# MeshDQN: A Deep Reinforcement Learning Framework for Improving Meshes in Computational Fluid Dynamics

arXiv: https://arxiv.org/abs/2212.01428 · HTML: https://mlsanity.com/paper/2212.01428/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 33 references that are in the map.
- First posted on arXiv 2 Dec 2022.

## Builds on (most cited first, 15 of 15)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [A Comprehensive Survey on Graph Neural Networks](https://mlsanity.com/paper/1901.00596/) (2019)
- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Deep Reinforcement Learning with Double Q-learning](https://mlsanity.com/paper/1509.06461/) (2015)
- [Transformer for Partial Differential Equations' Operator Learning](https://mlsanity.com/paper/2205.13671/) (2022)
- [Towards Sparse Hierarchical Graph Classifiers](https://mlsanity.com/paper/1811.01287/) (2018)
- [A review on Deep Reinforcement Learning for Fluid Mechanics](https://mlsanity.com/paper/1908.04127/) (2019)
- [Deep Learning for Reduced Order Modelling and Efficient Temporal Evolution of Fluid Simulations](https://mlsanity.com/paper/2107.04556/) (2021)
- [Graph Convolutional Neural Networks for Body Force Prediction](https://mlsanity.com/paper/2012.02232/) (2020)
- [Machine Learning-Based Optimal Mesh Generation in Computational Fluid Dynamics](https://mlsanity.com/paper/2102.12923/) (2021)
- [Data-driven Identification of 2D Partial Differential Equations using extracted physical features](https://mlsanity.com/paper/2010.10626/) (2020)
- [Deep Reinforcement Learning Optimizes Graphene Nanopores for Efficient Desalination](https://mlsanity.com/paper/2101.07399/) (2021)
- [Deep Learning Convective Flow Using Conditional Generative Adversarial Networks](https://mlsanity.com/paper/2005.06422/) (2020)

## Built on by (most cited first, 1 of 1)

- [Loop2Net: Data-Driven Generation and Optimization of Airfoil CFD Meshes from Sparse Boundary Coordinates](https://mlsanity.com/paper/2507.01057/) (2025)

## Abstract (arXiv)

Meshing is a critical, but user-intensive process necessary for stable and accurate simulations in computational fluid dynamics (CFD). Mesh generation is often a bottleneck in CFD pipelines. Adaptive meshing techniques allow the mesh to be updated automatically to produce an accurate solution for the problem at hand. Existing classical techniques for adaptive meshing require either additional functionality out of solvers, many training simulations, or both. Current machine learning techniques often require substantial computational cost for training data generation, and are restricted in scope to the training data flow regime. MeshDQN is developed as a general purpose deep reinforcement learning framework to iteratively coarsen meshes while preserving target property calculation. A graph neural network based deep Q network is used to select mesh vertices for removal and solution interpolation is used to bypass expensive simulations at each step in the improvement process. MeshDQN requires a single simulation prior to mesh coarsening, while making no assumptions about flow regime, mesh type, or solver, only requiring the ability to modify meshes directly in a CFD pipeline. MeshDQN successfully improves meshes for two 2D airfoils.