基于天气分型的风电功率预测方法.pdfVIP

38 卷 1 期 电 网 技 术 Vol. 38 No. 1 20 14 年1 月 Power System Technology Jan. 2014 文章编号：1000-3673 （2014 ）01-0093-06 中图分类号：TM 614 文献标志码：A 学科代码：470 4054 基于天气分型的风电功率预测方法 王勃，冯双磊，刘纯 （中国电力科学研究院，北京市海淀区 100192 ） Study on Weather Typing Based Wind Power Prediction WANG Bo, FENG Shuanglei, LIU Chun (China Electric Power Research Institute, Haidian District, Beijing 100192, China) ABSTRACT: It is an effective means to predict the output 有限；对总体样本而言，基于天气分型的预测方法较常规方 power of wind farms for ensuring secure and stable operation 法精度提高2% 以上。 of power grid concentratedly connected with large-scale wind 关键词：风电场；天气分型；功率；预测；数值天气预报 farms. A whether typing based wind power prediction DOI 10.13335/j.1000-3673.pst.2014.01.015 algorithm is proposed. Based on wind speed vector in 0 引言 numerical weather prediction (NWP) and daily atomospheric pressure and using principal component analysis, the proposed 风电属间歇式能源，其输出功率具有较大的不 algorithm performs dimensionality reduction for samples and 确定性，大规模风电集中接入将对电力系统的安全 classifies weather types by clustering analysis, and different 稳定运行带来挑战，风电功率预测是解决这一问题 prediction models are established respectively according to [1-2] 的有效手段之一 。通过功率预测可以提前掌握风 different whether types and compared with single prediction 电出力波动情况，有利于合理安排应对措施，提高 model. Research results show that using the method combining principal component analysis with clustering analysis the 电力系统的安全性和可靠性，根据预测结果制定科 weather phe