An Interpretable Machine Learning Model for Daily Global Solar Radiation Prediction

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 M...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Mohamed Chaibi, EL Mahjoub Benghoulam, Lhoussaine Tarik, Mohamed Berrada, Abdellah El Hmaidi
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
T
Acceso en línea:https://doaj.org/article/b747539a9fcd4b4291320360fb351d14
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Sumario: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 (<i>H</i>) 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 (R<sup>2</sup> = 0.9377, RMSE = 0.4827 kWh/m<sup>2</sup>, MAE = 0.3614 kWh/m<sup>2</sup>) 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 (<i>H</i><sub>0</sub>) and sunshine duration fraction (<i>SF</i>) are the two most important parameters that affect <i>H</i> 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 <i>H</i><sub>0</sub>, <i>SF</i>, and <i>RH</i> was better than the model with all features.