An Autoencoder-Based Deep Learning Approach for Load Identification in Structural Dynamics
In civil engineering, different machine learning algorithms have been adopted to process the huge amount of data continuously acquired through sensor networks and solve inverse problems. Challenging issues linked to structural health monitoring or load identification are currently related to big dat...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/12/4207 |
_version_ | 1797529546254188544 |
---|---|
author | Luca Rosafalco Andrea Manzoni Stefano Mariani Alberto Corigliano |
author_facet | Luca Rosafalco Andrea Manzoni Stefano Mariani Alberto Corigliano |
author_sort | Luca Rosafalco |
collection | DOAJ |
description | In civil engineering, different machine learning algorithms have been adopted to process the huge amount of data continuously acquired through sensor networks and solve inverse problems. Challenging issues linked to structural health monitoring or load identification are currently related to big data, consisting of structural vibration recordings shaped as a multivariate time series. Any algorithm should therefore allow an effective dimensionality reduction, retaining the informative content of data and inferring correlations within and across the time series. Within this framework, we propose a time series AutoEncoder (AE) employing inception modules and residual learning for the encoding and the decoding parts, and an extremely reduced latent representation specifically tailored to tackle load identification tasks. We discuss the choice of the dimensionality of this latent representation, considering the sources of variability in the recordings and the inverse-forward nature of the AE. To help setting the aforementioned dimensionality, the false nearest neighbor heuristics is also exploited. The reported numerical results, related to shear buildings excited by dynamic loadings, highlight the signal reconstruction capacity of the proposed AE, and the capability to accomplish the load identification task. |
first_indexed | 2024-03-10T10:15:23Z |
format | Article |
id | doaj.art-2f01705fa2ed408aa0017e419b813cf6 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T10:15:23Z |
publishDate | 2021-06-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-2f01705fa2ed408aa0017e419b813cf62023-11-22T00:49:04ZengMDPI AGSensors1424-82202021-06-012112420710.3390/s21124207An Autoencoder-Based Deep Learning Approach for Load Identification in Structural DynamicsLuca Rosafalco0Andrea Manzoni1Stefano Mariani2Alberto Corigliano3Dipartimento di Ingegneria Civile e Ambientale, Politecnico di Milano, Piazza L. da Vinci 32, 20133 Milano, ItalyMOX, Dipartimento di Matematica, Politecnico di Milano, Piazza L. da Vinci 32, 20133 Milano, ItalyDipartimento di Ingegneria Civile e Ambientale, Politecnico di Milano, Piazza L. da Vinci 32, 20133 Milano, ItalyDipartimento di Ingegneria Civile e Ambientale, Politecnico di Milano, Piazza L. da Vinci 32, 20133 Milano, ItalyIn civil engineering, different machine learning algorithms have been adopted to process the huge amount of data continuously acquired through sensor networks and solve inverse problems. Challenging issues linked to structural health monitoring or load identification are currently related to big data, consisting of structural vibration recordings shaped as a multivariate time series. Any algorithm should therefore allow an effective dimensionality reduction, retaining the informative content of data and inferring correlations within and across the time series. Within this framework, we propose a time series AutoEncoder (AE) employing inception modules and residual learning for the encoding and the decoding parts, and an extremely reduced latent representation specifically tailored to tackle load identification tasks. We discuss the choice of the dimensionality of this latent representation, considering the sources of variability in the recordings and the inverse-forward nature of the AE. To help setting the aforementioned dimensionality, the false nearest neighbor heuristics is also exploited. The reported numerical results, related to shear buildings excited by dynamic loadings, highlight the signal reconstruction capacity of the proposed AE, and the capability to accomplish the load identification task.https://www.mdpi.com/1424-8220/21/12/4207load/system identificationdeep learningstructural dynamicsautoencoderfalse nearest neighbor |
spellingShingle | Luca Rosafalco Andrea Manzoni Stefano Mariani Alberto Corigliano An Autoencoder-Based Deep Learning Approach for Load Identification in Structural Dynamics Sensors load/system identification deep learning structural dynamics autoencoder false nearest neighbor |
title | An Autoencoder-Based Deep Learning Approach for Load Identification in Structural Dynamics |
title_full | An Autoencoder-Based Deep Learning Approach for Load Identification in Structural Dynamics |
title_fullStr | An Autoencoder-Based Deep Learning Approach for Load Identification in Structural Dynamics |
title_full_unstemmed | An Autoencoder-Based Deep Learning Approach for Load Identification in Structural Dynamics |
title_short | An Autoencoder-Based Deep Learning Approach for Load Identification in Structural Dynamics |
title_sort | autoencoder based deep learning approach for load identification in structural dynamics |
topic | load/system identification deep learning structural dynamics autoencoder false nearest neighbor |
url | https://www.mdpi.com/1424-8220/21/12/4207 |
work_keys_str_mv | AT lucarosafalco anautoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics AT andreamanzoni anautoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics AT stefanomariani anautoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics AT albertocorigliano anautoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics AT lucarosafalco autoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics AT andreamanzoni autoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics AT stefanomariani autoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics AT albertocorigliano autoencoderbaseddeeplearningapproachforloadidentificationinstructuraldynamics |