期刊
RENEWABLE ENERGY
卷 127, 期 -, 页码 269-283出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.renene.2018.04.067
关键词
Improved support vector regression; Feature selection; Enhanced empirical mode decomposition; PV
资金
- Chongqing Big Data Engineering Laboratory for Children, Chongqing Electronics Engineering Technology Research Center for Interactive Learning
The critical role of photovoltaic (PV) energy as renewable sources in network can make some problems in power grids operation. Due to high volatility of PV signal, the prediction and its evaluation in planning and operation is very difficult. For this purpose, an accurate prediction approach is developed in this paper to tackle the mentioned problem. The proposed approach is based on enhanced empirical model decomposition (EEMD), a new feature selection method and hybrid forecast engine. The proposed feature selection is formulated by different criteria to select the best candidate inputs of forecast engine. And finally the hybrid forecast engine composed of improved support vector regression (ISVR) plus optimization algorithm to fine tune the related free parameters. Effectiveness of proposed method is applied over real-world engineering test cases through comparison with various prediction models. (C) 2018 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据