A composable autoencoder-based iterative algorithm for accelerating numerical simulations
arXiv:2110.03780
Abstract
Numerical simulations for engineering applications solve partial differential equations (PDE) to model various physical processes. Traditional PDE solvers are very accurate but computationally costly. On the other hand, Machine Learning (ML) methods offer a significant computational speedup but face challenges with accuracy and generalization to different PDE conditions, such as geometry, boundary conditions, initial conditions and PDE source terms. In this work, we propose a novel ML-based approach, CoAE-MLSim (Composable AutoEncoder Machine Learning Simulation), which is an unsupervised, lower-dimensional, local method, that is motivated from key ideas used in commercial PDE solvers. This allows our approach to learn better with relatively fewer samples of PDE solutions. The proposed ML-approach is compared against commercial solvers for better benchmarks as well as latest ML-approaches for solving PDEs. It is tested for a variety of complex engineering cases to demonstrate its computational speed, accuracy, scalability, and generalization across different PDE conditions. The results show that our approach captures physics accurately across all metrics of comparison (including measures such as results on section cuts and lines).
References in corpus (24)
- Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
- NSFnets (Navier-Stokes Flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations
- Fourier Neural Operator for Parametric Partial Differential Equations
- hp-VPINNs: Variational Physics-Informed Neural Networks With Domain Decomposition
- PhyGeoNet: Physics-Informed Geometry-Adaptive Convolutional Neural Networks for Solving Parameterized Steady-State PDEs on Irregular Domain
- Learning data driven discretizations for partial differential equations
- SciANN: A Keras/Tensorflow wrapper for scientific computations and physics-informed deep learning using artificial neural networks
- Nonlinear mode decomposition with convolutional neural networks for fluid dynamics
- Flow over an espresso cup: Inferring 3D velocity and pressure fields from tomographic background oriented schlieren videos via physics-informed neural networks
- Efficient training of physics-informed neural networks via importance sampling
- Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders
- Deep Equilibrium Models
- Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data
- Kohn-Sham equations as regularizer: building prior knowledge into machine-learned physics
- Distributed physics informed neural network for data-efficient solution to partial differential equations
- A physics-informed operator regression framework for extracting data-driven continuum models
- Learning to Control PDEs with Differentiable Physics
- Differentiable Molecular Simulations for Control and Learning
- NVIDIA SimNet^{TM}: an AI-accelerated multi-physics simulation framework
- Turbulence forecasting via Neural ODE
- An unsupervised learning approach to solving heat equations on chip based on Auto Encoder and Image Gradient
- One-shot learning for solution operators of partial differential equations
- Geometry encoding for numerical simulations
- A Latent space solver for PDE generalization