Estimating energy generation of a solar photovoltaic panel in Quezon City with a multi-layer perceptron model using meteorological parameters

Abstract

We train a multi-layer perceptron (MLP) model to estimate power generation of a Solar photovoltaic panel in Quezon City using meteorological parameters as input. The parameters used are temperature (maximum, mean, and minimum), relative humidity, rainfall, wind speed, wind direction, and cloud opacity from May 10, 2017, to May 9, 2021. The results of the study show that the model could explain 62.52% of the variance in the training set and 52.41% in the test set. Furthermore, we identify the parameters with the most influence on the solar panel's energy generation. maximum temperature, relative humidity, cloud opacity, and mean temperature emerge as the most influential parameters affecting solar energy generation.