An Interpretable Machine Learning Model for Daily Global Solar Radiation Prediction

El Mahjoub Benghoulam,Abdellah El Hmaidi,Mohamed Chaibi,Lhoussaine Tarik,Mohamed Berrada

doi:10.3390/en14217367

El Mahjoub Benghoulam, Abdellah El Hmaidi + Show 3 more

Open Access

https://doi.org/10.3390/en14217367

Copy DOI

Journal: Energies	Publication Date: Nov 5, 2021
Citations: 24	License type: CC BY 4.0

Affiliation: Université Moulay Ismail de Meknes

Abstract

Machine learning (ML) models are commonly used in solar modeling due to their high predictive accuracy. However, the predictions of these models are difficult to explain and trust. This paper aims to demonstrate the utility of two interpretation techniques to explain and improve the predictions of ML models. We compared first the predictive performance of Light Gradient Boosting (LightGBM) with three benchmark models, including multilayer perceptron (MLP), multiple linear regression (MLR), and support-vector regression (SVR), for estimating the global solar radiation (H) in the city of Fez, Morocco. Then, the predictions of the most accurate model were explained by two model-agnostic explanation techniques: permutation feature importance (PFI) and Shapley additive explanations (SHAP). The results indicated that LightGBM (R2 = 0.9377, RMSE = 0.4827 kWh/m2, MAE = 0.3614 kWh/m2) provides similar predictive accuracy as SVR, and outperformed MLP and MLR in the testing stage. Both PFI and SHAP methods showed that extraterrestrial solar radiation (H0) and sunshine duration fraction (SF) are the two most important parameters that affect H estimation. Moreover, the SHAP method established how each feature influences the LightGBM estimations. The predictive accuracy of the LightGBM model was further improved slightly after re-examination of features, where the model combining H0, SF, and RH was better than the model with all features.

Highlights

Introduction published maps and institutional affilRenewable energy transition will enormously benefit African countries by creating employment opportunities, protecting the environment, and promoting energy security [1].Morocco is regarded as one of the leading African countries in renewable energy, thanks to its policies encouraging investments in renewable energies
The performances are comparable or better than those obtained in [21,22,24,25,26]. They are worse than those reported in [23,27]. These findings proved the benefits of the interpretation techniques, the Shapley additive explanations (SHAP) method, for understanding the inner working of Machine learning (ML) models and boosting their predictive capability
The results revealed that the LightGBM had comparable performances with support-vector regression (SVR), and outperformed the multilayer perceptron (MLP) and multiple linear regression (MLR) models

Summary

Introduction

Renewable energy transition will enormously benefit African countries by creating employment opportunities, protecting the environment, and promoting energy security [1]. Morocco is regarded as one of the leading African countries in renewable energy, thanks to its policies encouraging investments in renewable energies. These sources of energy are expected to generate 52% of the country’s electricity by 2030 [2]. Solar energy is a sustainable energy source used widely for a variety of applications, including electricity generation, water pumping, air or water heating, and water desalination [4,5]. Global solar radiation information is critical for such applications.

Objectives

Methods

Discussion

Conclusion