Fully Differentiable Lagrangian Convolutional Neural Network for Physics-Informed Precipitation Nowcasting

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pavlík, Peter, Výboh, Martin, Ezzeddine, Anna Bou, Rozinajová, Viera
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908429041270784
author Pavlík, Peter
Výboh, Martin
Ezzeddine, Anna Bou
Rozinajová, Viera
author_facet Pavlík, Peter
Výboh, Martin
Ezzeddine, Anna Bou
Rozinajová, Viera
contents This paper presents a convolutional neural network model for precipitation nowcasting that combines data-driven learning with physics-informed domain knowledge. We propose LUPIN, a Lagrangian Double U-Net for Physics-Informed Nowcasting, that draws from existing extrapolation-based nowcasting methods. It consists of a U-Net that dynamically produces mesoscale advection motion fields, a differentiable semi-Lagrangian extrapolation operator, and an advection-free U-Net capturing the growth and decay of precipitation over time. Using our approach, we successfully implement the Lagrangian convolutional neural network for precipitation nowcasting in a fully differentiable and GPU-accelerated manner. This allows for end-to-end training and inference, including the data-driven Lagrangian coordinate system transformation of the data at runtime. We evaluate the model and compare it with other related AI-based models both quantitatively and qualitatively in an extreme event case study. Based on our evaluation, LUPIN matches and even exceeds the performance of the chosen benchmarks, opening the door for other Lagrangian machine learning models.
format Preprint
id arxiv_https___arxiv_org_abs_2402_10747
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fully Differentiable Lagrangian Convolutional Neural Network for Physics-Informed Precipitation Nowcasting
Pavlík, Peter
Výboh, Martin
Ezzeddine, Anna Bou
Rozinajová, Viera
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
I.2.1; J.2
This paper presents a convolutional neural network model for precipitation nowcasting that combines data-driven learning with physics-informed domain knowledge. We propose LUPIN, a Lagrangian Double U-Net for Physics-Informed Nowcasting, that draws from existing extrapolation-based nowcasting methods. It consists of a U-Net that dynamically produces mesoscale advection motion fields, a differentiable semi-Lagrangian extrapolation operator, and an advection-free U-Net capturing the growth and decay of precipitation over time. Using our approach, we successfully implement the Lagrangian convolutional neural network for precipitation nowcasting in a fully differentiable and GPU-accelerated manner. This allows for end-to-end training and inference, including the data-driven Lagrangian coordinate system transformation of the data at runtime. We evaluate the model and compare it with other related AI-based models both quantitatively and qualitatively in an extreme event case study. Based on our evaluation, LUPIN matches and even exceeds the performance of the chosen benchmarks, opening the door for other Lagrangian machine learning models.
title Fully Differentiable Lagrangian Convolutional Neural Network for Physics-Informed Precipitation Nowcasting
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
I.2.1; J.2
url https://arxiv.org/abs/2402.10747