Article
Construction & Building Technology
Xudong Jian, Ye Xia, Shouwang Sun, Limin Sun
Summary: This paper presents a novel bridge weigh-in-motion (BWIM) method that integrates deep-learning-based computer vision technique and bridge influence surface theory to tackle the complicated traffic problems. The computer vision technique is used to detect and track vehicles and axles, obtaining the spatio-temporal paths of vehicle loads on the bridge. The proposed method improves the existing BWIM technique with respect to complicated traffic scenarios.
STRUCTURAL CONTROL & HEALTH MONITORING
(2022)
Article
Engineering, Civil
Liangfu Ge, Danhui Dan, Zijia Liu, Xin Ruan
Summary: This paper proposes a TFL monitoring system that integrates the functions of machine vision and weigh-in-motion system. Using deep learning methods, the system accurately detects vehicles and wheels in videos and extracts key parameters for TFL modelling. The proposed method achieves accurate TFL simulation with lower measurement error and higher time measurement resolution compared to existing methods, and has great potential for engineering applications.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Daniel Cantero
Summary: This paper presents a revised theoretical interpretation for analyzing bridge response signals during vehicle passages. The theory is applicable for Bridge Weigh-in-Motion applications and has been validated numerically through examples and case studies.
ENGINEERING STRUCTURES
(2021)
Article
Engineering, Civil
Bence Szinyeri, Bence Kovari, Istvan Volgyi, Denes Kollar, Attila Laszlo Joo
Summary: This paper explores the application of deep learning in Bridge Weigh-In-Motion systems and provides a comprehensive dataset for benchmarking static algorithms and deep learning-based methods. It also proposes a deep learning-based solution that demonstrates promising results on the dataset and the applicability of the dataset.
ENGINEERING STRUCTURES
(2023)
Article
Engineering, Electrical & Electronic
Takaya Kawakatsu, Kenro Aihara, Atsuhiro Takasu, Tomonori Nagayama, Jun Adachi
Summary: The detection of heavy vehicles in the road system is an urgent issue for law enforcement and road health monitoring. Bridge Weigh-in-Motion (BWIM) is a cost-effective and easy-to-install technology that utilizes bridge components as weight scales. This article proposes a data-driven BWIM approach using a neural network, which optimizes model parameters through video analysis and vehicle identification to accurately estimate vehicle weights considering various traffic conditions.
IEEE SENSORS JOURNAL
(2023)
Article
Engineering, Civil
Liangfu Ge, Ki Young Koo, Miaomin Wang, James Brownjohn, Danhui Dan
Summary: This study validates a high-precision vision-based Displacement Influence Line (DIL) measurement system for detecting damages on bridges. The system combines two Computer Vision subsystems and weigh-in-motion (WIM) devices, and it successfully assesses damage existence and localization. The study also proposes a Chordwise Displacement Influence Line (cw-DIL) approach to compensate for the adverse effects of friction on boundary supports.
ENGINEERING STRUCTURES
(2023)
Article
Engineering, Civil
E. J. OBrien, J. M. W. Brownjohn, D. Hester, F. Huseynov, M. Casero
Summary: This study focuses on the application of Bridge Weigh-in-Motion systems for bridge condition assessment. A new method for detecting bridge damage using support rotation measurements is proposed. Results show that a rotation-based B-WIM system will continuously overestimate the weight of traversing vehicles in case of bridge damage.
JOURNAL OF CIVIL STRUCTURAL HEALTH MONITORING
(2021)
Article
Construction & Building Technology
Yufeng Zhang, Junxin Xie, Jiayi Peng, Hui Li, Yong Huang
Summary: In this study, a vehicle re-identification method based on Deep Learning and HardNet is proposed for structural health monitoring of bridge structures. Comparative experiments demonstrate the superior performance of this method on vehicle image data.
ADVANCES IN STRUCTURAL ENGINEERING
(2021)
Article
Environmental Sciences
Dongdong Zhao, Wei He, Lu Deng, Yuhan Wu, Hong Xie, Jianjun Dai
Summary: Traditional BWIM methods face challenges in solving the inverse problem, especially in situations where vehicles do not maintain a constant speed. The proposed new method improves accuracy and stability by associating bridge response and axle load with their accurate positions to estimate bridge influence line and axle weight.
Article
Engineering, Mechanical
Xiang Xiao, Dongping Pi, Qing Zhu
Summary: Accurate weight information of freight vehicles is crucial for the assessment and maintenance of railway infrastructures. This paper proposes a novel bridge weigh-in-motion (BWIM) algorithm for fast-passing railway freight vehicles, utilizing the dynamic coupling of the VB system. The proposed algorithm establishes an extended state-space model and utilizes an SPKF algorithm to identify mass parameters, thereby validating its effectiveness.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Multidisciplinary
Ning Hou, Limin Sun, Lin Chen
Summary: This study focuses on modeling vehicle load based on long-term weigh-in-motion data from a long-span cable-stayed bridge, highlighting the importance of using data from operational conditions for safety assessments of bridges. The established model found that concentrated load was significantly higher than values typically used in the design stage, emphasizing the need for more accurate vehicle load models in safety evaluations.
Article
Engineering, Civil
Jan Kalin, Ales Znidaric, Andrej Anzlin, Maja Kreslin
Summary: A new method for calculating the Dynamic Amplification Factor (DAF) has been developed, and data from 15 different bridges has been analyzed, showing a decrease in DAF with increasing axle load. This is significant for the assessment of existing structures, as it is beneficial to optimize structural analysis using measured parameters.
STRUCTURE AND INFRASTRUCTURE ENGINEERING
(2022)
Article
Construction & Building Technology
Yiqing Dong, Dalei Wang, Yue Pan, Yunlong Ma
Summary: This study proposes a vehicle load monitoring system for entire long-span bridge decks, which uses multivision image pre-processing, a modified YOLO-v4 model, a kinematics-enhanced vehicle tracking algorithm, and a data fusion method between vision and Weigh-In-Motion subsystems. The system was tested on a long-span bridge using six cameras, achieving vehicle monitoring of the entire deck with a multi-vehicle tracking precision of 99.28% and a mean Average Precision (mAP) of 96.2% based on the YOLO-v4 model. Comparative results show that the modified YOLO-v4 model outperforms state-of-the-art approaches, and our proposed tracking method surpasses other methods. Our proposed system offers a comprehensive solution for vehicle load monitoring on entire bridge decks, overcoming the limitations of existing methods. Future work could extend the system's capability to include complex traffic patterns.
AUTOMATION IN CONSTRUCTION
(2023)
Review
Engineering, Multidisciplinary
Debojyoti Paul, Koushik Roy
Summary: Bridge health monitoring (BHM) is essential for maintaining transportation networks worldwide. Traditional visual inspection methods have limitations such as bias, inaccessibility, and the need for physical presence. Vibration-based methods require artificial excitation. Therefore, researchers have explored the potential of bridge weigh-in-motion (B-WIM) systems as an alternative. B-WIM systems have advantages such as durability, portability, easy installation, and provide weight estimation and structural information, making them cost-effective compared to standalone BHM systems.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Civil
Ethan MacLeod, Kaveh Arjomandi
Summary: This study discusses the implementation of a novel acceleration-based vehicle identification method in a hybrid bridge weigh-in-motion system. Through a comprehensive evaluation using field study data, it was found that this method can provide more accurate velocity estimation, axle identification, and gross vehicle weight estimation.
JOURNAL OF BRIDGE ENGINEERING
(2022)
Article
Chemistry, Analytical
Ao Wang, Zongkai Zhang, Xiaoming Lei, Ye Xia, Limin Sun
Summary: Thermal energy exchange induces non-uniform temperature distribution on concrete bridge structures, affecting their static and dynamic properties. A systematic all-weather thermal simulation method was proposed to study the temperature distributions of concrete maglev bridges, accurately simulating solar shadow distribution. The simulation method showed higher accuracy under overcast or rainy weather, highlighting the importance of considering solar radiation effects in structural temperature prediction.
Article
Environmental Sciences
Ye Xia, Xiaoming Lei, Peng Wang, Limin Sun
Summary: This paper proposes an artificial intelligence-based methodology for condition assessment of regional bridges, which includes data integration, condition assessment, and maintenance optimization. The regional bridge condition deterioration model is established by neural networks, and the impact of the maintenance scheme on the future condition of bridges is quantified.
Article
Construction & Building Technology
Ye Xia, Xiaoming Lei, Peng Wang, Limin Sun
Summary: This study proposes a comprehensive data-driven framework for network-level bridge condition assessment. By analyzing periodic bridge inspection reports in China, the future condition of bridges can be predicted, and maintenance strategies can be optimized to meet economic and performance constraints.
STRUCTURAL CONTROL & HEALTH MONITORING
(2022)
Article
Construction & Building Technology
Fengyuan Wu, Chenglin Feng, Ye Xia
Summary: The Wufengshan Yangtze River Bridge, a thousand-meter high-speed railway suspension bridge, adopts innovative technologies to fill the gaps in the field and sets a model for the construction of high-speed railway suspension bridges in China and the world.
STRUCTURAL ENGINEERING INTERNATIONAL
(2022)
Article
Chemistry, Multidisciplinary
Fengzong Gong, Fei Han, Yingjie Wang, Ye Xia
Summary: The paper introduces a method to extract bridge damping values from a VBI system for evaluation. By simplifying the VBI system with a double-beam theoretical model and utilizing the extended dynamic stiffness method, a damping ratio equation for the simplified VBI system was obtained. Comparison with more complex models showed the accuracy of the simplified method in extracting bridge damping ratios.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Interdisciplinary Applications
Xiaoming Lei, Ye Xia, Lu Deng, Limin Sun
Summary: This study develops a deep reinforcement learning-based framework to optimize the life-cycle maintenance strategies of regional bridges. The framework is able to generate multiple optimal maintenance strategies that match budget constraints and maximize the cost-effectiveness of maintenance actions.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2022)
Article
Construction & Building Technology
Xudong Jian, Zhilu Lai, Ye Xia, Limin Sun
Summary: This paper aims to improve the performance of bridge weigh in motion (BWIM) in weighing individual axle loads. A novel robust BWIM algorithm is proposed based on regularized total least squares and constraints on the relationship between axles. Experimental results show that the proposed algorithm significantly outperforms other existing BWIM algorithms in terms of accuracy and robustness in identifying individual axle weight.
STRUCTURAL CONTROL & HEALTH MONITORING
(2022)
Article
Construction & Building Technology
Lanxin Luo, Ye Xia, Ao Wang, Xiaoming Lei, Xudong Jian, Limin Sun
Summary: This study proposes a new method for updating FE models in real time by directly applying massive structural and traffic data, achieving good results. Experimental results show that the updated FE model significantly improves in both dynamic and static aspects, with an average 75% reduction in static error indexes.
STRUCTURAL CONTROL & HEALTH MONITORING
(2022)
Article
Construction & Building Technology
Xudong Jian, Ye Xia, Shouwang Sun, Limin Sun
Summary: This paper presents a novel bridge weigh-in-motion (BWIM) method that integrates deep-learning-based computer vision technique and bridge influence surface theory to tackle the complicated traffic problems. The computer vision technique is used to detect and track vehicles and axles, obtaining the spatio-temporal paths of vehicle loads on the bridge. The proposed method improves the existing BWIM technique with respect to complicated traffic scenarios.
STRUCTURAL CONTROL & HEALTH MONITORING
(2022)
Article
Engineering, Mechanical
Xudong Jian, Ye Xia, Limin Sun
Summary: This study conducts a 3D simulation of vehicle-bridge interaction and proposes a novel frequency-domain method to identify bridge natural frequencies. The proposed method has the potential to fast inspect large quantities of bridges, making it valuable for indirect bridge structural identification.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Construction & Building Technology
Shiyang Zhang, Changzheng Feng, Yaoyao Fan, Ye Xia
Summary: Fair-faced concrete is a form of expression in architectural modernism, where no decoration is added to the surface of the structure and the natural color and texture of the concrete itself are used as the final design. While there are technical challenges in applying fair-faced concrete to complex structures, this report explores measures to improve its feasibility and economy in such structures.
STRUCTURAL ENGINEERING INTERNATIONAL
(2023)
Article
Engineering, Multidisciplinary
Xudong Jian, Ye Xia, Eleni Chatzi, Zhilu Lai
Summary: The identification of influence surfaces for bridge structures using calibration tests and deep learning methods is important for understanding traffic loads and assessing structural conditions.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Zhilu Lai, Wei Liu, Xudong Jian, Kiran Bacsa, Limin Sun, Eleni Chatzi
Summary: The article proposes a framework called neural modal ordinary differential equations to integrate physics-based modeling with deep learning for modeling the dynamics of monitored and high-dimensional engineered systems. The research focuses on linear or mildly nonlinear systems and combines dynamic variational autoencoders with physics-informed neural ODEs.
DATA-CENTRIC ENGINEERING
(2022)