Seismic Damage Identification Research Articles

Classification of Earthquake Damage of Buildings in Songyuan Area Based on Image Recognition

Since the emergence of human beings on the earth, various disasters have been accompanied. Among many natural disasters, the earthquake is undoubtedly one of the most threatening disasters. This project uses Res Net-50 model for deep learning and image recognition of building structural damage. Through the program to assess the local earthquake damage, given the feasible standards to facilitate a unified understanding of the earthquake situation, thereby improving the efficiency of disaster relief. Through experiments, the accuracy of the training set of the two classifications finally reached about 89.3 %, and the prediction accuracy of the test set finally reached about 71.4 %, Through the identification of post-earthquake building damage in Songyuan area, it can be learned that the accuracy of the software identification binary classification task is 73.21 %. Experiments show that taking photos can be used to predict the damage level of buildings in a certain area, and seismic damage identification can provide basis and support for post-disaster rescue and reconstruction and economic loss assessment.

Seismic damage identification of high arch dams based on an unsupervised deep learning approach

In actual engineering scenarios of arch dams, the incompleteness and nonstationarity of dynamic monitoring signals limit the accurate cognition of the health state. The effectiveness and robustness of damage characteristics in complex environments restrict the practical application of damage diagnosis theory. In this study, guided by the direct extraction of damage sensitivity features from the acceleration response signals of the arch dam, a seismic damage identification approach of high arch dams based on unsupervised learning is developed. Aiming at the problems of low measurement accuracy and poor identification robustness in the existing artificially designed damage-sensitive features, a denoising contractual sparse deep auto-encoder (DCS-DAE) model is proposed by exploring the mapping relationship between monitoring data and the structural state. This model integrates the advantages of denoising auto-encoder, compressive auto-encoder, and sparse auto-encoder. On this basis, based on the principle of reconstruction error and small probability, combined with box-plot and WKNN algorithm, a damage identification framework based on DCS-DAE is constructed. The effectiveness and noise resistance of the proposed method are verified by an extremely high arch dam. The results demonstrate that the damage identification framework based on multi-objective DCS-DAE constructed in this paper only requires the vibration information of the structure in the intact scenario, which provides a solution with higher stability and robustness for the seismic damage identification of high arch dams under strong noise pollution.

Seismic damage identification of moment frames based on random forest algorithm and enhanced gray wolf optimization

SummaryThe present study aims to identify damage in two‐dimensional (2‐D) moment frames using seismic responses by combining the random forest (RF) machine classifier and the enhanced gray wolf optimizer (EGWO) metaheuristic algorithm. First, a 2‐D moment frame for the dynamic analysis is simulated using the finite element method (FEM). Then, the placement of sensors is optimized using a proposed optimal sensor placement (POSP) method, which is a combination of the iterated improved reduced system (IIRS) and the binary differential evolution (BDE) optimization algorithm. The acceleration responses of the moment frame having damaged elements under 1995 Kobe earthquake are measured at the optimal sensor placement. Then, the natural frequencies and mode shapes of the structure are extracted using the auto‐regressive model with exogenous input method (ARX) as a system identification method. The natural frequencies are exploited to train an RF machine learning network that can find the damaged story of the moment frame. Then, EGWO is employed to accurately locate and quantify the damaged elements of the structure. The efficiency of the proposed method is assessed through considering a six‐story frame with 18 elements, a seven‐story frame with 49 elements, and the experimental data of an eight‐story frame for various conditions. The results show that the RF algorithm has an outstanding performance to correctly find a damaged story. Furthermore, the location and severity of damaged elements are precisely determined by EGWO algorithm. As a final outcome, it is demonstrated that the two‐step proposed method is very effective in seismically identifying damage to such structures.

Seismic Damage Identification of Composite Cable-Stayed Bridges Using Support Vector Machines and Wavelet Networks

A seismic damage identification method for composite cable-stayed bridges has been developed. The proposed method is based on a Support Vector Machine (SVM) and Wavelet Network (WN). A shaking table test of a composite cable-stayed bridge is employed to verify the identification accuracy of the WNSVM method; the test results show that the nonlinear Finite Element Model (FEM) can correctly simulate the single-tower cable-stayed bridge, and the learning samples of WNSVM can be produced based on the nonlinear FEM. The structural damage results identified by the WNSVM method are in good agreement with those obtained by the shaking table test, and the maximum error is less than 8%. Therefore, the WNSVM method can be used for the seismic damage identification of composite cable-stayed bridges.

Seismic Damage Identification Method for Curved Beam Bridges Based on Wavelet Packet Norm Entropy

Curved beam bridges, whose line type is flexible and beautiful, are an indispensable bridge type in modern traffic engineering. Nevertheless, compared with linear bridges, curved beam bridges have more complex internal forces and deformation due to the curvature; therefore, this type of bridge is more likely to suffer damage in strong earthquakes. The occurrence of damage reduces the safety of bridges, and can even cause casualties and property loss. For this reason, it is of great significance to study the identification of seismic damage in curved beam bridges. However, there is currently little research on curved beam bridges. For this reason, this paper proposes a damage identification method based on wavelet packet norm entropy (WPNE) under seismic excitation. In this method, wavelet packet transform is adopted to highlight the damage singularity information, the norm entropy of wavelet coefficient is taken as a damage characteristic factor, and then the occurrence of damage is characterized by changes in the damage index. To verify the feasibility and effectiveness of this method, a finite element model of Curved Continuous Rigid-Frame Bridges (CCRFB) is established for the purposes of numerical simulation. The results show that the damage index based on WPNE can accurately identify the damage location and characterize the severity of damage; moreover, WPNE is more capable of performing damage location and providing early warning than the method based on wavelet packet energy. In addition, noise resistance analysis shows that WPNE is immune to noise interference to a certain extent. As long as a series of frequency bands with larger correlation coefficients are selected for WPNE calculation, independent noise reduction can be achieved.

Life-cycle seismic damage identification and components damage sequences prediction for cable-stayed bridge based on fragility analyses

Life-cycle seismic damage identification and components damage sequences prediction for cable-stayed bridge based on fragility analyses

Seismic damage identification by fitting the nonlinear and hysteretic dynamic response of monitored buildings

The identification of hysteretic degrading systems exposed to nonstationary loading is a paramount research topic, especially in the case of structures subjected to ground motion excitations. In this paper, the data recorded by a masonry building are used to detect the presence of seismic damage. To this aim, a parametric nonlinear identification is performed by adopting a Bouc–Wen-type multiple oscillator model. Starting from the results of the identification process, a damage index based on the degrading stiffness matrix is defined. The damage index provides preliminary information to be used in post-event management. Finally, the identification procedure is applied to a monitored masonry structure, the Town Hall of Pizzoli (L’Aquila), which is part of the Italian Seismic Observatory of Structures (OSS). The records measured by accelerometers permanently installed on the building are used for the detection of damage occurred during a seismic event.

Seismic interferometry for earthquake-induced damage identification in historic masonry towers

The inherent vulnerability of masonry structures to seismic events makes Structural Health Monitoring of pivotal importance for the conservation of architectural heritage. In this regard, methods based on Operational Modal Analysis are becoming popular for damage identification. Nonetheless, these techniques may fail at detecting local damages with limited effects on the modal properties of the system. Recent studies report seismic interferometry to be a promising alternative for seismic damage identification of structures. This technique assesses the travelling times of propagating seismic waves between pairs of sensors, which are directly related to the local stiffness of the structure. Therefore, damage-induced degradation can be tracked through wave time delays. While some encouraging results have been reported on the application of acceleration-based seismic interferometry to reinforced-concrete structures, the number of works on masonry structures is far scarce. In this light, this paper is aimed at investigating the suitability of acceleration- and strain-based seismic interferometry for damage identification in historic masonry towers. To do so, an analytical layered Timoshenko beam model is devised for the wave propagation analysis of masonry towers under base motion. Parameter sensitivity analyses are first reported, with a special focus on the effects of dispersion upon system identification results. Secondly, a validation case study of a 41.6 m high masonry tower is presented. A realistic three-dimensional non-linear finite element model is built and subjected to seismic inputs causing increasing damage severities. The numerical results, used as pseudo-experimental data, demonstrate that it is possible to identify (detect, localize and quantify) earthquake-induced damages by wave propagation analysis of strain/acceleration records and inverse calibration of the proposed Timoshenko beam model. A particularly notable result is the possibility of detecting, localizing and, to some extent, quantifying earthquake-induced damage in a fully data-driven way by simply measuring wave travel times between pairs of sensors.

Automatic seismic damage identification of reinforced concrete columns from images by a region-based deep convolutional neural network

This paper proposed a modified faster region-based convolutional neural network (faster R-CNN) for the multitype seismic damage identification and localization (i.e., concrete cracking, concrete spalling, rebar exposure, and rebar buckling) of damaged reinforced concrete columns from images. Four hundred raw images containing different damages and complicated background information are taken by a consumer-grade camera in various locations and arbitrary perspectives to simulate the diverse situations where real-world postearthquake damaged structural images are taken by nonprofessionals. Rectangular bounding boxes are obtained to localize multitype structural damages along with the corresponding category labels and classification probabilities. Data augmentation is implemented by rotation at every 90°, vertical and horizontal flipping operations. An interactive labeling process for the ground-truth regions of the aforementioned damages is performed by a semiautomatic MATLAB program. A four-step alternating training procedure is adopted on the basis of the mini-batch stochastic gradient decent algorithm with momentum by backpropagation. Test results show that the trained faster R-CNN can automatically identify and localize the aforementioned multitype seismic damages and the overall average precision reaches 80%. The relative errors of coordinates of the left-top point obey minimum extreme value distributions, and those of width and height obey three-parameter lognormal distributions. The intersection ratio between the identification and ground truth has a mean value of 0.88, and the width–height ratio obeys a two-parameter lognormal distribution. Updated convolutional kernels in the first layer have shown trending, focusing, and line detectors for the feature extraction of multitype damages. Trending and focusing detectors contribute to the recognition of local damage regions, for example, concrete spalling and rebar exposure, whereas line detectors are more sensitive to the segmentation geometry, that is, concrete cracks.

Application of Bayesian statistical method in sensitivity-based seismic damage identification of structures: Numerical and experimental validation

Most of the developed sensitivity-based damage detection methods are based on the application of external excitations which could be prohibitive due to infeasible excitation of all structural degrees of freedom. In this regard, identification of damage properties using seismic structural response would be advantageous. In this research, sensitivity-based finite element model updating method is proposed to identify structural damage by earthquake response in the frequency domain and the transfer function of the structure due to ground excitation. The obtained sensitivity equation is solved by linear least square method through defining constraints on the design variables. Since the attainable measured data are restricted by limits on the instrumentations and preciseness of the measurements and due to the fact that only a few of the lower modes of a structure can generally be determined with confidence, a Bayesian statistical method is utilized to enhance the reliability of the predicted damage properties. The proposed technique is applied to a numerical frame model and an experimental six-story steel structure with various scenarios of story stiffness reduction. The results are indicative of the capability of the proposed method for identification of damage location and severity.

A neuro-wavelet technique for seismic damage identification of cantilever structures

A method based on artificial neural networks and wavelet transform is proposed for identifying seismic-induced damage of cantilever structures. In the proposed method, response accelerations are measured at strategically selected locations. To extract damage-induced sharp transitions from the measured signals, they are decomposed by continuous wavelet transform. The size of the decomposed signals is reduced by principal component analysis (PCA). Principal components obtained from PCA are fed to a set of neural networks to identify damage. The proposed algorithm is applied to a tall airport traffic control tower by means of numerical simulations. The obtained results show that the proposed method effectively identifies seismic-induced damage, and the noise intensity has a negligible effect on the predicted results. Moreover, the trained neural network system is able to predict the seismic-induced damage of unseen samples well.

Seismic damage identification for steel structures using distributed fiber optics

A distributed fiber optic monitoring methodology based on optic time domain reflectometry technology is developed for seismic damage identification of steel structures. Epoxy with a strength closely associated to a specified structure damage state is used for bonding zigzagged configured optic fibers on the surfaces of the structure. Sensing the local deformation of the structure, the epoxy modulates the signal change within the optic fiber in response to the damage state of the structure. A monotonic loading test is conducted on a steel specimen installed with the proposed sensing system using selected epoxy that will crack at the designated strain level, which indicates the damage of the steel structure. Then, using the selected epoxy, a varying degree of cyclic loading amplitudes, which is associated with different damage states, is applied on a second specimen. The test results show that the specimen's damage can be identified by the optic sensors, and its maximum local deformation can be recorded by the sensing system; moreover, the damage evolution can also be identified.

Seismic damage identification in buildings using neural networks and modal data

A seismic damage identification method intended for buildings with steel moment-frame structure is presented in this paper. The method has a statistical approach and is based on artificial neural networks and modal variables. It consists of two main stages. The initial one is devoted to the calibration of the undamaged structure and the final one to the identification of the damaged structure after an earthquake. The inputs of the nets are the first flexural modes (frequencies and mode shapes) at each principal direction of the structure and the outputs are the spatial variables (mass and stiffness). A damage index at each storey is determined by comparing the initial and final stiffness. A simplified finite element model was used to generate the data needed to train the nets. This model is consistent with available modal data and damage definition. The method was simulated on a 5-storey office building under conditions as close as possible to reality. The robustness of the method was verified with simulated data. Latter on, a sensitivity analysis of the mass variability was also carried out. Finally, the influence of modal error in the accuracy of damage predictions was statistically studied. Results are successful as concern as the robustness of the method. However, it is found that this approach is quite sensitive to modal errors.

Numerical simulation of a method for seismic damage identification in buildings

This article deals with a method for seismic damage identification in buildings with steel moment-frame structure. The damage identification is based on artificial neural networks and natural frequencies. A simplified finite element model is used to obtain the data needed for training the nets. The method is simulated on a four-storey building under conditions as close as possible to reality. The robustness of the method and its sensitivity to the variations of the mass in time and the influence of the data errors is addressed. The statistical analysis of the results is successful, but it reveals that the predictions are quite sensitive to the data errors.

Vibration-Based Seismic Damage Identification in Buildings

This article deals with a method for seismic damage identification in buildings with steel moment-frame structure. The damage identification is based on artificial neural networks and natural frequencies. A simplified finite element model is used to obtain the data needed for training the nets. The method is simulated on a four-storey building under conditions as close as possible to reality. The robustness of the method and its sensitivity to the variations of the mass with time and the influence of the data errors is addressed. The statistical analysis of the results is successful, but it reveals that the predictions are quite sensitive to the data errors.

Experimental investigation of seismic damage identification using PCA-compressed frequency response functions and neural networks

This paper presents an experimental investigation of seismic damage identification of a 38-storey tall building model using measured frequency response functions (FRFs) and neural networks (NNs). The 1:20 scale reinforced concrete structure is tested on a shaking table by exerting successively enhanced ground earthquake excitation to generate trifling, moderate, serious and complete (nearly collapsed) damage, respectively. After incurring the earthquake excitations at each level, a 20-min white-noise random excitation of low intensity is applied to the structure to produce ambient vibration response, from which FRFs are measured for post-earthquake damage detection by means of the NN technology. Principal component analysis (PCA) is pursued to the measured FRFs for dimensionality reduction and noise elimination, and then the PCA-compressed FRF data are used as input to NNs for damage identification. After a study on tolerance of PCA-reconstructed FRFs to measurement noise, different PCA configurations are designed for overall damage evaluation and damage location (distribution) identification, respectively. It is shown that the identification results by means of the FRF projections on a few principal components are much better than those directly using the measured FRF data, and agree fairly well with the visual inspection results of seismic damage during tests.