Adaptive learning of effective dynamics: Adaptive real-time, online modeling for complex systems
Preprint
Abstract
Predictive simulations are essential for applications ranging from weather forecasting to material design. The veracity of these simulations hinges on their capacity to capture the effective system dynamics. Massively parallel simulations predict the systems dynamics by resolving all spatiotemporal scales, often at a cost that prevents experimentation. On the other hand, reduced order models are fast but often limited by the linearization of the system dynamics and the adopted heuristic closures. We propose a novel systematic framework that bridges large scale simulations and reduced order models to extract and forecast adaptively the effective dynamics (AdaLED) of multiscale systems. AdaLED employs an autoencoder to identify reduced-order representations of the system dynamics and an ensemble of probabilistic recurrent neural networks (RNNs) as the latent time-stepper. The framework alternates between the computational solver and the surrogate, accelerating learned dynamics while leaving yet-to-be-learned dynamics regimes to the original solver. AdaLED continuously adapts the surrogate to the new dynamics through online training. The transitions between the surrogate and the computational solver are determined by monitoring the prediction accuracy and uncertainty of the surrogate. The effectiveness of AdaLED is demonstrated on three different systems - a Van der Pol oscillator, a 2D reaction-diffusion equation, and a 2D Navier-Stokes flow past a cylinder for varying Reynolds numbers (400 up to 1200), showcasing its ability to learn effective dynamics online, detect unseen dynamics regimes, and provide net speed-ups. To the best of our knowledge, AdaLED is the first framework that couples a surrogate model with a computational solver to achieve online adaptive learning of effective dynamics. It constitutes a potent tool for applications requiring many expensive simulations.Keywords
All Related Versions
This publication has 51 references indexed in Scilit:
- Unsteady flow prediction from sparse measurements by compressed sensing reduced order modelingComputer Methods in Applied Mechanics and Engineering, 2022
- POD-DL-ROM: Enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decompositionComputer Methods in Applied Mechanics and Engineering, 2022
- Model Order Reduction with Neural Networks: Application to Laminar and Turbulent FlowsSN Computer Science, 2021
- Machine-learning-based reduced-order modeling for unsteady flows around bluff bodies of various shapesTheoretical and Computational Fluid Dynamics, 2020
- Data-driven reduced order model with temporal convolutional neural networkComputer Methods in Applied Mechanics and Engineering, 2020
- Data-driven operator inference for nonintrusive projection-based model reductionComputer Methods in Applied Mechanics and Engineering, 2016
- Reduced models for binocular rivalryJournal of Computational Neuroscience, 2010
- Non‐linear model reduction for uncertainty quantification in large‐scale inverse problemsInternational Journal for Numerical Methods in Engineering, 2009
- A penalization method to take into account obstacles in incompressible viscous flowsNumerische Mathematik, 1999
- Generalization of backpropagation with application to a recurrent gas market modelNeural Networks, 1988