PIVONet: A Physically-Informed Variational Neuro ODE Model for Efficient Advection-Diffusion Fluid Simulation
By: Hei Shing Cheung, Qicheng Long, Zhiyue Lin
Potential Business Impact:
Simulates messy fluid motion with less math.
We present PIVONet (Physically-Informed Variational ODE Neural Network), a unified framework that integrates Neural Ordinary Differential Equations (Neuro-ODEs) with Continuous Normalizing Flows (CNFs) for stochastic fluid simulation and visualization. First, we demonstrate that a physically informed model, parameterized by CNF parameters θ, can be trained offline to yield an efficient surrogate simulator for a specific fluid system, eliminating the need to simulate the full dynamics explicitly. Second, by introducing a variational model with parameters φ that captures latent stochasticity in observed fluid trajectories, we model the network output as a variational distribution and optimize a pathwise Evidence Lower Bound (ELBO), enabling stochastic ODE integration that captures turbulence and random fluctuations in fluid motion (advection-diffusion behaviors).
Similar Papers
Physics-Informed Latent Neural Operator for Real-time Predictions of time-dependent parametric PDEs
Machine Learning (CS)
Makes computer models of physics faster and smarter.
Physics-Informed Neural Networks for Control of Single-Phase Flow Systems Governed by Partial Differential Equations
Machine Learning (CS)
Teaches computers to control fluid flow using physics.
Physics-Informed Neural Networks and Neural Operators for Parametric PDEs: A Human-AI Collaborative Analysis
Machine Learning (Stat)
Computers solve science problems much faster.