PINP: Physics-Informed Neural Predictor with latent estimation of fluid flows
This work addresses the problem of accurate fluid flow prediction for physical sciences, offering a novel method that integrates physics into deep learning to improve forecasting.
The paper tackles the challenge of predicting fluid dynamics by proposing a physics-informed learning approach that incorporates coupled physical quantities and discretized physical equations into the model architecture and loss function, achieving state-of-the-art performance in spatiotemporal prediction on numerical simulations and real-world extreme-precipitation nowcasting benchmarks.
Accurately predicting fluid dynamics and evolution has been a long-standing challenge in physical sciences. Conventional deep learning methods often rely on the nonlinear modeling capabilities of neural networks to establish mappings between past and future states, overlooking the fluid dynamics, or only modeling the velocity field, neglecting the coupling of multiple physical quantities. In this paper, we propose a new physics-informed learning approach that incorporates coupled physical quantities into the prediction process to assist with forecasting. Central to our method lies in the discretization of physical equations, which are directly integrated into the model architecture and loss function. This integration enables the model to provide robust, long-term future predictions. By incorporating physical equations, our model demonstrates temporal extrapolation and spatial generalization capabilities. Experimental results show that our approach achieves the state-of-the-art performance in spatiotemporal prediction across both numerical simulations and real-world extreme-precipitation nowcasting benchmarks.