小时风速的向量自回归模型及应用.pdf

第 28 卷 第 14 期 中 国 电 机 工 程 学 报 Vol.28 No.14 May 15, 2008 112 2008 年 5 月 15 日 Proceedings of the CSEE ©2008 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 (2008) 14-0112-06 中图分类号：TM 61 文献标识码：A 学科分类号：470⋅40 小时风速的向量自回归模型及应用 孙春顺，王耀南，李欣然 (湖南大学电气与信息工程学院，湖南省 长沙市 410082) A Vector Autoregression Model of Hourly Wind Speed and Its Applications in Hourly Wind Speed Forecasting SUN Chun-shun, WANG Yao-nan, LI Xin-ran (College of Electrical and Information Engineering, Hunan University, Changsha 410082, Hunan Province, China) ABSTRACT: Short-term wind speed forecasting is important 的可持续性能源，开发和利用风能的主要形式是现 to the operation of grid-connected wind power generation 代大规模并网风力发电。虽然风能是“免费的”，但 systems. This paper outlined the value and methods of 开发和利用风能的费用却很高昂。由于风能无法直 short-term wind speed forecasting. The characteristic of diurnal 接储存，因而风电不可调度。风电波动性大，而且 hourly wind speed variation has been thoroughly analyzed. A 很难准确预测。因此，并网风力发电会给电力系统 method named as vectorization of univariate hourly wind speed 带来额外的运行费用。 time series has been presented for eliminating diurnal nonstationary, and vectorized hourly wind speed was expressed 建造合适的蓄能设施可以解决风能无法直接储 as a vector autoregression (VAR) model. Finally, the VAR 存、风电波动性大等问题，但这意味着将要付出另 model was applied to forecast hourly wind speed. The results 一笔费用。风电预测也有助于解决此问题，而且， showed that the presented VAR model can yield satisfactory 几乎不需要额外费用。因此，人们对风电预测进行 hourly wind speed forecast as long as 72 h ahead under normal 了大量的研究。 weather conditions. The proposed method is suitable for 风电预测是要预测未来一段时间内风轮机(或 eliminati