A boundary enhancement and gaussian mixture model jointly optimized oversampling algorithm (BE-G-SMOTE) is proposed to improve diagnostic accuracy under imbalanced bearing fault data conditions. It is designed to solve the problem that the diversity of samples generated by the original SMOTE model is limited, as well as the deep learning model is limited by the size of training samples and processing speed. Firstly, a few bearing fault data are clustered by G to achieve cluster division. Secondly, according to the cluster density distribution function designed in this paper, determine the weights of different clusters and sample weights to achieve intra-class balance and improve data quality. Then, to take full advantage of the limited fault data, based on the sensitivity of the support vector machine (SVM) to imbalanced data, the enhanced boundary is established between generated data and the SVM classifier under different penalty factor (PF) values. According to the accuracy, the optimal PF is determined, and fault datasets satisfying diversity are obtained. To improve the classification accuracy, a convolutional neural network with an attention mechanism (CNN-AM) is built. Finally, analysis using two practical cases shows the effectiveness of the proposed method.
{"title":"A diagnosis method for imbalanced bearing data based on improved SMOTE model combined with CNN-AM","authors":"Zhenya Wang, Tao Liu, Xing Wu, Chang Liu","doi":"10.1093/jcde/qwad081","DOIUrl":"https://doi.org/10.1093/jcde/qwad081","url":null,"abstract":"\u0000 A boundary enhancement and gaussian mixture model jointly optimized oversampling algorithm (BE-G-SMOTE) is proposed to improve diagnostic accuracy under imbalanced bearing fault data conditions. It is designed to solve the problem that the diversity of samples generated by the original SMOTE model is limited, as well as the deep learning model is limited by the size of training samples and processing speed. Firstly, a few bearing fault data are clustered by G to achieve cluster division. Secondly, according to the cluster density distribution function designed in this paper, determine the weights of different clusters and sample weights to achieve intra-class balance and improve data quality. Then, to take full advantage of the limited fault data, based on the sensitivity of the support vector machine (SVM) to imbalanced data, the enhanced boundary is established between generated data and the SVM classifier under different penalty factor (PF) values. According to the accuracy, the optimal PF is determined, and fault datasets satisfying diversity are obtained. To improve the classification accuracy, a convolutional neural network with an attention mechanism (CNN-AM) is built. Finally, analysis using two practical cases shows the effectiveness of the proposed method.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"1 1","pages":"1930-1940"},"PeriodicalIF":4.9,"publicationDate":"2023-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79005274","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}
With the increasing expectations of clients and the growing complexity of the built environment, property management teams are facing constant pressure to effectively manage and rectify defects for improved building operational efficiency and performance. This study aims to develop and validate a defect correlation evaluation model for project and property management professionals by specifically (i) examining the defect detection and management mechanisms of residential buildings and (ii) quantifying the mechanical characteristics of defects by using association rules mining (ARM) techniques. In addressing the limitations of current evaluation approaches, this study proposed an ARM evaluation model that integrated, contextualized, and operationalized building defects into work type, location, elements, and defect type. The association between these classifications was explored and mapped. Among the resulting 123 meaningful rules, rules occurred at a rate of about 62% of the same work type in the linked work type, nearly 193% of the same element in another element, and about 23% of the close location in the far location. In conclusion, this study informs project and property management professionals of the key and complex associations between defects of different characteristics and highlights the most common occurrence defects in residential apartment buildings. Thus, this helps reduce the ambiguity and subjectivity of prioritization in defect management and facilitates maintenance and repair planning.
{"title":"Data-mining-based identification of post-handover defect association rules in apartment housings","authors":"Byeol Kim, B. Lim, B. Oo, Yonghan Ahn","doi":"10.1093/jcde/qwad080","DOIUrl":"https://doi.org/10.1093/jcde/qwad080","url":null,"abstract":"\u0000 With the increasing expectations of clients and the growing complexity of the built environment, property management teams are facing constant pressure to effectively manage and rectify defects for improved building operational efficiency and performance. This study aims to develop and validate a defect correlation evaluation model for project and property management professionals by specifically (i) examining the defect detection and management mechanisms of residential buildings and (ii) quantifying the mechanical characteristics of defects by using association rules mining (ARM) techniques. In addressing the limitations of current evaluation approaches, this study proposed an ARM evaluation model that integrated, contextualized, and operationalized building defects into work type, location, elements, and defect type. The association between these classifications was explored and mapped. Among the resulting 123 meaningful rules, rules occurred at a rate of about 62% of the same work type in the linked work type, nearly 193% of the same element in another element, and about 23% of the close location in the far location. In conclusion, this study informs project and property management professionals of the key and complex associations between defects of different characteristics and highlights the most common occurrence defects in residential apartment buildings. Thus, this helps reduce the ambiguity and subjectivity of prioritization in defect management and facilitates maintenance and repair planning.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"9 1","pages":"1838-1855"},"PeriodicalIF":4.9,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84371705","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}
Jongmin Park, Jinoh Yoo, Taehyung Kim, J. Ha, Byeng D. Youn
Fault diagnosis of rolling element bearings (REBs), one type of essential mechanical element, has been actively researched; recent research has focused on the use of deep-learning-based approaches. However, conventional deep-learning-based fault-diagnosis approaches are vulnerable to various operating speeds, which greatly affect the vibration characteristics of the system studied. To solve this problem, previous deep-learning-based studies have usually been carried out by increasing the complexity of the model or diversifying the task of the model. Still, limitations remain because the reason of increasing complexity is unclear and the roles of multiple tasks are not well-defined. Therefore, this study proposes a multi-head de-noising autoencoder-based multitask (MDAM) model for robust diagnosis of REBs under various speed conditions. The proposed model employs a multi-head de-noising autoencoder and multi-task learning strategy to robustly extract features under various speed conditions, while effectively disentangling the speed- and fault-related information. In this research, we evaluate the proposed method using the signals measured from bearing experiments under various speed conditions. The results of the evaluation study show that the proposed method outperformed conventional methods, especially when the training and test datasets have large discrepancies in their operating conditions.
{"title":"Multi-head de-noising autoencoder-based multi-task model for fault diagnosis of rolling element bearings under various speed conditions","authors":"Jongmin Park, Jinoh Yoo, Taehyung Kim, J. Ha, Byeng D. Youn","doi":"10.1093/jcde/qwad076","DOIUrl":"https://doi.org/10.1093/jcde/qwad076","url":null,"abstract":"\u0000 Fault diagnosis of rolling element bearings (REBs), one type of essential mechanical element, has been actively researched; recent research has focused on the use of deep-learning-based approaches. However, conventional deep-learning-based fault-diagnosis approaches are vulnerable to various operating speeds, which greatly affect the vibration characteristics of the system studied. To solve this problem, previous deep-learning-based studies have usually been carried out by increasing the complexity of the model or diversifying the task of the model. Still, limitations remain because the reason of increasing complexity is unclear and the roles of multiple tasks are not well-defined. Therefore, this study proposes a multi-head de-noising autoencoder-based multitask (MDAM) model for robust diagnosis of REBs under various speed conditions. The proposed model employs a multi-head de-noising autoencoder and multi-task learning strategy to robustly extract features under various speed conditions, while effectively disentangling the speed- and fault-related information. In this research, we evaluate the proposed method using the signals measured from bearing experiments under various speed conditions. The results of the evaluation study show that the proposed method outperformed conventional methods, especially when the training and test datasets have large discrepancies in their operating conditions.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"5 1","pages":"1804-1820"},"PeriodicalIF":4.9,"publicationDate":"2023-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90856648","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}
Iranian saffron products hold a unique position in the global market as the most highly valued agricultural and medicinal commodities. The various uses of saffron make it clear that there is a need for special attention to the supply chain network. Unfortunately, the absence of an integrated supply chain network within the saffron industry has resulted in significant challenges related to supply management and demand fulfillment. Addressing real-world uncertainties is paramount when developing models for optimization problems. Therefore, this research proposes a multi-objective optimization model for designing a saffron supply chain network under uncertainty. The model objectives are to decrease the total cost of the supply chain, increase job opportunities and economic development in regions, and improve the quality of products. The proposed mathematical model is solved using the interactive fuzzy method to deal with multiple functions. Furthermore, possibilistic chance constrained programming is employed to effectively manage uncertain variables such as demand, cost, and social parameters within the model. To demonstrate the applicability and validity of the proposed model and solution method, a real case study was conducted in Khorasan Razavi province, Iran. Additionally, because of the complexity of the proposed model in large-scale networks, NSGA-II and MOSA algorithms are proposed. Different parameters are analyzed to determine their impact on the results so that decision-makers can choose values more accurately. The sensitivity analysis and statistical tests performed on the results support the performance of the proposed model. Overall, the results demonstrate that the exact method and metaheuristic algorithms are capable of solving the problem in different dimensions. The computational results derived from this model offer invaluable managerial insights, empowering decision-makers to align their strategies and preferences more effectively.
{"title":"Metaheuristic algorithms for a sustainable saffron supply chain network considering government policies and product quality under uncertainty","authors":"F. Sogandi, M. Shiri","doi":"10.1093/jcde/qwad079","DOIUrl":"https://doi.org/10.1093/jcde/qwad079","url":null,"abstract":"\u0000 Iranian saffron products hold a unique position in the global market as the most highly valued agricultural and medicinal commodities. The various uses of saffron make it clear that there is a need for special attention to the supply chain network. Unfortunately, the absence of an integrated supply chain network within the saffron industry has resulted in significant challenges related to supply management and demand fulfillment. Addressing real-world uncertainties is paramount when developing models for optimization problems. Therefore, this research proposes a multi-objective optimization model for designing a saffron supply chain network under uncertainty. The model objectives are to decrease the total cost of the supply chain, increase job opportunities and economic development in regions, and improve the quality of products. The proposed mathematical model is solved using the interactive fuzzy method to deal with multiple functions. Furthermore, possibilistic chance constrained programming is employed to effectively manage uncertain variables such as demand, cost, and social parameters within the model. To demonstrate the applicability and validity of the proposed model and solution method, a real case study was conducted in Khorasan Razavi province, Iran. Additionally, because of the complexity of the proposed model in large-scale networks, NSGA-II and MOSA algorithms are proposed. Different parameters are analyzed to determine their impact on the results so that decision-makers can choose values more accurately. The sensitivity analysis and statistical tests performed on the results support the performance of the proposed model. Overall, the results demonstrate that the exact method and metaheuristic algorithms are capable of solving the problem in different dimensions. The computational results derived from this model offer invaluable managerial insights, empowering decision-makers to align their strategies and preferences more effectively.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"1964 1","pages":"1892-1929"},"PeriodicalIF":4.9,"publicationDate":"2023-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82730042","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 Production Scheduling (PS) problem is a challenging task that involves assigning manufacturing resources to jobs while ensuring that all constraints are satisfied. The key difficulty in PS is determining the appropriate order of operations. In this study, we propose a novel optimization algorithm called the Quantum-inspired African Vultures Optimization Algorithm with an Elite Mutation Strategy (QEMAVOA) to address this issue. QEMAVOA is an enhanced version of the African Vulture Optimization Algorithm (AVOA) that incorporates three new improvement strategies. Firstly, to enhance QEMAVOA's diversification ability, the population diversity is enriched by the introduction of Quantum Double-Chain Encoding (QDCE) in the initialization phase of QEMAVOA. Secondly, the implementation of the Quantum Rotating Gate (QRG) will balance QEMAVOA's diversification and exploitation capabilities, leading the vulture to a better solution. Finally, with the purpose of improving the exploitability of QEMAVOA, the Elite Mutation (EM) strategy is introduced. To evaluate the performance of QEMAVOA, we apply it to two benchmark scheduling problems: Flexible Job Shop Scheduling (FJSP) and Parallel Machine Scheduling (PMS). The results are compared to those of existing algorithms in the literature. The test results reveal that QEMAVOA surpasses comparison algorithms in accuracy, stability, and speed of convergence.
{"title":"Quantum-inspired African vultures optimization algorithm with elite mutation strategy for production scheduling problems","authors":"Bo Liu, Yongquan Zhou, Qifang Luo, Huajuan Huang","doi":"10.1093/jcde/qwad078","DOIUrl":"https://doi.org/10.1093/jcde/qwad078","url":null,"abstract":"\u0000 The Production Scheduling (PS) problem is a challenging task that involves assigning manufacturing resources to jobs while ensuring that all constraints are satisfied. The key difficulty in PS is determining the appropriate order of operations. In this study, we propose a novel optimization algorithm called the Quantum-inspired African Vultures Optimization Algorithm with an Elite Mutation Strategy (QEMAVOA) to address this issue. QEMAVOA is an enhanced version of the African Vulture Optimization Algorithm (AVOA) that incorporates three new improvement strategies. Firstly, to enhance QEMAVOA's diversification ability, the population diversity is enriched by the introduction of Quantum Double-Chain Encoding (QDCE) in the initialization phase of QEMAVOA. Secondly, the implementation of the Quantum Rotating Gate (QRG) will balance QEMAVOA's diversification and exploitation capabilities, leading the vulture to a better solution. Finally, with the purpose of improving the exploitability of QEMAVOA, the Elite Mutation (EM) strategy is introduced. To evaluate the performance of QEMAVOA, we apply it to two benchmark scheduling problems: Flexible Job Shop Scheduling (FJSP) and Parallel Machine Scheduling (PMS). The results are compared to those of existing algorithms in the literature. The test results reveal that QEMAVOA surpasses comparison algorithms in accuracy, stability, and speed of convergence.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"48 1","pages":"1767-1789"},"PeriodicalIF":4.9,"publicationDate":"2023-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80994839","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 proposes a differential evolution algorithm with improved crossover operation (ICRDE) to deal with combined heat and power dynamic economic dispatch (CHPDED) problems with wind power. First, the improved crossover operation is used to maintain the population diversity by using original individuals, first mutated individuals, and second mutated individuals. Second, the scaling factor and weighted factor are incorporated into the mutation operation to improve the convergence efficiency of the algorithm. Third, adaptive control parameters are introduced to balance local exploitation and global exploration. Moreover, after being updated by the mutation and crossover operation of ICRDE at each generation, the solutions of ICRDE will be further amended using a constraint handling method, which improves the chance of acquiring feasible solutions. Experimental results demonstrate that ICRDE has strong global optimization ability and surpasses the compared algorithms for the CEC2017 benchmark functions, the combined heat and power economic dispatch (CHPED) problems, and the CHPDED problem with and without wind power.
{"title":"Differential evolution algorithm with improved crossover operation for combined heat and power economic dynamic dispatch problem with wind power","authors":"Mengdi Li, D. Zou, H. Ouyang","doi":"10.1093/jcde/qwad077","DOIUrl":"https://doi.org/10.1093/jcde/qwad077","url":null,"abstract":"\u0000 This paper proposes a differential evolution algorithm with improved crossover operation (ICRDE) to deal with combined heat and power dynamic economic dispatch (CHPDED) problems with wind power. First, the improved crossover operation is used to maintain the population diversity by using original individuals, first mutated individuals, and second mutated individuals. Second, the scaling factor and weighted factor are incorporated into the mutation operation to improve the convergence efficiency of the algorithm. Third, adaptive control parameters are introduced to balance local exploitation and global exploration. Moreover, after being updated by the mutation and crossover operation of ICRDE at each generation, the solutions of ICRDE will be further amended using a constraint handling method, which improves the chance of acquiring feasible solutions. Experimental results demonstrate that ICRDE has strong global optimization ability and surpasses the compared algorithms for the CEC2017 benchmark functions, the combined heat and power economic dispatch (CHPED) problems, and the CHPDED problem with and without wind power.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"169 1","pages":"1821-1837"},"PeriodicalIF":4.9,"publicationDate":"2023-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80637026","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 recent years, the Sine Cosine Algorithm (SCA) has become one of the popular swarm intelligence algorithms due to its simple and convenient structure. However, the standard SCA tends to fall into the local optimum when solving complex multimodal tasks, leading to unsatisfactory results. Therefore, this study presents the SCA with communication and quality enhancement, called CCEQSCA. The proposed algorithm includes two enhancement strategies: the communication and collaboration strategy (CC) and the quality enhancement strategy (EQ). In the proposed algorithm, CC strengthens the connection of SCA populations by guiding the search agents closer to the range of optimal solutions. EQ improves the quality of candidate solutions to enhance the exploitation of the algorithm. Furthermore, EQ can explore potential candidate solutions in other scopes, thus strengthening the ability of the algorithm to prevent trapping in the local optimum. To verify the capability of CCEQSCA, 30 functions from the IEEE CEC2017 are analyzed. The proposed algorithm is compared with 5 advanced original algorithms and 10 advanced variants. The outcomes indicate that it is dominant over other comparison algorithms in global optimization tasks. The work in this paper is also utilized to tackle three typical engineering design problems with excellent optimization capabilities. It has been experimentally demonstrated that CCEQSCA works as an effective tool to tackle real issues with constraints and complex search space.
{"title":"Sine cosine algorithm with communication and quality enhancement: Performance design for engineering problems","authors":"Helong Yu, Zisong Zhao, Jing Zhou, Ali Asghar Heidari, Huiling Chen","doi":"10.1093/jcde/qwad073","DOIUrl":"https://doi.org/10.1093/jcde/qwad073","url":null,"abstract":"\u0000 In recent years, the Sine Cosine Algorithm (SCA) has become one of the popular swarm intelligence algorithms due to its simple and convenient structure. However, the standard SCA tends to fall into the local optimum when solving complex multimodal tasks, leading to unsatisfactory results. Therefore, this study presents the SCA with communication and quality enhancement, called CCEQSCA. The proposed algorithm includes two enhancement strategies: the communication and collaboration strategy (CC) and the quality enhancement strategy (EQ). In the proposed algorithm, CC strengthens the connection of SCA populations by guiding the search agents closer to the range of optimal solutions. EQ improves the quality of candidate solutions to enhance the exploitation of the algorithm. Furthermore, EQ can explore potential candidate solutions in other scopes, thus strengthening the ability of the algorithm to prevent trapping in the local optimum. To verify the capability of CCEQSCA, 30 functions from the IEEE CEC2017 are analyzed. The proposed algorithm is compared with 5 advanced original algorithms and 10 advanced variants. The outcomes indicate that it is dominant over other comparison algorithms in global optimization tasks. The work in this paper is also utilized to tackle three typical engineering design problems with excellent optimization capabilities. It has been experimentally demonstrated that CCEQSCA works as an effective tool to tackle real issues with constraints and complex search space.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"22 1","pages":"1868-1891"},"PeriodicalIF":4.9,"publicationDate":"2023-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89471275","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}
Hyungjung Kim, Hyunsub Lee, Semin Ahn, Woo-Kyun Jung, Sung-hoon Ahn
Sewing defect detection is an essential step in garment production quality control. Although sewing defects significantly influence the quality of clothing, they are yet to be studied widely compared to fabric defects. In this study, to address sewing defect detection and develop an appropriate method for small and labor-intensive garment companies, an on-machine broken stitch detection system is proposed. In hardware, a versatile mounting kit, including clamping, display, and adjustable linkage for a camera, is presented for easy installation on a typical industrial sewing machine and for placing the camera close to the sewing position. Additionally, a prototype is implemented using a low-cost single-board computer, Raspberry Pi 4 B, its camera, and Python language. For automated broken stitch detection, a method is proposed that includes removing the texture of the background fabric, image processing in the HSV color space, and edge detection for robust broken detection under various fabric and thread colors and lighting conditions. The proposed system demonstrates reasonable real-time detection accuracy. The maximum accuracy obtained on a sewing stitch dataset with 880 images and on-site tests of various industrial sewing machines is 82.5%, which is 12.1–34.6% higher than that of the two existing methods.
{"title":"Broken stitch detection system for industrial sewing machines using HSV color space and image processing techniques","authors":"Hyungjung Kim, Hyunsub Lee, Semin Ahn, Woo-Kyun Jung, Sung-hoon Ahn","doi":"10.1093/jcde/qwad069","DOIUrl":"https://doi.org/10.1093/jcde/qwad069","url":null,"abstract":"\u0000 Sewing defect detection is an essential step in garment production quality control. Although sewing defects significantly influence the quality of clothing, they are yet to be studied widely compared to fabric defects. In this study, to address sewing defect detection and develop an appropriate method for small and labor-intensive garment companies, an on-machine broken stitch detection system is proposed. In hardware, a versatile mounting kit, including clamping, display, and adjustable linkage for a camera, is presented for easy installation on a typical industrial sewing machine and for placing the camera close to the sewing position. Additionally, a prototype is implemented using a low-cost single-board computer, Raspberry Pi 4 B, its camera, and Python language. For automated broken stitch detection, a method is proposed that includes removing the texture of the background fabric, image processing in the HSV color space, and edge detection for robust broken detection under various fabric and thread colors and lighting conditions. The proposed system demonstrates reasonable real-time detection accuracy. The maximum accuracy obtained on a sewing stitch dataset with 880 images and on-site tests of various industrial sewing machines is 82.5%, which is 12.1–34.6% higher than that of the two existing methods.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"36 1","pages":"1602-1614"},"PeriodicalIF":4.9,"publicationDate":"2023-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88788818","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}
Yagiz Kayali, A. Gleadall, V. Silberschmidt, E. Demirci
Fibrous networks are abundant in nature and commonly used in industry. However, their geometrical modelling is challenging due to their complex microstructure. In this study, a novel method, called Fibre Placement Method (FPM), is developed. In contrast to the existing methods, the FPM has various advantages, such as a fully parametric definition of structure. Also, this method is superior in mimicking the stochastic microstructure of fibrous networks compared to other schemes. Various fibrous networks can be generated easily by employing a user-friendly graphical user interface (GUI). Also, the generated fibrous networks are compatible with analysis software such as computer-aided engineering (CAE) tools. Finally, this algorithm characterises various features of networks including uniformity, void area fraction, and average curliness.
{"title":"Advance algorithm for two-dimensional fibrous-network generation","authors":"Yagiz Kayali, A. Gleadall, V. Silberschmidt, E. Demirci","doi":"10.1093/jcde/qwad074","DOIUrl":"https://doi.org/10.1093/jcde/qwad074","url":null,"abstract":"\u0000 Fibrous networks are abundant in nature and commonly used in industry. However, their geometrical modelling is challenging due to their complex microstructure. In this study, a novel method, called Fibre Placement Method (FPM), is developed. In contrast to the existing methods, the FPM has various advantages, such as a fully parametric definition of structure. Also, this method is superior in mimicking the stochastic microstructure of fibrous networks compared to other schemes. Various fibrous networks can be generated easily by employing a user-friendly graphical user interface (GUI). Also, the generated fibrous networks are compatible with analysis software such as computer-aided engineering (CAE) tools. Finally, this algorithm characterises various features of networks including uniformity, void area fraction, and average curliness.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"37 1","pages":"1790-1803"},"PeriodicalIF":4.9,"publicationDate":"2023-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89619712","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}
Neural network (NN) ensembles can reduce large prediction variance of NN and improve prediction accuracy. For highly nonlinear problems with insufficient data set, the prediction accuracy of NN models becomes unstable, resulting in a decrease in the accuracy of ensembles. Therefore, this study proposes a prediction frequency-based ensemble that identifies core prediction values, which are core prediction members to be used in the ensemble and are expected to be concentrated near the true response. The prediction frequency-based ensemble classifies core prediction values supported by multiple NN models by conducting statistical analysis with a frequency distribution, which is a collection of prediction values obtained from various NN models for a given prediction point. The prediction frequency-based ensemble searches for a range of prediction values that contains prediction values above a certain frequency, and thus the predictive performance can be improved by excluding prediction values with low accuracy and coping with the uncertainty of the most frequent value. An adaptive sampling strategy that sequentially adds samples based on the core prediction variance calculated as the variance of the core prediction values is proposed to improve the predictive performance of the prediction frequency-based ensemble efficiently. Results of various case studies show that the prediction accuracy of the prediction frequency-based ensemble is higher than that of Kriging and other existing ensemble methods. In addition, the proposed adaptive sampling strategy effectively improves the predictive performance of the prediction frequency-based ensemble compared with the previously developed space-filling and prediction variance-based strategies.
{"title":"Adaptive neural network ensemble using prediction frequency","authors":"Ungki Lee, Namwoo Kang","doi":"10.1093/jcde/qwad071","DOIUrl":"https://doi.org/10.1093/jcde/qwad071","url":null,"abstract":"\u0000 Neural network (NN) ensembles can reduce large prediction variance of NN and improve prediction accuracy. For highly nonlinear problems with insufficient data set, the prediction accuracy of NN models becomes unstable, resulting in a decrease in the accuracy of ensembles. Therefore, this study proposes a prediction frequency-based ensemble that identifies core prediction values, which are core prediction members to be used in the ensemble and are expected to be concentrated near the true response. The prediction frequency-based ensemble classifies core prediction values supported by multiple NN models by conducting statistical analysis with a frequency distribution, which is a collection of prediction values obtained from various NN models for a given prediction point. The prediction frequency-based ensemble searches for a range of prediction values that contains prediction values above a certain frequency, and thus the predictive performance can be improved by excluding prediction values with low accuracy and coping with the uncertainty of the most frequent value. An adaptive sampling strategy that sequentially adds samples based on the core prediction variance calculated as the variance of the core prediction values is proposed to improve the predictive performance of the prediction frequency-based ensemble efficiently. Results of various case studies show that the prediction accuracy of the prediction frequency-based ensemble is higher than that of Kriging and other existing ensemble methods. In addition, the proposed adaptive sampling strategy effectively improves the predictive performance of the prediction frequency-based ensemble compared with the previously developed space-filling and prediction variance-based strategies.","PeriodicalId":48611,"journal":{"name":"Journal of Computational Design and Engineering","volume":"29 1","pages":"1547-1560"},"PeriodicalIF":4.9,"publicationDate":"2023-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79282737","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：