Summary of A Tvd Neural Network Closure and Application to Turbulent Combustion, by Seung Won Suh et al.
A TVD neural network closure and application to turbulent combustion
by Seung Won Suh, Jonathan F MacArt, Luke N Olson, Jonathan B Freund
First submitted to arxiv on: 6 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computational Engineering, Finance, and Science (cs.CE); Fluid Dynamics (physics.flu-dyn)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A new approach is introduced to improve the accuracy of neural networks (NN) used to solve governing equations. The method, called a constrained NN closure model, prevents spurious oscillations that can occur when small errors build up and violate physical reality. This is achieved by embedding the NN in the discretized equations and strictly enforcing a constraint during training, inspired by total variation diminishing (TVD) methods for hyperbolic conservation laws. The resulting model demonstrates improved performance in recovering linear and nonlinear hyperbolic phenomena, anti-diffusion, and non-oscillatory properties. The constrained NN closure model is also applied to subgrid-scale modeling of a turbulent reacting flow, successfully suppressing spurious oscillations that violate solution boundedness. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine using computers to help solve complex problems in science and engineering. A new way to do this uses special kinds of computer models called neural networks (NN). These models can get better at solving problems if they are set up correctly, but sometimes they can produce weird results that don’t make sense. This paper introduces a new method to prevent these weird results from happening. It works by adding rules to the model during training to ensure it produces reasonable solutions. The new approach is shown to be very effective in solving certain types of problems and could have many practical applications. |
Keywords
» Artificial intelligence » Diffusion » Embedding
