A Bayesian Model Committee Approach to Forecasting Global Solar Radiation

TL;DR

This paper introduces a Bayesian model committee that combines ARMA and Neural Network models for improved hourly global solar radiation forecasting, demonstrating better performance than persistence benchmarks.

Contribution

It presents a novel Bayesian approach to combine different forecasting models, enhancing accuracy in solar radiation prediction.

Findings

Bayesian model committee improves forecast accuracy.

The approach outperforms the persistence model.

Models fitted to one year of data show promising results.

Abstract

This paper proposes to use a rather new modelling approach in the realm of solar radiation forecasting. In this work, two forecasting models: Autoregressive Moving Average (ARMA) and Neural Network (NN) models are combined to form a model committee. The Bayesian inference is used to affect a probability to each model in the committee. Hence, each model's predictions are weighted by their respective probability. The models are fitted to one year of hourly Global Horizontal Irradiance (GHI) measurements. Another year (the test set) is used for making genuine one hour ahead (h+1) out-of-sample forecast comparisons. The proposed approach is benchmarked against the persistence model. The very first results show an improvement brought by this approach.