A Study on Spatial Average Insolation Forecast in Electric Utility Service Area using Meso-Scale Model Grid Point Value

Abstract

A photovoltaic power generation system (PVS) is one of the promising measures to develop a low carbon society. Because of the uncertain power output characteristics, a robust power output forecast method must be employed for realizing the high penetration of PVS into an electric power system. Although there are several researches focusing on the forecast of single point insolation, the research focusing on the forecast of spatial average insolation in electric utility service area is not enough. In this study, we developed method to forecast spatial average insolation using a meso-scale model grid point value (MSM-GPV) data computed by the meteorological simulation. The main results are as follows. In spite of very simple method using a linear function of relative humidity, low/medium/high cloud cover, and extraterrestrial insolation, %MAE of the proposed method using MSM-GPV delivered at 18: 00 is 15.7%, which is smaller by about 10% than that of single point forecast due to the so-called smoothing effect. However, there are still a few days with larger forecast error than 0.3kWh/m². Such the large error is hardly reduced even with the later delivered MSM-GPV, because of the essential inaccuracy of MSM-GPV.