Bridge Damage Detection and Quantification Under Environmental Effects by Principal Component Analysis

Over the decades, visual inspection has been adopted as a means to monitor infrastructure health. While visual inspection provides insights on a bridge’s condition, it has been generally agreed that it is insufficient and inefficient. This has called for the creation of autonomous, robust, continuous, and quantitative structural health monitoring (SHM) systems to detect potential deficiencies in an early stage, and monitor future condition. Various methods have been explored that associate changes in condition with changes in the structure’s vibration characteristics. These methods have been mostly tested on laboratory specimens experiencing simulated damage. There is need for extending validation of these SHM methods on in-situ structures experiencing real damage under operational and environmental conditions. This paper summarizes a full-scale experiment exploring bridge damage detection effectiveness under variable traffic loads. Three different types of damage were introduced into a full-scale, bridge deck mock-up. These included crash-induced bridge barrier damage, controlled barrier damage, and damage to the deck slab. At the end of each introduced damage case, the bridge’s response to the multiple passages was recorded using specific vehicles specifications. Data was extracted and analyzed to identify damage using principal component analysis (PCA) and independent component analysis (ICA) as damage-sensitive features. The extracted damage features were thereafter used as input for unsupervised learning (novelty detection). One interesting observation was how PCA revealed possibly significant damage after a crash, which under visual inspection appeared to be minor. Novelty detection using PCA as its damage feature was shown to provide robust damage detection irrespective of load, speed variation, and signal noise levels.

Download Full-text

Deep principal component analysis: An enhanced approach for structural damage identification

Structural Health Monitoring ◽

10.1177/1475921718799070 ◽

2018 ◽

Vol 18 (5-6) ◽

pp. 1444-1463 ◽

Cited By ~ 6

Author(s):

Moisés Silva ◽

Adam Santos ◽

Reginaldo Santos ◽

Eloi Figueiredo ◽

Claudomiro Sales ◽

...

Keyword(s):

Principal Component Analysis ◽

Structural Health Monitoring ◽

Environmental Factors ◽

Damage Detection ◽

Health Monitoring ◽

Principal Component ◽

Component Analysis ◽

Data Normalization ◽

Structural Health ◽

Detection And Quantification

The structural health monitoring relies on the continuous observation of a dynamic system over time to identify its actual condition, detect abnormal behaviors, and predict future states. The regular changes in environmental factors have been reported as one of the main challenges for the application of structural health monitoring systems. These influences in the structural responses are in general nonlinear, affecting the damage-sensitive features in the most varied forms. The usual process to remove these normal changes is referred to as data normalization. In that regard, principal component analysis is probably the most studied algorithm in structural health monitoring, having numerous versions to learn strong nonlinear normal changes. However, in most cases, not all variability is properly accounted for via the existing nonlinear principal component analysis approaches, resulting in poor damage detection and quantification performances. In this article, a new paradigm based on deep principal component analysis, rooted in the deep learning field, is presented to overcome these limitations. This approach extracts the most salient underlying feature distributions by stacking multiple feedforward neural networks trained to learn an identity mapping of the input variables, where the network inputs are reproduced into the outputs. Similar to the traditional nonlinear principal component analysis–based approach, our approach identifies a nonlinear output-only model of an undamaged structure by comprising modal features into an internal bottleneck layer, which implicitly represents the independent environmental factors. The proposed technique is validated through the application on a progressively damaged prestressed concrete bridge and a three-span suspension bridge. The experimental results demonstrate that capturing the most slight nonlinear variations in the data can lead to improved data normalization and, consequently, better damage detection and quantification performances.

Download Full-text

Research for Structural Damage Identification Method Based on Stable Time Series and Principal Component Analysis

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.578-579.1020 ◽

2014 ◽

Vol 578-579 ◽

pp. 1020-1023

Author(s):

Jing Zhou Lu ◽

Jia Chen Wang ◽

Xu Zhu

Keyword(s):

Principal Component Analysis ◽

Time Series ◽

Damage Detection ◽

Structural Damage ◽

Computing Time ◽

Principal Component ◽

Component Analysis ◽

Identification Accuracy ◽

Frame Structure ◽

Ar Model

In this paper, we introduce a set of techniques for time series analysis based on principal component analysis (PCA). Firstly, the autoregressive (AR) model is established using acceleration response data, and the root mean squared error (RMSE) of AR model is calculated based on PCA. Then a new damage sensitive feature (DSF) based on the AR coefficients is presented. To test the efficacy of the damage detection and localization methodologies, the algorithm has been tested on the analytical and experimental results of a three-story frame structure model of the Los Alamos National Laboratory. The result of the damage detection indicates that the algorithm is able to identify and localize minor to severe damage as defined for the structure. It shows that the suggested method can lead to less amount of computing time, high suitability and identification accuracy.

Download Full-text

Damage detection in nonlinear civil structures using kernel principal component analysis

Advances in Structural Engineering ◽

10.1177/1369433220913207 ◽

2020 ◽

Vol 23 (11) ◽

pp. 2414-2430

Author(s):

Khaoula Ghoulem ◽

Tarek Kormi ◽

Nizar Bel Hadj Ali

Keyword(s):

Principal Component Analysis ◽

Damage Detection ◽

Structural Model ◽

Principal Component ◽

Original Data ◽

Component Analysis ◽

Correlated Data ◽

Data Driven ◽

Kernel Principal Component Analysis ◽

Structural Systems

In the general framework of data-driven structural health monitoring, principal component analysis has been applied successfully in continuous monitoring of complex civil infrastructures. In the case of linear or polynomial relationship between monitored variables, principal component analysis allows generation of structured residuals from measurement outputs without a priori structural model. The principal component analysis has been widely used for system monitoring based on its ability to handle high-dimensional, noisy, and highly correlated data by projecting the data onto a lower dimensional subspace that contains most of the variance of the original data. However, for nonlinear systems, it could be easily demonstrated that linear principal component analysis is unable to disclose nonlinear relationships between variables. This has naturally motivated various developments of nonlinear principal component analysis to tackle damage diagnosis of complex structural systems, especially those characterized by a nonlinear behavior. In this article, a data-driven technique for damage detection in nonlinear structural systems is presented. The proposed method is based on kernel principal component analysis. Two case studies involving nonlinear cable structures are presented to show the effectiveness of the proposed methodology. The validity of the kernel principal component analysis–based monitoring technique is shown in terms of the ability to damage detection. Robustness to environmental effects and disturbances are also studied.

Download Full-text