SolarSeer: Ultrafast and accurate 24-hour solar irradiance forecasts outperforming numerical weather prediction across the USA

2025-08-05

AI Summary

Traditional numerical weather prediction (NWP) models suffer from high computational cost and low efficiency in solar irradiance forecasting. To address this, we propose SolarSeer—a fully end-to-end AI model that directly learns the nonlinear mapping between historical satellite imagery and surface irradiance, bypassing data assimilation and partial differential equation (PDE) solving. SolarSeer generates hourly 24-hour forecasts of cloud cover and irradiance at 5 km resolution across the contiguous United States. Evaluated on reanalysis data, it achieves a 27.28% reduction in root-mean-square error (RMSE); across 1,800 ground-based measurement sites, average RMSE decreases by 15.35%, significantly outperforming the High-Resolution Rapid Refresh (HRRR) system and better capturing short-term irradiance variability. Moreover, inference speed is accelerated by over 1,500× compared to NWP baselines. SolarSeer thus delivers both high accuracy and real-time capability, enabling efficient photovoltaic dispatch and enhanced power grid stability.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Accurate 24-hour solar irradiance forecasting is essential for the safe and economic operation of solar photovoltaic systems. Traditional numerical weather prediction (NWP) models represent the state-of-the-art in forecasting performance but rely on computationally costly data assimilation and solving complicated partial differential equations (PDEs) that simulate atmospheric physics. Here, we introduce SolarSeer, an end-to-end large artificial intelligence (AI) model for solar irradiance forecasting across the Contiguous United States (CONUS). SolarSeer is designed to directly map the historical satellite observations to future forecasts, eliminating the computational overhead of data assimilation and PDEs solving. This efficiency allows SolarSeer to operate over 1,500 times faster than traditional NWP, generating 24-hour cloud cover and solar irradiance forecasts for the CONUS at 5-kilometer resolution in under 3 seconds. Compared with the state-of-the-art NWP in the CONUS, i.e., High-Resolution Rapid Refresh (HRRR), SolarSeer significantly reduces the root mean squared error of solar irradiance forecasting by 27.28% in reanalysis data and 15.35% across 1,800 stations. SolarSeer also effectively captures solar irradiance fluctuations and significantly enhances the first-order irradiance difference forecasting accuracy. SolarSeer's ultrafast, accurate 24-hour solar irradiance forecasts provide strong support for the transition to sustainable, net-zero energy systems.

Problem

• Ultrafast 24-hour solar irradiance forecasting for PV systems
• Replacing costly numerical weather prediction with AI model
• Improving accuracy and speed of solar energy forecasts

Innovation

• AI model for solar irradiance forecasting
• Eliminates data assimilation and PDEs solving
• Operates 1,500 times faster than NWP