Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102922
Helin Li , Shufeng Zheng , Yonghao Shen , Minghai Han , Rui Zhang , Huadong Zhao
In the context of frequent accidents during hydro-steel structures (HSS) operations due to harsh environments and extended service conditions, a novel approach is proposed to reduce the frequency of structural failure incidents and ensure safe and reliable operation. The approach begins with introducing a comprehensive DT modeling framework. Subsequently, detailed DT modeling and DT-based SHM methods are developed. Finally, a platform with perception, interaction, analysis, and decision-making for intelligent health monitoring and maintenance of HSS is constructed and validated in China’s large-scale reservoir project, Luhun Reservoir. The platform includes functions of condition monitoring, fault feature recognition, health status assessment, and maintenance strategies optimization. The integration of DT technology has led to significant improvements in health monitoring and maintenance quality, which includes data collection, model optimization, comprehensive evaluation, and decision-making. This approach has also demonstrated its effectiveness by reducing the operation and maintenance response time and enhancing the overall efficiency and reliability.
{"title":"Hydro-steel structure digital twins: Application in structural health monitoring and maintenance of large-scale reservoir","authors":"Helin Li , Shufeng Zheng , Yonghao Shen , Minghai Han , Rui Zhang , Huadong Zhao","doi":"10.1016/j.aei.2024.102922","DOIUrl":"10.1016/j.aei.2024.102922","url":null,"abstract":"<div><div>In the context of frequent accidents during hydro-steel structures (HSS) operations due to harsh environments and extended service conditions, a novel approach is proposed to reduce the frequency of structural failure incidents and ensure safe and reliable operation. The approach begins with introducing a comprehensive DT modeling framework. Subsequently, detailed DT modeling and DT-based SHM methods are developed. Finally, a platform with perception, interaction, analysis, and decision-making for intelligent health monitoring and maintenance of HSS is constructed and validated in China’s large-scale reservoir project, Luhun Reservoir. The platform includes functions of condition monitoring, fault feature recognition, health status assessment, and maintenance strategies optimization. The integration of DT technology has led to significant improvements in health monitoring and maintenance quality, which includes data collection, model optimization, comprehensive evaluation, and decision-making. This approach has also demonstrated its effectiveness by reducing the operation and maintenance response time and enhancing the overall efficiency and reliability.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102922"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142586519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102931
Shuo Pan , Zhuo Liu , Hai Yan , Ning Chen , Xiaoxiong Zhao , Sichun Li , Frank Witlox
Urban streets exhibit a diverse range of characteristics, with some presenting significant challenges to ambulance passage, directly impacting the safety of residents. Thus, ensuring unimpeded passage for ambulances on streets is a key focus of urban renewal and street governance initiatives. However, the identification of bottlenecks for emergency vehicle passage on urban streets currently relies on labor-intensive and inefficient on-site manual audits. This study proposes a deep learning-based approach to achieve automatic identification of ambulance passage on urban streets. The Vision Transformer network is utilized to construct the classification model of Impassable Narrow Roads, Passable Narrow Roads, and Wide Roads based on street view images. To train and test the constructed models, a specialized dataset is established, consisting of street view images labeled by experienced ambulance drivers. Comparative experiments are conducted to confirm the optimal structure of the model and the necessity of semantic segmentation preprocessing for street view images. To confirm the superiority of the proposed approach, four commonly used deep learning methods, MobileNet, ShuffleNet, SuperViT and DualViT serve as the baseline tests. Experimental results reveal that the model with four-head and one sequential encoder achieves the highest evaluation accuracy at 75.65% among the proposed models on the original dataset, significantly outperforming benchmark models. Meanwhile, the segmentation of street view images improves accuracy to 77.42%, but it reduces computational efficiency from 0.01 to 3 seconds per image. Finally, the optimal model is applied to the area within the Second Ring Road of Beijing as an example to discuss how the deep learning-based approach proposed in this paper supports urban planning practice and emergency medical response. The proposed approach facilitates the rapid and large-scale identification of bottlenecks in urban streets for ambulances with very limited costs, making a significant contribution to the accurate identification of key areas for urban renewal and street governance efforts. The proposed method can further assist emergency vehicle dispatchers and drivers in identifying accessible routes with greater precision during operations, thereby enabling more timely transportation of patients to medical facilities.
{"title":"Automatic identification of bottlenecks for ambulance passage on urban streets: A deep learning-based approach","authors":"Shuo Pan , Zhuo Liu , Hai Yan , Ning Chen , Xiaoxiong Zhao , Sichun Li , Frank Witlox","doi":"10.1016/j.aei.2024.102931","DOIUrl":"10.1016/j.aei.2024.102931","url":null,"abstract":"<div><div>Urban streets exhibit a diverse range of characteristics, with some presenting significant challenges to ambulance passage, directly impacting the safety of residents. Thus, ensuring unimpeded passage for ambulances on streets is a key focus of urban renewal and street governance initiatives. However, the identification of bottlenecks for emergency vehicle passage on urban streets currently relies on labor-intensive and inefficient on-site manual audits. This study proposes a deep learning-based approach to achieve automatic identification of ambulance passage on urban streets. The Vision Transformer network is utilized to construct the classification model of Impassable Narrow Roads, Passable Narrow Roads, and Wide Roads based on street view images. To train and test the constructed models, a specialized dataset is established, consisting of street view images labeled by experienced ambulance drivers. Comparative experiments are conducted to confirm the optimal structure of the model and the necessity of semantic segmentation preprocessing for street view images. To confirm the superiority of the proposed approach, four commonly used deep learning methods, MobileNet, ShuffleNet, SuperViT and DualViT serve as the baseline tests. Experimental results reveal that the model with four-head and one sequential encoder achieves the highest evaluation accuracy at 75.65% among the proposed models on the original dataset, significantly outperforming benchmark models. Meanwhile, the segmentation of street view images improves accuracy to 77.42%, but it reduces computational efficiency from 0.01 to 3 seconds per image. Finally, the optimal model is applied to the area within the Second Ring Road of Beijing as an example to discuss how the deep learning-based approach proposed in this paper supports urban planning practice and emergency medical response. The proposed approach facilitates the rapid and large-scale identification of bottlenecks in urban streets for ambulances with very limited costs, making a significant contribution to the accurate identification of key areas for urban renewal and street governance efforts. The proposed method can further assist emergency vehicle dispatchers and drivers in identifying accessible routes with greater precision during operations, thereby enabling more timely transportation of patients to medical facilities.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102931"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102913
Akram Akbar , Chun Liu , Hangbin Wu , Shoujun Jia , Zeran Xu
Real 3D building models have become indispensable data sources for building spatial information bases for smart cities by leveraging structural correlations and rich semantic expressions of real-world scene entities. The essential prerequisite for real 3D reconstruction is real-time and dynamic detailed-level observations. Low-altitude multicopter UAV platforms are optimal for automatic and periodic building scene observations. However, there are still several challenges in UAV-based path planning for real 3D data capture while maintaining the overall fidelity of architectural details due to observational scale variations, surrounding uncertainties, structural complexity, and topological delicacy. We propose a scene information guided aerial photogrammetric mission recomposition method in response to this challenge. Depending on the architectural complexity, the two proposed observation patterns, parallel inspection and surface enveloping, can be recomposed to achieve UAV obstacle avoidance and complete coverage of individual buildings in a restricted space, capturing global surface detail with millimeter resolution and low texture distortion. The virtual simulation environment, which is constructed based on the semantics and elevation values of the surroundings, provides a basis for selecting the observation pattern and optimal flight parameters based on the reconstruction requirements of the building. In order to achieve quality control of 3D reconstruction models, this paper introduces a reconstruction quality assessment scheme consisting of four quantitative evaluation metrics, namely coverage, resolution distribution, texture distortion score, and geometric accuracy, which effectively establishes a close relationship between mission planning and 3D reconstruction. The observation capability of the proposed method is better than other typical observation patterns, obtaining a model of globally homogeneous resolution distribution over the main body of the building, reaching an average level of 7.01 mm and the highest level of 2.12 mm (façade region), which can provide high-quality data for the semantic extraction and instantiation of multiple surface elements of buildings.
{"title":"Scene information guided aerial photogrammetric mission recomposition towards detailed level building reconstruction","authors":"Akram Akbar , Chun Liu , Hangbin Wu , Shoujun Jia , Zeran Xu","doi":"10.1016/j.aei.2024.102913","DOIUrl":"10.1016/j.aei.2024.102913","url":null,"abstract":"<div><div>Real 3D building models have become indispensable data sources for building spatial information bases for smart cities by leveraging structural correlations and rich semantic expressions of real-world scene entities. The essential prerequisite for real 3D reconstruction is real-time and dynamic detailed-level observations. Low-altitude multicopter UAV platforms are optimal for automatic and periodic building scene observations. However, there are still several challenges in UAV-based path planning for real 3D data capture while maintaining the overall fidelity of architectural details due to observational scale variations, surrounding uncertainties, structural complexity, and topological delicacy. We propose a scene information guided aerial photogrammetric mission recomposition method in response to this challenge. Depending on the architectural complexity, the two proposed observation patterns, parallel inspection and surface enveloping, can be recomposed to achieve UAV obstacle avoidance and complete coverage of individual buildings in a restricted space, capturing global surface detail with millimeter resolution and low texture distortion. The virtual simulation environment, which is constructed based on the semantics and elevation values of the surroundings, provides a basis for selecting the observation pattern and optimal flight parameters based on the reconstruction requirements of the building. In order to achieve quality control of 3D reconstruction models, this paper introduces a reconstruction quality assessment scheme consisting of four quantitative evaluation metrics, namely coverage, resolution distribution, texture distortion score, and geometric accuracy, which effectively establishes a close relationship between mission planning and 3D reconstruction. The observation capability of the proposed method is better than other typical observation patterns, obtaining a model of globally homogeneous resolution distribution over the main body of the building, reaching an average level of 7.01 mm and the highest level of 2.12 mm (façade region), which can provide high-quality data for the semantic extraction and instantiation of multiple surface elements of buildings.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102913"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102928
Shifan Qiao , Haoyu Li , S. Thomas Ng , Junkun Tan , Yingyu Tang , Baoquan Cheng
Predicting ground settlement and identifying key influential factors during shield tunneling in dense karst areas presents a significant engineering challenge due to irregular geological conditions and the complex nonlinear interactions among multiple factors. Traditional computational methods and existing machine learning models often lack either accuracy or interpretability, limiting their practical application in such environments. To address this gap, a novel global sensitivity analysis (GSA) framework has been developed, specifically tailored for dense karst areas. This framework integrates eXtreme Gradient Boosting (XGBoost) as an interpretable metamodel enhanced with SHAP analysis and combines it with the Sobol method for comprehensive sensitivity quantification. In addition, this framework incorporates integrated detection methods and karst structural parameters to ensure its applicability in dense karst construction environments. By applying this framework to actual data from the Shenzhen Metro Line 14 project, key tunneling parameters such as synchronous grouting pressure, actual excavation volume, karst cross-section total area, and karst-to-tunnel distance were accurately identified as having a significant impact on ground settlement. This approach fills a critical research gap by providing an interpretable and accurate tool for shield tunneling in dense karst areas, ultimately improving safety and efficiency in these challenging environments.
{"title":"XGBoost-based global sensitivity analysis of ground settlement caused by shield tunneling in dense karst areas","authors":"Shifan Qiao , Haoyu Li , S. Thomas Ng , Junkun Tan , Yingyu Tang , Baoquan Cheng","doi":"10.1016/j.aei.2024.102928","DOIUrl":"10.1016/j.aei.2024.102928","url":null,"abstract":"<div><div>Predicting ground settlement and identifying key influential factors during shield tunneling in dense karst areas presents a significant engineering challenge due to irregular geological conditions and the complex nonlinear interactions among multiple factors. Traditional computational methods and existing machine learning models often lack either accuracy or interpretability, limiting their practical application in such environments. To address this gap, a novel global sensitivity analysis (GSA) framework has been developed, specifically tailored for dense karst areas. This framework integrates eXtreme Gradient Boosting (XGBoost) as an interpretable metamodel enhanced with SHAP analysis and combines it with the Sobol method for comprehensive sensitivity quantification. In addition, this framework incorporates integrated detection methods and karst structural parameters to ensure its applicability in dense karst construction environments. By applying this framework to actual data from the Shenzhen Metro Line 14 project, key tunneling parameters such as synchronous grouting pressure, actual excavation volume, karst cross-section total area, and karst-to-tunnel distance were accurately identified as having a significant impact on ground settlement. This approach fills a critical research gap by providing an interpretable and accurate tool for shield tunneling in dense karst areas, ultimately improving safety and efficiency in these challenging environments.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102928"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Automated vision defect detection is a crucial step in monitoring product quality in industrial production. Despite the widespread utilization of deep learning methods for surface defect identification, several challenges persist in the context of gear applications. Firstly, there is a lack of dedicated defect detection methods specifically tailored for gear tooth surfaces. As surface defects vary in size, the regular single-scale attention computation at each transformer layer tends to compromise spatial information. To address these challenges, we first propose a novel U-shaped spatial-attention transformer model for tooth surface detection. A shunted-window method is introduced to create a pyramid receptive field within a single self-attention layer. This method captures fine-grained features with a small window while preserving coarse-grained features with a larger window. Consequently, this technique enables effective multi-scale information fusion, accommodating objects of different sizes. We curate a dataset of defective samples collected under various working conditions using the CL-100 gear wear machine. Experimental results demonstrate that the proposed model outperforms the state-of-the-art (SOTA) U-shaped SwinUnet by +8.74% AP and +4.40% Sm, while surpassing the excellent defect detection method of ResT-UperNet by +0.63% AP and +4.69% Sm.
自动视觉缺陷检测是监控工业生产中产品质量的关键步骤。尽管深度学习方法已广泛应用于表面缺陷识别，但在齿轮应用方面仍存在一些挑战。首先，缺乏专门针对齿轮齿面的专用缺陷检测方法。由于表面缺陷大小不一，在每个变压器层进行常规的单尺度注意力计算往往会损害空间信息。为了应对这些挑战，我们首先提出了一种用于齿面检测的新型 U 形空间注意力变压器模型。我们引入了一种分流窗口方法，在单个自我注意层内创建一个金字塔形的感受野。这种方法用小窗口捕捉细粒度特征，同时用大窗口保留粗粒度特征。因此，这种技术能有效地进行多尺度信息融合，以适应不同大小的物体。我们利用 CL-100 磨齿机收集了不同工作条件下的缺陷样本数据集。实验结果表明，所提出的模型在 AP 和 Sm 方面分别比最先进（SOTA）的 U 型 SwinUnet 高出+8.74%和+4.40%，同时在 AP 和 Sm 方面分别比 ResT-UperNet 高出+0.63%和+4.69%。
{"title":"A Unet-inspired spatial-attention transformer model for segmenting gear tooth surface defects","authors":"Xin Zhou , Yongchao Zhang , Zhaohui Ren , Tianchuan Mi , Zeyu Jiang , Tianzhuang Yu , Shihua Zhou","doi":"10.1016/j.aei.2024.102933","DOIUrl":"10.1016/j.aei.2024.102933","url":null,"abstract":"<div><div>Automated vision defect detection is a crucial step in monitoring product quality in industrial production. Despite the widespread utilization of deep learning methods for surface defect identification, several challenges persist in the context of gear applications. Firstly, there is a lack of dedicated defect detection methods specifically tailored for gear tooth surfaces. As surface defects vary in size, the regular single-scale attention computation at each transformer layer tends to compromise spatial information. To address these challenges, we first propose a novel U-shaped spatial-attention transformer model for tooth surface detection. A shunted-window method is introduced to create a pyramid receptive field within a single self-attention layer. This method captures fine-grained features with a small window while preserving coarse-grained features with a larger window. Consequently, this technique enables effective multi-scale information fusion, accommodating objects of different sizes. We curate a dataset of defective samples collected under various working conditions using the CL-100 gear wear machine. Experimental results demonstrate that the proposed model outperforms the state-of-the-art (SOTA) U-shaped SwinUnet by +8.74% AP and +4.40% Sm, while surpassing the excellent defect detection method of ResT-UperNet by +0.63% AP and +4.69% Sm.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102933"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This research pioneers the application of precursor concepts to preemptively identify and prevent aviation safety incidents using Machine Learning (ML). Airlines and governing organizations, such as the Federal Aviation Administration (FAA) in the United States, have been trying to prevent safety incidents during routine operations. However, this task is challenging due to the lack of timestep-wise event annotation in flights and the complexity involved in the timely identification of incidents prior to their occurrence. To address these issues, we propose a real-time precursor identification methodology combining Multiple-Instance Learning (MIL) and feature-based Knowledge Distillation (KD) learning. Our two-stage approach, involving deep MIL for labeling and a KD-based model for real-time warnings, demonstrates state-of-the-art performance and a time delay of 2.99ms using a dataset of 23,549 real flights. Further experiments using t-distributed Stochastic Neighbor Embedding (t-SNE) and occlusion method confirm our model’s transparency, enabling the generation of reliable quantitative precursor scores and facilitating reasoning about the causes of safety incidents at the parameter level. Additionally, statistical analysis of precursors reveals varying evolution times for different safety events, which indicates that pilots have at least 8 s to react after receiving a warning. In conclusion, our research provides a theoretical foundation and technical support for the next generation of online risk warning systems, enhancing flight safety and offering a pathway towards more intelligent and secure flight operations.
本研究率先应用前兆概念，利用机器学习（ML）技术预先识别和预防航空安全事故。航空公司和管理机构，如美国联邦航空管理局（FAA），一直在努力预防日常运营中的安全事故。然而，由于缺乏按时间顺序排列的飞行事件注释，以及在事件发生前及时识别事件的复杂性，这项任务极具挑战性。为了解决这些问题，我们提出了一种结合多实例学习（MIL）和基于特征的知识蒸馏（KD）学习的实时前兆识别方法。我们的两阶段方法包括用于标记的深度 MIL 和用于实时警告的基于 KD 的模型，在使用 23,549 次真实航班的数据集时，表现出了一流的性能和 2.99 毫秒的时间延迟。使用 t 分布随机邻域嵌入（t-SNE）和闭塞方法进行的进一步实验证实了我们模型的透明度，能够生成可靠的定量前兆分数，并有助于在参数级别推理安全事故的原因。此外，对前兆的统计分析显示，不同安全事件的演变时间各不相同，这表明飞行员在收到警告后至少有 8 秒钟的反应时间。总之，我们的研究为下一代在线风险预警系统提供了理论基础和技术支持，提高了飞行安全，为实现更智能、更安全的飞行操作提供了途径。
{"title":"Real-time identification of precursors in commercial aviation using multiple-instance learning","authors":"Zhiwei Xiang , Zhenxing Gao , Yansong Gao , Yangyang Zhang , Runhao Zhang","doi":"10.1016/j.aei.2024.102856","DOIUrl":"10.1016/j.aei.2024.102856","url":null,"abstract":"<div><div>This research pioneers the application of precursor concepts to preemptively identify and prevent aviation safety incidents using Machine Learning (ML). Airlines and governing organizations, such as the Federal Aviation Administration (FAA) in the United States, have been trying to prevent safety incidents during routine operations. However, this task is challenging due to the lack of timestep-wise event annotation in flights and the complexity involved in the timely identification of incidents prior to their occurrence. To address these issues, we propose a real-time precursor identification methodology combining Multiple-Instance Learning (MIL) and feature-based Knowledge Distillation (KD) learning. Our two-stage approach, involving deep MIL for labeling and a KD-based model for real-time warnings, demonstrates state-of-the-art performance and a time delay of 2.99ms using a dataset of 23,549 real flights. Further experiments using t-distributed Stochastic Neighbor Embedding (t-SNE) and occlusion method confirm our model’s transparency, enabling the generation of reliable quantitative precursor scores and facilitating reasoning about the causes of safety incidents at the parameter level. Additionally, statistical analysis of precursors reveals varying evolution times for different safety events, which indicates that pilots have at least 8 s to react after receiving a warning. In conclusion, our research provides a theoretical foundation and technical support for the next generation of online risk warning systems, enhancing flight safety and offering a pathway towards more intelligent and secure flight operations.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102856"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142417049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102837
Jianbo Yu , Zhaomin Lv , Hang Ruan , Shijie Hu , Qingchao Jiang , Xuefeng Yan , Yuping Liu , Xiaofeng Yang
Due to the increasing complexity of variable relationships, fault detection has garnered significant attention, as it is crucial for ensuring industrial safety and engineering reliability. Traditional detection methods can be classified as twofold: global-based and local-based strategies, which respectively focus on mining macro- and micro-level information. However, our theoretical derivation and experiment results reveal that some spurious assumptions, such as local groups and their provided information are mutually independent are implicitly adhered to but are hardly satisfied in unsupervised fault detection under real industrial scenarios. Hence, this study introduces a novel mutual stacked autoencoder (M-SAE) which can be divided into three sub-networks: L-Net, R-Net, and M-Net. L-Net enriches local information learning through multiple local backbones by incorporating the unsupervised clustering algorithm. R-Net, employing a multi-scale attention mechanism, leverages complete local information for residual strength calculation and utilizes local features to capture residual information within the latent feature space. M-Net fuses the multi-scale local feature information to perform a reconstruction for each local. A multitask entropy-aided loss function is introduced to enrich local details, the global structure, and the residual associations. Finally, results on eleven datasets validate the high-performance of the proposed M-SAE and the ablation experiments demonstrate the efficacy of each component in M-SAE, confirming that this research effectively and accurately addresses multivariable industrial fault detection tasks, thereby enabling timely interventions that are crucial for maintaining operational safety in real-world scenarios.
{"title":"Mutual stacked autoencoder for unsupervised fault detection under complex multi-residual correlations","authors":"Jianbo Yu , Zhaomin Lv , Hang Ruan , Shijie Hu , Qingchao Jiang , Xuefeng Yan , Yuping Liu , Xiaofeng Yang","doi":"10.1016/j.aei.2024.102837","DOIUrl":"10.1016/j.aei.2024.102837","url":null,"abstract":"<div><div>Due to the increasing complexity of variable relationships, fault detection has garnered significant attention, as it is crucial for ensuring industrial safety and engineering reliability. Traditional detection methods can be classified as twofold: global-based and local-based strategies, which respectively focus on mining macro- and micro-level information. However, our theoretical derivation and experiment results reveal that some spurious assumptions, such as local groups and their provided information are mutually independent are implicitly adhered to but are hardly satisfied in unsupervised fault detection under real industrial scenarios. Hence, this study introduces a novel mutual stacked autoencoder (M-SAE) which can be divided into three sub-networks: L-Net, R-Net, and M-Net. L-Net enriches local information learning through multiple local backbones by incorporating the unsupervised clustering algorithm. R-Net, employing a multi-scale attention mechanism, leverages complete local information for residual strength calculation and utilizes local features to capture residual information within the latent feature space. M-Net fuses the multi-scale local feature information to perform a reconstruction for each local. A multitask entropy-aided loss function is introduced to enrich local details, the global structure, and the residual associations. Finally, results on eleven datasets validate the high-performance of the proposed M-SAE and the ablation experiments demonstrate the efficacy of each component in M-SAE, confirming that this research effectively and accurately addresses multivariable industrial fault detection tasks, thereby enabling timely interventions that are crucial for maintaining operational safety in real-world scenarios.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102837"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142417050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102877
Changmo Yeo, Sang-Uk Cheon, Seungeun Lim, Jun Hwan Park, Duhwan Mun
Design features refer to local shapes or regions within a part that perform specific functions such as fastening and force transmission. These design features must be identified from product design results to conduct design verification, manufacturing evaluation, and process planning. Design features are formed by combining various form features, which poses a challenge when using existing methods to retrieve individual features. Therefore, this study introduced a relational descriptor that describes the relational characteristics between topological elements to retrieve design features in boundary representation (B-rep) models. In addition, a method to retrieve design features by combining the relational descriptor with shape descriptors was proposed. Experiments were performed to identify specific design features to validate the proposed method. The experimental results successfully retrieved all the design features included in the B-rep model.
{"title":"Relational descriptors for retrieving design features in a B-rep model using the similarity-based retrieval approach","authors":"Changmo Yeo, Sang-Uk Cheon, Seungeun Lim, Jun Hwan Park, Duhwan Mun","doi":"10.1016/j.aei.2024.102877","DOIUrl":"10.1016/j.aei.2024.102877","url":null,"abstract":"<div><div>Design features refer to local shapes or regions within a part that perform specific functions such as fastening and force transmission. These design features must be identified from product design results to conduct design verification, manufacturing evaluation, and process planning. Design features are formed by combining various form features, which poses a challenge when using existing methods to retrieve individual features. Therefore, this study introduced a relational descriptor that describes the relational characteristics between topological elements to retrieve design features in boundary representation (B-rep) models. In addition, a method to retrieve design features by combining the relational descriptor with shape descriptors was proposed. Experiments were performed to identify specific design features to validate the proposed method. The experimental results successfully retrieved all the design features included in the B-rep model.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102877"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142442571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102821
Taehun Kim , Jin Uk Ko , Jinwook Lee , Yong Chae Kim , Joon Ha Jung , Byeng D. Youn
In the field of fault diagnosis for rotating machinery, where available fault data are limited, numerous studies have employed a generative adversarial network (GAN) for data generation. However, the limited fault data for training GAN exacerbate GAN’s inherent training instability and mode collapse issues, which are induced by adversarial training. Moreover, the stochastic nature of random sampling for latent vectors sampling often results in low-fidelity and poor diversity generation, which negatively affects the fault diagnosis models. To address these issues, this paper presents two novel approaches: a spectrum-guided GAN (SGAN) and density-directionality sampling (DDS). SGAN mitigates training instability and mode collapse through combinatorial data utilization, adversarial spectral loss, and a tailored model structure. DDS ensures the high-fidelity and high-diversity of the generated data by selectively sampling the latent vectors through two steps: density-based filtering and directionality-based sampling in the feature space. Validation on both rotor and rolling element bearing datasets demonstrates that SGAN-DDS considerably improves classification results under the limited fault data. Furthermore, fidelity and diversity analyses are conducted to validate DDS, which increase the credibility of the proposed method; and offer advancement toward the application of deep-learning and GAN in industrial fields.
在旋转机械故障诊断领域，由于可用的故障数据有限，许多研究都采用了生成式对抗网络（GAN）来生成数据。然而，用于训练 GAN 的故障数据有限，这加剧了 GAN 固有的训练不稳定性和模式崩溃问题，而这些问题都是由对抗训练引起的。此外，用于潜在向量采样的随机采样的随机性往往导致生成的数据保真度低、多样性差，从而对故障诊断模型产生负面影响。为了解决这些问题，本文提出了两种新方法：频谱引导 GAN（SGAN）和密度定向采样（DDS）。SGAN 通过组合数据利用、对抗性频谱损失和定制的模型结构来缓解训练不稳定性和模式崩溃。DDS 通过两个步骤对潜在向量进行选择性采样，确保生成数据的高保真和高多样性：在特征空间中进行基于密度的过滤和基于方向性的采样。转子和滚动轴承数据集的验证结果表明，在故障数据有限的情况下，SGAN-DDS 能显著改善分类结果。此外，还进行了保真度和多样性分析来验证 DDS，从而提高了所提方法的可信度，并推动了深度学习和 GAN 在工业领域的应用。
{"title":"Spectrum-guided GAN with density-directionality sampling: Diverse high-fidelity signal generation for fault diagnosis of rotating machinery","authors":"Taehun Kim , Jin Uk Ko , Jinwook Lee , Yong Chae Kim , Joon Ha Jung , Byeng D. Youn","doi":"10.1016/j.aei.2024.102821","DOIUrl":"10.1016/j.aei.2024.102821","url":null,"abstract":"<div><div>In the field of fault diagnosis for rotating machinery, where available fault data are limited, numerous studies have employed a generative adversarial network (GAN) for data generation. However, the limited fault data for training GAN exacerbate GAN’s inherent training instability and mode collapse issues, which are induced by adversarial training. Moreover, the stochastic nature of random sampling for latent vectors sampling often results in low-fidelity and poor diversity generation, which negatively affects the fault diagnosis models. To address these issues, this paper presents two novel approaches: a spectrum-guided GAN (SGAN) and density-directionality sampling (DDS). SGAN mitigates training instability and mode collapse through combinatorial data utilization, adversarial spectral loss, and a tailored model structure. DDS ensures the high-fidelity and high-diversity of the generated data by selectively sampling the latent vectors through two steps: density-based filtering and directionality-based sampling in the feature space. Validation on both rotor and rolling element bearing datasets demonstrates that SGAN-DDS considerably improves classification results under the limited fault data. Furthermore, fidelity and diversity analyses are conducted to validate DDS, which increase the credibility of the proposed method; and offer advancement toward the application of deep-learning and GAN in industrial fields.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102821"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142529668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-01DOI: 10.1016/j.aei.2024.102871
Yuming Liu , Wencai Yu , Qingyuan Lin , Wei Wang , Ende Ge , Aihua Su , Yong Zhao
Assembly is the final step in ensuring the precision and performance of mechanical products. Geometric variables, process variables, and other material or physical variables during the assembly process can all impact the assembly outcome. Therefore, the key for analyzing and predicting assembly results lies in establishing the mapping relationship between various assembly variables and the results. Traditional analysis methods typically consider the evolution of a single variable in relation to the assembly results and often focus on the value at a few nodes. Essentially, this approach constructs a value-to-value nonlinear mapping model, ignoring the coupling relationships between different variables. However, with the increase in assembly precision requirements and advancements in measurement equipment, assembly analysis has evolved from value-to-value prediction to field-to-field prediction. This shift necessitates the study of the assembly physical field results for specific regions rather than focusing on a few nodes. Therefore, this paper proposes an analysis framework, TF-F-GAN (Transformer-based- Field-Generative adversarial network), which is suitable for multi-source assembly variable inputs and physical field outputs. The framework draws inspiration from multimodal fusion and text-image generation models, leveraging the Vision Transformer (VIT) network to integrate multi-source heterogeneous data from the assembly process. The physical field data is color-mapped into a cloud image format, transforming the physical field prediction into a cloud image generation problem. The CFRP bolted joint structure assembly is used as a case study in this paper. Since assembly accuracy primarily focuses on geometric deformation, the deformation field of key regions in the CFRP bolted joint is taken as the output variable. In the case study, the geometric deviations of parts and mechanical behavior during the assembly process were considered. Data augmentation methods were used to construct the dataset. After training TF-F-GAN on this dataset, transfer learning was further conducted using experimental data. The final prediction error of TF-F-GAN relative to the experimental data was less than 15 %, with a computation time of less than 7 s. This prediction framework can serve as an effective tool for predicting the physical fields of general mechanical product assembly.
{"title":"TF-F-GAN: A GAN-based model to predict the assembly physical fields under multi-modal variables fusion on vision transformer","authors":"Yuming Liu , Wencai Yu , Qingyuan Lin , Wei Wang , Ende Ge , Aihua Su , Yong Zhao","doi":"10.1016/j.aei.2024.102871","DOIUrl":"10.1016/j.aei.2024.102871","url":null,"abstract":"<div><div>Assembly is the final step in ensuring the precision and performance of mechanical products. Geometric variables, process variables, and other material or physical variables during the assembly process can all impact the assembly outcome. Therefore, the key for analyzing and predicting assembly results lies in establishing the mapping relationship between various assembly variables and the results. Traditional analysis methods typically consider the evolution of a single variable in relation to the assembly results and often focus on the value at a few nodes. Essentially, this approach constructs a value-to-value nonlinear mapping model, ignoring the coupling relationships between different variables. However, with the increase in assembly precision requirements and advancements in measurement equipment, assembly analysis has evolved from value-to-value prediction to field-to-field prediction. This shift necessitates the study of the assembly physical field results for specific regions rather than focusing on a few nodes. Therefore, this paper proposes an analysis framework, TF-F-GAN (Transformer-based- Field-Generative adversarial network), which is suitable for multi-source assembly variable inputs and physical field outputs. The framework draws inspiration from multimodal fusion and text-image generation models, leveraging the Vision Transformer (VIT) network to integrate multi-source heterogeneous data from the assembly process. The physical field data is color-mapped into a cloud image format, transforming the physical field prediction into a cloud image generation problem. The CFRP bolted joint structure assembly is used as a case study in this paper. Since assembly accuracy primarily focuses on geometric deformation, the deformation field of key regions in the CFRP bolted joint is taken as the output variable. In the case study, the geometric deviations of parts and mechanical behavior during the assembly process were considered. Data augmentation methods were used to construct the dataset. After training TF-F-GAN on this dataset, transfer learning was further conducted using experimental data. The final prediction error of TF-F-GAN relative to the experimental data was less than 15 %, with a computation time of less than 7 s. This prediction framework can serve as an effective tool for predicting the physical fields of general mechanical product assembly.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"62 ","pages":"Article 102871"},"PeriodicalIF":8.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142532088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号