Vibration-based Structural Damage Detection Research Articles

Multi-task Framework for Vibration-Based Structural Damage Detection of Spatial Truss Structure Using Graph Learning

Multi-task Framework for Vibration-Based Structural Damage Detection of Spatial Truss Structure Using Graph Learning

Towards vibration-based damage detection of civil engineering structures: overview, challenges, and future prospects

AbstractIn this paper, we delve into the evolving landscape of vibration-based structural damage detection (SDD) methodologies, emphasizing the pivotal role civil structures play in society's wellbeing and progress. While the significance of monitoring the resilience, durability, and overall health of these structures remains paramount, the methodology employed is continually evolving. Our focus encompasses not just the transformation brought by the advent of artificial intelligence but also the nuanced challenges and future directions that emerge from this integration. We shed light on the inherent nonlinearities civil engineering structures face, the limitations of current validation metrics, and the conundrums introduced by inverse analysis. Highlighting machine learning's (ML) transformative role, we discuss how techniques such as artificial neural networks and support vector machine's have expanded the SDD's scope. Deep learning's (DL) contributions, especially the innovative capabilities of convolutional neural network in raw data feature extraction, are elaborated upon, juxtaposed with the potential pitfalls, like data overfitting. We propose future avenues for the field, such as blending undamaged real-world data with simulated damage scenarios and a tilt towards unsupervised algorithms. By synthesizing these insights, our review offers an updated perspective on the amalgamation of traditional SDD techniques with ML and DL, underlining their potential in fostering more robust civil infrastructures.

Vibration-based structural damage detection strategy using FRFs and machine learning classifiers

In this paper, a damage detection strategy for beam-type structures based on frequency response functions (FRFs) is presented. Five aluminum beams of equal nominal dimensions are used in an experimental program under laboratory conditions to obtain experimental FRF data from both undamaged and damaged conditions. Damages are induced by creating rectangular notches on the beams using saw-cutting. A stochastic finite element model of the beam is developed in MATLAB to construct several training datasets. The damages are modeled by reducing the cross-sectional area at the corresponding damaged elements. Simple damage indexes are proposed as damage-sensitive features. Decision Tree, Support Vector Machines, and Artificial Neural Networks classifiers are trained in the first stage to perform damage detection and localization. Single and multiple damages located in a single zone and more than one zone simultaneously are considered. In the second stage, experimental data not used for training are used for validation. The results from the first stage suggest that the proposed damage indexes can effectively detect and locate structural damage in beams. Among all classifiers, Artificial Neural Networks is the classifier that best performed. High accuracy is achieved to identify the presence of damage (99.3%) and detecting its location on the beam for some of the training datasets (from 80.0% to 97.1%). In the second stage of validation, accuracy, as expected, decreased. However, misclassifications occur mainly for FRF samples in the impact zone, which indicates that the proposed strategy can be efficient to detect damages at locations other than the excitation zone.

An FDD-based modal parameter-less proportional flexibility-resembling matrix for response-only damage detection

Modal flexibility-based methods are effective tools for vibration-based structural damage detection, including in the output-only case. These methods are typically characterized by two stages: first, the modal parameters are identified, thus obtaining a certain number of modes; second, these modal parameters are used to assemble the modal flexibility matrix. This paper proposes a method for estimating a matrix that approximates a proportional flexibility matrix, termed proportional flexibility-resembling (PFR) matrix, and shows that this matrix can be used for damage detection and localization purposes. This matrix is obtained through signal processing operations to be executed after applying the first steps of the frequency-domain decomposition (FDD) technique—i.e., after the singular value decomposition of the spectral density matrix. The defining aspect of the PFR matrix is that, differently from the traditional formulation of modal flexibility and proportional flexibility matrices, it can be assembled without the need of an explicit identification of the modal parameters. In fact, the matrix is estimated by processing all first singular vectors and also all first singular values in a selected frequency range. In the proposed method, the typical two stage approach of traditional modal flexibility methods is avoided, and the intervention of an operator is limited to setting the values of a few parameters in the initial phase of the process. Numerical simulations and experimental data from a testbed structure were used to show the effectiveness of the proposed approach, and the analyses were performed by considering structures with different damage scenarios and damping properties.

Development of Structural Damage Detection Method Working with Contaminated Vibration Data via Autoencoder and Gradient Boosting

Vibration-based structural damage detection is one of the most promising venues for building smart and automated structural health monitoring applications; however, its applicability is impeded by a large amount of collected vibration data, and the performance could be undermined by degraded data. Therefore, this study develops a robust framework, dubbed AutoBoost-SDD, that can effectively handle contaminated vibration data and provide reliable monitoring results within reasonable computational resources. The proposed method consists of three key components. Firstly, multi-domain feature extraction techniques are utilized to convert high-dimensional raw data into informative feature vectors. Secondly, the auto-encoder deep learning architecture is leveraged to refine feature vectors of contaminated data. Finally, a tree-based boosting machine learning algorithm, namely LightGBM, is employed to assess the structures’ operational states using learned output from the second step. The viability and performance of the proposed framework are illustrated via three case studies involving numerical data of a 5-degree of freedom system, a 2D frame structure, and experimental data of a large-scale 18-story frame structure from the literature. The results show that the AutoBoost-SDD framework is able to provide reasonable detection results despite the presence of various contaminations, including noisy, missing, and anomalous data.

Outlier analysis combined with Gaussian mixture model for structural damage detection

Damage diagnostic techniques based on changes in natural frequency has two major problems, limiting their applicability to practical structures. The limitations are low sensitivity to smaller-size damage and the influence of ambient and operational factors (temperature, humidity, wind, traffic, etc.). The change in natural frequency due to environmental and operational factors may be even larger than the change due to damage. A reliable vibration-based structural damage detection distinguishing damage and other confounding factors is needed to avoid false alarm of damage. This paper presents an enhanced approach combining the gaussian mixture model (GMM) and outlier analysis for damage detection considering the effect of changing environmental and operational conditions. In the proposed work, GMM is first applied to classify the feature vectors (first two natural frequencies) into different clusters. The natural frequencies obtained under similar environmental and operational conditions were classified into the same cluster by satisfying the same probability distribution. Outlier analysis is then performed for each different cluster using the Mahanabolis Distance (MD) measure. The MD is a measure of the distance between a point and a distribution. In the current study, the MD between the current data sample and the reference data of each mixture is computed and the minimum MD among all the clusters is used as the damage index (DI). The threshold to conclude the presence of damage (DI > threshold) is established based on Generalized Extreme Value Distribution using the datasets of the healthy structure with various environmental and operational conditions. The proposed method is verified using benchmark examples of numerically simulated simply supported beam and experimental wooden bridge structure. The results of the investigations concluded that the proposed DI considering the patterns in the data using GMM proves better in decoupling the damage and other confounding factors following the conventional outlier analysis of damage detection.

A data-driven sensor placement strategy for reconstruction of mode shapes by using recurrent Gaussian process regression

Current Optimal Sensor Placement (OSP) strategies for bridges mostly rely on data from a finite element model rather than from the real structure due to high cost in placing massive sensors for data collection. For large-scale bridges, however, it is difficult to formulate a precise model and thus the OSP strategies building upon a finite element model inevitably suffer from modelling errors. Besides, the finite element model cannot account for real measurement noise. Premised on the fact that it is not expensive to make in-situ trial measurements with a few sensors on a target bridge before deploying a structural health monitoring (SHM) system on it, a data-driven OSP strategy is proposed in this study which aims at accurately reconstructing mode shapes (to facilitate vibration-based structural damage detection) by using only a few vibration sensors to be included in the SHM system. The proposed OSP strategy is also applicable for the upgrade of a long-term SHM system currently deployed on a bridge, by using historical data collected from the current SHM system. To precisely reconstruct mode shapes, a two-stage OSP strategy in terms of Recurrent Gaussian Process Regression (RGPR) is developed, and its performance is validated on a simulation model and a real bridge. In the first stage, the greedy algorithm is leveraged to temporarily deploy sensors on the structure and train accurate RGPR models using the collected data, which are used to afford spatially complete mode shape data for optimization later. Starting from a few sensors temporarily deployed on the bridge, a one-by-one sensor adding procedure is performed to configure increasing sensors until the target is achieved. In the second stage, Cuckoo Search (CS) algorithm is pursued to obtain the globally optimal sensor placement solution, from which the temporarily deployed sensors can be re-configured to the optimum positions. Both the best sensor quantity and positions are obtained by the proposed OSP strategy.

Vibration-based structural damage detection using 1-D convolutional neural network and transfer learning

This paper presents a novel vibration-based structural damage detection approach by using a one-dimensional convolutional neural network (1-D CNN) and transfer learning (TL). The CNN can effectively extract structural damage information from the vibration signals. However, the CNN training needs enough samples, while some damage samples (scenarios) obtained from real structures are limited, which will compromise the CNN ability to detect structural damage. As a solution, the numerical models have potential to provide sufficient CNN training samples; meanwhile, the state-of-the-art TL technique can significantly shorten the network training time and improve the accuracy. Therefore, this paper proposes a new method to detect the damage of a bridge model. The 1-D CNN is firstly trained with the samples of the single damage scenarios of the numerical bridge model. And then it is transferred to the complex scenarios of multi-damage (double or triple simultaneously), random size structures, and experimental model. The results demonstrate that: with the TL, the accuracy of damage detection is increased by about 47% at most, and the convergence speed is increased by at least 50%; in particular, the TL can inhibit over-fitting, and for the real bridge case, the accuracy also increased by 44.4%. It is demonstrated that: the TL can effectively improve the damage detection accuracy and convergence effect, and the application of this method to the random size structures also proves its generalization.

Structural health monitoring of railway bridges using innovative sensing technologies and machine learning algorithms: a concise review

Abstract Railway bridges are a vital element of railway infrastructures, and their safety can directly affect the regional economy and commuter transportation. However, railway bridges are often subjected to severe loading and working conditions, caused by rising traffic levels and heavier vehicles, and increases in train running speeds makes the bridges extremely susceptible to degradation and failure. One of the promising tools for evaluating the overall safety and reliability of railway bridges is the bridge structural health monitoring (SHM) system, which not only monitors the structural conditions of bridges and maintains the safety of train operations, but also helps to expand the lifespan of bridges by enhancing their durability and reliability. While a multitude of review papers on SHM and vibration-based structural damage detection methods have been published in the past two decades, there is a paucity of literature that provides a review or overview on the SHM of railway bridges. Some of the review papers have become obsolete and do not reflect the state-of-the-art research. Therefore, the main goal of this article is to summarize state-of-the-art SHM techniques and methods that have been widely used and popular in recent years. First, two state-of-the-art SHM sensing technologies (i.e. fiber optic sensing (FOS) technology and computer vision-based (CV) technology) are reviewed, including the working principles of various sensors and their practical applications for railway bridge monitoring. Second, two state-of-the-art machine learning algorithms (i.e. convolutional neural networks (CNN) and transfer learning (TL)) and their applications for railway bridge structural condition assessment are exemplified. Third, the principle of digital twin (DT) and its applications for railway bridge monitoring are presented. Finally, issues related to the future direction and challenges of the monitoring technologies and condition assessment methods of railway bridges are highlighted.

Vibration-based structural damage detection via phase-based motion estimation using convolutional neural networks

Detection of structural damage is a major concern for engineers. In recent years, convolutional neural networks (CNNs) have been used for feature extraction and classification of vibration signals that reveal structural damage. Damage detection by CNNs greatly depends on high-quality learning data which are usually difficult to be obtained in actual engineering scenarios. To solve this problem, we combine phase-based motion estimation (PME) with the use of CNNs. By PME method, each pixel in a video can be regarded as a separate displacement sensor. Thus, it is possible to obtain millions of vibration signals from a single video, greatly facilitating CNN applications. We used a two-story steel structure for experimental validation. It was demonstrated that only one measured video sample obtained under each structural condition is possible to train a CNN model accurately detecting the location and severity of bolt looseness damage. This verified the outstanding performance of the proposed method.

Structural damage detection by integrating robust PCA and classical PCA for handling environmental variations and imperfect measurement data

This paper proposes a vibration-based structural damage detection approach considering the effects of uncertainties, including environmental variations and random errors that possibly stem from measurement and automatic modal identification. The existing methods that only employ the classical Principle Component Analysis (PCA) have been demonstrated effective to remove the effects of environmental variations while extremely sensitive to random errors. Therefore, the robust PCA is firstly introduced to remove the random errors, especially outliers, that significantly corrupt the low-rank property of the stacked damage sensitive feature (DSF) matrix. Then, the classical PCA is used to extract the environmental variation-free residues, which are inherently damage-dependent and can be used to detect the existence of damage. The problem of missing data is also considered in this study. It is tackled by adding virtual random errors to the locations of missing entities and thus can be addressed by the introduced robust PCA. The advantages of the proposed approach include: (1) Handling the random error-contaminated DSF data regardless of the error’s amplitude, which is an intractable problem for the existing classical PCA-based methods to consider the environmental effects; (2) Damage detection process can be automatic since the missing data can be automatically predicted and the random errors are not required to be manually distinguished. The effectiveness and performance of the proposed method are demonstrated on a numerical beam structure and an experimentally tested wooden bridge model.

A hybrid methodology for structural damage detection uniting FEM and 1D-CNNs: Demonstration on typical high-pile wharf

Vibration-based structural damage detection (SDD) has been a subject of intense research in structural health monitoring (SHM) for large civil engineering structures over the decades. The performance of the conventional SDD approaches predominantly relies on the rational choices of the damage feature and classifier. Hand-crafted features or fixed classifiers would not be the optimal choice for all structural damaged scenarios. This paper proposes a novel, quick and precise real-time SDD framework for high-pile wharf foundations using a combination of finite element modeling and 1D convolutional neural networks (CNNs). The distinct advantage of this method lies in extracting the damage-related features from the raw displacement response directly and automatically, and the computational complexity of the compact 1D CNNs is significantly lower because the data processing involves only simple 1D operations. The results show that the presented 1D CNNs have a superior ability to accurately identify the occurrence and location of damage in real time. In addition, the comprehensive performance of the CNNs trained by the displacement response dataset in component form is significantly better than that based on the dataset in absolute value form. The results also demonstrated that although the proposed CNNs are more sensitive to the longitudinal and lateral displacement responses of the high-pile wharf structure, the vertical component still has a positive effect on the improvement of the generalization and robustness of the CNNs.

A Framework of Structural Damage Detection for Civil Structures Using Fast Fourier Transform and Deep Convolutional Neural Networks

In the field of structural health monitoring (SHM), vibration-based structural damage detection is an important technology to ensure the safety of civil structures. By taking advantage of deep learning, this study introduces a data-driven structural damage detection method that combines deep convolutional neural networks (DCNN) and fast Fourier transform (FFT). In this method, the structural vibration data are fed into FFT method to acquire frequency information reflecting structural conditions. Then, DCNN is utilized to automatically extract damage features from frequency information to identify structural damage conditions. To verify the effectiveness of the proposed method, FFT-DCNN is carried out on a three-story building structure and ASCE benchmark. The experimental result shows that the proposed method achieves high accuracy, compared with classic machine-learning algorithms such as support vector machine (SVM), random forest (RF), K-Nearest Neighbor (KNN), and eXtreme Gradient boosting (xgboost).

An Investigation of Particle Swarm Optimization Topologies in Structural Damage Detection

In the past few decades, vibration-based structural damage detection (SDD) has attracted widespread attention. Using the response data of engineering structures, the researchers have developed many methods for damage localization and quantification. Adopting meta-heuristic algorithms, in which particle swarm optimization (PSO) is the most widely used, is a popular approach. Various PSO variants have also been proposed for improving its performance in SDD, and they are generally based on the Global topology. However, in addition to the Global topology, other topologies are also developed in the related literature to enhance the performance of the PSO algorithm. The effects of PSO topologies depend significantly on the studied problems. Therefore, in this article, we conduct a performance investigation of eight PSO topologies in SDD. The success rate and mean iterations that are obtained from the numerical simulations are considered as the evaluation indexes. Furthermore, the average rank and Bonferroni-Dunn’s test are further utilized to perform the statistic analysis. From these analysis results, the Four Clusters are shown to be the more favorable PSO topologies in SDD.

Automated Structural Damage Identification Using Data Normalization and 1-Dimensional Convolutional Neural Network

In the field of structural-health monitoring, vibration-based structural damage detection techniques have been practically implemented in recent decades for structural condition assessment. With the development of deep-learning networks that make automatic feature extraction and high classification accuracy possible, deep-learning-based structural damage detection has been gaining significant attention. The deep-learning neural networks come with fixed input and output size, and input data must be downsampled or cropped to the predetermined input size of the networks to obtain desired output of the network. However, the length of input data (i.e., sensing data) is associated with the excitation quality of a structure, adjusting the size of the input data while maintaining the excitation quality is critical to ensure high accuracy of the deep-learning-based structural damage detection. To address this issue, natural-excitation-technique-based data normalization and the use of 1-D convolutional neural networks for automated structural damage detection are presented. The presented approach converts input data to predetermined size using cross-correlation and uses convolutional network to extract damage-sensitive feature for automated structural damage identification. Numerical simulations were conducted on a simply supported beam model excited by random and traffic loadings, and the performance was validated under various scenarios. The proposed method successfully detected the location of damage on a beam under random and traffic loadings with accuracies of 99.90% and 99.20%, respectively.

A data-driven structural damage detection framework based on parallel convolutional neural network and bidirectional gated recurrent unit

With the extensive use of structural health monitoring technologies, vibration-based structural damage detection becomes a crucial task in both academic and industrial communities. Following the noteworthy trends of data-driven paradigms in recent years, some solutions have been released to identify, localize, and classify damages via deep neural networks. However, some deficiencies still exist for effective damage-intensive feature extraction and representation. To overcome such a problem, this paper proposes a novel end-to-end structural damage detection neural model by taking the advantages of the Convolutional Neural Network and Bidirectional Gated Recurrent Unit in parallel. The well-known IASC-ASCE benchmark and TCRF dataset are used for evaluation. The experimental results show that the proposed approach can achieve a better detecting effect than other existing manners.

Structural damage detection in steel frames using modal strain energy method and genetic algorithm

In recent years, the field of structural health monitoring (SHM) has been receiving the attention of many researchers. In particular, vibration-based structural damage detection methods have proven to be highly effective in civil engineering, mechanical engineering, aerospace engineering, … This paper presents a two-step procedure for damage detection in steel plane frame structures. In the first step, the damage locations in the frame are detected by Modal Strain Energy Index (MSEBI). This index is calculated on the difference from the modal strain energy value of every element before and after damage. In the second step, the damage extents are identified by Genetic Algorithm (GA). The objective function is established on the basis of the change of modal strain energy. A plane steel frame model is built by the finite element method. The results of the frame’s free vibration analysis are the input data of the proposed damage detection procedure. In order to improve the accuracy of the damage localization, the MSEBI is determined using the combination of the frame's first four bending mode shapes. The proposed method gives highly accurate results in detecting the damage’s location and extent for various damage scenarios.

Elliptical Method for Damage Identification in Carbon Fibre Reinforced Polymer Laminates

Oftentimes, researchers in the area of vibration-based structural health monitoring (SHM) and damage detection focused their attention on the global properties of structures, which are modal frequencies and modal damping factors. However, the effect on the local properties for SHM, that is, modal constants, has not been extremely explored. In this paper, the elliptical plane modal identification method is proposed to be used as a damage identification method itself. It is observed that when the receptance is plotted in the elliptical plane, the area of the ellipse formed close to the resonant frequencies (which depends on the modal constants) can be used to detect damage, namely, composite carbon fibre reinforced polymer (CFRP) rectangular plates. Although a mathematical correlation has not been established yet, results show that the method is sensitive to the presence of damage in the test plates, as the area of the ellipse changes with damage.

Output-only structural damage detection based on transmissibility measurements and kernel principal component analysis

Frequency response functions have been employed as damage-sensitive features in the vibration-based structural damage detection. However, the need for measuring the excitation forces arises as a remarkable limitation on the application of those features in real-world applications. As an alternative, transmissibility measurements can be explored as features with output-only nature, which implies the need for measuring only the response signals. In this paper, an output-only damage detection method is proposed, combining transmissibilities with kernel principal component analysis (KPCA). This technique is based on the pattern recognition paradigm for structural health monitoring, where feature extraction and feature classification phases are considered. In the first phase, the dimensionality of the transmissibilities is appropriately reduced by applying the KPCA algorithm. In the second phase, an outlier detection strategy is used to determine the condition of the instrumented structure. The possibility of clustering in the high-dimensional space mapped by KPCA is also reported and discussed. The proposed method is experimentally validated with transmissibilities acquired, under distinct structural conditions, from a laboratory steel beam instrumented with several accelerometers. The results demonstrate that the output-only method has high potential to be applied in a wide range of monitoring solutions, where economic issues and life-safety are primary motivations.

Structural Health Monitoring for Multi-story Shear Frames Based on Signal Processing Approach

Non-destructive vibration-based global structural health monitoring and damage detection methods are divided into modal-based and signal-based methods. There is much research in these areas, but each one has some advantages and disadvantages. The combination of these two areas can greatly overcome these disadvantages and thus can enhance the robustness of the damage detection methods. This article presents a signal-modal-based structural health monitoring technique for multi-story shear buildings subjected to an earthquake event. The proposed method utilizes the combination of wavelet transform and FFT as signal processing tools and also Hilbert transform with the aim of final modal parameters extraction, in order to determine the existence, location, and extent of damage in the multi-story shear frames. In fact, extraction of final modal parameters from the acceleration response signals is desired. To demonstrate the capabilities of the proposed algorithm, numerical simulations are performed on a four-story two-bay and ten-story two-bay shear frame with different damage scenarios using OpenSees. The results show that this method can detect the occurrence, location, and severity of the damage with good accuracy even in different levels of multi-damage scenarios and also in the presence of measurement noise.