期刊名称：IOP Conference Series: Earth and Environmental Science
印刷版ISSN：1755-1307
电子版ISSN：1755-1315
出版年度：2018
卷号：170
期号：4
页码：042016
DOI：10.1088/1755-1315/170/4/042016
语种：English
出版社：IOP Publishing
摘要：The Photovoltaic (PV) system output power has intermittency and randomness; thus, it is necessary to find an accurate method for PV system power prediction. This paper describes a practical approach to predict the output power for PV system based on Back Propagation (BP) neutral network and similar days. The global solar radiation intensity and output power data are used to classify the types of weather and similar days by using the Self-Organizing Map (SOM) and the Probabilistic Neural Network (PNN), respectively. The prediction models for different weather type are established. The models have been tested and evaluated by the measured data of the PV systems which is located in Yunnan, RMSE and MAPE values indicate that the method provides reliable output power prediction for PV system under different weather types.
