Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based Approaches
Micro-electro-mechanical-systems are complex structures, often involving nonlinearites of geometric and multiphysics nature, that are used as sensors and actuators in countless applications. Starting from full-order representations, we apply deep learning techniques to generate accurate, efficient,...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/6/3001 |
| _version_ | 1797609173693759488 |
|---|---|
| author | Giorgio Gobat Stefania Fresca Andrea Manzoni Attilio Frangi |
| author_facet | Giorgio Gobat Stefania Fresca Andrea Manzoni Attilio Frangi |
| author_sort | Giorgio Gobat |
| collection | DOAJ |
| description | Micro-electro-mechanical-systems are complex structures, often involving nonlinearites of geometric and multiphysics nature, that are used as sensors and actuators in countless applications. Starting from full-order representations, we apply deep learning techniques to generate accurate, efficient, and real-time reduced order models to be used for the simulation and optimization of higher-level complex systems. We extensively test the reliability of the proposed procedures on micromirrors, arches, and gyroscopes, as well as displaying intricate dynamical evolutions such as internal resonances. In particular, we discuss the accuracy of the deep learning technique and its ability to replicate and converge to the invariant manifolds predicted using the recently developed direct parametrization approach that allows the extraction of the nonlinear normal modes of large finite element models. Finally, by addressing an electromechanical gyroscope, we show that the non-intrusive deep learning approach generalizes easily to complex multiphysics problems. |
| first_indexed | 2024-03-11T05:56:51Z |
| format | Article |
| id | doaj.art-fa66bb7bd763492a9b582fd2946cdbe4 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T05:56:51Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-fa66bb7bd763492a9b582fd2946cdbe42023-11-17T13:44:35ZengMDPI AGSensors1424-82202023-03-01236300110.3390/s23063001Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based ApproachesGiorgio Gobat0Stefania Fresca1Andrea Manzoni2Attilio Frangi3Department of Civil and Environmental Engineering, Politecnico di Milano, Piazza Leonardo da Vinci, 32, 20133 Milano, ItalyMOX—Department of Mathematics, Politecnico di Milano, P.za Leonardo da Vinci 32, 20133 Milano, ItalyMOX—Department of Mathematics, Politecnico di Milano, P.za Leonardo da Vinci 32, 20133 Milano, ItalyDepartment of Civil and Environmental Engineering, Politecnico di Milano, Piazza Leonardo da Vinci, 32, 20133 Milano, ItalyMicro-electro-mechanical-systems are complex structures, often involving nonlinearites of geometric and multiphysics nature, that are used as sensors and actuators in countless applications. Starting from full-order representations, we apply deep learning techniques to generate accurate, efficient, and real-time reduced order models to be used for the simulation and optimization of higher-level complex systems. We extensively test the reliability of the proposed procedures on micromirrors, arches, and gyroscopes, as well as displaying intricate dynamical evolutions such as internal resonances. In particular, we discuss the accuracy of the deep learning technique and its ability to replicate and converge to the invariant manifolds predicted using the recently developed direct parametrization approach that allows the extraction of the nonlinear normal modes of large finite element models. Finally, by addressing an electromechanical gyroscope, we show that the non-intrusive deep learning approach generalizes easily to complex multiphysics problems.https://www.mdpi.com/1424-8220/23/6/3001deep learningreduced order modelingnonlinear dynamicsdata-driven modelinvariant manifolds |
| spellingShingle | Giorgio Gobat Stefania Fresca Andrea Manzoni Attilio Frangi Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based Approaches Sensors deep learning reduced order modeling nonlinear dynamics data-driven model invariant manifolds |
| title | Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based Approaches |
| title_full | Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based Approaches |
| title_fullStr | Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based Approaches |
| title_full_unstemmed | Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based Approaches |
| title_short | Reduced Order Modeling of Nonlinear Vibrating Multiphysics Microstructures with Deep Learning-Based Approaches |
| title_sort | reduced order modeling of nonlinear vibrating multiphysics microstructures with deep learning based approaches |
| topic | deep learning reduced order modeling nonlinear dynamics data-driven model invariant manifolds |
| url | https://www.mdpi.com/1424-8220/23/6/3001 |
| work_keys_str_mv | AT giorgiogobat reducedordermodelingofnonlinearvibratingmultiphysicsmicrostructureswithdeeplearningbasedapproaches AT stefaniafresca reducedordermodelingofnonlinearvibratingmultiphysicsmicrostructureswithdeeplearningbasedapproaches AT andreamanzoni reducedordermodelingofnonlinearvibratingmultiphysicsmicrostructureswithdeeplearningbasedapproaches AT attiliofrangi reducedordermodelingofnonlinearvibratingmultiphysicsmicrostructureswithdeeplearningbasedapproaches |