Model-Agnostic Hybrid Numerical Weather Prediction and Machine Learning Paradigm for Solar Forecasting in the Tropics

TL;DR

This paper introduces a hybrid approach combining numerical weather prediction and machine learning to improve solar irradiance forecasting in the tropics, reducing errors significantly without extensive model tuning.

Contribution

It proposes a site-agnostic, model-agnostic hybrid framework that enhances regional and global NWP models with ML error correction for better solar forecasting.

Findings

Normalized RMSE reduced by up to 50% with ML correction.

Different radiation models showed comparable performance after correction.

Sensor data integration shows promising results.

Abstract

Numerical weather prediction (NWP) and machine learning (ML) methods are popular for solar forecasting. However, NWP models have multiple possible physical parameterizations, which requires site-specific NWP optimization. This is further complicated when regional NWP models are used with global climate models with different possible parameterizations. In this study, an alternative approach is proposed and evaluated for four radiation models. Weather Research and Forecasting (WRF) model is run in both global and regional mode to provide an estimate for solar irradiance. This estimate is then post-processed using ML to provide a final prediction. Normalized root-mean-square error from WRF is reduced by up to 40-50% with this ML error correction model. Results obtained using CAM, GFDL, New Goddard and RRTMG radiation models were comparable after this correction, negating the need for WRF parameterization tuning. Other models incorporating nearby locations and sensor data are also evaluated, with the latter being particularly promising.