风电场短期风速预测方法研究  

Research of short-term wind speed prediction method

作　　者：王国权[1] 王森 刘华勇[1] 薛永端[2] 周平[1] 

机构地区：[1]国网重庆市电力公司经济技术研究院,重庆401121 [2]中国石油大学(华东)信息与控制工程学院,山东青岛266580

出　　处：《可再生能源》2014年第8期1134-1139,共6页Renewable Energy Resources

基　　金：国家"863"高技术基金项目(2012AA050213);国家自然科学基金项目(51077090);国网重庆市电力公司电力科学研究院规划评审中心KJ〔2013〕94号项目资助

摘　　要：随着风力发电的快速发展,并且风力发电系统的出力与风电场风速存在着的特殊关系,使得对风电场的风速实现较准确的预测已逐步成为研究的热点。该文先提出一种简单的的风速预测方法,即将指数平滑法应用到风速预测,并验证了指数平滑法预测风速的可行性。此外,为了提高预测精度,还提出了两种新的组合预测的方法,即基于指数平滑和灰色模型(GM)的组合预测方法、基于自回归滑动平均(ARMA)模型和灰色模型的组合预测方法。实例计算结果表明,组合预测方法比单独的用一种方法的预测效果要好,尤其是基于自回归滑动平均模型和灰色模型的组合预测方法更具有优势。With the rapid development of wind power,to achieve more accurate predictions of wind speed on wind farm has gradually become a hot research topic because of special relationship between output of wind pow prediction method has been presented in this paper,in which the exponential smoothing method is introduced into wind speed prediction, and then the feasibility of exponential smoothing method has been verified. In addition, in order to improve the prediction accuracy, two kinds of combination forecast methods are proposed. One is based on combination of exponential smoothing method and gray model, and the other combination forecasting method is based on auto-regressive moving average （ARMA） model and grey-model. Calculation examples show that the prediction effect of combined forecasting method is better than a single method, especially the combination of auto-regressive moving average model and grey-model.

关 键 词：短期风速 风力发电 指数平滑法 灰色模型 ARMA模型 组合预测 

分 类 号：TK81[动力工程及工程热物理—流体机械及工程]