MeshDQN: A Deep Reinforcement Learning Framework for Improving Meshes in Computational Fluid Dynamics
This addresses the problem of adaptive meshing for CFD practitioners by offering a more efficient and general-purpose method, though it appears incremental as it builds on existing machine learning techniques.
The paper tackles the bottleneck of user-intensive mesh generation in computational fluid dynamics by developing MeshDQN, a deep reinforcement learning framework that iteratively coarsens meshes while preserving target properties, requiring only a single simulation and successfully improving meshes for two 2D airfoils.
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.