Prediction of energy generation based on meteorological variables. Part I: Descriptive Analysis

Abstract

Photovoltaic solar energy generation depends directly on meteorological variables such as solar radiation, temperature, relative humidity, and wind speed, which exhibit high temporal variability. In this context, the present study, corresponding to Part I of the SOLAB project, aims to conduct a descriptive and exploratory analysis of the main climatic variables associated with solar energy generation in the municipality of La Paz, Cesar (Colombia). The purpose is to characterize their statistical behavior and internal relationships in order to establish a solid basis for the development of predictive models in later phases. To this end, descriptive statistics techniques, correlation analysis, and graphical visualizations with LOESS and polynomial fitting curves were applied, which made it possible to identify linear and nonlinear trends in the data. The results showed that solar radiation is the variable with the greatest direct influence on the energy generated, while temperature remains within a stable range between 24 °C and 36 °C. The Shapiro–Wilk test confirmed that the variables do not follow a normal distribution. Taken together, this analysis provides a comprehensive understanding of local meteorological behavior and constitutes an essential input for building more accurate predictive models in Part II of the SOLAB project.