Abstract: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.

Structural Response Recovery Based on Improved Multi-Scale Principal Component Analysis Considering Sensor Performance Degradation

Probabilistic Principal Component Analysis‐based Anomaly Detection for Structures with Missing Data

Structural Damage Detection by Integrating Robust PCA and Classical PCA for Handling Environmental Variations and Imperfect Measurement Data

Modal parameter extraction from improved principal component analysis and structural state identification

Bridge Condition Monitoring Using Fixed Moving Principal Component Analysis

Using a single sensor for bridge condition monitoring via moving embedded principal component analysis

On the Effectiveness of Principal Component Analysis for Decoupling Structural Damage and Environmental Effects in Bridge Structures

Structural performance assessment considering both observed and latent environmental and operational conditions: A Gaussian process and probability principal component analysis method

Subdomain principal component analysis for damage detection of structures subjected to changing environments

Identification of Sudden Stiffness Change in the Acceleration Response of a Nonlinear Hysteretic Structure

Structural Response Reconstruction Based on Empirical Mode Decomposition in Time Domain

Sensor Fault Diagnosis for Structural Health Monitoring using Weighted Principal Component Analysis *

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

Structural Dynamic Response Reconstruction and Virtual Sensing Using a Sequence to Sequence Modeling with Attention Mechanism

Multi-sensor data fusion reconstruction method for vibration dynamic responses of aerospace structures

Performance monitoring and diagnosis of bridge structures based on time multiscale divide and conquer strategy

Framework for long-term structural health monitoring by computer vision and vibration-based model updating

Structural damage detection based on stochastic subspace identification and statistical pattern recognition: II. Experimental validation under varying temperature

Structural Damage Detection Using Wavelet Packet Transform Combining with Principal Component Analysis

Reconstruction of full-field dynamic responses for large-scale structures using optimal sensor placement

An Improved Frequency-Domain Regression Method for Structural Damage Detection in Wireless Sensor Network