Development of Power Predictions Models of the Photovoltaic Systems
이준우, 김경재, 윤상철, 장세동
대한설비공학회 2017년도 동계학술발표대회 논문집, 2017.11, 454-457 (4 pages)
This paper reports power generation prediction methods of photovoltaic systems for simulation based power system operations. To predict power generation of photovoltaic systems, 1) The photovoltaic system modeling methods and 2) the statistic data driven modeling methods such as Linear Regression(LR), Artificial Neural Network(ANN), and Gaussian Process Regression Model(GPRM) are analysed. Developed model is verified through annual measured data of existing photovoltaic power generation system. The error rate of GPRM is 5.5%, which has the highest accuracy compare to 6.3% of LR or 7.8% of ANN. This highly accurate prediction model contributes to improve the power quality and stability of the power management system.