Power forecast of regional large-scale photovoltaics by SV method model with snow loss

Abstract

With a large installation of a photovoltaic (PV) power plant, forecasting PV generation is crucial for stable power grid operation. Forecasting power drop due to snow covering on PV array is one of an important issue particularly in cold area. However, it is difficult to always measure all snow conditions on the PV array. In this paper, the estimation of PV power generation in the Hokkaido region has been performed by using the model of Sophisticated Verification (SV) method with including the snow-covered loss, in which system configuration and meteorological data are used as input data. To describe the snow-covered loss, the average snow cover amount is defined on the PV array and its variation is represented as a function of snowfall and solar radiation. The proposed model expresses snow covering and melting naturally and contributes to the improvement of forecast accuracy.