.
{"title":"The applications of hybrid approach combining exact method and evolutionary algorithm in combinatorial optimization","authors":"Suling Duan, Shan Jiang, H. Dai, Luping Wang, Zhenan He","doi":"10.1093/jcde/qwad029","DOIUrl":"https://doi.org/10.1093/jcde/qwad029","url":null,"abstract":"\u0000 Combinatorial optimization problems have very important applications in information technology, transportation, economics, management, network communication, and other fields. Since the problem size in real-scenario application is in large-scale, the demand for real-time and efficient solving approaches increases rapidly. The traditional exact methods guarantee the optimality of the final solution, but these methods can hardly solve the problem in acceptable time due to extremely high computational costs. Heuristic approaches can find feasible solutions in a limited time, while these approaches cannot meet the demand of solution quality. In recent years, hybrid algorithms based on exact methods and heuristic algorithms show outstanding performance in solving large-scale combinatorial optimization problems. The hybridization not only overcomes the shortcomings from single algorithm but also fully utilizes the search ability for population-based approaches as well as the interpretability in exact methods, which promotes the application of combinatorial optimization in real-world problems. This paper reviews existing studies on hybrid algorithms combining exact method and evolutionary computation, summarizes the characteristics of the existing algorithms, and directs the future research.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"21 1","pages":"934-946"},"PeriodicalIF":4.9,"publicationDate":"2023-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73876075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hyo-Jin Kim, Seung-Hoon Um, Y. Kang, Minwoo Shin, H. Jeon, Beop-Min Kim, Deukhee Lee, K. Yoon
This study aimed to develop a simulation model that accounts for skin-specific properties in order to predict photothermal damage during skin laser treatment. To construct a computational model, surface geometry information was obtained from an optical coherence tomography image, and the absorption coefficient of the skin was determined through spectrophotometry. The distribution of the internal light dose inside the skin medium was calculated using the light propagation model based on the Monte Carlo method. The photothermal response due to the absorption of laser light was modeled by a finite difference time domain model to solve the bio-heat transfer equation. The predicted depth and area of the damaged lesions from the simulation model were compared to those measured in ex vivo porcine skin. The present simulation model gave acceptable predictions with differences of approximately ∼10% in both depth and area.
{"title":"Laser-tissue interaction simulation considering skin-specific data to predict photothermal damage lesions during laser irradiation","authors":"Hyo-Jin Kim, Seung-Hoon Um, Y. Kang, Minwoo Shin, H. Jeon, Beop-Min Kim, Deukhee Lee, K. Yoon","doi":"10.1093/jcde/qwad033","DOIUrl":"https://doi.org/10.1093/jcde/qwad033","url":null,"abstract":"\u0000 This study aimed to develop a simulation model that accounts for skin-specific properties in order to predict photothermal damage during skin laser treatment. To construct a computational model, surface geometry information was obtained from an optical coherence tomography image, and the absorption coefficient of the skin was determined through spectrophotometry. The distribution of the internal light dose inside the skin medium was calculated using the light propagation model based on the Monte Carlo method. The photothermal response due to the absorption of laser light was modeled by a finite difference time domain model to solve the bio-heat transfer equation. The predicted depth and area of the damaged lesions from the simulation model were compared to those measured in ex vivo porcine skin. The present simulation model gave acceptable predictions with differences of approximately ∼10% in both depth and area.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"5 1","pages":"947-958"},"PeriodicalIF":4.9,"publicationDate":"2023-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84265961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Presenting a robust intelligent model capable of making accurate reliability forecasts has been an attractive topic to most industries. This study mainly aims to develop an approach by utilizing back propagation neural network (BPNN) to predict the reliability of engineering systems, such as industrial robot systems and turbochargers, with reasonable computing speed and high accuracy. Boxing Match Algorithm (BMA), as an evolutionary meta-heuristic algorithm with a new weight update strategy, is proposed to bring about performance improvements of the ANN in reliability forecast. Consequently, the hybrid model of BMA-BPNN has been provided to gain a significant level of accuracy in optimizing the weight and bias of BPNN using three sets of function approximation data to benchmark the proposed approach's performance. Then, the BMA is utilized to improve reliability forecasting accuracy in engineering problems. The obtained results reveal that the presented algorithm delivers exceptional performance in function approximation, and its performance in forecasting engineering systems' reliability is about 20% better than further compared algorithms. Similarly, rapid convergence rate, reasonable computing time, and well-performing are additional characteristics of the presented algorithm. Given the BMA-BPNN characteristics and the acquired findings, we can conclude that the proposed algorithm can be applicable in forecasting engineering problems' reliability.
对于大多数行业来说，提出一个能够做出准确可靠性预测的稳健智能模型一直是一个有吸引力的话题。本研究的主要目的是开发一种利用反向传播神经网络(BPNN)对工业机器人系统和涡轮增压器等工程系统进行可靠性预测的方法，该方法具有合理的计算速度和较高的精度。拳击匹配算法(Boxing Match Algorithm, BMA)作为一种进化元启发式算法，采用一种新的权值更新策略，提高了人工神经网络在可靠性预测方面的性能。因此，提出了BMA-BPNN的混合模型，在优化BPNN的权重和偏差方面获得了显著的精度，使用三组函数逼近数据来衡量所提出方法的性能。然后，利用BMA来提高工程问题的可靠性预测精度。结果表明，该算法在函数逼近方面具有优异的性能，在预测工程系统可靠性方面的性能比进一步比较的算法提高约20%。同样，快速的收敛速度、合理的计算时间和良好的性能是该算法的附加特点。结合BMA-BPNN的特点和所获得的结果，我们可以得出结论，该算法可以应用于工程问题的可靠性预测。
{"title":"Optimization of backpropagation neural network models for reliability forecasting using the boxing match algorithm: electro-mechanical case","authors":"M. Tanhaeean, S. Ghaderi, M. Sheikhalishahi","doi":"10.1093/jcde/qwad032","DOIUrl":"https://doi.org/10.1093/jcde/qwad032","url":null,"abstract":"\u0000 Presenting a robust intelligent model capable of making accurate reliability forecasts has been an attractive topic to most industries. This study mainly aims to develop an approach by utilizing back propagation neural network (BPNN) to predict the reliability of engineering systems, such as industrial robot systems and turbochargers, with reasonable computing speed and high accuracy. Boxing Match Algorithm (BMA), as an evolutionary meta-heuristic algorithm with a new weight update strategy, is proposed to bring about performance improvements of the ANN in reliability forecast. Consequently, the hybrid model of BMA-BPNN has been provided to gain a significant level of accuracy in optimizing the weight and bias of BPNN using three sets of function approximation data to benchmark the proposed approach's performance. Then, the BMA is utilized to improve reliability forecasting accuracy in engineering problems. The obtained results reveal that the presented algorithm delivers exceptional performance in function approximation, and its performance in forecasting engineering systems' reliability is about 20% better than further compared algorithms. Similarly, rapid convergence rate, reasonable computing time, and well-performing are additional characteristics of the presented algorithm. Given the BMA-BPNN characteristics and the acquired findings, we can conclude that the proposed algorithm can be applicable in forecasting engineering problems' reliability.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"99 1","pages":"918-933"},"PeriodicalIF":4.9,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75835836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this manuscript, we propose a new effective method for eigenpair reanalysis of large-scale finite element (FE) models. Our method utilizes the matrix block-partitioning algorithm in the Rayleigh-Ritz approach and expresses the Ritz basis matrix using thousands of block matrices of very small size. To avoid significant computational costs from the projection procedure, we derive a new formulation that uses tiny block computations instead of global matrix computations. Additionally, we present an algorithm that recognizes which blocks are changed in the modified FE model to achieve computational cost savings when computing new eigenpairs. Through selective updating for the recognized blocks, we can effectively construct the new Ritz basis matrix and the new reduced mass and stiffness matrices corresponding to the modified FE model. To demonstrate the performance of our proposed method, we solve several practical engineering problems and compare the results with those of the combined approximation (CA) method, the most well-known eigenpair reanalysis method, and ARPACK, an eigenvalue solver embedded in many numerical programs.
{"title":"Block-partitioned Rayleigh-Ritz method for efficient eigenpair reanalysis of large-scale finite element models","authors":"Yeon-Ho Jeong, Seung-Hwan Boo, S. Yim","doi":"10.1093/jcde/qwad030","DOIUrl":"https://doi.org/10.1093/jcde/qwad030","url":null,"abstract":"\u0000 In this manuscript, we propose a new effective method for eigenpair reanalysis of large-scale finite element (FE) models. Our method utilizes the matrix block-partitioning algorithm in the Rayleigh-Ritz approach and expresses the Ritz basis matrix using thousands of block matrices of very small size. To avoid significant computational costs from the projection procedure, we derive a new formulation that uses tiny block computations instead of global matrix computations. Additionally, we present an algorithm that recognizes which blocks are changed in the modified FE model to achieve computational cost savings when computing new eigenpairs. Through selective updating for the recognized blocks, we can effectively construct the new Ritz basis matrix and the new reduced mass and stiffness matrices corresponding to the modified FE model. To demonstrate the performance of our proposed method, we solve several practical engineering problems and compare the results with those of the combined approximation (CA) method, the most well-known eigenpair reanalysis method, and ARPACK, an eigenvalue solver embedded in many numerical programs.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"56 1","pages":"959-978"},"PeriodicalIF":4.9,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73483240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hye-A Kim, Chan Hee Park, Chaehyun Suh, Minseok Chae, Heonjun Yoon, Byeng D. Youn
Multi-scale convolutional neural network structures consisting of parallel convolution paths with different kernel sizes have been developed to extract features from multiple temporal scales and applied for fault diagnosis of rotating machines. However, when the extracted features are used to the same extent regardless of the temporal scale inside the network, good diagnostic performance may not be guaranteed due to the influence of the features of certain temporal scale less related to faults. Considering this issue, this paper presents a novel architecture called a multi-scale path attention residual network to further enhance the feature representational ability of a multi-scale structure. Multi-scale path attention residual network adopts a path attention module after a multi-scale dilated convolution layer, assigning different weights to features from different convolution paths. In addition, the network is composed of a stacked multi-scale attention residual block structure to continuously extract meaningful multi-scale characteristics and relationships between scales. The effectiveness of the proposed method is verified by examining its application to a helical gearbox vibration dataset and a permanent magnet synchronous motor current dataset. The results show that the proposed multi-scale path attention residual network can improve the feature learning ability of the multi-scale structure and achieve better fault diagnosis performance.
{"title":"MPARN: multi-scale path attention residual network for fault diagnosis of rotating machines","authors":"Hye-A Kim, Chan Hee Park, Chaehyun Suh, Minseok Chae, Heonjun Yoon, Byeng D. Youn","doi":"10.1093/jcde/qwad031","DOIUrl":"https://doi.org/10.1093/jcde/qwad031","url":null,"abstract":"\u0000 Multi-scale convolutional neural network structures consisting of parallel convolution paths with different kernel sizes have been developed to extract features from multiple temporal scales and applied for fault diagnosis of rotating machines. However, when the extracted features are used to the same extent regardless of the temporal scale inside the network, good diagnostic performance may not be guaranteed due to the influence of the features of certain temporal scale less related to faults. Considering this issue, this paper presents a novel architecture called a multi-scale path attention residual network to further enhance the feature representational ability of a multi-scale structure. Multi-scale path attention residual network adopts a path attention module after a multi-scale dilated convolution layer, assigning different weights to features from different convolution paths. In addition, the network is composed of a stacked multi-scale attention residual block structure to continuously extract meaningful multi-scale characteristics and relationships between scales. The effectiveness of the proposed method is verified by examining its application to a helical gearbox vibration dataset and a permanent magnet synchronous motor current dataset. The results show that the proposed multi-scale path attention residual network can improve the feature learning ability of the multi-scale structure and achieve better fault diagnosis performance.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"32 1","pages":"860-872"},"PeriodicalIF":4.9,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84914747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Combining topology optimization and additive manufacturing (AM) is a promising approach to breaking through the limitations of conventional design and developing innovative structures with high performance. However, the unique manufacturing constraints in AM should be considered when developing the topology optimization algorithms for AM. Material anisotropy is one of the common characteristics of AM materials due to the layer-by-layer manufacturing techniques. The present work proposes a topology optimization approach for AM with strength constraints considering anisotropy. The Hoffman failure criterion is adopted to represent the anisotropic strength behaviors of AM materials. Based on the Hoffman failure criteria and the p-norm measure aggregation function, a global strength constraint formulation is established. Under the framework of solid isotropic material with penalization (SIMP), we develop a topology optimization methodology to minimize the structural weight or volume fraction subject to structural stiffness and strength constraints. Several 2D or 3D numerical test cases are performed to validate the effectiveness and performance of the developed method. The results indicated that the proposed method could make full use of material properties by considering anisotropic strength. Besides, the topological optimization considering strength anisotropy could be combined with build direction optimization to further reduce the structural weight.
{"title":"Topology optimization for additive manufacturing with strength constraints considering anisotropy","authors":"Jun Zou, Xiaoyu Xia","doi":"10.1093/jcde/qwad028","DOIUrl":"https://doi.org/10.1093/jcde/qwad028","url":null,"abstract":"\u0000 Combining topology optimization and additive manufacturing (AM) is a promising approach to breaking through the limitations of conventional design and developing innovative structures with high performance. However, the unique manufacturing constraints in AM should be considered when developing the topology optimization algorithms for AM. Material anisotropy is one of the common characteristics of AM materials due to the layer-by-layer manufacturing techniques. The present work proposes a topology optimization approach for AM with strength constraints considering anisotropy. The Hoffman failure criterion is adopted to represent the anisotropic strength behaviors of AM materials. Based on the Hoffman failure criteria and the p-norm measure aggregation function, a global strength constraint formulation is established. Under the framework of solid isotropic material with penalization (SIMP), we develop a topology optimization methodology to minimize the structural weight or volume fraction subject to structural stiffness and strength constraints. Several 2D or 3D numerical test cases are performed to validate the effectiveness and performance of the developed method. The results indicated that the proposed method could make full use of material properties by considering anisotropic strength. Besides, the topological optimization considering strength anisotropy could be combined with build direction optimization to further reduce the structural weight.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"46 1","pages":"892-904"},"PeriodicalIF":4.9,"publicationDate":"2023-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82442108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study examines the visual-search and information-gathering behavior of architects in the early architectural design phase in relation to varied media tools. The study proposes the idea that navigation skills in online media help designers discover more creative solution areas during their design process. In continuation of our research, conceptual conclusions are made based on the results obtained from the field study and the literature review. In this context, we discuss the concept constituting the habitus of digital architecture. We re-evaluated our conceptual proposal by applying design experiments to examine the phenomena contained in the habitus of design-oriented research. We have discussed the results from the experiments in this article in detail; focusing on whether correlation exists between the interviewees’ expressions and designers’ practices. We then adapted field theory, as elaborated by Pierre Bourdieu in 1984, to the digital habitus of architecture. Afterward, by taking the process of design-oriented knowledge production into account, we have identified two fields of design-oriented digital habitus: online and offline. The fields forming the habitus of digital architecture and the possible advantages that may occur based on these fields have been identified. Finally, the meaning of having digital privilege for architects has been evaluated in terms of the future of architecture.
{"title":"Mimarlığın Dijital Habitusu: Tasarım odaklı İnternet Kullanım Pratiği","authors":"Hanife Sümeyye Taşdelen, Leman Figen Gül","doi":"10.53710/jcode.1236623","DOIUrl":"https://doi.org/10.53710/jcode.1236623","url":null,"abstract":"This study examines the visual-search and information-gathering behavior of architects in the early architectural design phase in relation to varied media tools. The study proposes the idea that navigation skills in online media help designers discover more creative solution areas during their design process. In continuation of our research, conceptual conclusions are made based on the results obtained from the field study and the literature review. In this context, we discuss the concept constituting the habitus of digital architecture. We re-evaluated our conceptual proposal by applying design experiments to examine the phenomena contained in the habitus of design-oriented research. We have discussed the results from the experiments in this article in detail; focusing on whether correlation exists between the interviewees’ expressions and designers’ practices. We then adapted field theory, as elaborated by Pierre Bourdieu in 1984, to the digital habitus of architecture. Afterward, by taking the process of design-oriented knowledge production into account, we have identified two fields of design-oriented digital habitus: online and offline. The fields forming the habitus of digital architecture and the possible advantages that may occur based on these fields have been identified. Finally, the meaning of having digital privilege for architects has been evaluated in terms of the future of architecture.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"71 6 1","pages":""},"PeriodicalIF":4.9,"publicationDate":"2023-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89170397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xingyu Fu, Fengfeng Zhou, Dheeraj Peddireddy, Zhengyang Kang, M. Jun, V. Aggarwal
In this work, we present a Boundary Oriented Graph Embedding (BOGE) approach for the Graph Neural Network (GNN) to assist in rapid design and digital prototyping. The cantilever beam problem has been solved as an example to validate its potential of providing physical field results and optimized designs using only 10ms. Providing shortcuts for both boundary elements and local neighbor elements, the BOGE approach can embed unstructured mesh elements into the graph and performs an efficient regression on large-scale triangular-mesh-based FEA results, which cannot be realized by other machine-learning-based surrogate methods. It has the potential to serve as a surrogate model for other boundary value problems. Focusing on the cantilever beam problem, the BOGE approach with 3-layer DeepGCN model achieves the regression with MSE of 0.011706 (2.41% MAPE) for stress field prediction and 0.002735 MSE (with 1.58% elements having error larger than 0.01) for topological optimization. The overall concept of the BOGE approach paves the way for a general and efficient deep-learning-based FEA simulator that will benefit both industry and CAD design-related areas.
{"title":"An finite element analysis surrogate model with boundary oriented graph embedding approach for rapid design","authors":"Xingyu Fu, Fengfeng Zhou, Dheeraj Peddireddy, Zhengyang Kang, M. Jun, V. Aggarwal","doi":"10.1093/jcde/qwad025","DOIUrl":"https://doi.org/10.1093/jcde/qwad025","url":null,"abstract":"\u0000 In this work, we present a Boundary Oriented Graph Embedding (BOGE) approach for the Graph Neural Network (GNN) to assist in rapid design and digital prototyping. The cantilever beam problem has been solved as an example to validate its potential of providing physical field results and optimized designs using only 10ms. Providing shortcuts for both boundary elements and local neighbor elements, the BOGE approach can embed unstructured mesh elements into the graph and performs an efficient regression on large-scale triangular-mesh-based FEA results, which cannot be realized by other machine-learning-based surrogate methods. It has the potential to serve as a surrogate model for other boundary value problems. Focusing on the cantilever beam problem, the BOGE approach with 3-layer DeepGCN model achieves the regression with MSE of 0.011706 (2.41% MAPE) for stress field prediction and 0.002735 MSE (with 1.58% elements having error larger than 0.01) for topological optimization. The overall concept of the BOGE approach paves the way for a general and efficient deep-learning-based FEA simulator that will benefit both industry and CAD design-related areas.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"57 1","pages":"1026-1046"},"PeriodicalIF":4.9,"publicationDate":"2023-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79372623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The survivability of a naval ship is defined as its ability to evade or withstand a hostile environment while performing a given mission. Stealth technology, which reduces the probability of detection by enemy detection equipment using a highly advanced detection system, is one of the most important technologies to improve the survivability of naval ships. Moreover, radar cross section (RCS) reduction is a very important factor in stealth technology because a small RCS, which is the main parameter determining susceptibility, improves the ability of ships to evade enemy detection equipment. In this study, an automated topology design for reducing susceptibility was developed by combining geometric deep learning and topology optimization. A convolutional neural network model was used as the geometric deep-learning model, and the triangular meshes of the naval ship models and equipment models were used as datasets. To compensate for the lack of training data, randomly generated meshes were additionally used as datasets. To express the feature data of the mesh as a matrix, points at equal intervals were projected orthogonally and the distance between the plane and point was set as a matrix value. The label data were defined as the highest RCS values excluding the cardinal points. After realizing the topology design for reducing susceptibility using the developed system, verification was performed through RCS analysis of the original model and the topology-designed model.
{"title":"Automated topology design to improve the susceptibility of naval ships using geometric deep learning","authors":"Joon-Tae Hwang, Suk-Yoon Hong, Jee-hun Song","doi":"10.1093/jcde/qwad023","DOIUrl":"https://doi.org/10.1093/jcde/qwad023","url":null,"abstract":"\u0000 The survivability of a naval ship is defined as its ability to evade or withstand a hostile environment while performing a given mission. Stealth technology, which reduces the probability of detection by enemy detection equipment using a highly advanced detection system, is one of the most important technologies to improve the survivability of naval ships. Moreover, radar cross section (RCS) reduction is a very important factor in stealth technology because a small RCS, which is the main parameter determining susceptibility, improves the ability of ships to evade enemy detection equipment. In this study, an automated topology design for reducing susceptibility was developed by combining geometric deep learning and topology optimization. A convolutional neural network model was used as the geometric deep-learning model, and the triangular meshes of the naval ship models and equipment models were used as datasets. To compensate for the lack of training data, randomly generated meshes were additionally used as datasets. To express the feature data of the mesh as a matrix, points at equal intervals were projected orthogonally and the distance between the plane and point was set as a matrix value. The label data were defined as the highest RCS values excluding the cardinal points. After realizing the topology design for reducing susceptibility using the developed system, verification was performed through RCS analysis of the original model and the topology-designed model.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"16 1","pages":"794-808"},"PeriodicalIF":4.9,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87867163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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：