Bridge Health Monitoring System Research Articles

Explainable machine learning model for load-deformation correlation in long-span suspension bridges using XGBoost-SHAP

The deformation of long-span suspension bridges in multiple loads is an important indictor to reflect their operation state. However, the correlation between multiple loads and structural deformation is difficult to quantify. Therefore, this study proposes an explainable machine learning model for the load-deformation correlation in long-span suspension bridges using eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP). Firstly, the structural health monitoring system for a suspension bridge was used to construct the dataset for the training and testing of XGBoost model. Herein, temperature, wind and vehicle loads were used as the input variables, while midspan deflections and expansion joint displacements were treated as outputs. Subsequently, the hyperparameters of XGBoost model were optimized using grid search and 5-fold cross-validation to ensure its prediction performance. Then, the prediction results were compared with other four machine learning methods (i.e., linear regression, artificial neural networks, gradient boosted decision trees and CatBoost). Finally, the correlation between different loads and displacement responses were explained by the SHAP method to identify the contribution of the loads on deformation. The results show that the XGBoost model has the highest prediction accuracy. Compared to vehicle and wind loads, temperature significantly affects the deformation of long-span suspension bridges during daily operation. The effects of temperature and wind on bridge deformation are independent, and there is no significant interaction between these two factors.

Anomaly detection for bridge health monitoring data based on multiple encoded images and convolutional neural network

In bridge health monitoring systems, abnormal monitoring data inevitably appear due to the complex operational conditions, and these anomalous data will seriously affect the accuracy and reliability of the results of bridge health assessment. Hence, it is vital to detect the abnormal data before further processing. However, the pattern diversity and sample scarcity usually make the accuracy of data anomaly detection unsatisfactory. A novel anomaly detection method for bridge health monitoring data based on encoded images and convolutional neural network (CNN) is proposed in this paper. First, the raw time series are encoded into images to highlight the intrinsic features in the anomaly data, and multiple encoding techniques are adopted to represent the data in different perspectives. Then, the CNN is utilized to establish the mapping between the encoded images and the anomaly patterns for achieving the data anomaly detection. At last, the proposed method is verified with the acceleration data measured from an actual large span cable-stayed bridge. It shows that the proposed method can significantly improve the performance of data anomaly detection while the results are hardly affected by the small sample sizes, and a detection accuracy of over 95% is achieved based on the multiple encoded images.

Deep learning-based approach for identifying vortex-induced vibrations in stay cables

Due to variations in wind speed profiles along the length of bridge stay cables, vortex-induced vibrations (VIV) exhibit multimodal characteristics, presenting challenges for VIV identification. Currently, the VIV identification is concentrated on the stable stage of VIV, lacking an available early warning system for detecting the initial developing stage of VIV. In this study, a deep learning-based approach that integrates energy distribution ratio features derived from frequency band wavelet packet decomposition to recognize VIV of stay cable was proposed. Firstly, vibration characteristics induced by vortices in cable-stayed bridges were analyzed based on field monitoring data from the bridge health monitoring system, aiming to propose suitable feature indicators for VIV identification. Secondly, using root mean square as label classification, a deep learning model was constructed, incorporating convolutional neural networks, long short-term memory networks, and attention mechanisms. Finally, four different stages in the evolution of stay cables VIV were identified utilizing field monitoring datasets to analyze the optimal parameter. Meanwhile, effective early warning recognition was achieved through the classification and recognition of confusion matrix. This study provides technical support for early warning systems and structural condition assessment concerning bridge stay cable VIV.

Hybrid WARIMA-WANN algorithm for data prediction in bridge health monitoring system

Application of data-driven algorithm to model and predict the future trend of the structural health monitoring (SHM) data serves as important references for the future safety evaluation and decision-making of the bridges. An innovative hybrid WARIMA-WANN algorithm combining the Wavelet model, the autoregressive moving average (ARIMA) model and the artificial neural network (ANN) model is proposed to predict massive monitoring data in the bridge SHM system. The WPT serves as the decomposition function for the linear and nonlinear components of the hybrid model. The separated linear/nonlinear components are individually predicted with the ARIMA/ANN models and then combined to obtain the prediction values of the future data. The hybrid method is applied to predict the deflection values of a single-tower cable-stayed bridge. Multiple parameters in individual models are discussed and selected to increase the credibility of the prediction results for the example bridge. The results show that the proposed hybrid algorithm successfully captures the linear and nonlinear coupling relationships inside the SHM data from the frequency perspective of view and shows the best prediction performance when compared with the other four existing models. Different prediction steps obviously influence prediction performance due to the periodic nature of the bridge SHM data.

Enhancing acceleration data cleaning through time–frequency feature integration using deep learning

In recent years, the safety issues of bridge has garnered significant attention from various sectors of society. Bridge structural health monitoring (SHM) systems have emerged as essential tools for real-time monitoring and assessing the operational safety of bridges. Despite the substantial implementation and operation of monitoring systems, challenges have arisen from abnormal data caused by equipment malfunctions, power outages, and environmental interference. These anomalies are challenging to manually remove within the vast dataset. To address the challenge of automated detection and classification of anomalies, this paper focuses on acceleration data and proposes a data cleaning method that integrates time–frequency features and one-dimensional convolutional neural network (1D CNN). The method utilizes extreme value envelopes and wavelet scattering transforms to extract time–frequency features from acceleration data, thereby achieving data dimensionality reduction and feature extraction goals. Subsequently, these features are employed to train and test a 1D CNN. The proposed method’s validation is conducted using acceleration monitoring data obtained from a large-span arch bridge. Through extensive testing, the model demonstrates a classification accuracy exceeding 98% on both the test set and unseen data, showcasing high classification precision and generalization capabilities. This approach provides an effective solution for automated data cleaning and anomaly identification within the context of bridge health monitoring systems.

ANN-Based Bridge Support Fixity Quantification Using Thermal Response Data from Real-Time Wireless Sensing.

Bridges are critical infrastructures that support our economic activities and daily lives. Aging bridges have been a major issue for decades, prompting researchers to improve resilience and performance through structural health monitoring. While most research focuses on superstructure damage, the majority of bridge failures are associated with support or joint damages, indicating the importance of bridge support. Indeed, bridge support affects the performance of both the substructure and superstructure by maintaining the load path and allowing certain movements to mitigate thermal and other stresses. The support deterioration leads to a change in fixity in the superstructure, compromising the bridge's integrity and safety. Hence, a reliable method to determine support fixity level is essential to detecting bearing health and enhancing the accuracy of the bridge health monitoring system. However, such research is lacking because of its complexity. In this study, we developed a support fixity quantification method based on thermal responses using an Artificial Neural Network (ANN) model. A finite element (FE) model of a representative highway bridge is used to derive thermal displacement data under different bearing stiffnesses, superstructure damage, and thermal loading. The thermal displacement behavior of the bridge under different support fixity conditions is presented, and the model is trained on the simulated response. The performance of the developed FE model and ANN was validated with field monitoring data collected from two in-service bridges in Connecticut using a real-time Wireless Sensor Network (WSN). Finally, the support stiffnesses of both bridges were predicted using the ANN model for validation.

Data Processing of Bridge IoT Monitoring Based on Task Scheduling of Cloud Service Listening Signal

For the long-term continuous monitoring of bridge-related indicators, it is necessary to arrange relatively perfect acquisition equipment on the bridge, which can feedback various information parameters of the bridge. However, there are many parameters to feedback the bridge information, which leads to the complex and overstaffed structure of the monitoring system. Furthermore, the huge amount of data collected and the complex calculation process also increase the difficulty of the operation of the monitoring system. In this regard, we should choose more scientific and reasonable indicators, lightweight data structure, stable data transmission, and analysis programs to improve the accuracy of continuous monitoring. To establish a stable and efficient bridge monitoring system, we use the distance coefficient-effective independent algorithm to optimize. Then, we calculate the relevant information of the strain environment with the help of a neural network model, strengthen the training of deep learning through the YOLOv5s model, and improve the task scheduling strategy of attention concentration. Through that, we solve the problem of embedded systems with relatively low computing power. Different weights are assigned to each fused feature map, and the nodes at the highest level and the lowest level are deleted so that a concise and efficient lightweight network model is constructed. Multiple iterations are performed to achieve deeper feature fusion. Therefore, the complexity of the model is effectively reduced, and the monitoring performance can be effectively improved. Finally, through the experimental analysis, it is proved that compared with the traditional fusion model, the number of parameters of the improved fusion network structure in bridge health monitoring is reduced by 7.37%. The detection speed is increased by 18.2%. The amount of computation is reduced by 42.92%, and the average detection accuracy is required to reach 95.33%. It is verified that the proposed method can effectively improve the accuracy and risk control ability of the detection data by learning from the samples with small labels. It also has great practical significance and market value for the design and optimization of the bridge health monitoring system, which is suitable for the monitoring data of large-scale construction projects.

Health monitoring and maintenance of highway and bridge

In recent years, due to the rapid growth of traffic volume, the use frequency and load of Bridges have been significantly increased, and the service state of the bridge structure has deteriorated sharply. At present, there is no corresponding standard specification for the health monitoring and maintenance of Bridges in China, so the safety, durability and normal performance of Bridges are still the focus of people's attention. This paper summarizes the research results of health monitoring and maintenance at home and abroad, summarizes and analyzes the highway bridge health monitoring and maintenance, and introduces the bridge health monitoring system in China.

Whole-process analysis and implementation of a self-powered wireless health monitoring system for railway bridges: Theory, simulation and experiment

In this work, a self-powered wireless health monitoring system (SP-WHMS) combined with the Internet-of-Things (IoT) is proposed and implemented for a long-term and long-distance health monitoring of railway bridges. This system frees the wireless sensor nodes (WSNs) from their dependence on chemical batteries by using the piezoelectric energy harvester (PEH) to harvest the vibration energy from bridges to power the WSNs. The whole-process of the SP-WHMS including energy conversion, energy storage, energy management and application is analyzed at the first time through theory, simulation and experiment. For the energy conversion, a reported PEH model with high energy conversion efficiency (called D-M PEH) is refabricated, which can achieve a maximum average output power of 0.9 W under unit harmonic acceleration excitation. For the energy storage, a high-precision iteration analysis procedure (IAP) is proposed to predict the charging process of the strong-coupled PEHs and the charging process under non-harmonic excitations. The comparison results show that the IAP’s prediction deviates less than 3% from the experiment result and even less than 1% from the simulation result. For the energy management, an AP64500 chip with two adjustable input voltage thresholds and a stable output voltage ensures the normal operation of the SP-WHMS and makes the SP-WHMS suitable for different types of WSNs. For the application, the synergies between energy conversion, storage and management are considered, and the effectiveness of the SP-WHMS used on railway bridges is tested through an activation experiment. This work provides a technical guidance and framework for the implementation of the self-powered wireless monitoring on railway bridges.

Jellyfish-inspired bistable piezoelectric-triboelectric hybrid generator for low-frequency vibration energy harvesting

In pursuit of enhancing the output power of energy harvesters, this paper introduces a novel jellyfish-inspired bistable piezoelectric-triboelectric hybrid generator (JIB-PTHG) with low potential energy barrier characteristics and low-frequency, broadband energy harvesting capabilities. The JIB-PTHG comprises two flexible beams, and two rigid links, connected via a spring. Utilizing the harmonic balance method and Runge-Kutta fourth-order algorithm, a comprehensive analysis of the mechanical properties of the low-barrier bistable structure is conducted. Static equilibrium bifurcation and dynamic response analysis are performed. Subsequently, an electromechanical coupling model incorporating piezoelectricity and triboelectricity is developed to theoretically predict the output voltage of the two modules, revealing a positive correlation between displacement and output voltage. Experiment validation demonstrated that adjusting the length of the rigid links can effectively increase the deformation of the piezoelectric beam and the relative displacement of the triboelectric layers, thus improving the generator's output performance. Specifically, the TENG module exhibited exceptional energy harvesting performance within the frequency range of 3.5–6.5 Hz, achieving a voltage of 1050 V, a power of 1.84 mW, and the ability to illuminate 70 LED lights under an excitation amplitude of 17.5 mm and a frequency of 5.5 Hz. The PEG module, displayed high output performance within the frequency range of 4-7 Hz, reaching a voltage of 6.2 Vs and a power of 12.8 μW at an excitation frequency of 7 Hz and an amplitude of 17.5 mm. Both theoretical and experimental findings indicate that the JIB-PTHG has achieved improved output power within the frequency range of 3.5–6.5 Hz. Furthermore, the JIB-PTHG has the potential to harness bridge vibration energy and convert it into useful electrical energy for powering wireless sensor nodes in bridge health monitoring systems.

Review and Prospect of Research on Structural Health Monitoring Technology for Bridges

As a crucial infrastructure in the transport system, the safe operation of bridges is directly related to all aspects of people’s daily lives. The development of bridge structural health monitoring technology and its application play an important role in ensuring the safety and extending the service life of bridges. This paper carries out in-depth research and analysis on the related technology of bridge structural health monitoring. Firstly, the existing monitoring technologies at home and abroad are sorted out, and the advantages and problems of various methods are compared and analyzed, including nondestructive testing, stress measurement, vibration characteristic identification, and other commonly used monitoring technologies. Secondly, the key technologies and equipment in the bridge health monitoring system, such as sensor technology, data acquisition, and processing technology, are introduced in detail. Finally, the development trend in the field of bridge health monitoring is prospected from both theoretical research and technical application. In the future, with the development of emerging technologies such as big data, cloud computing, and the Internet of Things, it is expected that bridge health monitoring with intelligent and systematic features will be more widely applied to provide a stronger guarantee for the safe and efficient operation of bridges.

IoT Based Monitoring and Maintenance of Highway Bridges using WSN

This paper presents a WSN-based IOT-based bridge monitoring system. The automated real-time bridge health monitoring system has been made possible by sophisticated modifications to sensor technology; this system will aid in disaster management. The technology used in this development is a wireless sensor network (WSN). This system uses a variety of sensors to continually monitor the state of the bridge. These sensors include an accelerometer to identify jerks in the bridge, a vibration sensor to identify vibrations occurring on the bridge, a flex sensor to identify bends in the bridge, and a water level sensor to identify the water level. To determine the weight on the bridge, use the load sensor. The barrier is detected by an IR sensor. Only the buzzer will activate when these are high. Additionally, a microcontroller processes the data from many sensors and transmits it to a server and management system so that real-time bridge condition monitoring via a mobile device using the GSM model is possible

Smart Bridge Damage Assessment through Integrated Multi-Sensor Fusion Vehicle Monitoring

Abstract. This study explores the efficacy of vehicle-assisted monitoring for bridge damage assessment, emphasizing the integration of diverse sensor data sources. A novel method utilizing a deep neural network is proposed, enabling the fusion of fixed sensors on bridges and onboard vehicle sensors for damage assessment. The network offers scalability, robustness, and implementability, accommodating various measurement types while handling noise and dynamic loading conditions. The main novel aspect of our work is its ability to extract damage-sensitive features without signal preprocessing for future bridge health monitoring systems. Through numerical evaluations, considering realistic operational conditions, the proposed method demonstrates the capability to detect subtle damage under varying traffic conditions. Findings underscore the importance of integrating vehicle and bridge sensor data for reliable damage assessment, recommending strategies for optimal monitoring implementation by road authorities and bridge owners.

Optimization Of Structural Health Monitoring For Steel Bridges Using Wireless Sensor Networks and Machine Learning Algorithms

This study presents an advanced structural health monitoring (SHM) system for steel bridges based on wireless sensor networks (WSN) integrated with machine learning algorithms. The proposed system monitors and predicts structural integrity under various load conditions. The research focuses on developing a machine learning model capable of real-time anomaly detection, allowing for early warnings of potential failures. Experimental results from both simulation and field tests demonstrate the system’s effectiveness in prolonging bridge lifespan while reducing maintenance costs.

A bridge damage monitoring and assessment strategy based on Gaussian mixture modelling

For the bridges equipped with health monitoring systems, it is essential to precisely assess the health status of the bridge from the data with noise and outliers. An improved Gaussian mixture model clustering algorithm is used to process the obtained bridge strain sensor monitoring data and generate the characteristic data of clustering in order to improve the accuracy of data analysis of bridge health monitoring system. The cluster and feature data are obtained by Expectation Maximization process, and the isolated clusters are filtered out by the threshold parameters of weight and the Euclidean distance between clusters center. Based on the feature data, a scoring strategy is established to assess the bridge health and sensor status. The proposed strategy is used to analyze the monitoring data of a bridge in western China, and the analysis results show the availability of this strategy. Compared with the directly collected data, the processed bridge health scoring curve variation is smooth, which can reduce the influence of noise and abnormal data. The sensor status scoring curve can track changes in collected data and respond to different transformation scenarios. This shows that the proposed strategy can evaluate the state of this bridge and the sensors. provided a reference for bridge maintenance decision during the operational period.

Estimating Bridge Natural Frequencies Based on Modal Analysis of Vehicle-Bridge Synchronized Vibration Data.

This paper presents a method for accurately estimating the natural frequencies of bridges by simultaneously measuring the acceleration vibration data of vehicles and bridges and applying modal analysis theory. Vibration sensors synchronized with GPS timing were installed on both vehicles and bridges, achieving stable and high-precision time synchronization. This enabled the computation of the bridge's Frequency Response Functions (FRFs) for each mode, leading to a refined estimation of natural frequencies. The validity of the theory was confirmed through numerical simulations and experimental tests. The simulations confirmed its effectiveness, and similar trends were observed in actual bridge measurements. Consequently, this method significantly enhances the feasibility of bridge health monitoring systems. The proposed method is suitable for road bridges with spans ranging from short- to medium-span length, where the vehicle is capable of exciting the bridge.

Mapping model between lateral responses of bridges under random traffic loads based on data-driven approach

Abstract The lateral distribution characteristic of a bridge is one of the important features reflecting its in-service state. Conventional deterministic indicators often struggle to consider this time-varying feature. In this study, a data-driven approach is employed to establish the mapping model among responses at the different lateral positions using the eXtreme Gradient Boosting model optimized by the Bayesian optimization algorithm. The proposed method is validated based on both data from a bridge health monitoring system equipped on an actual bridge and data from numerical simulation. At different lateral positions of the actual bridge, the overall coefficient of determination (R2) of the strain response mapped by the model is above 0.991, and the overall root mean squared error (RMSE) of the strain response mapped by the model are below 1.159 με. The numerical simulation method is used to consider a variety of working conditions with different road surface roughness and different traffic densities. Under different working conditions, the overall R2 of the deflection response mapped by the model is still above 0.961, and the overall RMSE of the deflection response mapped by the model is below 0.249 mm. These indicate that the proposed model can consider time-varying mapping relationships among responses at different lateral positions, and has good accuracy and applicability.

A method for suspenders tension identification of bridges based on the spatio‐temporal correlation between the girder strain and suspenders tension

AbstractIn the actual structural health monitoring system of suspension bridges, only part of suspenders tension can be monitored, but not all the suspenders tension can be obtained. To solve this problem, a method for suspenders tension identification of bridges based on the spatio‐temporal correlation between the girder strain and suspenders tension is proposed. By using actual monitoring data of vehicle loads, a spatio‐temporal correlation model of the girder strain and tension forces of all suspenders is constructed based on the combined application of stacked denoising autoencoder and convolutional neural networks‐long short‐term memory model, so as to realize the preliminary identification of suspenders tension. On this basis, by using the actual monitoring data of suspenders tension and the strain monitoring data obtained through the distributed optical fiber sensors, the delicate identification of tension forces of all suspenders is realized based on the error interpolation of preliminary identification results. The results of the example bridge show that the method in this paper can effectively identify tension forces of all suspenders of the suspension bridge, and identification results are more accurate than the method using only the monitoring data of suspenders.

An IoT-Based Road Bridge Health Monitoring and Warning System.

Recent earthquakes worldwide have led to significant loss of life and structural damage to infrastructure, especially road bridges. Existing bridge monitoring systems have limitations, including restricted detection capabilities, subjectivity, human error, labor-intensive inspections, limited access to remote areas, and high costs. Aging infrastructures pose a critical concern for organizations and government funding policies, showing signs of decay and impending structural failure. To address these challenges, this research proposes an IoT-based bridge health status monitoring and warning system that is wireless, low-cost, durable, and user-friendly. The proposed system builds upon engineering standards and guidelines to classify bridge health status into categories ranging from excellent to collapse condition. It incorporates deflection, vibration, temperature, humidity, and infrared sensors, combined with IoT and a fuzzy logic algorithm. The primary objective is to reduce bridge maintenance costs, extend lifespans, and enhance transportation safety through an early warning system via a mobile application. Additionally, a Google Maps interface has been developed to display bridge conditions along with real-time traffic video. To validate the proposed system, a 3-D prototype model was constructed and tested. Practical testing of the fuzzy logic algorithm aligned with the simulation outcomes, demonstrating expected accuracy in determining bridge health status.

Harvest more bridge vibration energy by nonlinear multi-stable piezomagnetoelastic harvester

The idea of using bridge vibration energy to power node sensors is promising in solving the power problems of bridge health monitoring systems. This study proposes a piezomagnetoelastic harvester that tries utilizing nonlinear multi-stability to improve the harvesting performance of bridge vibration energy. The harvester can exhibit multi-stable behaviors due to the introduced nonlinear forces. The potential energy functions are derived and corresponding analyzes are carried out. The results show that the tri-stable state of harvester can be obtained by adjusting the parameters related to magnetic forces. Its potential barriers are low and the distance between potential wells is large, so it can execute snap-through motion easily. The experimental study was carried out for different vehicle speeds. The results verify that the energy harvester in a tri-stable state can generate the largest electric output over a wide range of vehicle speeds. Moreover, it is proved that the tri-stable state can protect the harvester from excessive deformation and subsequent damage. Therefore, the tri-stable configuration has a potential advantage in bridge vibration energy harvesting and providing power for self-powered sensors.