Active learning structural model updating of a multisensory system based on Kriging method and Bayesian inference

Model updating techniques are often applied to calibrate the numerical models of bridges using structural health monitoring data. The updated models can facilitate damage assessment and prediction of responses under extreme loading conditions. Some researchers have adopted surrogate models, for example, Kriging approach, to reduce the computations, while others have quantified uncertainties with Bayesian inference. It is desirable to further improve the efficiency and robustness of the Kriging-based model updating approach and analytically evaluate its uncertainties. An active learning structural model updating method is proposed based on the Kriging method. The expected feasibility learning function is extended for model updating using a Bayesian objective function. The uncertainties can be quantified through a derived likelihood function. The case study for verification involves a multisensory vehicle-bridge system comprising only two sensors, with one installed on a vehicle parked temporarily on the bridge and another mounted directly on the bridge. The proposed algorithm is utilized for damage detection of two beams numerically and an aluminum model beam experimentally. The proposed method can achieve satisfactory accuracy in identifying damage with much less data, compared with the general Kriging model updating technique. Both the computation and instrumentation can be reduced for structural health monitoring and model updating.     

钢筋混凝土斜拉桥寿命期内整体极限承载力分析

提出基于定期检测和健康监测技术的斜拉桥结构服役期内主梁整体失效模式及其极限承载力演化规律的分析方法,为结构的全寿命设计以及结构安全预警的分级提供理论依据。首先,利用结构的检测/监测数据,建立与实际结构状态相符的有限元模型,并分析构件的耐久性损伤,在此基础上计算考虑抗力退化的构件的极限承载力;然后,在有限元模型上施加逐级渐进的车辆荷载,分析各构件的内力状态;最后,将荷载效应超过抗力的主梁截面设置为塑性铰,改变结构体系,继续施加车辆荷载,直至结构承受的车辆荷载达到最大值,对应的变结构体系和车辆荷载分别为最终主梁整体失效模式和极限承载力。利用上述方法对我国北方某座干线斜拉桥在运营期内的4个典型状态进行计算分析,研究该桥在运营期内的失效模式及其极限承载力的演化规律。结果表明,所提出的方法可以获得在车辆荷载作用下斜拉桥主梁整体失效模式和相应的极限承载力;斜拉桥在服役过程中主梁开裂、钢筋锈蚀、边界条件变化等导致桥梁结构损伤、甚至结构体系及其相应的力学模型发生较大的变化,使结构失效模式及其极限承载力发生较大的变化。     

Structural health monitoring of a cable-stayed bridge with Bayesian neural networks

In recent years, there has been an increasing interest in permanent observation of the dynamic behaviour of bridges for long-term monitoring purpose. This is due not only to the ageing of a lot of structures, but also for dealing with the increasing complexity of new bridges. The long-term monitoring of bridges produces a huge quantity of data that need to be effectively processed. For this purpose, there has been a growing interest on the application of soft computing methods. In particular, this work deals with the applicability of Bayesian neural networks for the identification of damage of a cable-stayed bridge. The selected structure is a real bridge proposed as benchmark problem by the Asian-Pacific Network of Centers for Research in Smart Structure Technology (ANCRiSST). They shared data coming from the long-term monitoring of the bridge with the structural health monitoring community in order to assess the current progress on damage detection and identification methods with a full-scale example. The data set includes vibration data before and after the bridge was damaged, so they are useful for testing new approaches for damage detection. In the first part of the paper, the Bayesian neural network model is discussed; then in the second part, a Bayesian neural network procedure for damage detection has been tested. The proposed method is able to detect anomalies on the behaviour of the structure, which can be related to the presence of damage. In order to obtain a confirmation of the obtained results, in the last part of the paper, they are compared with those obtained by using a traditional approach for vibration-based structural identification.     

Fatigue stress monitoring and analyses for steel box girder of Runyang Suspension Bridge

The monitoring of fatigue stress of steel bridge is a key issue of bridge health monitoring and safety assessment. This paper aims to find out the strain history features of the girder components of Runyang Suspension Bridge (RYSB) under vehicle and environmental loading during its first year of service based on the strain-history data recorded by structural health monitoring system installed in the bridge. The fatigue stress spectrums of steel box girders under normal traffic load, heavy trucks and typhoon loads were studied as well as the correlation between varying strain and temperature based on real-time monitoring of temperature histories. The results show that, monitoring on local strain in health monitoring systems of RYSB can effectively provide the information on fatigue stresses. The range of the equivalent fatigue stress on the monitoring location of the bridge is currently low, mainly due to varying traffic loading and temperature. There exists significant correlation between varying temperature and mean value of fatigue stress. Effect of the passing of heavy trucks on bridge fatigue is quite significant since the value of the cumulative fatigue damage generated by heavy trucks is 10 to 100 times larger than that by normal traffic. Therefore, the effect on fatigue cumulative damage due to heavy trucks or overloaded vehicles needs to be monitored and paid much attention to.     

温度对斜拉桥线型的影响研究

桥梁的线型变化分析是桥梁技术状态评估的重要手段之一,然而实测桥梁线型不可避免地受到温度差异的影响,从而影响对结构技术状态的评估。文中通过桥梁健康监测的实测数据,阐述温度对斜拉桥线型变化的实际影响,为今后的桥梁线型分析提供技术参考。     

A filtering-based bridge weigh-in-motion system on a continuous multi-girder bridge considering the influence lines of different lanes

A real-time vehicle monitoring is crucial for effective bridge maintenance and traffic management because overloaded vehicles can cause damage to bridges, and in some extreme cases, it will directly lead to a bridge failure. Bridge weigh-in-motion (BWIM) system as a high performance and cost-effective technology has been extensively used to monitor vehicle speed and weight on highways. However, the dynamic effect and data noise may have an adverse impact on the bridge responses during and immediately following the vehicles pass the bridge. The fast Fourier transform (FFT) method, which can significantly purify the collected structural responses (dynamic strains) received from sensors or transducers, was used in axle counting, detection, and axle weighing technology in this study. To further improve the accuracy of the BWIM system, the field-calibrated influence lines (ILs) of a continuous multi-girder bridge were regarded as a reference to identify the vehicle weight based on the modified Moses algorithm and the least squares method. In situ experimental results indicated that the signals treated with FFT filter were far better than the original ones, the efficiency and the accuracy of axle detection were significantly improved by introducing the FFT method to the BWIM system. Moreover, the lateral load distribution effect on bridges should be considered by using the calculated average ILs of the specific lane individually for vehicle weight calculation of this lane.     

Frontier of continuous structural health monitoring system for short & medium span bridges and condition assessment

It is becoming an important social problem to make maintenance and rehabilitation of existing short and medium span(10-20 m) bridges because there are a huge amount of short and medium span bridges in service in the world. The kernel of such bridge management is to develop a method of safety(condition) assessment on items which include remaining life and load carrying capacity. Bridge health monitoring using information technology and sensors is capable of providing more accurate knowledge of bridge performance than traditional strategies. The aim of this paper is to introduce a state-of-the-art on not only a rational bridge health monitoring system incorporating with the information and communication technologies for lifetime management of existing short and medium span bridges but also a continuous data collecting system designed for bridge health monitoring of mainly short and medium span bridges. In this paper, although there are some useful monitoring methods for short and medium span bridges based on the qualitative or quantitative information, mainly two advanced structural health monitoring systems are described to review and analyse the potential of utilizing the long term health monitoring in safety assessment and management issues for short and medium span bridge. The first is a special designed mobile in-situ loading device(vehicle) for short and medium span road bridges to assess the structural safety(performance) and derive optimal strategies for maintenance using reliability based method. The second is a long term health monitoring method by using the public buses as part of a public transit system (called bus monitoring system) to be applied mainly to short and medium span bridges, along with safety indices, namely, “characteristic deflection” which is relatively free from the influence of dynamic disturbances due to such factors as the roughness of the road surface, and a structural anomaly parameter.     

Fatigue stress monitoring and analyses for steel box girder of Runyang Suspension Bridge

The monitoring of fatigue stress of steel bridge is a key issue of bridge health monitoring and safety assessment. This paper aims to find out the strain history features of the girder components of Runyang Suspension Bridge (RYSB) under vehicle and environmental loading during its first year of service based on the strain-history data recorded by structural health monitoring system installed in the bridge. The fatigue stress spectrums of steel box girders under normal traffic load, heavy trucks and typhoon loads were studied as well as the correlation between varying strain and temperature based on real-time monitoring of temperature histories. The results show that, monitoring on local strain in health monitoring systems of RYSB can effectively provide the information on fatigue stresses. The range of the equivalent fatigue stress on the monitoring location of the bridge is currently low, mainly due to varying traffic loading and temperature. There exists significant correlation between varying temperature and mean value of fatigue stress. Effect of the passing of heavy trucks on bridge fatigue is quite significant since the value of the cumulative fatigue damage generated by heavy trucks is 10 to 100 times larger than that by normal traffic. Therefore, the effect on fatigue cumulative damage due to heavy trucks or overloaded vehicles needs to be monitored and paid much attention to. __________ Translated from Journal of Southeast University (Natural Science Edition), 2007, 37(2): 280–286 [译自: 东南大学学报(自然科学版)]     

Critical issues,condition assessment and monitoring of heavy movable structures: emphasis on movable bridges

In this paper, a relatively less studied class of structures is presented based on the research conducted on Florida's movable bridges over the last several years. Movable bridges consist of complex structural, mechanical and electrical systems that provide versatility to these bridges, but at the same time, create intermittent operational and maintenance challenges. Movable bridges have been designed and constructed for some time; however, there are fewer studies in the literature on movable bridges as compared to other bridge types. In addition, none of these studies provide a comprehensive documentation of issues related to the condition of movable bridge populations in conjunction with possible monitoring applications specific to these bridges. This paper characterises and documents these issues related to movable bridges considering both the mechanical and structural components. Considerations for designing a monitoring system for movable bridges are also presented based on inspection reports and expert opinions. The design and implementation of a monitoring system for a representative bascule bridge are presented along with long-term monitoring data. Various movable bridge characteristics such as opening/closing torque, bridge balance and friction are shown since these are critical for maintenance applications on mechanical components. Finally, the impact of environmental effects (such as wind and temperature) on bridge mechanical characteristics is demonstrated by analysing monitoring data for more than 1000 opening/closing events.     

Damage detection of a cable-stayed bridge using feature extraction and selection methods

Abstract

In this paper, a novel approach is presented to detect damage in a cable-stayed bridge based on feature extraction and selection. In the first part of the paper, several features are used in time domain, frequency domain, and time-frequency domain to detect damage. Next, several spectral parameters are introduced as effective features to reduce false alarms. Support vector machine (SVM) technique is employed as a classifier to compare the detection ability of the proposed and conventional features. In the second part of the paper, several feature selection techniques including forward selection, backward elimination, the minimum redundancy and max relevancy, ReliefF and SVM approach based on recursive feature elimination (SVM-RFE) algorithm are employed to improve feature extraction accuracy through selecting and ranking the most discriminative feature as an optimal feature subset. The capability of the proposed approach to detect damage is investigated by a data set obtained from a health monitoring system of a real benchmark problem. The data consists of vibration acceleration of the Yonghe bridge in both healthy and damaged states. Results show that the proposed procedure can effectively reduce false alarms.     

Effect of Road Surface,Vehicle, and Device Characteristics on Energy Harvesting from Bridge–Vehicle Interactions

Energy harvesting to power sensors for structural health monitoring (SHM) has received huge attention worldwide. A number of practical aspects affecting energy harvesting and the possibility of health monitoring directly from energy harvesters is investigated here. The key idea is the amount of power received from a damaged and an undamaged structure varying and the signature of such variation can be used for SHM. For this study, a damaged bridge and an undamaged bridge are considered with harvesters located at different positions and the power harvested is accessed numerically to determine how energy harvesting can act as a damage detector and monitor. Bridge–vehicle interaction is exploited to harvest energy. For a damaged bridge, a bilinear breathing crack is considered. Variable surface roughness according to ISO 8606:1995(E) is considered such that the real values can be considered in the simulation. The possibility of a drive‐by type health monitoring using energy harvesting is highlighted and the effects of road surface on such monitoring are identified. The sensitivity of the harvester health monitoring to locations and extents of crack damage are reported. This study investigates the effects of multiple harvesters and the effects of vehicular parameters on the harvested power. Continuous harvesting over a length of the bridge is considered semianalytically. A comparison among the numerical simulations, detailed finite element analysis, and experimental results emphasizes the feasibility of the proposed method.     

Integration of structural health monitoring in a system performance based life-cycle bridge management framework

In this paper, an approach for integrating the information obtained from structural health monitoring in a life-cycle bridge management framework is proposed. The framework is developed on the basis of life-cycle system performance concepts that are also presented in this paper. The performance of the bridge is quantified by incorporating prior knowledge and information obtained from structural health monitoring using Bayesian updating concepts. This performance is predicted in the future using extreme value statistics. Advanced modelling tools and techniques are used for the lifetime reliability computations, including incremental nonlinear finite element analyses, quadratic response surface modelling using design of experiments concepts, and Latin hypercube sampling, among other techniques. The methodology is illustrated on an existing bridge in the state of Wisconsin.     

Integrating reliability and structural health monitoring in the fatigue assessment of concrete bridge decks

The overall objective of this article is to demonstrate how the concepts of reliability and structural health monitoring (SHM) can be integrated to create bridge assessment and decision systems. The steel-free concrete bridge deck system was chosen as a specific case study providing tangible focus for the research. The bridge assessment model therefore focused on fatigue cracking issues associated with wheel loads due to heavy truck traffic. The bridge assessment model required five components: a vehicle load model, a fatigue damage accumulation model, a residual strength model, a reliability model and an SHM decision model. Each of these components is discussed within the article with specific reference to previous research on steel-free bridge deck systems. The proposed model is used to develop a decision threshold for monitoring the oldest steel-free bridge deck in service.     

基于环境激励的连续刚构桥模态参数识别及有限元模型修正

一个合理准确的有限元模型是大型桥梁健康监测系统中损伤识别及安全评估、预警的前提。以一座三跨预应力混凝土连续刚构桥为例,根据设计资料,利用ANSYS建立其初始三维有限元模型,利用其健康监测系统采集得到的加速度时程数据,采用随机子空间法对实际结构的模态频率进行识别,依据识别结果,在参数灵敏度分析的基础上,根据最优化理论对该桥的有限元模型进行修正。研究结果表明,经修正后的有限元模型能真实准确地反映该桥梁结构的真实动力特性,可以作为健康监测中损伤识别、安全评估等研究工作的基准模型。     

大型桥梁健康监测研究动态与发展趋势

大型桥梁健康监测是近些年来的研究热点之一。本文首先阐述了桥梁健康监测的内涵及意义,对健康监测系统的组成做了总结,然后从传感器的优化布设、信号处理与信息提取技术、损伤识别技术和安全评估方面对研究现状做了综述,最后针对目前存在的问题指出了未来的研究方向。     

用人工神经网络方法估计桥梁在温度作用下的挠度行为

桥梁结构的挠度变化除了与车辆、人群等荷载有关之外,还与环境的作用密切相关,在大跨径桥梁结构安全监测系统中进行结构安全状态评估时需要知道桥梁结构在环境作用下的行为,以便分离环境与例如汽车、人群等荷载的不同影响。挠度变化是桥梁结构安全监测系统中反应桥梁安全状态最直观的参数之一,荷载与环境的作用、结构材料的变异都可以通过挠度的变化表现出来。但是通常所测得的挠度是桥梁在环境温度、车辆荷载等综合作用下的总响应,而不同荷载条件作用可以使得结构产生同样的挠度,由于不同的荷载条件可以引起同样的挠度但是产生不同的应力,因此将挠度分类是十分必要的。本文尝试用神经网络方法通过实测值来模拟温度与挠度之间的非线性关系,并用它来预测桥梁由温度所产生的挠度变化,从而可以将这温度产生的挠度值从实时的总挠度中分离出来,以便进行其他部分挠度评估的分析。     

Effect of tuned mass damper on the interaction of a quarter car model with a damaged bridge

This paper considers the effects of a tuned mass damper (TMD) on damaged bridge–accelerating quarter car vehicle interaction. The damage of the bridge is considered to be an open crack. The incorporation of a TMD to control the vibration response of the bridge and the quarter car vehicle model has been investigated from different aspects. A simplified form for the tuning ratio of the TMD is proposed. The vibration mitigation of the peak displacement, velocity and acceleration of the damaged bridge and the accelerating quarter car vehicle model using such a tuning is observed, along with the effects of possible detuning of the TMD due to the progressive deterioration of the bridge. A detailed parametric study is performed on the system with the TMD, considering the effects of quarter car vehicle model velocity, acceleration and the severity of the damage of the bridge.     

基于BP-LSTM混合模型的钢-混组合桥面系空间温度场及温度效应实时评估及预测

考虑到传统健康监测系统与有限元方法的不足,为实现钢-混组合桥面系温度及温度效应的实时评估及预测,提出基于多层前馈——长短期记忆(BP-LSTM)混合模型的实时评估及预测方法。研究过程中,以某桥为工程背景,采用其健康监测系统数据,并利用有限元法热力耦合分析扩充数据,构建包含“结构特征、时间特征、环境特征——温度、温度效应”映射的样本库,并通过实测数据验证有限元计算的准确性;设计BP-LSTM混合模型结构,对LSTM网络以考虑时间权重的均方误差(MSE)损失函数加以改进,读入样本库数据进行训练、验证、测试,从而获得输出精度较高的混合模型;最后,分别给出利用BP-LSTM混合模型进行温度及温度效应实时评估与预测的方法。结果表明:基于气象数据的结构温度场有限元计算结果与实测基本一致,“结构特征、时间特征、环境特征——温度、温度效应”映射具有高度非线性,单独BP网络或LSTM网络效果不佳,而BP-LSTM混合模型精度较高;根据本文提出的实时评估及预测方法,混合模型输出值与目标值的决定系数最低为0.930,最高为0.993,精度较高,可用于钢-混组合桥面系空间温度场及温度效应实时评估及预测。     

基于WIM的典型城市桥梁车辆荷载状况分析

基于安装在杭州石祥路留石高架上的动态称重系统（WIM）采集的交通流和车辆荷载数据,通过对交通流量、车辆构成、轴重、车辆总重等进行分析,获取了其概率统计特性及分布规律,建立了轴载谱和车辆总重谱,并与相关文献、规范进行了比较。结果表明：一天中各时段的交通流量具有很强的潮汐规律性;2.86%的车辆轴重超过了10 t的超载标准,最大的单轴重量达39.5 t,最大车辆总重达115 t,后半夜车辆恶性超载严重;2轴车车辆总重表现为单峰分布,3轴以上车辆的总重均表现为多峰分布,5轴车和6轴车的总重表现为3峰分布;轴重大的车辆主要是超载的2轴、3轴车辆,这部分车辆对桥梁等基础设施的危害较大。这些车辆荷载特征的获取有助于区域既有桥梁安全性的合理评估,为超重车辆治理和桥梁安全管理提供帮助。     

Optimization of bridge maintenance strategies based on structural health monitoring information

Highway bridges are subjected to strength degradation processes. Under budget constraints, it is important to determine the best maintenance strategies. Optimized strategies, based on prediction models, are already considered for the maintenance and operation of highway bridges. Prediction models are updated both in space and time by using non-destructive testing methods. Nevertheless, there is an urgent need for the efficient inclusion of structural health monitoring (SHM) data in structural assessment and prediction models. Indeed, SHM allows keeping strength degradation processes under control and should be included in life-cycle cost models. The lifetime reliability of structures is characterized by survivor functions. The SHM data enable to update the probability density function of time to failure through a Bayesian process. The aim of this paper is threefold: (a) to include SHM data in a bridge life-cycle cost analysis, (b) to determine optimal maintenance strategies based on monitoring information, and (c) to show the benefits of SHM. Optimal strategies are determined considering the cases without and with including monitoring results; the benefit of monitoring is then highlighted. The proposed concepts are applied to the I-39 Northbound Bridge over the Wisconsin River in Wisconsin, USA. A monitoring program of that bridge was performed by the ATLSS Engineering Research Center at Lehigh University.