A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep Learning
Recent advances in sensor technologies coupled with the development of machine/deep learning strategies are opening new frontiers in Structural Health Monitoring (SHM). Dealing with structural vibrations recorded with pervasive sensor networks, SHM aims at extracting meaningful damage-sensitive feat...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-04-01
|
| Series: | Proceedings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-3900/42/1/67 |
| _version_ | 1797570112672235520 |
|---|---|
| author | Luca Rosafalco Alberto Corigliano Andrea Manzoni Stefano Mariani |
| author_facet | Luca Rosafalco Alberto Corigliano Andrea Manzoni Stefano Mariani |
| author_sort | Luca Rosafalco |
| collection | DOAJ |
| description | Recent advances in sensor technologies coupled with the development of machine/deep learning strategies are opening new frontiers in Structural Health Monitoring (SHM). Dealing with structural vibrations recorded with pervasive sensor networks, SHM aims at extracting meaningful damage-sensitive features from the data, shaped as multivariate time series, and taking real-time decisions concerning the safety level. Within this context, we discuss an approach able to detect and localize a structural damage avoiding any pre-processing of the acquired data. The method takes advantage of the capability of Deep Learning of Fully Convolutional Networks, trained during an offline SHM phase. As a hybrid model- and data-based solution is looked for, Reduced Order Models are also built in the offline phase to reduce the computational burden of the whole monitoring approach. Through a numerical benchmark test, we show how the proposed method can recognize and localize different damage states. |
| first_indexed | 2024-03-10T20:20:23Z |
| format | Article |
| id | doaj.art-7a85ef62dd204a338fe985361c2edf61 |
| institution | Directory Open Access Journal |
| issn | 2504-3900 |
| language | English |
| last_indexed | 2024-03-10T20:20:23Z |
| publishDate | 2020-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Proceedings |
| spelling | doaj.art-7a85ef62dd204a338fe985361c2edf612023-11-19T22:14:15ZengMDPI AGProceedings2504-39002020-04-014216710.3390/ecsa-6-06585A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep LearningLuca Rosafalco0Alberto Corigliano1Andrea Manzoni2Stefano Mariani3Dipartimento 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, 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, ItalyRecent advances in sensor technologies coupled with the development of machine/deep learning strategies are opening new frontiers in Structural Health Monitoring (SHM). Dealing with structural vibrations recorded with pervasive sensor networks, SHM aims at extracting meaningful damage-sensitive features from the data, shaped as multivariate time series, and taking real-time decisions concerning the safety level. Within this context, we discuss an approach able to detect and localize a structural damage avoiding any pre-processing of the acquired data. The method takes advantage of the capability of Deep Learning of Fully Convolutional Networks, trained during an offline SHM phase. As a hybrid model- and data-based solution is looked for, Reduced Order Models are also built in the offline phase to reduce the computational burden of the whole monitoring approach. Through a numerical benchmark test, we show how the proposed method can recognize and localize different damage states.https://www.mdpi.com/2504-3900/42/1/67structural health monitoringfully convolutional networksdamage localizationtime series analysisdeep learning |
| spellingShingle | Luca Rosafalco Alberto Corigliano Andrea Manzoni Stefano Mariani A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep Learning Proceedings structural health monitoring fully convolutional networks damage localization time series analysis deep learning |
| title | A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep Learning |
| title_full | A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep Learning |
| title_fullStr | A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep Learning |
| title_full_unstemmed | A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep Learning |
| title_short | A Hybrid Structural Health Monitoring Approach Based on Reduced-Order Modelling and Deep Learning |
| title_sort | hybrid structural health monitoring approach based on reduced order modelling and deep learning |
| topic | structural health monitoring fully convolutional networks damage localization time series analysis deep learning |
| url | https://www.mdpi.com/2504-3900/42/1/67 |
| work_keys_str_mv | AT lucarosafalco ahybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning AT albertocorigliano ahybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning AT andreamanzoni ahybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning AT stefanomariani ahybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning AT lucarosafalco hybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning AT albertocorigliano hybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning AT andreamanzoni hybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning AT stefanomariani hybridstructuralhealthmonitoringapproachbasedonreducedordermodellinganddeeplearning |