Journal Article10.1088/1361-665x/ac50f4
High-dimensional data analytics in structural health monitoring and non-destructive evaluation: a review paper
31
TL;DR: In this paper , a review of high-dimensional data analytic (HDDA) methods for structural health monitoring (SHM) and non-destructive evaluation (NDE) applications is presented.
read more
Abstract: Abstract This paper aims to review high-dimensional data analytic (HDDA) methods for structural health monitoring (SHM) and non-destructive evaluation (NDE) applications. High-dimensional data is a type of data in which the number of features for each observation is much larger than the number of all observations. High-dimensional data may violate assumptions of the classic methods for statistical modeling and data analysis. Then, classic statistical modeling will no longer be applicable. HDDA methods were developed to overcome this challenge and analyze these types of data. In the field of SHM/NDE, there are several sources of high-dimensionality. Examples include a large number of data points in continuous waves/signals or high-resolution images/videos. HDDA methods are used as a dimension-reduction tool to preprocess data for further analysis, or they are directly implemented for damage detection and localization. This paper reviews six HDDA methods as well as existing and potential applications in SHM/NDE. Particularly, this paper discusses the vast range of implemented SHM/NDE applications from crack detection to missing data imputation. Furthermore, experimental and simulated datasets have been used to show the application of HDDA methods as hands-on examples. It is shown that the potential of HDDA for SHM/NDE studies is significantly more than the existing studies in the literature, and these methods can be used as a powerful tool that provides vast opportunities in SHM/NDE.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
The State of the Art of Artificial Intelligence Approaches and New Technologies in Structural Health Monitoring of Bridges
Raffaele Zinno,Sina Shaffiee Haghshenas,Giuseppe Guido,Kaveh Rashvand,Alessandro Vitale,Ali Sarhadi +5 more
TL;DR: In this paper , an outline of the role AI and other technologies will play in structural health monitoring (SHM) systems of bridges in the future was provided, including conceptual frameworks, benefits and problems, and existing methods.
Task-aware meta-learning paradigm for universal structural damage segmentation using limited images
TL;DR: Zhang et al. as discussed by the authors proposed a task-aware meta-learning paradigm using limited images for universal structural damage segmentation, where an interpretable task generation strategy instead of random sampling is designed based on feature density clustering, and a synthetical metric of Jaccard distance and Euclidean distance is established to measure the feature similarity and discover the class separability in the high-level feature space.
36
State-of-the-art AI-based computational analysis in civil engineering
TL;DR: In this article , a state-of-the-art review of the research on material and structural analyses using AI technology in civil engineering was performed to provide a general introduction to the current progress.
33
A locally unsupervised hybrid learning method for removing environmental effects under different measurement periods
TL;DR: In this paper , a locally unsupervised hybrid learning method based on an innovative discriminative reconstruction-based dictionary learning (DRDL) algorithm is proposed for structural health monitoring (SHM).
24
A multi-scale robotic approach for precise crack measurement in concrete structures
TL;DR: A multi-scale robotic approach measures concrete crack width, length, and shape using a CNN, high-resolution laser scanner, and LiDAR, outperforming vision-based methods and providing vital data for assessing civil infrastructure conditions through simulations and physical experiments.
19
References
Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography
TL;DR: New results are derived on the minimum number of landmarks needed to obtain a solution, and algorithms are presented for computing these minimum-landmark solutions in closed form that provide the basis for an automatic system that can solve the Location Determination Problem under difficult viewing.
Regularization and variable selection via the elastic net
Hui Zou,Trevor Hastie +1 more
TL;DR: It is shown that the elastic net often outperforms the lasso, while enjoying a similar sparsity of representation, and an algorithm called LARS‐EN is proposed for computing elastic net regularization paths efficiently, much like algorithm LARS does for the lamba.
Model selection and estimation in regression with grouped variables
Ming Yuan,Yi Lin +1 more
TL;DR: In this paper, instead of selecting factors by stepwise backward elimination, the authors focus on the accuracy of estimation and consider extensions of the lasso, the LARS algorithm and the non-negative garrotte for factor selection.
The adaptive lasso and its oracle properties
TL;DR: A new version of the lasso is proposed, called the adaptive lasso, where adaptive weights are used for penalizing different coefficients in the ℓ1 penalty, and the nonnegative garotte is shown to be consistent for variable selection.
Robust principal component analysis
TL;DR: In this paper, the authors prove that under some suitable assumptions, it is possible to recover both the low-rank and the sparse components exactly by solving a very convenient convex program called Principal Component Pursuit; among all feasible decompositions, simply minimize a weighted combination of the nuclear norm and of the e1 norm.