초록 열기/닫기 버튼
Meteorological data measured at the location of the system is commonly used to improve the accuracy of predicting the power output of a solar power system. If there is no observation or data is missing, public data such as an automated synoptic observing system (ASOS) and automatic weather station (AWS) can be used. However, since the public observatories are far from the PV system, the uncertainty of the prediction due to the distance difference is expected to increase. To solve this problem, we propose a multiple regression analysis technique that predicts power output based on inverse distance weighted (IDW) interpolation. This spatial statistics technique can estimate the value of a location where data is not observed. The proposed method was demonstrated through a case study, and the prediction accuracy of solar power output was reviewed. The prediction accuracy varies depending on the case, but the accuracy of the case, which is within 15 km from the public observatory, was good because the mean absolute percentage error (MAPE) of the prediction of power output was less than 4.2%.
