Quantification of Solar Energy Grid Disturbances in the United States

Abstract

Solar energy penetration levels have been increasing steadily in recent years, becoming a significant factor in the energy system in the United States and the world. The increase in photovoltaic generation leads to a potential increase in grid disturbances. This study analyzes the relationship between solar energy generation and error (difference between the demand and the forecast energy demand) in seven subregions of the United States. Three types of errors were calculated mean error (ME), mean absolute error (MAE) and mean squared error (MSE). Correlation analysis was performed between solar energy generation and the three types of errors. Furthermore, graphical comparisons were made between the percentage of energy generated and the three types of errors for each of the seven subregions. As a result, negative correlations were found between the generation of solar energy and ME in five analyzed subregions. When analyzing the correlation between solar energy generation with MAE and MSE, a correlation was found in all the subregions. Besides, ME, MAE, and MSE values decrease in the subregions with the highest solar energy generation percentage. The results indicate that the generation of solar energy impacts the demand forecast. Also, Balancing Authorities (BAs) with a high percentage of solar energy consider the solar generation's effects on demand forecasting. Due to these results, even BAs with low solar generation should consider solar energy as a relevant factor to improve the demand forecast accuracy. Additionally, it is necessary to incorporate methodological standards across all BAs that consider solar generation effects in demand forecasting.