用物理驱动的神经微分方程提升天气预报精度，解决时间离散误差问题。
Mitigating Time Discretization Challenges with WeatherODE: A Sandwich Physics-Driven Neural ODE for Weather Forecasting
- 基于波动方程与时变源模型构建物理驱动的神经ODE，统一建模大气动态。
- 在全局和区域预报上分别降低40.0%和31.8%的均方根误差。
- 适合需要高精度短期天气预测的研究者与气象应用开发者。
在天气预报领域，传统模型常受时间离散化误差与时变源不一致问题制约，影响预测性能。本文提出WeatherODE，一种新型单阶段物理驱动的常微分方程（ODE）模型，旨在提升天气预报准确性。通过融合波动方程理论与时变源模型，WeatherODE有效缓解了时间离散化误差及动态大气过程建模难题。此外，设计了CNN-ViT-CNN夹心结构，以适应平流方程估计中不同优化偏置的复杂任务。大量实验表明，WeatherODE在全局与区域天气预报任务中均表现优异，相较于最新方法，均方根误差（RMSE）分别降低超过40.0%和31.8%。代码已开源：https://github.com/DAMO-DI-ML/WeatherODE。
原文摘要 · Abstract (English)
In the field of weather forecasting, traditional models often grapple with discretization errors and time-dependent source discrepancies, which limit their predictive performance. In this paper, we present WeatherODE, a novel one-stage, physics-driven ordinary differential equation (ODE) model designed to enhance weather forecasting accuracy. By leveraging wave equation theory and integrating a time-dependent source model, WeatherODE effectively addresses the challenges associated with time-discretization error and dynamic atmospheric processes. Moreover, we design a CNN-ViT-CNN sandwich structure, facilitating efficient learning dynamics tailored for distinct yet interrelated tasks with varying optimization biases in advection equation estimation. Through rigorous experiments, WeatherODE demonstrates superior performance in both global and regional weather forecasting tasks, outperforming recent state-of-the-art approaches by significant margins of over 40.0\% and 31.8\% in root mean square error (RMSE), respectively. The source code is available at \url{https://github.com/DAMO-DI-ML/WeatherODE}.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。