Estimation of Invisible Distributed PV Power Generation from Bus Load

Abstract

With the growing influx of solar power on the demand side, the estimation of photovoltaic (PV) power generation plays an increasingly significant role in bus load forecasting. This paper proposes an algorithm for identifying distributed PV power generation from bus load. Firstly, the net load of bus line is decomposed by Prophet algorithm, and the correlation between each decomposed item and the solar radiation intensity is calculated by Kendall and Spearman correlation coefficients respectively. The irrelevant items of the net load are filtered out, while the remaining part of the load is taken as the filtered load. Then, the natural days are divided into different day sets according to the solar radiation intensity, and the average centralization PV generation curves as well as the average net load curves of each set are calculated separately. Finally, the equivalent installation scale of distributed PV is calculated based on the ratio of the area between the PV generation curves and the area between the filtered load curves. The case studies with data from both China and the United States show that the proposed method can effectively identify the distributed PV generation from bus load.