Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Predictive Monitoring of Large-Scale Engineering Assets Using Machine Learning Techniques and Reduced-Order Modeling

Bigoni, Caterina ; Guo, Mengwu LU and Hesthaven, Jan S. (2022) In Structural Integrity 21. p.185-205
Abstract

Structural health monitoring techniques aim at providing an automated solution to the threat of unsurveilled aging of structures that can have tremendous consequences in terms of fatalities, environmental pollution, and economic loss. To assess the state of damage of a complex structure, this paper proposes to fully characterize its behavior under multiple environmental and operational scenarios and compare new sensor measurements with the baseline behavior. However, the repeated simulations of a nonlinear, time-dependent structural model with high-dimensional input parameters represent a severe computational bottleneck for large-scale engineering assets. This chapter presents how to use efficient reduced-order modeling techniques to... (More)

Structural health monitoring techniques aim at providing an automated solution to the threat of unsurveilled aging of structures that can have tremendous consequences in terms of fatalities, environmental pollution, and economic loss. To assess the state of damage of a complex structure, this paper proposes to fully characterize its behavior under multiple environmental and operational scenarios and compare new sensor measurements with the baseline behavior. However, the repeated simulations of a nonlinear, time-dependent structural model with high-dimensional input parameters represent a severe computational bottleneck for large-scale engineering assets. This chapter presents how to use efficient reduced-order modeling techniques to mitigate the computational effort of many-query simulations without jeopardizing the accuracy. To compare new sensor measurements with the natural behavior of synthetic solutions, the proposed methodology uses hierarchical semi-supervised learning algorithms on a small amount of extracted damage-sensitive features, thus allowing one to assess the state of damage in real time. Using the inexpensive simulations, one can also optimally place sensors to maximize the observability of discriminant features. The all-round methodology is validated on a numerical example.

(Less)
Please use this url to cite or link to this publication:
author
; and
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Digital twins, One-class classification, Reduced order models, Sensor placement, Simulation-based anomaly detection, Structural health monitoring
host publication
Structural Health Monitoring Based on Data Science Techniques
series title
Structural Integrity
volume
21
pages
21 pages
publisher
Springer
external identifiers
  • scopus:85117955283
ISSN
2522-5618
2522-560X
ISBN
978-3-030-81716-9
978-3-030-81715-2
DOI
10.1007/978-3-030-81716-9_9
language
English
LU publication?
no
additional info
Publisher Copyright: © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
id
4b53736c-f5b3-4da4-800b-ad420d089ba9
date added to LUP
2024-03-19 12:16:42
date last changed
2024-06-25 18:05:39
@inbook{4b53736c-f5b3-4da4-800b-ad420d089ba9,
  abstract     = {{<p>Structural health monitoring techniques aim at providing an automated solution to the threat of unsurveilled aging of structures that can have tremendous consequences in terms of fatalities, environmental pollution, and economic loss. To assess the state of damage of a complex structure, this paper proposes to fully characterize its behavior under multiple environmental and operational scenarios and compare new sensor measurements with the baseline behavior. However, the repeated simulations of a nonlinear, time-dependent structural model with high-dimensional input parameters represent a severe computational bottleneck for large-scale engineering assets. This chapter presents how to use efficient reduced-order modeling techniques to mitigate the computational effort of many-query simulations without jeopardizing the accuracy. To compare new sensor measurements with the natural behavior of synthetic solutions, the proposed methodology uses hierarchical semi-supervised learning algorithms on a small amount of extracted damage-sensitive features, thus allowing one to assess the state of damage in real time. Using the inexpensive simulations, one can also optimally place sensors to maximize the observability of discriminant features. The all-round methodology is validated on a numerical example.</p>}},
  author       = {{Bigoni, Caterina and Guo, Mengwu and Hesthaven, Jan S.}},
  booktitle    = {{Structural Health Monitoring Based on Data Science Techniques}},
  isbn         = {{978-3-030-81716-9}},
  issn         = {{2522-5618}},
  keywords     = {{Digital twins; One-class classification; Reduced order models; Sensor placement; Simulation-based anomaly detection; Structural health monitoring}},
  language     = {{eng}},
  pages        = {{185--205}},
  publisher    = {{Springer}},
  series       = {{Structural Integrity}},
  title        = {{Predictive Monitoring of Large-Scale Engineering Assets Using Machine Learning Techniques and Reduced-Order Modeling}},
  url          = {{http://dx.doi.org/10.1007/978-3-030-81716-9_9}},
  doi          = {{10.1007/978-3-030-81716-9_9}},
  volume       = {{21}},
  year         = {{2022}},
}