Virtual twins of nonlinear vibrating multiphysics microstructures: physics-based versus deep learning-based approaches

TL;DR

This paper develops deep learning-based reduced order models as virtual twins for complex nonlinear microstructures, demonstrating high accuracy and generalization in simulating their intricate dynamics and multiphysics interactions.

Contribution

It introduces a physics-based versus deep learning-based approach for creating real-time virtual twins of nonlinear microstructures, validated on various micromechanical systems.

Findings

Deep learning models accurately replicate nonlinear dynamics.

Models converge to invariant manifolds predicted by nonlinear normal modes.

Approach generalizes well to complex multiphysics systems like gyroscopes.

Abstract

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 as virtual twin for the simulation and optimization of higher-level complex systems. We extensively test the reliability of the proposed procedures on micromirrors, arches and gyroscopes, also displaying intricate dynamical evolutions like 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 extracting 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