Analysis of Meteorological Variables Used for Solar Power Generation Forecast

As photovoltaic solar energy depends on meteorological variables such as irradiance, air temperature and wind speed, they are used in artificial intelligence models to carry out generation forecasts. However, it is necessary to evaluate the quality of the data and its relationship with the target variable for prediction to find the best data source to be used in model training. This work aims to graphically and statistically evaluate four sources of meteorological data. Using Pearson and Spearman correlation and scatter plots to understand the behavior of irradiance, air temperature and wind speed with photovoltaic solar generation in relation to the active power at the inverter output. In the end, there are valuable insights into the variables and data sources analyzed, being a step prior to modeling the forecast of solar photovoltaic generation.