JAX-SPH: A Differentiable Smoothed Particle Hydrodynamics Framework

• URL: http://arxiv.org/abs/2403.04750v2
• Date: Sun, 7 Jul 2024 17:53:28 GMT
• Title: JAX-SPH: A Differentiable Smoothed Particle Hydrodynamics Framework
• Authors: Artur P. Toshev, Harish Ramachandran, Jonas A. Erbesdobler, Gianluca Galletti, Johannes Brandstetter, Nikolaus A. Adams,
• Abstract summary: Particle-based fluid simulations have emerged as a powerful tool for solving the Navier-Stokes equations. Recent addition of machine learning methods to the toolbox for solving such problems is pushing the boundary of the quality vs. speed tradeoff. We lead the way to Lagrangian fluid simulators compatible with deep learning frameworks, and propose JAX-SPH.
• Score: 8.977530522693444
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Particle-based fluid simulations have emerged as a powerful tool for solving the Navier-Stokes equations, especially in cases that include intricate physics and free surfaces. The recent addition of machine learning methods to the toolbox for solving such problems is pushing the boundary of the quality vs. speed tradeoff of such numerical simulations. In this work, we lead the way to Lagrangian fluid simulators compatible with deep learning frameworks, and propose JAX-SPH - a Smoothed Particle Hydrodynamics (SPH) framework implemented in JAX. JAX-SPH builds on the code for dataset generation from the LagrangeBench project (Toshev et al., 2023) and extends this code in multiple ways: (a) integration of further key SPH algorithms, (b) restructuring the code toward a Python package, (c) verification of the gradients through the solver, and (d) demonstration of the utility of the gradients for solving inverse problems as well as a Solver-in-the-Loop application. Our code is available at https://github.com/tumaer/jax-sph.

Related papers

• LagrangeBench: A Lagrangian Fluid Mechanics Benchmarking Suite [0.0]
  We present LagrangeBench, the first benchmarking suite for Lagrangian particle problems. Our contribution is: (a) seven new fluid mechanics datasets (four in 2D and three in 3D) generated with the Smoothed Particle Hydrodynamics (SPH) method. We also go beyond established position errors and introduce physical metrics like kinetic energy MSE and Sinkhorn distance for the particle distribution.
  arXiv  Detail & Related papers  (2023-09-28T11:03:23Z)
• Differentiable Turbulence II [0.0]
  We develop a framework for integrating deep learning models into a generic finite element numerical scheme for solving the Navier-Stokes equations. We show that the learned closure can achieve accuracy comparable to traditional large eddy simulation on a finer grid that amounts to an equivalent speedup of 10x.
  arXiv  Detail & Related papers  (2023-07-25T14:27:49Z)
• QUBO.jl: A Julia Ecosystem for Quadratic Unconstrained Binary Optimization [0.0]
  QUBO.jl is an end-to-end Julia package for working with QUBO instances. QUBO.jl allows its users to interface with the aforementioned hardware, sending QUBO models in various file formats and retrieving results for subsequent analysis.
  arXiv  Detail & Related papers  (2023-07-05T18:20:31Z)
• Learning Generic Solutions for Multiphase Transport in Porous Media via the Flux Functions Operator [0.0]
  DeepDeepONet has emerged as a powerful tool for accelerating rendering fluxDEs. We use Physics-In DeepONets (PI-DeepONets) to achieve this mapping without any input paired-output observations.
  arXiv  Detail & Related papers  (2023-07-03T21:10:30Z)
• In Situ Framework for Coupling Simulation and Machine Learning with Application to CFD [51.04126395480625]
  Recent years have seen many successful applications of machine learning (ML) to facilitate fluid dynamic computations. As simulations grow, generating new training datasets for traditional offline learning creates I/O and storage bottlenecks. This work offers a solution by simplifying this coupling and enabling in situ training and inference on heterogeneous clusters.
  arXiv  Detail & Related papers  (2023-06-22T14:07:54Z)
• FluidLab: A Differentiable Environment for Benchmarking Complex Fluid Manipulation [80.63838153351804]
  We introduce FluidLab, a simulation environment with a diverse set of manipulation tasks involving complex fluid dynamics. At the heart of our platform is a fully differentiable physics simulator, providing GPU-accelerated simulations and gradient calculations. We propose several domain-specific optimization schemes coupled with differentiable physics.
  arXiv  Detail & Related papers  (2023-03-04T07:24:22Z)
• Deep Random Vortex Method for Simulation and Inference of Navier-Stokes Equations [69.5454078868963]
  Navier-Stokes equations are significant partial differential equations that describe the motion of fluids such as liquids and air. With the development of AI techniques, several approaches have been designed to integrate deep neural networks in simulating and inferring the fluid dynamics governed by incompressible Navier-Stokes equations. We propose the emphDeep Random Vortex Method (DRVM), which combines the neural network with a random vortex dynamics system equivalent to the Navier-Stokes equation.
  arXiv  Detail & Related papers  (2022-06-20T04:58:09Z)
• Lettuce: PyTorch-based Lattice Boltzmann Framework [0.0]
  The lattice Boltzmann method (LBM) is an efficient simulation technique for computational fluid mechanics and beyond. Here, we introduce Lettuce, a PyTorch-based LBM code with a threefold aim.
  arXiv  Detail & Related papers  (2021-06-24T11:44:21Z)
• PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics [89.81550748680245]
  We introduce a new differentiable physics benchmark called PasticineLab. In each task, the agent uses manipulators to deform the plasticine into the desired configuration. We evaluate several existing reinforcement learning (RL) methods and gradient-based methods on this benchmark.
  arXiv  Detail & Related papers  (2021-04-07T17:59:23Z)
• DiffPD: Differentiable Projective Dynamics with Contact [65.88720481593118]
  We present DiffPD, an efficient differentiable soft-body simulator with implicit time integration. We evaluate the performance of DiffPD and observe a speedup of 4-19 times compared to the standard Newton's method in various applications.
  arXiv  Detail & Related papers  (2021-01-15T00:13:33Z)
• A Near-Optimal Gradient Flow for Learning Neural Energy-Based Models [93.24030378630175]
  We propose a novel numerical scheme to optimize the gradient flows for learning energy-based models (EBMs) We derive a second-order Wasserstein gradient flow of the global relative entropy from Fokker-Planck equation. Compared with existing schemes, Wasserstein gradient flow is a smoother and near-optimal numerical scheme to approximate real data densities.
  arXiv  Detail & Related papers  (2019-10-31T02:26:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.