Damage Identification Strategy Research Articles

A general approach to assessing SHM reliability considering sensor failures based on information theory

Structural health monitoring systems (SHM) involve implementing damage identification strategies to determine the health state of structures. However, it is important to pay close attention to the system degradation, especially the effect of sensor degradation on the SHM system reliability. This paper aims to formulate a general framework for evaluating SHM reliability that takes sensor failures into account. The framework involves modelling sensor network degradation processes using Petri nets (PNs) and calculating the expected information gain of the sensor network. The PNs allow for identifying the location and number of sensor failures. Kullback–Leibler (KL) divergence with Bayesian inversion is used to calculate the expected information loss due to sensor failure. Two case studies are used to illustrate the methodology: (i) a damage localization scheme using an ellipse-based time-of-flight (ToF) model and (ii) a damage identification scheme using a guided waves damage interaction model. The proposed framework is demonstrated by both numerical and physical experimental case studies. Whereas the case studies are specific to an ultrasonic guided wave monitoring system, the proposed approach is generic. The proposed model is able to predict the health condition state and utility of SHM, which can potentially help in constructing asset management models in various industries.

ARX-Naïve Bayes based structural damage identification

ABSTRACT The nature of damage identification is close to a pattern recognition process that classifies different damage patterns. The naïve Bayes classifier (NBC) can effectively handle multiple-classification problems by choosing patterns with high probabilities. Therefore, by absorbing the autoregressive model with exogenous inputs (the ARX model), an ARX-Naïve Bayes damage identification strategy has been proposed in which the autoregressive coefficients of the ARX model are taken as the sensitive damage feature. The classification training and test sample datasets are then built on these coefficients corresponding to various damage scenarios. The model order of an AR model is first determined for the subsequent order selection of the ARX model, whose autoregressive coefficients are further used to construct the NBC. This procedure can enhance the pattern recognition robustness to uncertainties such as measurement noises. Different damage patterns are determined by calculating the sum of logarithmic likelihoods of testing samples. The effectiveness of the proposed method has been verified against a bridge benchmark model having different damage scenarios under the noise pollution. In addition, an experimental five-story shear frame structure was adopted for validation, showing that compared with the SVM algorithm suitable for handling binary classification problems, the proposed method excels in multi-classification of damage patterns.

Structural damage identification with output-only strain measurements and swarm intelligence algorithms: a comparative study

The identification of structural damage with the unavailability of input excitations is highly desired but challenging since structural dynamic responses are affected by the coupling effect of structural parameters and external excitations. To deal with this issue, in this paper, an output-only damage identification strategy based on swarm intelligence algorithms and correlation functions of strain responses is proposed to identify structures subjected to single or multiple unknown white noise excitations. In the proposed strategy, four different population-based optimization algorithms—particle swarm optimization, the butterfly optimization algorithm, the tree seed algorithm, and a micro search Jaya (MS-Jaya)—are employed and compared. The micro search mechanism is integrated into a basic Jaya algorithm to improve its computational efficiency and accuracy by eliminating some damage variables with small values for the identified best solution after several iterations. The objective function is established based on the proposed auto/cross-correlation function of strain responses and a penalty function. The effectiveness of the proposed method is verified with numerical studies on a simply supported beam structure and a steel grid benchmark structure under ambient excitation. In addition, the effect of the reference point, number of sensors, and arrangement of strain gauges on the performance of the proposed method are discussed in detail. The investigated results demonstrate that the proposed approach can accurately detect, locate, and quantify structural damage with limited sensors and 20% noise-polluted strain responses. In particular, the proposed MS-Jaya algorithm presents a more superior capacity in solving the optimization-based damage identification problem than the other three algorithms.

Damage assessment on an asymmetric sandwich composite made of natural materials using strain gauges

Sandwich composites have been a growing subject of interest to various industries, thus it has become essential to monitor these structures. There is a need to implement damage identification and characterization strategies. This work aims to study the reliability of the implementation of strain gauge sensors, the mechanical properties of an asymmetric sandwich composite made by a core of cork agglomerate, two fibre-reinforced epoxy skins and a natural stone on top. Initially, destructive bending tests were conducted to establish the maximum load for the specimens. This served as a crucial reference point for subsequently employing strain gauges to investigate the mechanical properties of the composite. Finally, the destructive bending tests were performed using strain gauges in order to assess the performance of strain gauges to detect damage within the composite. The study provides comprehensive insights into various mechanical properties, including bending strength, tensile, compressive, and flexural moduli, as well as the determination of the neutral line location within the material. Notably, the results from the destructive bending tests employing strain gauges demonstrated their effectiveness in detecting catastrophic events within the composite. This research contributes to the advancement of reliable structural integrity monitoring and characterization techniques for sandwich composites.

Application of Vibration Data Mining and Deep Neural Networks in Bridge Damage Identification

The aim of this paper is to mine the information contained in the bridge health monitoring data as well as to improve the shortcomings of traditional identification methods. In this paper, a bridge damage identification method based on the combination of data mining and deep neural networks is introduced. Firstly, a noise reduction method based on parameter optimisation of wavelet threshold decomposition is proposed, which further removes the noise signal by introducing two adjustment parameters in the threshold function to adapt to different wavelet decomposition layers. Furthermore, the Fast Fourier Transform is used to analyse the feature pattern of the original signal in the frequency domain, and the modal frequency features that exhibit the difference in damage categories are extracted from the spectrogram through sliding windows. Finally, a large number of irrelevant variables with small weight contributions are discarded by principal component analysis, and only the sensitive features with the most informative categories are retained as the input to the deep neural networks. The experimental results show that the new metrics after the feature engineering process improve the ability of damage identification and have stronger robustness, while our damage identification scheme achieves a good balance between the model computation and recognition accuracy. Furthermore, the recognition accuracy of the deep neural networks reaches over 93% with only three feature dimensions retained.

Damage Quantification and Identification in Structural Joints through Ultrasonic Guided Wave-Based Features and an Inverse Bayesian Scheme.

In this paper, defect detection and identification in aluminium joints is investigated based on guided wave monitoring. Guided wave testing is first performed on the selected damage feature from experiments, namely, the scattering coefficient, to prove the feasibility of damage identification. A Bayesian framework based on the selected damage feature for damage identification of three-dimensional joints of arbitrary shape and finite size is then presented. This framework accounts for both modelling and experimental uncertainties. A hybrid wave and finite element approach (WFE) is adopted to predict the scattering coefficients numerically corresponding to different size defects in joints. Moreover, the proposed approach leverages a kriging surrogate model in combination with WFE to formulate a prediction equation that links scattering coefficients to defect size. This equation replaces WFE as the forward model in probabilistic inference, resulting in a significant enhancement in computational efficiency. Finally, numerical and experimental case studies are used to validate the damage identification scheme. An investigation into how the location of sensors can impact the identified results is provided as well.

Model-informed deep learning strategy with vision measurement for damage identification of truss structures

Structural damage identification approaches can be divided into two categories, i.e. data-driven approaches via statistical pattern recognition and model-based approaches via finite element (FE) model updating. These two approaches have their own merits, and their main shortcomings can be remedied by each other’s merits. Therefore, this study proposed a deep learning-based damage identification strategy involving both data-driven and model-based approaches, termed as model-informed deep learning (MIDL)-based strategy. This strategy first proposes a vision-based displacement estimation approach to extract structural displacement responses from video data. This approach reduces the displacement drift induced by conventional optical flow approaches and improves the tracking accuracy of feature points. Then, a calibrated FE model is built to construct data sets with different damage levels via FE model updating and time-history analysis. Following this, a one-dimensional convolutional neural network (1D CNN) is established to detect and localize structural damage by using direct displacement responses. Finally, FE model updating is performed again to quantify structural damage level with constrained targets. A truss structure is further used to evaluate the accuracy of the proposed strategy experimentally. Results illustrate that the proposed MIDL strategy, which uses time series to localize the structural damage, achieves a global location accuracy of 86.09% and avoids the feature extraction process. Meanwhile, with the known damage location, the efficiency and accuracy of structural damage quantification via rerunning model updating can also be significantly improved. In addition, the displacements estimated by the proposed approach have a good match with ground truth values with error standard deviations of less than 0.3 mm.

Time-Domain Structural Damage Identification Using Ensemble Bagged Trees and Evolutionary Optimization Algorithms

This paper presents a two-step vibration-based strategy for damage identification of framed structures using ensemble bagged trees known as a well-known supervised machine learning (ML) paradigm in conjunction with evolutionary optimization algorithms. The proposed model incorporates the actual response, wavelet coefficients, and wavelet energy to extract damage-sensitive features from the time-domain of the measured and simulated signals. Unlike available studies in this scope, the key objective of this research is to identify damage with a localization precision down to a single structural member, rather than limiting the evaluation to the group of elements. In order to increase the training performance in contributing to extremely large datasets with numerous class labels, the proposed strategy involves the artificial generation of features. Additionally, a modified genetic algorithm is proposed for fast damage localization. It is deduced that the damage locations are confidently detected within a fast computational time. Subsequently, damage identification is followed by the application of evolutionary optimization algorithms. For comparison purpose, the employment of the water cycle optimization algorithm (WCA) is comparatively investigated with the other three state-of-the-art optimizers, i.e., particle swarm optimization (PSO), imperialist competitive algorithm (ICA), and differential evolution algorithm (DE). The numerical and experimental validation studies evidence satisfactorily reliable identification results with no false detection in dealing with multiple damage scenarios in large-scale and real-world applications. It is concluded that developing the most damage-sensitive features and using the proposed data fusion strategy lead to informative features with a reasonably small size and significantly improve the ML performance.

Combined digital twin and hierarchical deep learning approach for intelligent damage identification in cable dome structure

Accurate identification of structural damage is the most critical step in structural health monitoring. Traditional damage identification strategies are easily interfered by environmental and human factors, resulting in low-accuracy of identification. In this light, a combined digital twin (DT) and hierarchical deep learning (DL) approach for intelligent damage identification in cable dome structures is proposed in this paper. Based on actual engineering cases, a DT model that accurately maps the physical structure of the cable dome is constructed using APDL based on data. A cable dome structure damage sample database is then automatically established through the large-scale finite element analysis of DT. Finally, the damage features of the data samples are extracted using the hierarchical DL framework proposed in this study. Accuracy verification based on cable force confirms that the established DT model can accurately reflect the mechanical state of the physical structure. The identification results of the trained network on a test set demonstrate that the proposed framework can intelligently identify the damage type, damage location, and damage degree in the cable dome structure with a high accuracy and strong robustness. The proposed intelligent damage identification approach is feasible and reliable and can provide a new basis for structural damage identification with broad application prospects.

Output-only structural damage identification using hybrid Jaya and differential evolution algorithm with reference-free correlation functions

To solve the optimization-based structural damage identification problem, a novel hybrid algorithm based on Jaya and differential evolution algorithm (HJDEA) is proposed to detect, locate and quantify structural damages by effectively incorporating the powerful local exploitation capacity of Jaya algorithm and global exploration capability of differential evolution. Meanwhile, Hammersley sequence initialization and Lévy flight search mechanism are introduced into HJDEA to further improve convergence rate and refining the quality of the best solution. Four different algorithms, genetic algorithm, particle swarm optimization, Jaya and the proposed HJDEA are employed for comparative study. In addition, the objective function is established by adjacent acceleration correlation function so as to avoid false identification caused by defining improper reference point. The performance of the proposed damage identification strategy based on HJDEA and adjacent acceleration correlation function is investigated with numerical examples involving an 8-DOF lumped mass model and a cantilever beam, as well as an experimental study of the ASCE benchmark structure under white noise excitation. Results show that the proposed hybrid identification method is accurate, efficient and robust in the identification of the damage existence, location and severity of stiffness and mass parameters even with partial output-only responses and 20% noise-polluted measurements.

Guided waves-based damage identification in plates through an inverse Bayesian process

The use of guided waves to identify damage has become a popular method due to its robustness and fast execution, as well as the advantage of being able to inspect large areas and detect minor structural defects. When a travelling wave on a plate interacts with a defect, it generates a scattered field that will depend on the defects geometry. By analysing the scattered field, one can thus characterize the type and size of the plate damage. A Bayesian framework based on a guided waves interaction model for damage identification of infinite plate for the first time is presented here. A semi-analytical approach based on the lowest order plate theories is adopted to obtain the scattering features for damage geometries with circular symmetry, resulting in an efficient inversion procedure. Subsequently, ultrasound experiments are performed on a large aluminium plate with a circular indentation to generate wave reflection and transmission coefficients. With the aid of signal processing techniques, the effectiveness and efficiency of the proposed approach are verified. A full finite element model is used to test the damage identification scheme. Finally, the scattering coefficients are reconstructed, reliably matching the experimental results. The framework supports digital twin technology of structural health monitoring.

Visible Particle Series Search Algorithm and Its Application in Structural Damage Identification.

Identifying structural damage is an essential task for ensuring the safety and functionality of civil, mechanical, and aerospace structures. In this study, the structural damage identification scheme is formulated as an optimization problem, and a new meta-heuristic optimization algorithm, called visible particle series search (VPSS), is proposed to tackle that. The proposed VPSS algorithm is inspired by the visibility graph technique, which is a technique used basically to convert a time series into a graph network. In the proposed VPSS algorithm, the population of candidate solutions is regarded as a particle series and is further mapped into a visibility graph network to obtain visible particles. The information captured from the visible particles is then utilized by the algorithm to seek the optimum solution over the search space. The general performance of the proposed VPSS algorithm is first verified on a set of mathematical benchmark functions, and, afterward, its ability to identify structural damage is assessed by conducting various numerical simulations. The results demonstrate the high accuracy, reliability, and computational efficiency of the VPSS algorithm for identifying the location and the extent of damage in structures.

Damage Identification of Pipeline Based on Ultrasonic Guided Wave and Wavelet Denoising

Guided wave technology shows excellent capacity in the damage identification of the pipeline. In practical inspection, the detection ability is greatly reduced due to the interference echo signal from the feature structure in the pipeline. In this paper, the propagation models of guided wave in straight pipe and pipe with feature structures are established, and the effect of feature structure on the guided wave echo signal is analyzed, including pipe with an elbow, ring joint, and accessory structure, and the corresponding damage identification strategies for pipe with different forms of features are proposed. The damage index of the echo signal difference and radial displacement signal is proposed to extract the signal from damage. Besides, to eliminate the noise effect caused by the actual detection situation, a novel guided wave denoising method based on the wavelet analysis is studied, and the analysis results indicate that the damage detection strategy based on the guided wave can accurately locate the damage region in both straight pipe and pipe with features and that the proposed denoising method can effectively eliminate the noise in the echo signal.

Structural damage identification of offshore wind turbines: A two‐step strategy via FE model updating

This study proposes a two-step strategy for damage identification of offshore wind turbines via Finite Element (FE) model updating. In this methodology, the first step identifies the damaged component(s) using a verified global mathematical model of the offshore wind turbine under operational wind and wave loading. Damaged components can be the tower, a certain blade, or the foundation. Natural frequencies and mode shapes obtained from operational modal analysis are used to formulate the objective function for deterministic model updating. With the identified damaged component(s) in the first step, the second step of the strategy focuses on exact damage localization and quantification of the damaged component(s). To this end, an FE model of a baseline 5 MW monopile offshore wind turbine is established using ANSYS. Wind and wave loading are applied to simulate the structural responses under different damage scenarios. In additional to natural frequencies and mode shapes, mode shape curvatures obtained in the second step are included in the objective function. Results indicate that the proposed methodology can correctly identify most structural damage on the offshore wind turbines using operational data. The proposed two-step damage identification strategy can circumvent the requirement of intensive computation due to full model updating of offshore wind turbine models and thus greatly improve the efficiency in real application.

Static strain-based identification of extensive damages in thin-walled structures

Interest has been expressed during the past few years toward incorporating structural health monitoring (SHM) systems in ship hull structures for detecting damages that cause significant load-carrying reductions and subsequent load redistributions. The guiding principle of the damage identification strategy considered in this work is based upon measuring, through a limited number of sensors, the static strain redistributions caused by an extensive damage. The problem is tackled as a statistical pattern recognition one, and therefore, methods sourcing from machine learning (ML) are applied. The SHM strategy is both virtually and experimentally applied to a thin-walled prismatic geometry that represents an idealized hull form solely subjected to principal bending stresses (sagging/hogging). Damage modes causing extensive stress redistribution, are abstractly represented by a circular discontinuity. The damage identification problem is treated in a hierarchical order, initialized by damage detection and moving to an increasingly more localized prediction of the damage location. Training datasets for the ML tools are generated from numerical finite element simulations. Measurement uncertainty is propagated in the theoretical strains by information inferred from experimental data. Two different sensor architectures were assessed. An experimental programme is performed for testing the accuracy of the proposed damage identification strategy, yielding promising results and providing valuable insights.

A two-step interval structural damage identification approach based on model updating and set-membership technique

External conditions unavoidably cause deterioration to structures in service, and the degeneration in material property may accumulate and lead to huge accidents. Meanwhile, practical engineering includes a variety of uncertainties from responses, which can negatively influence the damage identification result. In this study, a two-step interval structural damage identification approach is proposed with consideration of response uncertainty. Deterministic damage identification strategy by which damage index can be solved is firstly introduced. Then, damage locations are identified as the first step based on model updating. For the second step, uncertain structural dynamic responses are viewed as interval variables. Then, the bounds of the damage extents of the identified elements are estimated using Chebyshev polynomial approximation and interval mathematics. For a tighter solution, set-membership technique is used to obtain the fusion interval damage index. Two numerical verifications and an experimental verification show the use of the presented two-step identification method. In addition, the Taylor expansion and interval combination methods are also analyzed for comparable purpose.

Combined two-level damage identification strategy using ultrasonic guided waves and physical knowledge assisted machine learning

Structural Health Monitoring of composite structures is one of the significant challenges faced by the aerospace industry. A combined two-level damage identification viz damage detection and localization is performed in this paper for a composite panel using ultrasonic guided waves. A novel physical knowledge-assisted machine learning technique is proposed in which domain knowledge and expert supervision is utilized to assist the learning process. Two supervised learning-based convolutional neural networks are trained for damage detection (binary classification) and localization (multi-class classification) on an experimental benchmark dataset. The performance of the trained models is evaluated using loss curve, accuracy, confusion matrix, and receiver-operating characteristics curve. It is observed that incorporating physical knowledge helps networks perform better than a direct deep learning approach. In this work, a combined damage identification strategy is proposed for a real-time application. In this strategy, the damage detection model works in an outer-loop and predicts the state of the structure (undamaged or damaged), whereas an inner-loop predicts the location of the damage only if the outer-loop detects damage. It is seen that the proposed technique offers advantages in terms of accuracy (above 99% for both detection and localization), computational time (prediction time per signal in milliseconds), sensor optimization, in-situ monitoring, and robustness towards the noise.

A combined finite element and hierarchical Deep learning approach for structural health monitoring: Test on a pin-joint composite truss structure

Structural Health Monitoring (SHM) is an emerging field of engineering with a wide range of applications. The most common SHM strategies operate on structural responses through vibration measurements and focus on training mathematical classifiers which are used after to identify damage in unknown responses. Classifiers may additionally locate damage when adequate labeled damaged data is available. In the present work, a novel SHM method is presented where labeled damaged data is generated through FE models for a pin-joint composite truss structure employing a model-based approach for the problem of data acquisition. The truss is made of carbon fiber reinforced polymer (CFRP) members joint on aluminum connections forming a complex and large FE problem. A Deep Learning (DL) Convolutional Neural Network (CNN) classifier is trained on the FE generated vibration data combined with a hierarchical multiple damage identification and location scheme. The numerically trained CNN is after validated on experimental statuses of the truss in both damage detection and location, proving to be robust and accurate for the considered test case. The potential of hierarchical CNNs with FE based SHM data for multiple damages is investigated in this work and a comparison is given between hierarchical and direct multiclass CNNs. The large performance gains of the former are proven for the studied experimental case highlighting also the importance of SHM system architectures with CNNs.

A novel structural damage identification scheme based on deep learning framework

Deep learning algorithm can autonomously mine the representative information which is hidden in the data and provides a new idea for damage identification of building structures. Keeping in view that a structural damage identification based on time series data has low accuracy, a new method for structural damage identification based on IASC-ASCE SHM benchmark is proposed which combines the advantages of Hilbert-Huang transform (HHT) and deep neural network. The damage signal of the Benchmark model is first analyzed by HHT. After that the obtained time-frequency graph and the marginal spectrum of the signal are used as the input of the convolutional neural network. In addition, the structural parameters of CNN model are adaptively optimized by particle swarm optimization (PSO) algorithm to ensure better performance of CNN. Compared with other traditional methods (ANN, SVM), the experimental results show that the newly proposed damage identification method has significant performance advantages. Accuracy of the proposed CNN model is improved more than 10% after getting optimized by PSO in comparison with non-adaptive CNN model. CNN model also showed better robustness against noise interference.

A multiscale reconstructed attractors-based method for identification of structural damage under impact excitations

Vibrational responses-based damage identification has been widely investigated in recent decades. Different from existing studies that process vibrational responses in time, frequency, or time-frequency domains, a newly emerged reconstructed attractor-based damage identification is becoming a novel strategy for damage identification from vibrational responses. In this aspect, reported studies to date demonstrate the dependence of the efficacy of this method on the tuned chaotic excitation quantified by the Lyapunov exponents. Nevertheless, the tuned chaotic excitation is not always actualized for a complicated engineering structure in reality, seriously hindering the application of the method. To conquer this problem, this study retrofits the conventional reconstructed attractor by introducing multiscale analysis offered by the stationary wavelet transform into phase space reconstruction of attractors. Such a retrofit leads to a novel quantity of multiscale reconstructed attractors (MRAs) that consist of a set of attractors reconstructed on individual scales of vibrational responses. The MRAs characterize structural damage from the compound variations of a set of attractors topological trajectories, termed the MRAs-based prediction error. Compared with the conventional reconstructed attractor, the MRAs-based prediction error features detecting structural damage with great accuracy and strong noise immunity, and especially no need for tuned chaotic excitation. The feasibility of the MRAs in damage identification are numerically verified and its effectiveness are experimentally validated on cantilever beams. The proposed method expands the applications of reconstructed attractors from vibrational responses on structural health monitoring.