Computer vision-based vehicle load monitoring methods could obtain spatiotemporal data of vehicle loads, which is important for bridge monitoring and operation. However, during the process of vehicle detection and tracking, current research usually focuses on the vehicle as a whole, and there is a lack of research on the accurate positioning of vehicle wheel loads. For the fatigue analysis of orthotropic steel deck, stress at the structural details belongs to the typical third-class system, and related research requires accurate wheel load position. Based on the principle of camera imaging, this study proposes an innovative vehicle wheel load location method based on vehicle license plate detection and multiple projective planes, and the accurate positioning of the vehicle center is achieved by the projective relationship matrix of different planes. Then, accurate measurement of the lateral wheelbase is achieved through secondary detection and projective transformation. Further, accurate wheel load tracking for fatigue research is achieved by the multi-objective tracking algorithm. Based on theoretical analysis and practical application results, the effectiveness and accuracy of this method have been verified. Different from traditional positioning methods based on vehicle detection boxes and 3D reconstruction boxes, the proposed method has higher accuracy and will play a fundamental role in the use of vehicle load spatiotemporal data for more accurate analysis such as fatigue research.
{"title":"Vehicle wheel load positioning method based on multiple projective planes","authors":"Kai Sun, Xu Jiang, Xuhong Qiang","doi":"10.1111/mice.13432","DOIUrl":"https://doi.org/10.1111/mice.13432","url":null,"abstract":"Computer vision-based vehicle load monitoring methods could obtain spatiotemporal data of vehicle loads, which is important for bridge monitoring and operation. However, during the process of vehicle detection and tracking, current research usually focuses on the vehicle as a whole, and there is a lack of research on the accurate positioning of vehicle wheel loads. For the fatigue analysis of orthotropic steel deck, stress at the structural details belongs to the typical third-class system, and related research requires accurate wheel load position. Based on the principle of camera imaging, this study proposes an innovative vehicle wheel load location method based on vehicle license plate detection and multiple projective planes, and the accurate positioning of the vehicle center is achieved by the projective relationship matrix of different planes. Then, accurate measurement of the lateral wheelbase is achieved through secondary detection and projective transformation. Further, accurate wheel load tracking for fatigue research is achieved by the multi-objective tracking algorithm. Based on theoretical analysis and practical application results, the effectiveness and accuracy of this method have been verified. Different from traditional positioning methods based on vehicle detection boxes and 3D reconstruction boxes, the proposed method has higher accuracy and will play a fundamental role in the use of vehicle load spatiotemporal data for more accurate analysis such as fatigue research.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"31 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143050875","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}
X. Tan, W. Wei, C. Liu, K. Cheng, Y. Wang, Z. Yao, Q. Huang
This paper addresses the challenge of real-time, continuous trajectory planning for autonomous excavation. A hybrid method combining particle swarm optimization (PSO) and reinforcement learning (RL) is proposed. First, three types of excavation trajectories are defined for different geometric shapes of the digging area. Then, an excavation trajectory optimization method based on the PSO algorithm is established, resulting in optimal trajectories, the sensitive parameters, and the corresponding variation ranges. Second, an RL model is built, and the optimization results obtained offline are used as training samples. The RL-based method can be applied for continuous digging tasks, which is beneficial for improving the overall efficiency of the autonomous operation of the excavator. Finally, simulation experiments were conducted in four distinct conditions. The results demonstrate that the proposed method effectively accomplishes excavation tasks, with trajectory generation completed within 0.5 s. Comprehensive performance metrics remained below 0.14, and the excavation rate exceeded 92%, surpassing or matching the performance of the optimization-based method and PINN-based method. Moreover, the proposed method produced consistently balanced trajectory performance across all sub-tasks. These results underline the method's effectiveness in achieving real-time, multi-objective, and continuous trajectory planning for autonomous excavators.
{"title":"Reinforcement learning-based trajectory planning for continuous digging of excavator working devices in trenching tasks","authors":"X. Tan, W. Wei, C. Liu, K. Cheng, Y. Wang, Z. Yao, Q. Huang","doi":"10.1111/mice.13428","DOIUrl":"https://doi.org/10.1111/mice.13428","url":null,"abstract":"This paper addresses the challenge of real-time, continuous trajectory planning for autonomous excavation. A hybrid method combining particle swarm optimization (PSO) and reinforcement learning (RL) is proposed. First, three types of excavation trajectories are defined for different geometric shapes of the digging area. Then, an excavation trajectory optimization method based on the PSO algorithm is established, resulting in optimal trajectories, the sensitive parameters, and the corresponding variation ranges. Second, an RL model is built, and the optimization results obtained offline are used as training samples. The RL-based method can be applied for continuous digging tasks, which is beneficial for improving the overall efficiency of the autonomous operation of the excavator. Finally, simulation experiments were conducted in four distinct conditions. The results demonstrate that the proposed method effectively accomplishes excavation tasks, with trajectory generation completed within 0.5 s. Comprehensive performance metrics remained below 0.14, and the excavation rate exceeded 92%, surpassing or matching the performance of the optimization-based method and PINN-based method. Moreover, the proposed method produced consistently balanced trajectory performance across all sub-tasks. These results underline the method's effectiveness in achieving real-time, multi-objective, and continuous trajectory planning for autonomous excavators.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"1 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143050874","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}
Gaining insights into pedestrian flow patterns in public spaces can greatly benefit decision-making processes related to infrastructure planning. Interestingly, even pedestrians are unfamiliar with one another, they often follow others, drawing on positive information and engaging in a spontaneous collective behavior of pedestrians. To model this collective behavior, this paper proposed a social force-based technique characterized by a loosely defined leader–follower structure. First, a complex field-based phase transfer entropy (PTE) method was applied to measure the difference in information flow between pedestrians. Setting the detecting threshold with the 3 sigma principle, the radial basis function (RBF) was utilized to identify the leader in the collective. Integrating the PTE, RBF, and social force model (SFM), a comprehensive model (PTE-RBF-SFM) was developed to simulate collective behavior. Some bidirectional pedestrian flow data, collected from Fairground Düsseldorf, were used to validate the model in a real-world setting. The results showed that the proposed model provided more realistic trajectories than benchmark models, and the spontaneous leader–follower structure was found to change over time and stable with time interval prolonging.
{"title":"Modeling the collective behavior of pedestrians with the spontaneous loose leader–follower structure in public spaces","authors":"Jie Xu, Dengyu Xu, Jing Wu, Xiaowei Shi","doi":"10.1111/mice.13429","DOIUrl":"https://doi.org/10.1111/mice.13429","url":null,"abstract":"Gaining insights into pedestrian flow patterns in public spaces can greatly benefit decision-making processes related to infrastructure planning. Interestingly, even pedestrians are unfamiliar with one another, they often follow others, drawing on positive information and engaging in a spontaneous collective behavior of pedestrians. To model this collective behavior, this paper proposed a social force-based technique characterized by a loosely defined leader–follower structure. First, a complex field-based phase transfer entropy (PTE) method was applied to measure the difference in information flow between pedestrians. Setting the detecting threshold with the 3 sigma principle, the radial basis function (RBF) was utilized to identify the leader in the collective. Integrating the PTE, RBF, and social force model (SFM), a comprehensive model (PTE-RBF-SFM) was developed to simulate collective behavior. Some bidirectional pedestrian flow data, collected from Fairground Düsseldorf, were used to validate the model in a real-world setting. The results showed that the proposed model provided more realistic trajectories than benchmark models, and the spontaneous leader–follower structure was found to change over time and stable with time interval prolonging.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"34 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143044282","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}
Precise three-dimensional (3D) instance segmentation of indoor scenes plays a critical role in civil engineering, including reverse engineering, size detection, and advanced structural analysis. However, existing methods often fall short in accurately segmenting complex indoor environments due to challenges of diverse material textures, irregular object shapes, and inadequate datasets. To address these limitations, this paper introduces StructNet3D, a point cloud neural network specifically designed for instance segmentation in indoor components including ceilings, floors, and walls. StructNet3D employs a novel multi-scale 3D U-Net backbone integrated with ArchExtract, which designed to capture both global context and local structural details, enabling precise segmentation of diverse indoor environments. Compared to other methods, StructNet3D achieved an AP50 of 87.7 on the proprietary dataset and 68.6 on the S3DIS dataset, demonstrating its effectiveness in accurately segmenting and classifying major structural components within diverse indoor environments.
{"title":"Enhanced three-dimensional instance segmentation using multi-feature extracting point cloud neural network","authors":"Hongxu Wang, Jiepeng Liu, Dongsheng Li, Tianze Chen, Pengkun Liu, Han Yan, Yadong Wu","doi":"10.1111/mice.13430","DOIUrl":"https://doi.org/10.1111/mice.13430","url":null,"abstract":"Precise three-dimensional (3D) instance segmentation of indoor scenes plays a critical role in civil engineering, including reverse engineering, size detection, and advanced structural analysis. However, existing methods often fall short in accurately segmenting complex indoor environments due to challenges of diverse material textures, irregular object shapes, and inadequate datasets. To address these limitations, this paper introduces StructNet3D, a point cloud neural network specifically designed for instance segmentation in indoor components including ceilings, floors, and walls. StructNet3D employs a novel multi-scale 3D U-Net backbone integrated with ArchExtract, which designed to capture both global context and local structural details, enabling precise segmentation of diverse indoor environments. Compared to other methods, StructNet3D achieved an AP50 of 87.7 on the proprietary dataset and 68.6 on the S3DIS dataset, demonstrating its effectiveness in accurately segmenting and classifying major structural components within diverse indoor environments.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"58 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143026362","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}
The cover image is based on the article Modeling of spatially embedded networks via regional spatial graph convolutional networks by Jürgen Hackl et al., https://doi.org/10.1111/mice.13286.
{"title":"Cover Image, Volume 40, Issue 4","authors":"","doi":"10.1111/mice.13426","DOIUrl":"https://doi.org/10.1111/mice.13426","url":null,"abstract":"<b>The cover image</b> is based on the article <i>Modeling of spatially embedded networks via regional spatial graph convolutional networks</i> by Jürgen Hackl et al., https://doi.org/10.1111/mice.13286.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"120 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143020912","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}
Particle morphology influences the mechanical behavior of granular soils. Generating particles with realistic shapes for discrete element method simulations is gaining popularity. However, it is still challenging to efficiently generate very angular particles with less computational cost. Addressing this challenge, this paper introduces a novel noise‐based framework for generating realistic soil particle geometry. Noise algorithms are utilized to apply random variations with certain morphological patterns on the surface of the base geometry (e.g., a sphere), thereby generating a variety of particles with morphological patterns ranging from very angular to rounded. In addition, the base geometry can be replaced with other geometries including real particle scans, allowing rapid generation of realistic particles with morphological characteristics of the base geometry. The framework stands out for its simplicity, the wide range of particle morphologies generated, reducing the need for extensive computation and scanning, and provides a new idea for the granular soil behavior simulations.
{"title":"A noise‐based framework for randomly generating soil particle with realistic geometry","authors":"Chen‐Xi Tong, Jia‐Jun Li, Quan Sun, Sheng Zhang, Wan‐Huan Zhou, Daichao Sheng","doi":"10.1111/mice.13424","DOIUrl":"https://doi.org/10.1111/mice.13424","url":null,"abstract":"Particle morphology influences the mechanical behavior of granular soils. Generating particles with realistic shapes for discrete element method simulations is gaining popularity. However, it is still challenging to efficiently generate very angular particles with less computational cost. Addressing this challenge, this paper introduces a novel noise‐based framework for generating realistic soil particle geometry. Noise algorithms are utilized to apply random variations with certain morphological patterns on the surface of the base geometry (e.g., a sphere), thereby generating a variety of particles with morphological patterns ranging from very angular to rounded. In addition, the base geometry can be replaced with other geometries including real particle scans, allowing rapid generation of realistic particles with morphological characteristics of the base geometry. The framework stands out for its simplicity, the wide range of particle morphologies generated, reducing the need for extensive computation and scanning, and provides a new idea for the granular soil behavior simulations.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"44 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142989103","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}
Gyalwang Dhundup, Jianing Zhou, Michael Bekoe, Lijun Sun, Sheng Mao, Yu Yan
Cracks impact the performance and durability of asphalt pavements, necessitating a comprehensive understanding of the mixture cracking behavior. While discrete element modeling has been implemented, many studies oversimplify the simulation of asphalt mortar, a critical component affecting mixture cracking resistance. This study proposes a mortar model that is applicable to both two‐dimensional (2D) and, to a preliminary extent, three‐dimensional (3D) simulations. The model incorporates a geometric representation of mortar distribution and a mechanical softening model to simulate damage accumulation and fracture. Laboratory and virtual Superpave indirect tensile tests were performed on asphalt mixtures with varying gradations at different aging levels. The virtual simulations successfully mirrored indoor test results in volumetric parameters, load–displacement behavior, and stress distribution. Minor differences in strength, strain, and fracture energy between virtual and indoor tests confirmed the accuracy of the mortar model. Notably, the 3D simulations provided a more accurate reconstruction of the cracking process, showing smaller discrepancies between virtual and indoor results, compared to the 2D simulations, with errors in stress, strain, and fracture energy of 5.6%, 5.7%, and 4.7%, respectively. Employing the mortar model in discrete element simulation revealed insights into fracture angle distribution and tendencies, enabling meticulous analysis of mixture damage characteristics and cracking behavior. This allows for the improved design of mixtures with excellent cracking performance and contributes to advancing computational methods that could complement laboratory testing.
{"title":"Integrating a mortar model into discrete element simulation for enhanced understanding of asphalt mixture cracking","authors":"Gyalwang Dhundup, Jianing Zhou, Michael Bekoe, Lijun Sun, Sheng Mao, Yu Yan","doi":"10.1111/mice.13425","DOIUrl":"https://doi.org/10.1111/mice.13425","url":null,"abstract":"Cracks impact the performance and durability of asphalt pavements, necessitating a comprehensive understanding of the mixture cracking behavior. While discrete element modeling has been implemented, many studies oversimplify the simulation of asphalt mortar, a critical component affecting mixture cracking resistance. This study proposes a mortar model that is applicable to both two‐dimensional (2D) and, to a preliminary extent, three‐dimensional (3D) simulations. The model incorporates a geometric representation of mortar distribution and a mechanical softening model to simulate damage accumulation and fracture. Laboratory and virtual Superpave indirect tensile tests were performed on asphalt mixtures with varying gradations at different aging levels. The virtual simulations successfully mirrored indoor test results in volumetric parameters, load–displacement behavior, and stress distribution. Minor differences in strength, strain, and fracture energy between virtual and indoor tests confirmed the accuracy of the mortar model. Notably, the 3D simulations provided a more accurate reconstruction of the cracking process, showing smaller discrepancies between virtual and indoor results, compared to the 2D simulations, with errors in stress, strain, and fracture energy of 5.6%, 5.7%, and 4.7%, respectively. Employing the mortar model in discrete element simulation revealed insights into fracture angle distribution and tendencies, enabling meticulous analysis of mixture damage characteristics and cracking behavior. This allows for the improved design of mixtures with excellent cracking performance and contributes to advancing computational methods that could complement laboratory testing.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"9 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142989527","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}
Quantifying tiny cracks is crucial for assessing structural conditions. Traditional non-contact measurement technologies often struggle to accurately measure tiny crack widths, especially in hard-to-access areas. To address these challenges, this study introduces an image-based, handheld parallel laser line-camera (PLLC) system designed for automated tiny crack localization and width measurement from multiple angles and safe distances. Established by processing parallel laser strips, the camera coordinate system addresses crack positioning and pixel scale distortion challenges typical in non-perpendicular photography. The determined pixel scale enables accurate width measurement. An improved U-Net model automatically identifies crack pixels, enhancing detection accuracy. Additionally, the newly developed Equal Area algorithm enables the sub-pixel width measurement of tiny cracks. Comprehensive laboratory and field testing demonstrates the system's accuracy and feasibility across various conditions. This PLLC system achieves quantitative tiny crack detection in one shot, significantly enhancing the efficiency and utility of on-site inspections.
{"title":"Automatic tiny crack positioning and width measurement with parallel laser line-camera system","authors":"Chaobin Li, R. K. L. Su","doi":"10.1111/mice.13420","DOIUrl":"https://doi.org/10.1111/mice.13420","url":null,"abstract":"Quantifying tiny cracks is crucial for assessing structural conditions. Traditional non-contact measurement technologies often struggle to accurately measure tiny crack widths, especially in hard-to-access areas. To address these challenges, this study introduces an image-based, handheld parallel laser line-camera (PLLC) system designed for automated tiny crack localization and width measurement from multiple angles and safe distances. Established by processing parallel laser strips, the camera coordinate system addresses crack positioning and pixel scale distortion challenges typical in non-perpendicular photography. The determined pixel scale enables accurate width measurement. An improved U-Net model automatically identifies crack pixels, enhancing detection accuracy. Additionally, the newly developed Equal Area algorithm enables the sub-pixel width measurement of tiny cracks. Comprehensive laboratory and field testing demonstrates the system's accuracy and feasibility across various conditions. This PLLC system achieves quantitative tiny crack detection in one shot, significantly enhancing the efficiency and utility of on-site inspections.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"33 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142988135","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}
Yonghui An, Siyuan Gong, Zhongzheng Wang, Wei Shen, Zhihao Wang, Jinping Ou
Suspender cables/hangers occupy a crucial role during the whole service life of suspension bridges/arch bridges/space structures, and their long-term repeated vibration under corrosion and high-stress service state will cause fatigue damage and even induce fatigue failure. To mitigate the vibration of the vertical suspender cables/hangers, a four-wire pendulum tuned mass damper (FWPTMD) is proposed. It mainly consists of the cross bracket, four pendulum ropes, the moving mass, and four universal rotating ball hinges that can rotate in any direction and provide damping. A suspender cable in a real suspension bridge is selected as the research object. First, the design procedure and effect of the modal mass ratio are provided; the optimization design method for parameters of optimal frequency ratio and optimal damping ratio is investigated in detail. Second, simulations are conducted to illustrate its feasibility, and results show excellent vibration mitigation effect. Third, the optimal FWPTMD is designed and fabricated; its performance is further validated by field experiments, and the results are very close to those in simulation. The FWPTMD has the advantages of simple structural form, convenient installation, low cost, easy tuning, easy maintenance, and so forth. Therefore, it can play an obvious vibration mitigation role in the life-cycle of the suspender cable/hanger, and it has a positive meaning to retard fatigue damage, extend the service life, and assure traffic safety under extreme weather.
{"title":"Theoretical analysis, simulation, and field experiment for vibration mitigation of suspender cables/hangers using the four-wire pendulum tuned mass damper","authors":"Yonghui An, Siyuan Gong, Zhongzheng Wang, Wei Shen, Zhihao Wang, Jinping Ou","doi":"10.1111/mice.13412","DOIUrl":"https://doi.org/10.1111/mice.13412","url":null,"abstract":"Suspender cables/hangers occupy a crucial role during the whole service life of suspension bridges/arch bridges/space structures, and their long-term repeated vibration under corrosion and high-stress service state will cause fatigue damage and even induce fatigue failure. To mitigate the vibration of the vertical suspender cables/hangers, a four-wire pendulum tuned mass damper (FWPTMD) is proposed. It mainly consists of the cross bracket, four pendulum ropes, the moving mass, and four universal rotating ball hinges that can rotate in any direction and provide damping. A suspender cable in a real suspension bridge is selected as the research object. First, the design procedure and effect of the modal mass ratio are provided; the optimization design method for parameters of optimal frequency ratio and optimal damping ratio is investigated in detail. Second, simulations are conducted to illustrate its feasibility, and results show excellent vibration mitigation effect. Third, the optimal FWPTMD is designed and fabricated; its performance is further validated by field experiments, and the results are very close to those in simulation. The FWPTMD has the advantages of simple structural form, convenient installation, low cost, easy tuning, easy maintenance, and so forth. Therefore, it can play an obvious vibration mitigation role in the life-cycle of the suspender cable/hanger, and it has a positive meaning to retard fatigue damage, extend the service life, and assure traffic safety under extreme weather.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"53 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142981902","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}
Particle morphology is a crucial factor influencing the mechanical properties of granular materials particularly in infrastructure construction processes where accurate shape descriptors are essential. Accurately measuring three-dimensional (3D) morphology has significant theoretical and practical value for exploring the multiscale mechanical properties of civil engineering materials. This study proposes a novel approach using multiview (two-dimensional [2D]) particle images to efficiently predict 3D morphology, making real-time aggregate quality analysis feasible. A 3D convolutional neural network (CNN) model is developed, which combines Monte Carlo dropout and attention mechanisms to achieve uncertainty-evaluated predictions of 3D morphology. The model incorporates a convolutional block attention module, involving a two-stage attention mechanism with channel attention and spatial attention, to further optimize feature representation and enhance the effectiveness of the attention mechanism. A new dataset comprising 18,000 images of 300 natural gravel and 300 blasted rock fragment particles is used for model training. The prediction accuracy and uncertainty of the proposed model are benchmarked against a range of alternative models including 2D CNN, 3D CNN, and 2D CNN with attention, in particular, to the influence of the number of input multiview particle images on the performance of the models for predicting various morphological parameters is explored. The results indicate that the proposed 3D CNN model with the attention mechanism achieves high prediction accuracy with an error of less than 10%. Whilst it exhibits initially greater uncertainty compared to other models due to its increased complexity, the model shows significant improvement in both accuracy and uncertainty as the number of training images is increased. Finally, residual challenges associated with the prediction of more complex particle angles and irregular shapes are also discussed.
{"title":"Automatic determination of 3D particle morphology from multiview images using uncertainty-evaluated deep learning","authors":"Hongchen Liu, Huaizhi Su, Brian Sheil","doi":"10.1111/mice.13421","DOIUrl":"https://doi.org/10.1111/mice.13421","url":null,"abstract":"Particle morphology is a crucial factor influencing the mechanical properties of granular materials particularly in infrastructure construction processes where accurate shape descriptors are essential. Accurately measuring three-dimensional (3D) morphology has significant theoretical and practical value for exploring the multiscale mechanical properties of civil engineering materials. This study proposes a novel approach using multiview (two-dimensional [2D]) particle images to efficiently predict 3D morphology, making real-time aggregate quality analysis feasible. A 3D convolutional neural network (CNN) model is developed, which combines Monte Carlo dropout and attention mechanisms to achieve uncertainty-evaluated predictions of 3D morphology. The model incorporates a convolutional block attention module, involving a two-stage attention mechanism with channel attention and spatial attention, to further optimize feature representation and enhance the effectiveness of the attention mechanism. A new dataset comprising 18,000 images of 300 natural gravel and 300 blasted rock fragment particles is used for model training. The prediction accuracy and uncertainty of the proposed model are benchmarked against a range of alternative models including 2D CNN, 3D CNN, and 2D CNN with attention, in particular, to the influence of the number of input multiview particle images on the performance of the models for predicting various morphological parameters is explored. The results indicate that the proposed 3D CNN model with the attention mechanism achieves high prediction accuracy with an error of less than 10%. Whilst it exhibits initially greater uncertainty compared to other models due to its increased complexity, the model shows significant improvement in both accuracy and uncertainty as the number of training images is increased. Finally, residual challenges associated with the prediction of more complex particle angles and irregular shapes are also discussed.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"22 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142975485","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}
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：