Structural Damage Localization Research Articles

Locating and Quantifying Damage in Deck Type Arch Bridges Using Frequency Response Functions and Artificial Neural Networks

Vibration-based methods can be used to detect damage in a structure as its vibration characteristics change with physical changes in the structure. Arch bridge is a popular type of bridge with rather complex vibration characteristics which pose a challenge for using existing vibration-based methods to detect damage in the bridge. Further, its particular geometry with a curved arch rib and vertical members (either in compression or tension) to support the horizontal deck makes the process of damage quantification using vibration-based methods harder and challenging. This paper develops and presents a vibration-based method that utilizes damage pattern changes in frequency response functions (FRFs) and artificial neural networks (ANNs) to locate and quantify damage in the rib of deck-type arch bridge, which is the most important load bearing component in the bridge. Principal component analysis, which is performed to reduce the dimension of original FRF data series and to obtain limited principal component analysis (PCA)-compressed FRF data is used in the development of the proposed method. FRF change, which is the difference in the FRF data between the intact and the damaged structure, is compressed to a few principal components and fed to ANNs to predict the location and severity of structural damage. The process and the hierarchy of developed ANN systems are presented, including the “fusion network” concept, which individually analyses FRF-based damage indicators separated by sensor locations. Finally, results obtained for many tested damage cases (inverse problems) are presented, which demonstrate the applicability of the proposed method for locating and quantifying damage in the rib of deck type arch bridge.

Model-based damage identification of railway bridges using genetic algorithms

The assessment of structural integrity via numerical model updating has been drawing attention in several areas of engineering over the last years. Basically, it consists in an optimization process based on the minimization of the residuals between measured and estimated numerical responses. In such methodologies, several factors influence the success of both localization and quantification of structural damage, such as: the damage features used in the formulation of the objective function, the optimization algorithm and the adopted updating parameters. Many existing studies using these methods are applied to simple structural systems, e.g., beams, frames and trusses. However, few studies applied to large and complex structures are found in the literature. In this context, this work proposes to assess the performance of a genetic algorithm-based approach applied to two case studies. The first case refers to a two-dimensional model of a hypothetical railway bridge, where the efficiency and robustness of five different indicators are assessed considering three damage scenarios. In the second case, a real railway bridge is considered. The results obtained show that the proposed approach is able to detect, locate and quantify multiple damage with several updating parameters and few target responses.

Two-step structural damage detection method for shear frame structures using FRF and Neumann series expansion

In this paper, a novel damage detection method for shear frame structures was presented. The proposed approach was derived from frequency response function (FRF) measurements before and after damage. Using a Neumann series technique, the damaged FRF data were expanded as functions of a truncated intact FRF matrix. The truncated approximate series equations with multiple frequency points were solved using the constrained optimization method. Nevertheless, a correlation of the damage detection solution vector was defined as the structural damage location index. Unlike existing methods, the effect of noise on damage detection was greatly suppressed by simultaneously increasing the number of equations and reducing the unknown coefficients. The proposed approach was validated numerically and experimentally with a 12-story shear frame structure model and a 7-story steel-frame structure, respectively. The results of different damage scenarios showed that the proposed method has good noise immunity and high accuracy.

Experimental and Numerical Impact Analysis of Automotive Bumper Brackets Made of 2D Triaxially Braided CFRP Composites.

The impact behavior of carbon fiber epoxy bumper brackets reinforced with 2D biaxial and 2D triaxial braids was experimentally and numerically analyzed. For this purpose, a phenomenological damage model was modified and implemented as a user material in ABAQUS. It was hypothesized that all input parameters could be determined from a suitable high-speed test program. Therefore, novel impact test device was designed, developed and integrated into a drop tower. Drop tower tests with different impactor masses and impact velocities at different bumper bracket configurations were conducted to compare the numerically predicted deformation and damage behavior with experimental evidence. Good correlations between simulations and tests were found, both for the global structural deformation, including fracture, and local damage entities in the impact zone. It was proven that the developed phenomenological damage models can be fully applied for present-day industrial problems.

Structural Damage Identification Based on the Wavelet Transform and Improved Particle Swarm Optimization Algorithm

A method based on the wavelet transform and improved particle swarm optimization (WIPSO) algorithm is proposed to identify the microdamage of structures. First, the singularity of wavelet coefficients is used to identify the structural damage location, and then, the improved particle swarm optimization (IPSO) algorithm is used to calculate the optimal solution of the objective function of the structural damage location to determine the structural damage severity. To study the performance of WIPSO, the structural microdamage severity is set within 10%, and a numerical simulation and experimental structure under different damage scenarios are considered. In addition, the ability of wavelet coefficients to identify the location of the structural damage under different noise levels is studied. To evaluate the performance of IPSO, the standard particle swarm optimization algorithm with an inertia weight factor of 0.8 (0.8PSO), the genetic algorithm (GA), and the bat algorithm (BA) are also considered. The results show that WIPSO can effectively and accurately identify the structural damage location and severity. Wavelet transform is very robust to the structural damage location. Compared with the standard 0.8PSO and other mainstream algorithms, IPSO has good convergence and performs more stable and more accurate in the identification of structural damage severity.

DEEP CASCADED NEURAL NETWORKS FOR AUTOMATIC DETECTION OF STRUCTURAL DAMAGE AND CRACKS FROM IMAGES

Abstract. In this paper, two different convolutional neural networks (CNNs) are applied on images for automated structural damage detection (SDD) in earthquake damaged structures and cracking localization (e.g., detection of cracks, their widths and distributions) at various scales, such as pixel level, object level, and structural level. The proposed method has two main steps: 1) diagnosis, and 2) localization of cracking or other damage. At first a residual CNN with transfer learning is employed to classify the damage in the structures and structural components. This step performs damage detection using two public datasets. The second step uses another CNN with U-Net structure to locate the cracking on low resolution images. The implementations using public and self-collected datasets show promising performance for a problem that had remained a challenge in the structure engineering field for a long time and indicate that the proposed approach can perform detection and localization of structural damage with an acceptable accuracy.

Oxidized-LDL inhibits testosterone biosynthesis by affecting mitochondrial function and the p38 MAPK/COX-2 signaling pathway in Leydig cells

Abnormal lipid/lipoprotein metabolism induced by obesity may affect spermatogenesis by inhibiting testosterone synthesis in Leydig cells. It is crucial to determine which components of lipoproteins inhibit testosterone synthesis. Circulating oxidized low-density lipoprotein (oxLDL), the oxidized form of LDL, has been reported to be an independent risk factor for decreased serum testosterone levels. However, whether oxLDL has a damaging effect on Leydig cell function and the detailed mechanisms have been rarely studied. This study first showed the specific localization of oxLDL and mitochondrial structural damage in testicular Leydig cells of high-fat diet-fed mice in vivo. We also found that oxLDL reduced the mitochondrial membrane potential (MMP) by disrupting electron transport chain and inhibited testosterone synthesis-related proteins and enzymes (StAR, P450scc, and 3β‑HSD), which ultimately led to mitochondrial dysfunction and decreased testosterone synthesis in Leydig cells. Further experiments demonstrated that oxLDL promoted lipid uptake and mitochondrial dysfunction by inducing CD36 transcription. Meanwhile, oxLDL facilitated COX2 expression through the p38 MAPK signaling pathway in Leydig cells. Blockade of COX-2 attenuated the oxLDL-induced decrease in StAR and P450scc. Our clinical results clarified that the increased serum oxLDL level was associated with a decline in circulating testosterone levels. Our findings amplify the damaging effects of oxLDL and provide the first evidence that oxLDL is a novel metabolic biomarker of male-acquired hypogonadism caused by abnormal lipid metabolism.

Damage localization in plate-like structures using time-varying feature and one-dimensional convolutional neural network

Lamb wave-based SHM technology for damage detection and localization in plate-like structures has typically relied on post-processing of ultrasonic guided waves. Traditionally, the damage localization is realized using the time-of-flight (TOF) of damage-scattered waves. However, this method often requires the identification of a pure mode from the wave signal which is difficult in many cases. Damage index (DI) based methods offer another type of approaches that do not need such singal explanation. Since DI alone doesn’t contain temporal information, data fusion of signals from multiple actuator-sensor pairs must be performed for localization. As a result, a relatively dense actuator-sensor network is needed, and localization can only be realized within the region covered by the network. Realizing that temporal information contained in the wave signal is extremely important to damage localization, we propose a time-varying DI feature that preserves the temporal information to improve localization accuracy. In addition, we propose to use one-dimensional convolutional neural network (1-D CNN) to correlate the time-varying DI directly with the damage location. The equivariance property of CNN preserves the temporal information. The efficiency and feature extraction capability of the CNN help to build a neural network model with certain generalization capability, and thus the model trained on one plate can be applicable to a new plate. The performance of the proposed method was demonstrated in three cases: localization in the same plate with different damage locations, localization in a new plate with the same damage locations, and localization in a new plate but with different damage locations. Despite that only four transducers were used, and limited experimental data for training were available, good results have been obtained. Performance comparison with several other existing methods was also conducted.

Damage detection in girder bridges using modal curvatures gapped smoothing method and Convolutional Neural Network: Application to Bo Nghi bridge

This paper addresses a damage detection method based on changes in modal curvature combined with Convolutional Neural Network (CNN). The use of modal curvature for damage detection is very well known. Some methods require modal data of healthy structures as a reference, others do not. Gapped Smoothing Method (GSM) is a vibration-based damage detection methodology, which makes use of modal curvature for identifying the location of structural damage and does not require information of the intact structure. The Bo Nghi bridge is used as an illustrative example. This bridge consists of four T-shaped concrete simply supported girders. One single beam with the same length and cross-section of the bridge girder is modeled and used to extract numerical data to train the CNN. A CNN is trained by using images from the damage index of the GSM to classify the damage location in the numerical beam model. Finally, the finite element model of the bridge is built and used to model damage scenarios to test the trained CNN. The results indicate that the combination of GSM and CNN can be used for damage detection and localization.

Model-free damage localization of structures using wavelet based grey relational analysis

Damage detection of real-life civil structures is generally performed under a condition of poor information, mainly referring to our insufficient knowledge of structures, uncertainties and limited sensors. Moreover, most damage detection methods require the numerical or mathematical models of a structure, whose stiffness parameters are sought by inverse solutions such as model updating. However, ill-conditioning problems are often found when updating complex structures. Hence, a model-free damage localization strategy has been proposed incorporating wavelet packet transform with grey relation analysis. First acceleration response signals are decomposed into different frequency subbands using wavelet packet transform. The wavelet packet energies of specific frequency subbands corresponding to fundamental modal frequencies are extracted and then normalized by the total energy of all the accelerometers. After that, wavelet packet energy curvatures are constructed and incorporated with grey relation analysis to propose a wavelet grey correlation coefficient. Damage locations are determined according to magnitude changes in these coefficients. Lastly, the feasibility of the proposed method has been successfully verified against an experimental steel beam. The validation also found that direct damage localization based on mode shapes or modal curvatures could not provide effective judgment.

An Improved Multiple Damage Identification Method of Plate Structure Using Discrete Wavelet Transform

In order to improve the identification accuracy of multiple damage of the plate structures, a method based on a two-dimensional discrete wavelet transform is proposed. Firstly, the structural characteristic parameters of the undamaged and damaged plate structures are obtained through the finite element modal analysis. The discrete wavelet transform is used to decompose and evaluate the modal node displacement data of the rectangular thin plate, and the high frequency information and singular values of the structural damage location are calculated. Secondly, the quantitative analysis of damage is realized by the change rate of the maximum value of the wavelet coefficient. Finally, according to the proposed algorithm, the anti-noise performance of the algorithm is verified through different ratios of noise. The results are shown that the vibration-based damage identification method is effective to detect the location and severity of the rectangular thin plate structures, which can be used to solve the damage identification problems of complex structures.

Rapid post-earthquake damage localization and quantification in masonry structures through multidimensional non-linear seismic IDA

This paper presents a methodology aimed at addressing the rapid post-earthquake damage localization and quantification tasks in heritage masonry structures, based on non-linear Incremental Dynamic Analysis (IDA). The proposed methodology relies on pre-run multidimensional non-linear IDA simulations carried out using a numerical Finite Element model together with vibration data recorded during an earthquake. The IDA curves are built with reference to different portions of the structure, relating meaningful local damage parameters to relevant seismic intensity measures. The selection of appropriate intensity parameters is crucial and a study on this aspect is carried out. The methodology is validated through application to a numerical model of a reduced-scale masonry structure, called Brick House, which is a well-known international benchmark tested on the LNEC-3D shaking table. The obtained results demonstrate that, if the set of IDA curves has been previously constructed using a suitable model, the proposed methodology yields an immediate and accurate estimation of damage conditions after an earthquake is recorded.

Study on brain structure network of patients with delayed encephalopathy after carbon monoxide poisoning: based on diffusion tensor imaging.

To analyze the network alteration characteristics of brain structure network in patients with delayed encephalopathy after CO poisoning (DEACMP) based on diffusion tensor imaging (DTI), and to explore the structural correlation neuroimaging mechanism of DEACMP cognitive impairment. DTI scanning was performed in 33 patients with DEACMP and 25 healthy controls (HCs) who were matched in age and sex. The whole brain was divided into 90 regions by automated anatomical marker templates. The continuous tracing method was used to reconstruct the brain fiber bundle connection and construct the brain structure weighted network. The global and regional properties were computed by graph theoretical analysis. To compare the brain network regional properties between the DEACMP group and the HCs group, two-sample t test (false discovery rate correction, P<0.05) was utilized. The correlations between the brain structural network properties and clinical parameters were further analyzed. Both of the two groups were found to follow the efficient small-world characteristics. The shortest path length of the DEACMP group increased (Lp = 0.86 ± 0.05), whereas global efficiency (Eglob = 9.60 ± 2.65) and local efficiency (Eloc = 17.98 ± 3.89) decreased. Moreover, the core nodes of the DEACMP group's default network, highlighting network, central execution network, and visual area, were decreased (P < 0.05, FDR correction). The left amygdala node degree of DEACMP group was positively correlated with MMSE and MoCA scores of the clinical scale (r = 0.863, P = 0.001, r = 0.525, P = 0.021). The node degree value of the left lingual gyrus was positively correlated with MoCA score (r = 0.406, P = 0.019) and negatively correlated with CDR score (r = -0.563, P = 0.016). The efficiency value of the right dorsolateral superior frontal gyrus in the DEACMP group was negatively correlated with the CDR score (r = -0.377, P = 0.031). By comparing the differences and changes in the topological properties and nodes of the brain structure network between DEACMP group and HCs group, the degree of related brain regions, especially the damage of higher brain functions in DEACMP patients, was verified, which was helpful to understand the cognitive damage caused by CO poisoning and to predict the efficacy of late remodeling. Small-worldness is a dynamic reorganization of the small-world topology and its community structure from the brain network to provide system-wide flexibility and adaptability (Barbey in Trends Cogn Sci 22(1):8-20, 2018). The combination with DTI is helpful for the accurate localization of brain structural damage, especially in DEACMP patients.

Inverse structural damage identification problem in CFRP laminated plates using SFO algorithm based on strain fields

Damage detection methods are an important field of engineering and crucial in terms of structural safety. However, in many practical cases, the process of monitoring and identifying damage is extremely difficult or even impractical due to the conditions of access and operation of a given component/structure. In this study, an inverse algorithm based on strain fields for damage identification in composite plate structures is presented. The inverse analyses combine experimental tests and digital image correlation (DIC) with numerical models based on finite element update method with great advantage of being a non-contact method. The proposed technique identifies the location and dimension of damages in a CFRP plate using static strains formulated as an objective function to be minimized. By model updating, the discrepancies between the experimental and the numerical results are minimized. For the success of the model updating, the efficiency of the optimization algorithm is essential. A powerful new metaheuristic sunflower optimization (SFO) is employed to update the unknown model parameters. Experimental results showed the excellent efficiency in the combined use of DIC, numerical modeling and SFO optimization to accurately identify the location of damage in numerical and experimental tests. The obtained results indicate that the proposed method can be used to determine efficiently the location and dimension of structural damages in mechanical structures.

A novel damage index for damage detection and localization of plate-type structures using twist derivatives of laser-measured mode shapes

It is important to detect and locate local damage in plate-type structures before such damage develops to a significant degree, jeopardizing the integrity and safety of structures. To address this problem, approaches using derivatives of laser-measured mode shapes have become the research focus during the recent decade. In a physical sense, damage can cause discontinuities in shear strains; conversely, such discontinuities can be reflected in twist derivatives of mode shapes, whereby the presence of the damage can be manifested. To address the susceptibility of twist derivatives to noise interference, this study formulates a new concept of multi-resolution twist derivative by integrating the multi-resolution analysis into twist derivatives, which features higher robustness against noise interference. A novel damage index using multi-resolution twist derivatives is established for detecting and locating damage in plate-type structures. In particular, a strategy of choosing mode shapes and a scheme of sparsely sampling them are proposed. By numerical simulation using the finite element method, the capability of the approach is numerically verified on a plate-type structure with a square notch. The applicability of the method is experimentally validated by detecting and locating a notch on an aluminum plate, whose mode shapes are acquired through non-contact measurement using a scanning laser vibrometer. The numerical and experimental results show that the approach can accurately characterize the presence, location, and size of the damage, and is robust against noise interference, suitable for detecting and locating damage of plate-type structures under noisy conditions.

Data-Driven Structural Health Monitoring Approach Using Guided Lamb Wave Responses

AbstractIn this paper, a data-driven structural health monitoring (SHM) approach is proposed to conduct in situ evaluation of the structural health state, i.e., damage location and extent, in which...

Sparse Bayesian learning approach for propagation distance recognition and damage localization in plate-like structures using guided waves

In ultrasonic guided wave–based damage detection, the propagation distance recognition of wave packets is an essential step. However, it is difficult to perform direct distance extraction from guided wave signals since the multimode, mode conversion, and dispersion effects typically lead to wave packet overlapping and distortion. In addition, the identified damage location may be incorrect due to inevitable uncertainties in the procedure of propagation distance recognition and damage localization. Motivated by these difficulties, a novel two-stage approach for propagation distance recognition and damage localization is proposed based on sparse Bayesian learning framework. In the first stage, prior knowledge of a small number of wave packets contained in a signal is exploited to sparsely represent the guided wave signal and then the corresponding propagation distance and amplitude information of each wave packet can be obtained. In the next stage, only a small number of damages occurring in a structure are exploited and a vector consisting of the propagation distances extracted from the previous stage is used to match the atoms in a pre-defined over-complete distance dictionary matrix, to achieve our goal of localizing structural damage. Both procedures of the two stages are realized by the sparse Bayesian learning algorithm, which obtains the most probable value and the corresponding uncertainty. A sampling strategy is presented to transfer the uncertainty of the propagation distance recognition to the subsequent damage localization. Finally, the effectiveness of the proposed method is validated using numerical simulation and experimental investigation on aluminum plates. The proposed method is only valid for single damage localization in the present form, but it has the potential to be extended for multiple damage localization.

Structural Damage Localization and Quantification Based on a CEEMDAN Hilbert Transform Neural Network Approach: A Model Steel Truss Bridge Case Study.

Vibrations of complex structures such as bridges mostly present nonlinear and non-stationary behaviors. Recently, one of the most common techniques to analyze the nonlinear and non-stationary structural response is Hilbert–Huang Transform (HHT). This paper aims to evaluate the performance of HHT based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) technique using an Artificial Neural Network (ANN) as a proposed damage detection methodology. The performance of the proposed method is investigated for damage detection of a scaled steel-truss bridge model which was experimentally established as the case study subjected to white noise excitations. To this end, four key features of the intrinsic mode function (IMF), including energy, instantaneous amplitude (IA), unwrapped phase, and instantaneous frequency (IF), are extracted to assess the presence, severity, and location of the damage. By analyzing the experimental results through different damage indices defined based on the extracted features, the capabilities of the CEEMDAN-HT-ANN model in detecting, addressing the location and classifying the severity of damage are efficiently concluded. In addition, the energy-based damage index demonstrates a more effective approach in detecting the damage compared to those based on IA and unwrapped phase parameters.

An effective deep feedforward neural networks (DFNN) method for damage identification of truss structures using noisy incomplete modal data

Structural damage assessment is a challenging problem of study due to lack of information in data measurement and the difficulty of extracting noisy features from the structural responses. Therefore, this paper proposes an effective deep feedforward neural networks (DFNN) method for damage identification of truss structures based on noisy incomplete modal data. In the proposed approach, incomplete datasets are randomly generated by a reducing finite element (FE) model. Based on the collected data, the DFNN model is constructed to predict damage position and severity of structures. To obtain a better performance of the network, the new ReLu activation function and Adadelta algorithm are employed in this work. In addition, the state-of-the-art mini-batch and dropout techniques are adopted to speed up the training process and avoid the over-fitting issue in training networks. Various hyperparameters such as number of hidden units, layers and epoches are surveyed to built a good training model. In order to demonstrate the efficiency and stability of the proposed method, a 31-bar planar truss structure and a 52-bar dome-like space truss structure are investigated with various damage scenarios. Moreover, the performance of the DFNN method is not only illustrated with the noise free input data but also with noisy input data. Different noise levels of the input data are taken into account in this study. To accurately predict the damage location and severity of the structures, 10000 and 20000 data samples corresponding to the 31-bar planar truss and the 52-bar dome-like space truss are randomly created in term of quantity of damage members, damage locations and damage severity of the structures for training the DFNN models. The results predicted by the DFNN using incomplete modal data are compared with those of the complete and actual models. The obtained results indicate that the DFNN is a promising method in damage localization and quantification of civil engineering structures.

An Effective Approach for Damage Identification in Beam-Like Structures Based on Modal Flexibility Curvature and Particle Swarm Optimization

In this paper, a computationally simple approach for damage localization and quantification in beam-like structures is proposed. This method is in consonance with applying modal flexibility curvature (MFC) and particle swarm optimization (PSO) algorithm. Analytical studies in the literature have revealed that changes in the modal flexibility curvature can be considered as a sensitive and suitable criterion for identifying damage in the beam-like structures. Modal flexibility curvature can be calculated utilizing central difference approximation, based on entries of the modal flexibility matrix. The PSO algorithm, as a powerful optimization tool, is employed in order to minimize the error function which is formulated as an error function between the measured modal flexibility curvatures of the damaged structure and those calculated from the analytical structure. To demonstrate the efficiency of the method, two beam-like structures under different damage scenarios are studied. In addition, the robustness of presented method is investigated only when the first several modal data are available. It is observed that the proposed approach is able to localize and quantify various damage cases only by a few lower vibrational modes and also, it is low-sensitive to measurement noise.