Wei Zhu, Zhihui Li, Hang Su, Lei Liu, Ali Asgher Heidari, Huiling Chen, Guoxi Liang
In mining mineral resources, it is vital to monitor the stability of the rock body in real time, reasonably regulate the area of ground pressure concentration, and guarantee the safety of personnel and equipment. The microseismic signals generated by monitoring the rupture of the rock body can effectively predict the rock body disaster, but the current microseismic monitoring technology is not ideal. In order to address the issue of microseismic monitoring in deep wells, this research suggests a machine learning-based model for predicting microseismic phenomena. First, this work presents the random spare, double adaptive weight, and Gaussian-Cauchy fusion strategies as additions to the multi-verse optimizer (MVO) and suggests an enhanced MVO algorithm (RDGMVO). Subsequently, the RDGMVO-FKNN microseismic prediction model is presented by combining it with the Fuzzy K-Nearest Neighbours (FKNN) classifier. The experimental section compares twelve traditional and recently enhanced algorithms with RDGMVO, demonstrating the latter's excellent benchmark optimization performance and remarkable improvement effect. Next, the FKNN comparison experiment, the classical classifier experiment, and the microseismic dataset feature selection experiment confirm the precision and stability of the RDGMVO-FKNN model for the microseismic prediction problem. According to the results, the RDGMVO-FKNN model has an accuracy above 89%, indicating that it is a reliable and accurate method for classifying and predicting microseismic occurrences. Code has been available at https://github.com/GuaipiXiao/RDGMVO.
{"title":"Optimizing Microseismic Monitoring: A Fusion of Gaussian-Cauchy and Adaptive Weight Strategies","authors":"Wei Zhu, Zhihui Li, Hang Su, Lei Liu, Ali Asgher Heidari, Huiling Chen, Guoxi Liang","doi":"10.1093/jcde/qwae073","DOIUrl":"https://doi.org/10.1093/jcde/qwae073","url":null,"abstract":"\u0000 In mining mineral resources, it is vital to monitor the stability of the rock body in real time, reasonably regulate the area of ground pressure concentration, and guarantee the safety of personnel and equipment. The microseismic signals generated by monitoring the rupture of the rock body can effectively predict the rock body disaster, but the current microseismic monitoring technology is not ideal. In order to address the issue of microseismic monitoring in deep wells, this research suggests a machine learning-based model for predicting microseismic phenomena. First, this work presents the random spare, double adaptive weight, and Gaussian-Cauchy fusion strategies as additions to the multi-verse optimizer (MVO) and suggests an enhanced MVO algorithm (RDGMVO). Subsequently, the RDGMVO-FKNN microseismic prediction model is presented by combining it with the Fuzzy K-Nearest Neighbours (FKNN) classifier. The experimental section compares twelve traditional and recently enhanced algorithms with RDGMVO, demonstrating the latter's excellent benchmark optimization performance and remarkable improvement effect. Next, the FKNN comparison experiment, the classical classifier experiment, and the microseismic dataset feature selection experiment confirm the precision and stability of the RDGMVO-FKNN model for the microseismic prediction problem. According to the results, the RDGMVO-FKNN model has an accuracy above 89%, indicating that it is a reliable and accurate method for classifying and predicting microseismic occurrences. Code has been available at https://github.com/GuaipiXiao/RDGMVO.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141920304","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}
Qiuxuan Wu, Zikai Zhao, Mingming Chen, Xiaoni Chi, Botao Zhang, Jian Wang, Anton A. Zhilenkov, S. A. Chepinskiy
The optimization of numerical functions with multiple independent variables was a significant challenge with numerous practical applications in process control systems, data fitting, and engineering designs. Although RNA genetic algorithms offer clear benefits in function optimization, including rapid convergence, they have low accuracy and can easily become trapped in local optima. To address these issues, a new heuristic algorithm was proposed, a gradient descent-based RNA genetic algorithm. Specifically, adaptive moment estimation (Adam) was employed as a mutation operator to improve the local development ability of the algorithm. Additionally, two new operators inspired by the inner-loop structure of RNA molecules were introduced: an inner-loop crossover operator and an inner-loop mutation operator. These operators enhance the global exploration ability of the algorithm in the early stages of evolution and enable it to escape from local optima. The algorithm consists of two stages: a pre-evolutionary stage that employs RNA genetic algorithms to identify individuals in the vicinity of the optimal region and a post-evolutionary stage that applies a adaptive gradient descent mutation to further enhance the solution's quality. When compared with the current advanced algorithms for solving function optimization problems, Adam RNA-GA produced better optimal solutions. In comparison with RNA Genetic Algorithm (RNA-GA) and Genetic Algorithm (GA) across 17 benchmark functions, Adam RNA-GA ranked first with the best result of an average rank of 1.58 according to the Friedman test. In the set of 29 functions of the CEC2017 suite, compared with heuristic algorithms such as African Vulture Optimization Algorithm (AVOA), Dung Beetle Optimization (DBO), Whale Optimization Algorithm (WOA), and Grey Wolf Optimizer (GWO), Adam RNA-GA ranked first with the best result of an average rank of 1.724 according to the Friedman test. Our algorithm not only achieved significant improvements over RNA-GA but also performed excellently among various current advanced algorithms for solving function optimization problems, achieving high precision in function optimization.
{"title":"An RNA Evolutionary Algorithm Based on Gradient Descent for Function Optimization","authors":"Qiuxuan Wu, Zikai Zhao, Mingming Chen, Xiaoni Chi, Botao Zhang, Jian Wang, Anton A. Zhilenkov, S. A. Chepinskiy","doi":"10.1093/jcde/qwae068","DOIUrl":"https://doi.org/10.1093/jcde/qwae068","url":null,"abstract":"\u0000 The optimization of numerical functions with multiple independent variables was a significant challenge with numerous practical applications in process control systems, data fitting, and engineering designs. Although RNA genetic algorithms offer clear benefits in function optimization, including rapid convergence, they have low accuracy and can easily become trapped in local optima. To address these issues, a new heuristic algorithm was proposed, a gradient descent-based RNA genetic algorithm. Specifically, adaptive moment estimation (Adam) was employed as a mutation operator to improve the local development ability of the algorithm. Additionally, two new operators inspired by the inner-loop structure of RNA molecules were introduced: an inner-loop crossover operator and an inner-loop mutation operator. These operators enhance the global exploration ability of the algorithm in the early stages of evolution and enable it to escape from local optima. The algorithm consists of two stages: a pre-evolutionary stage that employs RNA genetic algorithms to identify individuals in the vicinity of the optimal region and a post-evolutionary stage that applies a adaptive gradient descent mutation to further enhance the solution's quality. When compared with the current advanced algorithms for solving function optimization problems, Adam RNA-GA produced better optimal solutions. In comparison with RNA Genetic Algorithm (RNA-GA) and Genetic Algorithm (GA) across 17 benchmark functions, Adam RNA-GA ranked first with the best result of an average rank of 1.58 according to the Friedman test. In the set of 29 functions of the CEC2017 suite, compared with heuristic algorithms such as African Vulture Optimization Algorithm (AVOA), Dung Beetle Optimization (DBO), Whale Optimization Algorithm (WOA), and Grey Wolf Optimizer (GWO), Adam RNA-GA ranked first with the best result of an average rank of 1.724 according to the Friedman test. Our algorithm not only achieved significant improvements over RNA-GA but also performed excellently among various current advanced algorithms for solving function optimization problems, achieving high precision in function optimization.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141798538","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}
Guanghui Li, Taihua Zhang, Chieh-Yuan Tsai, Yao Lu, Jun Yang, Liguo Yao
Crayfish optimization algorithm (COA) is a novel, bionic, metaheuristic algorithm with high convergence speed and solution accuracy. However, in some complex optimization problems and real application scenarios, the performance of COA is not satisfactory. In order to overcome the challenges encountered by COA, such as being stuck in the local optimal and insufficient search range, this paper proposes four improvement strategies: search-hide, adaptive spiral elite greedy opposition-based learning (ASEG-OBL), competition-elimination, and chaos mutation. To evaluate the convergence accuracy, speed, and robustness of the modified crayfish optimization algorithm (MCOA), some simulation comparison experiments of 10 algorithms are conducted. Experimental results show that the MCOA achieved the minor Friedman test (FT) value in 23 test functions, CEC2014, and CEC2020, and achieved average superiority rates of 80.97%, 72.59%, and 71.11% in the Wilcoxon rank sum test (WT) respectively. In addition, MCOA shows high applicability and progressiveness in five engineering problems in actual industrial field. Moreover, MCOA achieved 80% and 100% superiority rate against COA on CEC2020 and the fixed-dimension function of 23 benchmark test functions. Finally, MCOA owns better convergence and population diversity.
{"title":"Modified Crayfish Optimization Algorithm with Adaptive Spiral Elite Greedy Opposition-based Learning and Search-hide Strategy for Global Optimization","authors":"Guanghui Li, Taihua Zhang, Chieh-Yuan Tsai, Yao Lu, Jun Yang, Liguo Yao","doi":"10.1093/jcde/qwae069","DOIUrl":"https://doi.org/10.1093/jcde/qwae069","url":null,"abstract":"\u0000 Crayfish optimization algorithm (COA) is a novel, bionic, metaheuristic algorithm with high convergence speed and solution accuracy. However, in some complex optimization problems and real application scenarios, the performance of COA is not satisfactory. In order to overcome the challenges encountered by COA, such as being stuck in the local optimal and insufficient search range, this paper proposes four improvement strategies: search-hide, adaptive spiral elite greedy opposition-based learning (ASEG-OBL), competition-elimination, and chaos mutation. To evaluate the convergence accuracy, speed, and robustness of the modified crayfish optimization algorithm (MCOA), some simulation comparison experiments of 10 algorithms are conducted. Experimental results show that the MCOA achieved the minor Friedman test (FT) value in 23 test functions, CEC2014, and CEC2020, and achieved average superiority rates of 80.97%, 72.59%, and 71.11% in the Wilcoxon rank sum test (WT) respectively. In addition, MCOA shows high applicability and progressiveness in five engineering problems in actual industrial field. Moreover, MCOA achieved 80% and 100% superiority rate against COA on CEC2020 and the fixed-dimension function of 23 benchmark test functions. Finally, MCOA owns better convergence and population diversity.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141799850","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 activities of the traffic department mainly contribute to the generation of greenhouse gas (GHG) emissions. The swift expansion of the traffic department results in a significant increase in global pollution levels, adversely affecting human health. To address GHG emissions and propose impactful solutions for reducing fuel consumption in vehicles, environmental considerations are integrated with the core principles of the Vehicle Routing Problem (VRP). This integration gives rise to the Pollution Routing Problem (PRP), which aims to optimize routing decisions with a focus on minimizing environmental impact. At the same time, the retail distribution system explores the use of an Omni-channel approach as a transportation mode adopted in this study. The objectives of this research include minimizing total travel costs and fuel consumption while aiming to reduce GHG emissions, promote environmental sustainability, and enhance the convenience of shopping and pickup for customers through the integration of online and offline modes. This problem is NP-Hard; therefore, the Non-dominated Sorting Simplified Swarm Optimization (NSSO) algorithm is employed. NSSO combines the non-dominated technique of Non-dominated Sorting Genetic Algorithm II (NSGA-II) with the update mechanism of SSO to obtain a set of Pareto optimal solutions. Moreover, the NSSO, a multi-objective evolutionary algorithm, is adopted to address multi-objective problems. The PRP benchmark dataset is utilized, and the results are compared with two other multi-objective evolutionary algorithms: NSGA-II and Non-dominated Sorting Particle Swarm Optimization (NSPSO). The findings of the study confirm that NSSO exhibits feasibility, provides good solutions, and achieves faster convergence compared to the other two algorithms, NSGA-II and NSPSO.
{"title":"Non-dominated sorting simplified swarm optimization for multi-objective omni-channel of pollution routing problem","authors":"Wenbo Zhu, Tzu-Ching Liang, Wei-Chang Yeh, Guangyi Yang, Shi-Yi Tan, Zhenyao Liu, Chia-Ling Huang","doi":"10.1093/jcde/qwae062","DOIUrl":"https://doi.org/10.1093/jcde/qwae062","url":null,"abstract":"\u0000 The activities of the traffic department mainly contribute to the generation of greenhouse gas (GHG) emissions. The swift expansion of the traffic department results in a significant increase in global pollution levels, adversely affecting human health. To address GHG emissions and propose impactful solutions for reducing fuel consumption in vehicles, environmental considerations are integrated with the core principles of the Vehicle Routing Problem (VRP). This integration gives rise to the Pollution Routing Problem (PRP), which aims to optimize routing decisions with a focus on minimizing environmental impact. At the same time, the retail distribution system explores the use of an Omni-channel approach as a transportation mode adopted in this study. The objectives of this research include minimizing total travel costs and fuel consumption while aiming to reduce GHG emissions, promote environmental sustainability, and enhance the convenience of shopping and pickup for customers through the integration of online and offline modes. This problem is NP-Hard; therefore, the Non-dominated Sorting Simplified Swarm Optimization (NSSO) algorithm is employed. NSSO combines the non-dominated technique of Non-dominated Sorting Genetic Algorithm II (NSGA-II) with the update mechanism of SSO to obtain a set of Pareto optimal solutions. Moreover, the NSSO, a multi-objective evolutionary algorithm, is adopted to address multi-objective problems. The PRP benchmark dataset is utilized, and the results are compared with two other multi-objective evolutionary algorithms: NSGA-II and Non-dominated Sorting Particle Swarm Optimization (NSPSO). The findings of the study confirm that NSSO exhibits feasibility, provides good solutions, and achieves faster convergence compared to the other two algorithms, NSGA-II and NSPSO.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141823154","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 introduces a novel approach to architectural visualization using generative artificial intelligence (AI), particularly emphasizing text-to-image (txt2img) technology, to remarkably improve the visualization process right from the initial design phase within the architecture, engineering, and construction industry. By creating >10,000 images incorporating an architect's personal style and characteristics into a residential house model, the effectiveness of base AI models. Furthermore, various architectural styles were integrated to enhance the visualization process. This method involved additional training for styles with low similarity rates, which required extensive data preparation and their integration into the base AI model. Demonstrated to be effective across multiple scenarios, this technique markedly enhances the efficiency and speed of production of architectural visualization images. Highlighting the vast potential of AI in design visualization, our study emphasizes the technology's shift toward facilitating more user-centered and personalized design applications.
{"title":"Generative Early Architectural Visualizations: Incorporating Architect's Style-trained Models","authors":"Jin-Kook Lee, Y. Yoo, Seung Hyun Cha","doi":"10.1093/jcde/qwae065","DOIUrl":"https://doi.org/10.1093/jcde/qwae065","url":null,"abstract":"\u0000 This study introduces a novel approach to architectural visualization using generative artificial intelligence (AI), particularly emphasizing text-to-image (txt2img) technology, to remarkably improve the visualization process right from the initial design phase within the architecture, engineering, and construction industry. By creating >10,000 images incorporating an architect's personal style and characteristics into a residential house model, the effectiveness of base AI models. Furthermore, various architectural styles were integrated to enhance the visualization process. This method involved additional training for styles with low similarity rates, which required extensive data preparation and their integration into the base AI model. Demonstrated to be effective across multiple scenarios, this technique markedly enhances the efficiency and speed of production of architectural visualization images. Highlighting the vast potential of AI in design visualization, our study emphasizes the technology's shift toward facilitating more user-centered and personalized design applications.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141642626","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}
Designing distributed filter circuits (DFCs) is complex and time-consuming, involving setting and optimizing multiple hyperparameters. Traditional optimization methods, such as using the commercial finite element solver HFSS (High-Frequency Structure Simulator) to enumerate all parameter combinations with fixed steps and then simulate each combination, are not only time-consuming and labor-intensive but also rely heavily on the expertise and experience of electronics engineers, making it difficult to adapt to rapidly changing design requirements. Additionally, these commercial tools struggle with precise adjustments when parameters are sensitive to numerical changes, resulting in limited optimization effectiveness. This study proposes a novel end-to-end automated method for DFC design. The proposed method harnesses reinforcement learning (RL) algorithms, eliminating the dependence on the design experience of engineers. Thus, it significantly reduces the subjectivity and constraints associated with circuit design. The experimental findings demonstrate clear improvements in design efficiency and quality when comparing the proposed method with traditional engineer-driven methods. Furthermore, the proposed method achieves superior performance when designing complex or rapidly evolving DFCs, highlighting the substantial potential of RL in circuit design automation. In particular, compared to the existing DFC automation design method CircuitGNN, our method achieves an average performance improvement of 8.72%. Additionally, the execution efficiency of our method is 2000 times higher than CircuitGNN on the CPU and 241 times higher on the GPU.
{"title":"Automated design and optimization of distributed filter circuits using reinforcement learning","authors":"Peng Gao, Tao Yu, Fei Wang, Ruyue Yuan","doi":"10.1093/jcde/qwae066","DOIUrl":"https://doi.org/10.1093/jcde/qwae066","url":null,"abstract":"\u0000 Designing distributed filter circuits (DFCs) is complex and time-consuming, involving setting and optimizing multiple hyperparameters. Traditional optimization methods, such as using the commercial finite element solver HFSS (High-Frequency Structure Simulator) to enumerate all parameter combinations with fixed steps and then simulate each combination, are not only time-consuming and labor-intensive but also rely heavily on the expertise and experience of electronics engineers, making it difficult to adapt to rapidly changing design requirements. Additionally, these commercial tools struggle with precise adjustments when parameters are sensitive to numerical changes, resulting in limited optimization effectiveness. This study proposes a novel end-to-end automated method for DFC design. The proposed method harnesses reinforcement learning (RL) algorithms, eliminating the dependence on the design experience of engineers. Thus, it significantly reduces the subjectivity and constraints associated with circuit design. The experimental findings demonstrate clear improvements in design efficiency and quality when comparing the proposed method with traditional engineer-driven methods. Furthermore, the proposed method achieves superior performance when designing complex or rapidly evolving DFCs, highlighting the substantial potential of RL in circuit design automation. In particular, compared to the existing DFC automation design method CircuitGNN, our method achieves an average performance improvement of 8.72%. Additionally, the execution efficiency of our method is 2000 times higher than CircuitGNN on the CPU and 241 times higher on the GPU.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141643875","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}
Semin Ahn, Jinoh Yoo, Kyu-Wha Lee, B. D. Youn, Sung-Hoon Ahn
A frequency-focused sound data generator was developed for the in-situ fault sound diagnosis of industrial robot reducers. The sound data generator, based on a conditional generative adversarial network, selects a target frequency range without relying on domain knowledge. A sound dataset of normal and faulty harmonic drive rotations of in-situ industrial robots was collected using an attachable wireless sound sensor. The generated sound data were evaluated based on the fault diagnosis accuracy of a simple classifier trained using the generated data and tested using real data. The proposed method well-defined the frequency feature clusters and produced high-quality data, exhibiting up to 16.0% higher precision score on normal and 13.0% higher accuracy on weak-fault harmonic drive compared to the conventional methods, achieving fault diagnosis accuracy of 95.6% even in situations of fault data comprising only 5% of the normal data.
{"title":"Frequency-Focused Sound Data Generator for Fault Diagnosis in Industrial Robots","authors":"Semin Ahn, Jinoh Yoo, Kyu-Wha Lee, B. D. Youn, Sung-Hoon Ahn","doi":"10.1093/jcde/qwae061","DOIUrl":"https://doi.org/10.1093/jcde/qwae061","url":null,"abstract":"\u0000 A frequency-focused sound data generator was developed for the in-situ fault sound diagnosis of industrial robot reducers. The sound data generator, based on a conditional generative adversarial network, selects a target frequency range without relying on domain knowledge. A sound dataset of normal and faulty harmonic drive rotations of in-situ industrial robots was collected using an attachable wireless sound sensor. The generated sound data were evaluated based on the fault diagnosis accuracy of a simple classifier trained using the generated data and tested using real data. The proposed method well-defined the frequency feature clusters and produced high-quality data, exhibiting up to 16.0% higher precision score on normal and 13.0% higher accuracy on weak-fault harmonic drive compared to the conventional methods, achieving fault diagnosis accuracy of 95.6% even in situations of fault data comprising only 5% of the normal data.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141677469","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}
Recently, interest in functional safety has surged because vehicle technology increasingly relies on electronics and automation. Failure of certain system components can endanger driver safety and is costly to address. The detection of abnormal data is crucial for enhancing the reliability, safety, and efficiency. This study introduces a novel anomaly detection method of designable generative adversarial network anomaly detection (DGANomaly). DGANomaly combines the data augmentation method of a designable generative adversarial network (DGAN) with a GANomaly data classification technique. DGANomaly not only generates virtual data that are challenging to obtain or simulate but also produces a range of statistical design variables for normal and abnormal data. This approach enables the specific identification of normal and abnormal design variables. To demonstrate its effectiveness, the DGANomaly method was applied to an electric power steering (EPS) model when multiple degradations of gear stiffness, gear friction, and rack displacement were considered. An EPS model was constructed and validated using simulation programs such as Prescan, Amesim, and Simulink. Consequently, DGANomaly exhibited a higher classification accuracy than the other methods, allowing for more accurate detection of abnormal data. Additionally, a clearer range of statistical designs can be obtained for normal data. These results indicate that the statistical design variables that are less likely to fail can be obtained using minimal data.
{"title":"Diagnosis-based design of electric power steering system considering multiple degradations: role of designable generative adversarial network anomaly detection","authors":"Jeongbin Kim, Dabin Yang, Jongsoo Lee","doi":"10.1093/jcde/qwae056","DOIUrl":"https://doi.org/10.1093/jcde/qwae056","url":null,"abstract":"\u0000 Recently, interest in functional safety has surged because vehicle technology increasingly relies on electronics and automation. Failure of certain system components can endanger driver safety and is costly to address. The detection of abnormal data is crucial for enhancing the reliability, safety, and efficiency. This study introduces a novel anomaly detection method of designable generative adversarial network anomaly detection (DGANomaly). DGANomaly combines the data augmentation method of a designable generative adversarial network (DGAN) with a GANomaly data classification technique. DGANomaly not only generates virtual data that are challenging to obtain or simulate but also produces a range of statistical design variables for normal and abnormal data. This approach enables the specific identification of normal and abnormal design variables. To demonstrate its effectiveness, the DGANomaly method was applied to an electric power steering (EPS) model when multiple degradations of gear stiffness, gear friction, and rack displacement were considered. An EPS model was constructed and validated using simulation programs such as Prescan, Amesim, and Simulink. Consequently, DGANomaly exhibited a higher classification accuracy than the other methods, allowing for more accurate detection of abnormal data. Additionally, a clearer range of statistical designs can be obtained for normal data. These results indicate that the statistical design variables that are less likely to fail can be obtained using minimal data.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.9,"publicationDate":"2024-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141340912","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}
Minseok Chae, Hye-A Kim, Hye Jun Oh, Chan Hee Park, Chaehyun Suh, Heonjun Yoon, Byeng D. Youn
The fault diagnosis of permanent magnet synchronous motor is of vital importance in industrial fields to ensure user safety and minimize economic losses from accidents. However, recent fault diagnosis methods, particularly the methods using deep learning, require a massive amount of labeled data, which may not be available in industrial fields. Few-shot learning has been recently applied in fault diagnosis for rotary machineries, to alleviate the data deficiency and/or to enable unseen fault diagnosis. However, two major obstacles still remain, specifically: a) the limited ability of the models to be generalized for use under new operating conditions and b) insufficient discriminative features to precisely diagnose fault types. To address these limitations, this study proposes a Prototype-assisted dual-Contrastive learning with Depthwise separable Convolutional neural network (PCDC) for few-shot fault diagnosis for permanent magnet synchronous motors under new working conditions. Operation-robust fault features are extracted to reinforce generalization of PCDC under new operating conditions by extracting fault-induced amplitude and frequency modulation features and by eliminating the influence of operating conditions from the motor stator current signals. Prototype-assisted dual-contrastive learning is proposed to clearly distinguish the fault categories even when the fault features are similar to each other by learning both local- and global-similarity features, which increases the instance-discrimination ability while alleviating an overfitting issue. Experimental results show that the proposed PCDC outperforms the comparison models in few-shot fault diagnosis tasks under new operating conditions.
{"title":"PCDC: Prototype-assisted dual-contrastive learning with depthwise separable convolutional neural network for few-shot fault diagnosis of permanent magnet synchronous motors under new operating conditions","authors":"Minseok Chae, Hye-A Kim, Hye Jun Oh, Chan Hee Park, Chaehyun Suh, Heonjun Yoon, Byeng D. Youn","doi":"10.1093/jcde/qwae052","DOIUrl":"https://doi.org/10.1093/jcde/qwae052","url":null,"abstract":"\u0000 The fault diagnosis of permanent magnet synchronous motor is of vital importance in industrial fields to ensure user safety and minimize economic losses from accidents. However, recent fault diagnosis methods, particularly the methods using deep learning, require a massive amount of labeled data, which may not be available in industrial fields. Few-shot learning has been recently applied in fault diagnosis for rotary machineries, to alleviate the data deficiency and/or to enable unseen fault diagnosis. However, two major obstacles still remain, specifically: a) the limited ability of the models to be generalized for use under new operating conditions and b) insufficient discriminative features to precisely diagnose fault types. To address these limitations, this study proposes a Prototype-assisted dual-Contrastive learning with Depthwise separable Convolutional neural network (PCDC) for few-shot fault diagnosis for permanent magnet synchronous motors under new working conditions. Operation-robust fault features are extracted to reinforce generalization of PCDC under new operating conditions by extracting fault-induced amplitude and frequency modulation features and by eliminating the influence of operating conditions from the motor stator current signals. Prototype-assisted dual-contrastive learning is proposed to clearly distinguish the fault categories even when the fault features are similar to each other by learning both local- and global-similarity features, which increases the instance-discrimination ability while alleviating an overfitting issue. Experimental results show that the proposed PCDC outperforms the comparison models in few-shot fault diagnosis tasks under new operating conditions.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.9,"publicationDate":"2024-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141271405","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 paper introduces a sequential approximate optimization method that combines the finite element method (FEM), dynamic differential evolution (DDE), and artificial neural network (ANN) surrogate models. The developed method is applied to address two optimization problems. The first involves metamaterial design optimization for metal sheet architecture with binary design variables. The second pertains to optimizing process parameters in multi-stage metal forming, where the discrete nature arises owing to changing tool geometries across stages. This process is highly nonlinear, accumulating contact, geometric, and material nonlinear effects discretely through forming stages. The efficacy of the proposed optimization method, utilizing ANN surrogate models, is compared with traditionally used polynomial response surface (PRS) surrogate models, primarily based on low-order polynomials. Efficient learning of ANN surrogate models is facilitated through the FEM and Python integration framework. Initial data for surrogate model training is collected via Latin hypercube sampling and FEM simulations. DDE is employed for sequential approximate optimization, optimizing ANN or PRS surrogate models to determine optimal design variables. PRS surrogate models encounter challenges in dealing with nonlinear changes in sequential approximate optimization concerning discrete characteristics such as binary design variables and discrete nonlinear behavior found in multi-stage metal forming processes. Owing to the discrete nature, PRS surrogate models require more data and iterations for optimal design variables. In contrast, ANN surrogate models adeptly predict nonlinear behavior through the activation function's characteristics. In the optimization problem of Metal Sheet Architecture for design target C, the ANN surrogate model required an average of 4.6 times fewer iterations to satisfy stopping criteria compared to the PRS surrogate model. Furthermore, in the optimization of multi-stage deep drawing processes, the ANN surrogate model required an average of 6.1 times fewer iterations to satisfy stopping criteria compared to the PRS surrogate model. As a result, the sequential global optimization method utilizing ANN surrogate models achieves optimal design variables with fewer iterations than PRS surrogate models. Further confirmation of the method's efficiency is provided by comparing Pearson correlation coefficients and locus plots.
本文介绍了一种结合有限元法（FEM）、动态微分进化法（DDE）和人工神经网络（ANN）代理模型的顺序近似优化方法。所开发的方法被用于解决两个优化问题。第一个问题涉及具有二进制设计变量的金属板结构的超材料设计优化。第二个问题涉及多阶段金属成型工艺参数的优化，由于各阶段的工具几何形状不断变化，因此产生了离散性。这一过程是高度非线性的，通过成型阶段离散地积累接触、几何和材料非线性效应。利用 ANN 代理模型的拟议优化方法与传统使用的多项式响应面 (PRS) 代理模型（主要基于低阶多项式）的功效进行了比较。通过有限元和 Python 集成框架，促进了 ANN 代理模型的高效学习。通过拉丁超立方采样和有限元模拟收集用于代用模型训练的初始数据。采用 DDE 进行顺序近似优化，优化 ANN 或 PRS 代理模型，以确定最佳设计变量。PRS 代用模型在处理顺序近似优化中的非线性变化时遇到了挑战，这些非线性变化涉及离散特性，如二进制设计变量和多阶段金属成型过程中的离散非线性行为。由于离散性，PRS 代用模型需要更多的数据和迭代来优化设计变量。相比之下，ANN 代理模型能通过激活函数的特性来预测非线性行为。在针对设计目标 C 的金属板材结构优化问题中，与 PRS 代理模型相比，ANN 代理模型为满足停止标准所需的迭代次数平均减少了 4.6 倍。此外，在多阶段拉深工艺的优化中，与 PRS 代理模型相比，ANN 代理模型为满足停止标准所需的迭代次数平均减少了 6.1 倍。因此，与 PRS 代用模型相比，利用 ANN 代用模型的顺序全局优化方法以更少的迭代次数实现了最优设计变量。通过比较皮尔逊相关系数和定位图，进一步证实了该方法的效率。
{"title":"Artificial neural network-based sequential approximate optimization of metal sheet architecture and forming process","authors":"Seong-Sik Han, Heung-Kyu Kim","doi":"10.1093/jcde/qwae049","DOIUrl":"https://doi.org/10.1093/jcde/qwae049","url":null,"abstract":"\u0000 This paper introduces a sequential approximate optimization method that combines the finite element method (FEM), dynamic differential evolution (DDE), and artificial neural network (ANN) surrogate models. The developed method is applied to address two optimization problems. The first involves metamaterial design optimization for metal sheet architecture with binary design variables. The second pertains to optimizing process parameters in multi-stage metal forming, where the discrete nature arises owing to changing tool geometries across stages. This process is highly nonlinear, accumulating contact, geometric, and material nonlinear effects discretely through forming stages. The efficacy of the proposed optimization method, utilizing ANN surrogate models, is compared with traditionally used polynomial response surface (PRS) surrogate models, primarily based on low-order polynomials. Efficient learning of ANN surrogate models is facilitated through the FEM and Python integration framework. Initial data for surrogate model training is collected via Latin hypercube sampling and FEM simulations. DDE is employed for sequential approximate optimization, optimizing ANN or PRS surrogate models to determine optimal design variables. PRS surrogate models encounter challenges in dealing with nonlinear changes in sequential approximate optimization concerning discrete characteristics such as binary design variables and discrete nonlinear behavior found in multi-stage metal forming processes. Owing to the discrete nature, PRS surrogate models require more data and iterations for optimal design variables. In contrast, ANN surrogate models adeptly predict nonlinear behavior through the activation function's characteristics. In the optimization problem of Metal Sheet Architecture for design target C, the ANN surrogate model required an average of 4.6 times fewer iterations to satisfy stopping criteria compared to the PRS surrogate model. Furthermore, in the optimization of multi-stage deep drawing processes, the ANN surrogate model required an average of 6.1 times fewer iterations to satisfy stopping criteria compared to the PRS surrogate model. As a result, the sequential global optimization method utilizing ANN surrogate models achieves optimal design variables with fewer iterations than PRS surrogate models. Further confirmation of the method's efficiency is provided by comparing Pearson correlation coefficients and locus plots.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":null,"pages":null},"PeriodicalIF":4.9,"publicationDate":"2024-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141098841","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：