Finite volume method network for acceleration of unsteady computational fluid dynamics: non-reacting and reacting flows

Abstract

Despite rapid improvements in the performance of central processing unit (CPU), the calculation cost of simulating chemically reacting flow using CFD remains infeasible in many cases. The application of the convolutional neural networks (CNNs) specialized in image processing in flow field prediction has been studied, but the need to develop a neural netweork design fitted for CFD is recently emerged. In this study, a neural network model introducing the finite volume method (FVM) with a unique network architecture and physics-informed loss function was developed to accelerate CFD simulations. The developed network model, considering the nature of the CFD flow field where the identical governing equations are applied to all grids, can predict the future fields with only two previous fields unlike the CNNs requiring many field images (>10,000). The performance of this baseline model was evaluated using CFD time series data from non-reacting flow and reacting flow simulation; counterflow and hydrogen flame with 20 detailed chemistries. Consequently, we demonstrated that (1) the FVM-based network architecture provided improved accuracy of multistep time series prediction compared to the previous MLP model (2) the physic-informed loss function prevented non-physical overfitting problem and ultimately reduced the error in time series prediction (3) observing the calculated residuals in an unsupervised manner could indirectly estimate the network accuracy. Additionally, under the reacting flow dataset, the computational speed of this network model was measured to be about 10 times faster than that of the CFD solver.

References in corpus (19)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Practical Bayesian Optimization of Machine Learning Algorithms
• The role of artificial intelligence in achieving the Sustainable Development Goals
• NSFnets (Navier-Stokes Flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations
• Surrogate Modeling for Fluid Flows Based on Physics-Constrained Deep Learning Without Simulation Data
• Machine learning accelerated computational fluid dynamics
• Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networks
• Physics-informed neural networks for solving Reynolds-averaged Navier$\unicode{x2013}$Stokes equations
• Deep Neural Networks for Data-Driven Turbulence Models
• Data-driven prediction of unsteady flow fields over a circular cylinder using deep learning
• Accelerating Eulerian Fluid Simulation With Convolutional Networks
• Synthetic turbulent inflow generator using machine learning
• CFDNet: a deep learning-based accelerator for fluid simulations
• Extreme learning machine for reduced order modeling of turbulent geophysical flows
• A Baseline Model for Software Effort Estimation
• On the Impact of the Activation Function on Deep Neural Networks Training
• FiniteNet: A Fully Convolutional LSTM Network Architecture for Time-Dependent Partial Differential Equations
• Finite Volume Neural Network: Modeling Subsurface Contaminant Transport
• Predicting the near-wall region of turbulence through convolutional neural networks

Cited by in corpus (2)

• Enhancing Computational Fluid Dynamics with Machine Learning
• Residual-based physics-informed transfer learning: A hybrid method for accelerating long-term CFD simulations via deep learning