Renewable Energy Prediction through Machine Learning Algorithms

This paper aims to implement an efficient renewable energy selection (either solar or wind) based on the chosen geographic location of Aguascalientes, Mexico through a Machine Learning (ML) method. Likewise, the information listed below will provide both a critical analysis and review of the state-of-the-art applications for ML Algorithms such as Support Vector Machines (SVM), Linear Regression (LR) and Neural Network Models (NNM). Rigorous data measurements taken over a period of six months, including those of solar irradiance, temperature, wind speed and wind direction to name a few, were the inputs used in different algorithms in order to find t he o ne t hat could most accurately predict future weather conditions. Based on the obtained results, the best ML Algorithm ended up being Random Forest; an approach that is capable of building an accurate prediction through the calculation of two crucial parameters; Mean Square Error (MSE) and Mean Absolute Error (MAE).