F. Necati Catbas

Structural identification of constructed systems : approaches, methods, and technologies for effective practice of St-Id

Structural Identification of Constructed Systems: Approaches, Methods, and Technologies for Effective Practice of St-Id offers an overview of nearly 20 years of research directed at bridging the gap in structural engineering between models and real structural systems. Structural identification, known as St-Id, can be defined as the process of creating and updating a model of a structure (for instance, a finite element model) using experimental observations and data. By developing reliable estimates of the performance and vulnerability of structural systems, St-Id produces improved simulations that, in turn, assist in decision making and the transition to performance-based civil engineering. Drawing upon contributions from experts in the field, this report focuses on defining the most critical considerations of St-Id, which include: modelling, both analytical and numerical experimentation, including observations, sensing, and monitoring data processing, including error screening and feature extraction model calibration, including comparisons of models and experimental data, model updating, and model selection decision support, such as scenario analyses and risk assessment Two appendixes present case studies demonstrating the St-Id of buildings and of bridges. Structural engineers, educators, and researchers working in the areas of structural modelling, health monitoring, assessment, forensics, performance evaluation, predictive analysis, and decision making will find this book useful in covering critical and practical aspects of these concepts.

Damage Assessment with Ambient Vibration Data Using a Novel Time Series Analysis Methodology

In this study, a novel approach using a modified time series analysis methodology is used to detect and locate structural changes by using ambient vibration data. In addition, it is shown that the level of the damage feature gives important information about the relative change of the damage severity, although direct damage quantification is not achieved. In this methodology, random decrement (RD) is used to obtain pseudofree response data from the ambient vibration time histories. Autoregressive models with exogenous input (ARX models) are created for different sensor clusters by using the pseudofree response of the structure. The output of each sensor in a cluster is used as an input to the ARX model to predict the output of the reference channel of that sensor cluster. After creating ARX models for the healthy structure for each sensor cluster, these models are used for predicting the data from the damaged structure. The difference between the fit ratios is used as the damage feature. The methodology i...

Nonparametric analysis of structural health monitoring data for identification and localization of changes: Concept, lab, and real-life studies

Structural health monitoring systems integrate novel experimental technologies, analytical methods, and information technologies for a number of objectives such as detecting structural changes and damage as well as assessing the condition, safety, and serviceability of the monitored structure. The objective of this article is to present a correlation-based methodology as an effective nonparametric data analysis approach for detecting and localizing structural changes using strain data under operational loading conditions. While several methods have been explored in the literature, the focus of this article is to explore a practical and cost-effective (in terms of sensor, data acquisition, and analysis) methodology to identify structural problems. The methodology presented here is based on tracking correlation coefficients between strain time histories at different locations. After discussing the background, the effectiveness of the methodology is first demonstrated on a laboratory test structure. A unique contribution of this study is the validation of the methodology on a real-life bridge, which was monitored before damage was induced, during the bridge was damaged, and after damage was repaired. It is shown that structural changes can be detected and located for both the laboratory test structure and the real-life bridge using the variations in the correlation matrices. Since the real-life bridge was monitored under different conditions, the effectiveness of the bridge repair is also presented in comparative fashion with respect to before damage conditions. Some of the critical issues such as signal processing, data length, and level of data separation for change detection are also discussed. The correlation-based data analysis methodology is computationally efficient and easy to use, especially for handling large amounts of monitoring data. The results show that this methodology has the potential to be easily applied by engineers to different kinds of civil infrastructure for condition monitoring and maintenance.

Sensor Networks, Computer Imaging, and Unit Influence Lines for Structural Health Monitoring: Case Study for Bridge Load Rating

In this paper, a novel methodology for structural health monitoring of a bridge is presented with implementations for bridge load rating using sensor and video image data from operating traffic. With this methodology, video images are analyzed by means of computer vision techniques to detect and track vehicles crossing the bridge. Traditional sensor data are correlated with computer images to extract unit influence lines (UILs). Based on laboratory studies, UILs can be extracted for a critical section with different vehicles by means of synchronized video and sensor data. The synchronized computer vision and strain measurements can be obtained for bridge load rating under operational traffic. For this, the following are presented: a real life bridge is instrumented and monitored, and the real-life data are processed under a moving load. A detailed finite-element model (FEM) of the bridge is also developed and presented along with the experimental measurements to support the applicability of the approach for load rating using UILs extracted from operating traffic. The load rating of the bridges using operational traffic in real life was validated with the FEM results of the bridge and the simulation of the operational traffic on the bridge. This approach is further proven with different vehicles captured with video and measurements. The UILs are used for load rating by multiplying the UIL vector of the critical section with the load vector from the HL-93 design truck. The load rating based on the UIL is compared with the FEM results and indicates good agreement. With this method, it is possible to extract UILs of bridges under regular traffic and obtain load rating efficiently.

Structural health monitoring using video stream, influence lines, and statistical analysis:

Civil infrastructure systems experience damage, overloading, aging due to normal operations, severe environmental conditions, and extreme events. These effects change the structural behavior and performance. Novel structural health monitoring (SHM) strategies are increasingly becoming more important to objectively determine the actual condition and these changes. The main objective of this study is to demonstrate the integration of video images and sensor data as promising techniques for the safety of bridges in the context of SHM. The UCF 4-span bridge model is used to demonstrate the method. Image and sensing data are analyzed to obtain unit influence line (UIL) as an index for monitoring the bridge behavior under loading conditions identified using computer vision techniques. UILs are extracted for several different moving loads. In addition to the analysis of UILs in a comparative fashion, a new method based on statistical outlier detection from UIL vector sets is proposed and demonstrated. The new method is applied to detect and identify some of the most common damage scenarios for bridges such as changes in boundary conditions and loss of connectivity between composite sections. Successful results are obtained from the experimental studies.

Structural Identification for Performance Prediction Considering Uncertainties: Case Study of a Movable Bridge

AbstractStructural identification (St-Id) can be described simply as estimating the properties of a structural system based on a correlation of inputs and outputs for decision making. For a complete St-Id process, establishing the decision-making needs, developing analytical and numerical models, and conducting field measurements, along with parameter identification using the experimental data for model calibration, are carried out. One important consideration is evaluation of the limitations and adequacy of using a single calibrated model before leveraging it for decision making, such as the reliability of the structural system for the remainder of its design life. The uncertainties in the data collected, determined by means of intermittent testing or monitoring; the limitations of the models; and the nonstationary nature of structural behavior need to be considered. These uncertainties can be incorporated by using of a family of parent and offspring models. The objective of this paper is to illustrate t...

Investigation of Uncertainty Changes in Model Outputs for Finite-Element Model Updating Using Structural Health Monitoring Data

This article aims to investigate the effect of uncertainties on the predicted response of structures using updated finite-element models (FEMs). Modeling uncertainties are quantified by fuzzy numbers and are incorporated into the fuzzy FEM updating procedure. The impact of the amount and types of data used on the performance of the updated model is investigated. In order to perform the complex FEM updating calculations, which generally take too much time for complex models, a Gaussian process (GP) is used as a surrogate model. The central composite design (CCD) method is used to sample the input parameter space for more accurate GP models. Genetic algorithms (GA) are employed to solve the inverse fuzzy model updating problem. Additional constraints are presented to capture the variation space of the uncertain response parameters. The University of Central Florida benchmark test structure, which is designed to represent short-span to medium-span bridges, is used in the scope of uncertainty quantification study. Static and dynamic experimental test data obtained from the benchmark structure under different loadings and conditions are used for the demonstration. A damage case, in which the stiffness reduction in boundaries is simulated by using flexible pads, is considered. The results show that appropriate data sets, which contain the least uncertainty, should be generated instead of involving the entire set of measurements obtained from different tests. Nevertheless, uncertainty quantification should be employed to find the variation range of uncertain responses predicted by simplified FEM models.

Computer vision-based displacement and vibration monitoring without using physical target on structures

Abstract Although vision-based methods for displacement and vibration monitoring have been used in civil engineering for more than a decade, most of these techniques require physical targets attached to the structures. This requirement makes computer vision-based monitoring for real-life structures cumbersome due to need to access certain critical locations. In this study, a non-target computer vision-based method for displacement and vibration measurement is proposed by exploring a new type of virtual markers instead of physical targets. The key points of measurement positions obtained using a robust computer vision technique named scale-invariant feature transform show a potential ability to take the place of classical targets. To calculate the converting ratio between pixel-based displacement and engineering unit (millimetre), a practical camera calibration method is developed to convert pixel-based displacements to engineering unit since a calibration standard (a target) is not available. Methods and approaches to handle challenges such as low contrast, changing illumination and outliers in matching key points are also presented. The proposed method is verified and demonstrated on the UCF four-span bridge model and on a real-life structure, with excellent results for both static and dynamic behaviour of the two structures. Finally, the method requires a simple, less complicated and more cost-effective hardware compared to conventional displacement and vibration monitoring measuring technologies.

Structural Identification of Constructed Systems: Collective Effort toward an Integrated Approach That Reduces Barriers to Adoption

Structural identification (St-Id) is a powerful tool that bridges the gap between constructed systems and themodels used in their design and assessment. Although St-Id has attracted the attention of numerous researchers worldwide over the last several decades, it unfortunately has not experienced widespread adoption in practice. The ASCE Structural Engineering Institute Structural Identification of Constructed Systems Committee is seeking to reverse this trend by enhancing advocacy toward and promoting implementation of St-Id within the public and private sectors. The committee’s first action on this front was the development of a comprehensive report that benchmarks the current state of the art in St-Id, with special attention to case studies of its successful implementation. To organize the diverse paradigm of St-Id, the committee adopted a six-step cycle that spansmodeling through experimentation and ultimately to decision support. This forum paper overviews the report with the first six chapters dedicated to this cycle, as well as the report’s closing two chapters dedicated to case studies that exemplify the implementation of St-Id to various buildings and bridges around the world.

Use of FBG Sensors to Detect Damage from Large Amount of Dynamic Measurements

Fiber Optic Sensors (FOS) offer several promising features for long term Structural Health Monitoring such as distributed sensing capability, durability, stability and immunity to electrical noise. There are different FOS technologies with a wide range of performance metrics that define their suitability for different applications. One of the most widely used fiber optic sensing technologies are point sensors (Fiber Bragg Gratings-FBG). If the data collected with the FBG sensors are analyzed and handled effectively, important information about the behaviour of the structure can be obtained. In this study, a recently developed damage detection method based on strain correlation analysis is employed using strain data collected with FBG sensors. In pursuing these objectives, different damage scenarios have been designed and tested on a 4-span bridge model in the laboratory environment. The efficiency of both FBG sensors and correlation analysis method for detection and localizing damage is explored. It is shown that damage can be clearly identified and localized for most of the cases under investigation.