Estimation of the Monthly Average Daily Solar Radiation using Geographic Information System and Advanced Case-Based Reasoning

详细信息    查看全文

• 作者：Choongwan Koo ; Taehoon Hong ; Minhyun Lee ; Hyo Seon Park
• 刊名：Environmental Science & Technology (ES&T)
• 出版年：2013
• 出版时间：May 7, 2013
• 年：2013
• 卷：47
• 期：9
• 页码：4829-4839
• 全文大小：438K
• 年卷期：v.47,no.9(May 7, 2013)
• ISSN：1520-5851

文摘

The photovoltaic (PV) system is considered an unlimited source of clean energy, whose amount of electricity generation changes according to the monthly average daily solar radiation (MADSR). It is revealed that the MADSR distribution in South Korea has very diverse patterns due to the country鈥檚 climatic and geographical characteristics. This study aimed to develop a MADSR estimation model for the location without the measured MADSR data, using an advanced case based reasoning (CBR) model, which is a hybrid methodology combining CBR with artificial neural network, multiregression analysis, and genetic algorithm. The average prediction accuracy of the advanced CBR model was very high at 95.69%, and the standard deviation of the prediction accuracy was 3.67%, showing a significant improvement in prediction accuracy and consistency. A case study was conducted to verify the proposed model. The proposed model could be useful for owner or construction manager in charge of determining whether or not to introduce the PV system and where to install it. Also, it would benefit contractors in a competitive bidding process to accurately estimate the electricity generation of the PV system in advance and to conduct an economic and environmental feasibility study from the life cycle perspective.