Time Series Analysis and Forecasting of Solar Generation in Spain Using eXtreme Gradient Boosting: A Machine Learning Approach

Abstract

The rapid expansion of solar photovoltaic (PV) generation has established its pivotal role in the shift toward sustainable energy systems. This study conducts an in-depth analysis of solar generation data from 2015 to 2018 in Spain, with a specific emphasis on temporal patterns, excluding weather data. Employing the powerful eXtreme gradient boosting (XGBoost) algorithm for modeling and forecasting, our research underscores its exceptional efficacy in capturing solar generation trends, as evidenced by a remarkable root mean squared error (RMSE) of 11.042, a mean absolute error (MAE) of 5.621, an R-squared (R²) of 0.999, and a minimal mean absolute percentage error (MAPE) of 0.046. These insights hold substantial implications for grid management, energy planning, and policy development, reaffirming solar energy’s promise as a dependable and sustainable contributor to the electrical power system’s evolution. This research contributes to the growing body of knowledge aimed at optimizing renewable energy integration and enhancing energy sustainability for future generations.

Application of Methods Based on Artificial Intelligence and Optimisation in Power Engineering—Introduction to the Special Issue

“…The authors compared the results obtained using both models, and it resulted that the BP method obtained better prediction results than the GR method. In [49], Saigustia and Pijarski used the eXtreme gradient boosting over decision trees (XGBoost) algorithm to forecast photovoltaic generation trends in Spain. The model proposed by the authors makes it easier to optimise the operation of the power system; by adapting generation to the periods of peak customer demand, the flexibility and reliability of the network increases.…”

“…Table 2 presents a summary of the artificial intelligence techniques used in forecasting generation and load in the power system. Boosting [30,49,52] Expert system [58,59] Fuzzy logic [78,82]…”