Vibration-based Damage Identification Methods Research Articles

A comparative analysis of damage identification methods based on multi-channel data fusion and convolutional neural network for frame structures

Vibration-based damage identification methods are crucial for structural health monitoring. In recent years, deep learning methods, such as convolutional neural networks (CNNs), have been widely used for damage identification due to their superior performance. To enhance the accuracy of damage identification for complex frame structures and effectively utilize data measured from multiple sensors, this study proposes and compares four different damage identification methods based on CNN and four different kinds of multi-channel data fusion approaches, i.e., data splicing, matrix reconstruction, data dimension elevation, and sub-model integration. Firstly, the original one-dimensional vibration signals of the frame structures are measured. Then, multi-channel data fusion is performed on the signals to construct datasets used as input to train the CNN model. Finally, the sensitive damage features are automatically extracted using the trained CNN models after parameter tuning, and the structural damage patterns are identified. The effectiveness of these damage identification methods is evaluated using the numerical model of IASC-ASCE Benchmark structure, the Qatar University Grandstand Simulator test, and a three-layer stainless steel frame structure test. The efficiency of damage identification methods using multi-channel data is compared to those using only single-channel data. Additionally, the impact of noise on damage identification accuracy is also analyzed. The results demonstrate that the proposed four different damage identification methods based on CNN and multi-channel data fusion, outperform the methods that solely rely on single-channel data. Furthermore, these methods significantly enhance the accuracy of damage identification.

Improving damage detection by combining multiple classifiers in different feature spaces

Vibration-based damage identification methods have been explored in several engineering fields. However, the effective damage-sensitivity of vibration-based features and the balance between false alarms and readiness of the anomaly detection are still open issues. Here, a cost-effective damage detection methodology based on a Deterministically Generated Negative Selection Algorithm is implemented and tested over a wide dataset generated through the numerical simulation of a three-storey concrete building in different damage scenarios. The approach allows training the classifiers only based on undamaged samples and ensures a prompt response to damage outbreak by fusing the classification conducted in parallel on different feature spaces, exploiting their distinct sensitivity to damage. A strategy to improve the reliability of the alarm is implemented by introducing counters and thresholds on consecutive classification outputs. A satisfactory trade-off between false and true alarms is achieved by comparing their probabilities, obtained through Markov chains, for different values of such thresholds.

Baseline-free structural damage detection using PCA- Hilbert transform with limited sensors

Conventional vibration-based damage identification methods need a large number of sensors on the bridge to identify the structural modal information and rely on the response data of bridges under a non-damage state for comparison, which restricts their implementations in the health monitoring of bridge structures in service. This paper proposes a baseline-free structural damage detection method based on limited sensor information. The proposed method uses the Hilbert transform combined with principal component analysis (PCA) to locate the damage. Firstly, PCA is performed on the displacement response data matrix of the bridge under a moving load to obtain the principal component matrix. According to the physical interpretation of PCA of the responses of a beam-like bridge under a moving load, each column of the principal component matrix, namely the principal component, composes of a mode shape and the dynamic information at this modal frequency of the bridge. The dynamic information of the modal frequency is indirectly obtained by filtering out the mode shape by low pass filter. Then the Hilbert transform is used to process the dynamic component information to receive the instantaneous frequency which is regarded as the damage index. When the structure is damaged, the first instantaneous frequency at the damaged location changes abruptly. To demonstrate the effectiveness and performance of the proposed method, numerical simulations and experimental validations on beam bridge models are conducted. The results show that the method is free from the requirement of the information of bridges under the undamaged state, and only limited sensors are needed to locate the damage.

Railway track support condition assessment — Methodology validation using numerical simulations

A novel methodology to evaluate railway track support conditions is currently being developed. This methodology is based on modal analysis of the multi-element system composed by the railway infrastructure and an instrumented railway vehicle. It belongs to the group of vibration-based structural damage identification methods and is focused on observing the characteristic frequencies of this multi-element system, which can be related with changes in the physical properties of the railway infrastructure. As opposed to other vehicle-based track monitoring methods, the proposed methodology should enable condition assessment of the railway infrastructure subgrade, an element that is often neglected during railway monitoring operations. Since this methodology is vehicle-based, it can be used to assess the entire extension of a railway infrastructure to gather data regarding the support conditions of the track. It can also assess the evolution of track support conditions over time by comparing different rides over the same railway stretch. An important aspect of this methodology that still lacks validation is the topic of characterizing track support conditions of a railway infrastructure through its natural frequencies. The suitability of the theoretical model behind this methodology is another aspect that needs to be assessed. Numerical simulations tests using the multibody simulation software Simpack® are currently being performed to address both topics. This paper describes some of the simulations performed in this context, including a description of the numerical models used. The obtained results support the selected theoretical model and the overall validity of the proposed methodology.

Damage identification in beams using burst video-records during free-vibration

In Structural Health Monitoring, computer vision has emerged as a promising approach for dynamic measurement; especially for its full-field characteristic. However, real-world implementations are still sparse due to several reasons. One of them is that vibration-based damage identification methods usually rely on Output-Only Modal-Analysis which, to be accurate, requires very long records with stochastic excitation or relatively long free-vibration records. The first option hinders video taking, storage and processing; specially with high-speed cameras. The second is statistically difficult under common operating conditions. To overcome such difficulties, this paper presents a methodology for damage identification in beams; which is vibration-based and Output-Only, but does not rely on Modal (or other Fourier-based) Analysis. Therefore, it can be fed with very short free-vibration displacement time histories; e.g., extracted from burst video records. The key idea is using the recently developed Synthesis of the Healthy-Structure Model Response, but interpreting the accelerations in free-vibration as an inertial pseudo-loading. The likely problem of high levels of noise is addressed by separating damage detection and localisation (using a statistical approach) from damage quantification (which is physics based). In an experimental example, the methodology showed outstanding detection and localisation performance with low quantification error.

The Current Development of Structural Health Monitoring for Bridges: A Review

The health monitoring system of a bridge is an important guarantee for the safe operation of the bridge and has always been a research hotspot in the field of civil engineering. This paper reviews the latest progressions in bridge health monitoring over the past five years. This paper is organized according to the various links of the bridge health monitoring system. Firstly, the literature on monitoring technology is divided into two categories, sensor technology and computer vision technology, for review. Secondly, based on the obtained monitoring data, the data processing methods including preprocessing, noise reduction, and reconstruction are summarized. Then, the technical literature on abnormal data early warning systems is summarized. The recent advances in vibration-based and non-destructive testing-based damage identification methods are reviewed in the next section. Finally, the advantages and disadvantages of the existing research and the future research directions are summarized. This review aims to provide a clear framework and some reliable methods for future research.

Damage identification in steel frames using dual-criteria vibration-based damage detection method and artificial neural network

Vibration-based damage identification methods have proven to be effective in identifying damage in various structures by employing modal parameters. This paper presents and develops a two-stage multi-criteria damage detection method based on modified damage indices (DIs), specifically modal flexibility and modal strain energy-based damage indices, and artificial neural network (ANN) to locate and quantify damage in steel frames. Only information from the first three bending modes was used to calculate the DIs. The location of damages is detected using DIs in the first stage, and two separate artificial neural networks are trained based on the DIs as input parameters in the second stage. The feasibility and performance of the proposed method were assessed by applying different single and multiple damage scenarios to a 3D finite element model of a simply supported steel beam and a steel frame of an industrial shed. The results demonstrate the efficiency and competency of the proposed non-destructive method for damage detection in such steel frames.

A review on recent developments in vibration-based damage identification methods for laminated composite structures: 2010–2022

Laminated composites received great attention in the structural applications due to innumerable advantages over conventional materials. These structures are subjected to environmental and mechanical loading conditions which leads to structural damage. The damage identification at early stage is crucial to overcome catastrophic failures, economic loss, human loss, and its disastrous consequences. Therefore, various structural damage identification techniques have been developed to identify the structural damages stages. These techniques provide the damage information about the systems, including structural health, serviceability, integrity, and safety. Over the past few decades, several review articles have been published on damage detection methods employing vibration characteristics for conventional and composite structures together. There has not been a study where complete concentration is focused on the laminated composite structures alone. This article concentrates on damage identification of laminated composite structures using vibration-based methods integrated with finite element methods, optimization techniques, signal processing methods, and machine learning algorithms over the past thirteen years. In addition, the limitations and applications of various methods have been discussed. This review article will provide the researchers and engineers an insight into the damage detection methods for laminated composite structures.

Railway track support condition assessment—Initial developments on a vehicle-based monitoring solution through modal analysis

Nowadays, there are multiple initiatives showing a renewed interest on railway transport of goods and passengers around the world. Thus, an efficient management of railway infrastructures, both at the operational level and in terms of economic profitability, is not only desirable but also corresponds to an area of ongoing research. In order to contribute to these efforts, an alternative and novel methodology to evaluate railway track support conditions is presented here, based on modal analysis of the characteristic frequencies of the multi-element system composed by a railway infrastructure and an instrumented vehicle moving over it. This methodology belongs to the group of vibration-based structural damage identification methods, and is focused on observing the characteristic frequencies of this multi-element system, which can be correlated with changes in the physical properties of the railway infrastructure under analysis. An important feature of the proposed methodology is that it should enable the collection of information regarding the conditions of the substructure of a railway infrastructure. By performing this assessment of a railway infrastructure over its length, and over time by comparing different rides over the same railway stretch, important information can be gathered regarding the support conditions of the track. This paper presents a complete description on the current stage of development of the proposed methodology, along with the theoretical model that serves as the basis to interpret the collected data. Preliminary verification of this methodology is performed through the analysis of two case studies regarding the passage of an instrumented vehicle over two underpasses. The results obtained so far show that the proposed methodology can provide relevant information regarding the support conditions of railway tracks.

Review on Structural Health Monitoring of Offshore Platform

The structural health monitoring (SHM) and damage detection of offshore platforms, one of the most common marine structures operating in a hostile environment, have gained global attention in recent years. This paper presents a review of vibration-based damage identification methods used for SHM of offshore platforms. The application and progress of these methods are discussed and some case studies are analyzed. The challenges and future work for vibration-based damage identification are summarized.

Structural Damage Identification Based on Integrated Utilization of Inverse Finite Element Method and Pseudo-Excitation Approach.

The attempt to integrate the applications of conventional structural deformation reconstruction strategies and vibration-based damage identification methods is made in this study, where, more specifically, the inverse finite element method (iFEM) and pseudo-excitation approach (PE) are combined for the first time, to give rise to a novel structural health monitoring (SHM) framework showing various advantages, particularly in aspects of enhanced adaptability and robustness. As the key component of the method, the inverse finite element method (iFEM) enables precise reconstruction of vibration displacements based on measured dynamic strains, which, as compared to displacement measurement, is much more adaptable to existing on-board SHM systems in engineering practice. The PE, on the other hand, is applied subsequently, relying on the reconstructed displacements for the identification of structural damage. Delamination zones in a carbon fibre reinforced plastic (CFRP) laminate are identified using the developed method. As demonstrated by the damage detection results, the iFEM-PE method possesses apparently improved accuracy and significantly enhanced noise immunity compared to the original PE approach depending on displacement measurement. Extensive parametric study is conducted to discuss the influence of a variety of factors on the effectiveness and accuracy of damage identification, including the influence of damage size and position, measurement density, sensor layout, vibration frequency and noise level. It is found that different factors are highly correlated and thus should be considered comprehensively to achieve optimal detection results. The application of the iFEM-PE method is extended to better adapt to the structural operational state, where multiple groups of vibration responses within a wide frequency band are used. Hybrid data fusion is applied to process the damage index (DI) constructed based on the multiple responses, leading to detection results capable of indicating delamination positions precisely.

Damage Sensitive PCA-FRF Feature in Unsupervised Machine Learning for Damage Detection of Plate-Like Structures

Damage detection is important in maintaining the integrity and safety of structures. The vibration-based Structural Health Monitoring (SHM) methods have been explored and applied extensively by researchers due to its non-destructive manner. The damage sensitivity of features used can significantly affect the accuracy of the vibration-based damage identification methods. The Frequency Response Function (FRF) was used as a damage sensitive feature in several works due to its rich yet compact representation of dynamic properties of a structure. However, utilizing the full size of FRFs in damage assessment requires high processing and computational time. A novel reduction technique using Principal Component Analysis (PCA) and peak detection on raw FRFs is proposed to extract the main damage sensitive feature while maintaining the dynamic characteristics. A rectangular Perspex plate with ground supports, simulating an automobile, was used for damage assessment. The damage sensitivity of the extracted feature, i.e. PCA-FRF is then evaluated using unsupervised [Formula: see text]-means clustering results. The proposed method is found to exaggerate the shift of damaged data from undamaged data and improve the repeatability of the PCA-FRF. The PCA-FRF feature is shown to have higher damage sensitivity compared to the raw FRFs, in which it yielded well-clustered results even for low damage conditions.

A Comparative Study on Vibration-Based Damage Identification and Localization Methods

This paper outlines various vibration-based methods for detecting and locating damage in mechanical and civil structural systems. The basic premises of these methods are that any changes in physical characteristics (mass and stiffness) of the structure will induce detectable changes in the modal properties of the structure (natural frequency, mode shape and damping). A comparative study of eight extensively used damage identification methods is done on a simulated simply supported beam to demonstrate the validity and effectiveness. The outcomes of various methods are compared in terms of damage detection and localization under single and multiple damage scenarios.

Damage identification in steel girder bridges using modal strain energy-based damage index method and artificial neural network

Health monitoring of infrastructures is of utmost significance, as they play a vital role in transportation. Identifying damages at the onset is essential to ensure the structure’s health state. The vibration-based damage identification methods use the vibration characteristics of structures to detect damages. Also, artificial neural networks (ANN) have been used to estimate damage magnitude with varied success. In this study, a two-stage damage identification technique was proposed to locate and estimate damages in steel girder bridges. The performance and feasibility of the proposed method were evaluated by applying several single and multiple damage scenarios to a validated finite element (FE) model of I-40 Bridge. Initially, the damage location was determined using the modal strain energy-based damage index method. For this purpose, the damage index was calculated separately for the first three bending modes of the bridge. The calculated damage vectors were combined along the length of girders. The peak value of the combined damage index showed the damage location along the length of steel plate girders. The results revealed that all the damage locations were detected with appropriate accuracy even if multiple damage scenarios existed in plate girders. In the second step, an ANN was used for estimation of damage severity. The damage index was used as input parameters to train the ANN. Then, the trained ANN could predict the severity of structural damages. The results indicated appropriate accuracy proposed by ANN for estimating the damage magnitude as well as proper performance of the proposed method.

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.

Sealant Delamination Detection of Structural Sealant Glazing Systems Based on Driving-Point Accelerance

The structural sealant glazing system is widely used in glass curtain wall worldwide. However, due to the aging of the sealants, cracking may form along the sides of the glass. If the panel of curtain wall arises, delamination may occur and induce failure or loss of the system. In order to detect the delamination of the sealant, in this paper, a new vibration-based damage identification method is proposed by using the difference between the predamage and postdamage driving-point accelerance (DA). This detection method would require only one acceleration sensor on a glass panel and a rubber hammer, which operates very conveniently. When the rubber hammer taps the glass panel near the acceleration sensor, the glass curtain wall panel would generate acceleration response which can be measured by the acceleration sensor. The measured acceleration responses will be used to calculate the DA and lead to a new delamination index, called relative accumulative difference of DA, which will give indications on the potential delamination of the sealant. In addition, the influence of the acceleration sensor installation position was analysed by the finite element method, and the optimal sensor location was determined to be at the intersection of the long-side quarter and the short-side quarter. Nine cases on various delamination severities were identified by the new method, as well as the natural frequency reduction and modal assurance criterion. The laboratory experiments showed that the relative accumulative difference of DA is extremely sensitive to sealant delamination. Even if delamination severity was only 6.39%, the relative accumulative difference of DA would be larger than 18%. As the delamination of the sealant progresses, the relative accumulative difference of DA increases, resulting in an effective detection method for sealant delamination of the structural sealant glazing system. The fundamental frequency reduction is suitable to identify relatively large delamination, and the modal assurance criterion of higher modes is also sensitive to sealant delamination.

A Data-Driven Damage Identification Framework Based on Transmissibility Function Datasets and One-Dimensional Convolutional Neural Networks: Verification on a Structural Health Monitoring Benchmark Structure.

Vibration-based data-driven structural damage identification methods have gained large popularity because of their independence of high-fidelity models of target systems. However, the effectiveness of existing methods is constrained by critical shortcomings. For example, the measured vibration responses may contain insufficient damage-sensitive features and suffer from high instability under the interference of random excitations. Moreover, the capability of conventional intelligent algorithms in damage feature extraction and noise influence suppression is limited. To address the above issues, a novel damage identification framework was established in this study by integrating massive datasets constructed by structural transmissibility functions (TFs) and a deep learning strategy based on one-dimensional convolutional neural networks (1D CNNs). The effectiveness and efficiency of the TF-1D CNN framework were verified using an American Society of Civil Engineers (ASCE) structural health monitoring benchmark structure, from which dynamic responses were captured, subject to white noise random excitations and a number of different damage scenarios. The damage identification accuracy of the framework was examined and compared with others by using different dataset types and intelligent algorithms. Specifically, compared with time series (TS) and fast Fourier transform (FFT)-based frequency-domain signals, the TF signals exhibited more significant damage-sensitive features and stronger stability under excitation interference. The utilization of 1D CNN, on the other hand, exhibited some unique advantages over other machine learning algorithms (e.g., traditional artificial neural networks (ANNs)), particularly in aspects of computation efficiency, generalization ability, and noise immunity when treating massive, high-dimensional datasets. The developed TF-1D CNN damage identification framework was demonstrated to have practical value in future applications.

A Sequence-Based Damage Identification Method for Composite Rotors by Applying the Kullback–Leibler Divergence, a Two-Sample Kolmogorov–Smirnov Test and a Statistical Hidden Markov Model

Composite structures undergo a gradual damage evolution from initial inter-fibre cracks to extended damage up to failure. However, most composites could remain in service despite the existence of damage. Prerequisite for a service extension is a reliable and component-specific damage identification. Therefore, a vibration-based damage identification method is presented that takes into consideration the gradual damage behaviour and the resulting changes of the structural dynamic behaviour of composite rotors. These changes are transformed into a sequence of distinct states and used as an input database for three diagnostic models, based on the Kullback–Leibler divergence, the two-sample Kolmogorov–Smirnov test and a statistical hidden Markov model. To identify the present damage state based on the damage-dependent modal properties, a sequence-based diagnostic system has been developed, which estimates the similarity between the present unclassified sequence and obtained sequences of damage-dependent vibration responses. The diagnostic performance evaluation delivers promising results for the further development of the proposed diagnostic method.

Probability distribution of decay rate: a statistical time-domain damping parameter for structural damage identification

This article proposes a novel vibration-based damage identification method, named the probability distribution of decay rate. By introducing a statistical framework, the probability distribution of decay rate method estimates the damage-induced changes in overall damping behaviour of a free-vibration dynamic system. Utilising free-vibration impulse response data, a one-dimensional data set of local maxima–minima points is constructed. A statistical analysis of this data set is then performed to derive damage-sensitive parameters. It is demonstrated that through the use of a statistical analysis framework, a number of enhancements are attained in terms of both robustness and leniency in estimating the significantly nonlinear property of overall damping. An impact hammer test is conducted in the laboratory to verify the efficacy of the proposed probability distribution of decay rate method. The test was performed on a scale-model steel Warren-truss bridge structure, subjected to bolt-connection failures. The comparison results between the probability distribution of decay rate method and the standard experimental modal analysis method confirm that the former is effective for damage identification of complex structures, particularly with real experimental data and steel-frame structure assemblies.

Structural Health Monitoring for Performance Assessment of Bridges under Flooding and Seismic Actions

Bridges can be subjected to damaging environmental actions due to flooding and seismic hazards. Flood actions that result in scour are a leading cause of bridge failure, while seismic actions that induce lateral forces may lead to high ductility demand that exceeds pier capacity. When combined, seismic actions and scour can lead to effects that depend on the governing scour condition affecting a bridge. Loss of stiffness under scour can reduce the ductility capacity of a bridge but can also lead to an increase in flexibility that may reduce seismic inertial forces. Conversely, increased flexibility can lead to deck collapse due to support loss, so there exists some uncertainty about the combined effect of both phenomena. A necessary step towards the performance assessment of bridges under flooding and seismic actions is to calibrate numerical models that can reproduce structural responses under different actions. A further step is verifying the achievement of performance goals defined by codes. Structural health monitoring (SHM) techniques allow the computation of performance parameters that are useful for calibrating numerical models and performing direct checks of performance goal compliance. In this paper, various strategies employed to monitor bridge health against scour and seismic actions are discussed, with a particular focus on vibration-based damage identification methods.