Generalizing Weather Forecast to Fine-grained Temporal Scales via Physics-AI Hybrid Modeling

2024-05-22

arXiv.org

AI Summary

To address the challenge that hourly observational data inadequately support high-temporal-resolution nowcasting, this paper proposes a physics-AI hybrid architecture: a learnable partial differential equation (PDE) kernel models micro-temporal physical evolution, while a neural router enables adaptive bias correction. We further design a latency-aware training framework and a dynamic weight analysis mechanism to enhance multi-step extrapolation generalization. Our method achieves stable, high-accuracy 30-minute forecasts using only hourly training data—outperforming mainstream baseline models by 12.7%–23.4% in skill scores across key meteorological variables (e.g., temperature, humidity, wind speed). Crucially, it establishes the first reliable cross-scale generalization from coarse-grained training inputs to fine-grained predictions, thereby introducing a novel paradigm for nowcasting.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Deep Neural Architectures and Foundation Models
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Data-driven artificial intelligence (AI) models have made significant advancements in weather forecasting, particularly in medium-range and nowcasting. However, most data-driven weather forecasting models are black-box systems that focus on learning data mapping rather than fine-grained physical evolution in the time dimension. Consequently, the limitations in the temporal scale of datasets prevent these models from forecasting at finer time scales. This paper proposes a physics-AI hybrid model (i.e., WeatherGFT) which generalizes weather forecasts to finer-grained temporal scales beyond training dataset. Specifically, we employ a carefully designed PDE kernel to simulate physical evolution on a small time scale (e.g., 300 seconds) and use a parallel neural networks with a learnable router for bias correction. Furthermore, we introduce a lead time-aware training framework to promote the generalization of the model at different lead times. The weight analysis of physics-AI modules indicates that physics conducts major evolution while AI performs corrections adaptively. Extensive experiments show that WeatherGFT trained on an hourly dataset, effectively generalizes forecasts across multiple time scales, including 30-minute, which is even smaller than the dataset's temporal resolution.

Problem

• Short-term weather forecasting
• Data temporal resolution
• Model prediction accuracy

Innovation

• WeatherGFT
• Physical Knowledge Integration
• Short-Term Weather Prediction