The Forecast of the Electrical Energy Generated by the Photovoltaic System in Huilong using Neural Network Method

• Main Authors: Hsiao-tse Chang, 張孝澤
• Other Authors: Ting-chung Yu
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/99033210362720185208

碩士 === 龍華科技大學 === 電機工程研究所 === 99 === Currently, with the rapid industry development the energy demands are growing day by day. However, the storage of the traditional fossil energy is decreasing owing to the reason that the people did not treasure them. On the other hand, the environmental protection questions and the greenhouse effects caused by carbon dioxide emissions lead to the global warming phenomenon more and more serious, even threaten the survival of certain living creatures. These environmental protection phenomena and question make the people to value the development of renewable energy, especially solar energy. Since the applications of photovoltaic generation technologies have the advantages of energy saving and environmental protection, therefore, they have the priority to be developed by most of the countries. The purpose of this paper is to forecast the electrical energy generated by photovoltaic system in Huilong using neural network. A database is established in advance by the parameters which influence the generated electrical energy of the photovoltaic system and the previously measured electrical energy in order to use in electrical energy forecast. The Matlab software was used in this paper to establish the neural network model, and the method of back-propagation network was also used to forecast the generated electrical energy of the photovoltaic system. After observing the results of electrical energy forecast, it can be found that the proposed neural network model can accurately forecast the generated electrical energy and output current under different weather conditions, and the feasibility of the proposed model is also validated.