Wind power day-ahead prediction with cluster analysis of NWP

详细信息    查看全文

• 作者：Lei Dong^a ; ^{leidong@bit.edu.cn" class="auth_mail" title="E-mail the corresponding author} ; Lijie Wang^b ; ^{wanglijie_0203@126.com" class="auth_mail" title="E-mail the corresponding author} ; Shahnawaz Farhan Khahro^a ; ^{shahnawazfarhan@gmail.com" class="auth_mail" title="E-mail the corresponding author} ; Shuang Gao^a ; ^{gao19820808@163.com" class="auth_mail" title="E-mail the corresponding author} ; Xiaozhong Liao^a ; ^{liaoxiaozhong@bit.edu.cn" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Wind power prediction ; Numerical weather prediction ; Cluster analysis ; Modeling ; Daily similarity
• 刊名：Renewable and Sustainable Energy Reviews
• 出版年：2016
• 出版时间：July 2016
• 年：2016
• 卷：60
• 期：Complete
• 页码：1206-1212
• 全文大小：1620 K

文摘

The selection of training data for establishing a model directly affects the prediction precision. Wind power has the characteristic of daily similarity. The corresponding meteorological data also has the characteristic of daily similarity. This paper proposes a new model with cluster analysis of the numerical weather prediction information. The similar day with the predicted day is searched as training sample to a generalized regression neural network model. The numerical weather prediction data and actual wind power data from a wind farm are used in this study to test the model. The prediction results show that correct cluster analysis method is a useful tool in day-ahead wind power prediction.