Probabilistic Solar

Probabilistic solar forecasting aims to predict solar energy production, not just as a single value, but as a range of possibilities reflecting inherent uncertainties. Current research focuses on improving forecast accuracy and reliability through advanced techniques like post-processing of numerical weather prediction models, employing machine learning algorithms (including neural networks and ensemble methods), and leveraging generative AI to simulate diverse future sky conditions. These advancements are crucial for enhancing the integration of solar power into electricity grids, improving the efficiency of space-based operations affected by solar activity, and providing more reliable energy resource planning.

Papers

June 6, 2024

Improving Model Chain Approaches for Probabilistic Solar Energy Forecasting through Post-processing and Machine Learning
Nina Horat, Sina Klerings, Sebastian Lerch
Machine Learning Post Processing Solar Power Model Driven Ensemble Weather Solar Generation Probabilistic Solar

June 20, 2023

SkyGPT: Probabilistic Short-term Solar Forecasting Using Synthetic Sky Videos from Physics-constrained VideoGPT
Yuhao Nie, Eric Zelikman, Andea Scott, Quentin Paletta, Adam Brandt
Synthetic Video Solar Forecasting Probabilistic Solar

June 3, 2023

Probabilistic Solar Proxy Forecasting with Neural Network Ensembles
Joshua D. Daniell, Piyush M. Mehta
Ensemble Method Space Weather Thermospheric Density Probabilistic Solar

Probabilistic Solar

Papers

Improving Model Chain Approaches for Probabilistic Solar Energy Forecasting through Post-processing and Machine Learning

SkyGPT: Probabilistic Short-term Solar Forecasting Using Synthetic Sky Videos from Physics-constrained VideoGPT

Probabilistic Solar Proxy Forecasting with Neural Network Ensembles