基于信息粒化和支持向量机的风电场功率变化趋势和变化空间的预测算法.pdf

20 13 年 电 工 技 术 学 报 Vol.2 8 Sup.1 第28 卷增刊 1 TRANSACTIONS OF CHINA ELECTROTECHNICAL SOCIETY 2013 基于信息粒化和支持向量机的风电场功率 变化趋势和变化空间的预测算法 陈 伟 赵庆堂 赵锦苹 （兰州理工大学电气工程与信息工程学院 兰州 730050 ） 摘要 由于风力发电具有波动性、间歇性、随机性，准确的预测风电场发电功率的变化趋势 和变化空间对风电场与电力系统的协调运行具有重要的意义。首先通过对风电场发电功率的时间 序列进行模糊信息粒化，然后对粒化后的数据，利用支持向量机回归预测模型进行回归预测，最 后用甘肃酒泉地区某风电场的实测数据进行了仿真验证，仿真结果表明本模型可以有效地预测未 来风电功率的变化趋势和变化空间，验证了该算法的可行性和有效性。 关键词：风力发电 信息粒化 支持向量机 功率预测 中图分类号：TM614 Combined Model Based on the Information Granulation and SVM for Predicting Change Trend and Change Space of Wind Power Chen Wei Zhao Qingtang Zhao Jinping （Lanzhou University of Technology Lanzhou 730050 China ） Abstract It is significant to predict the change trend and change space of the wind power accurately for wind farm and the power system stable operation as wind power has the characteristics of volatility, intermittent and randomness. This paper conducted the time series of daily active power with fuzzy information granulation, and did the regression forecasting of the data after granulation with the support vector machine(SVM) prediction model, then simulated and tested this algorithm using the data from the wind farm located in the Gansu Jiu Quan wind farm of China. The results show that the new method has better prediction about the change trend and change space of the future wind power, and prove the feasibility and effectivene ss of the model. Keywords ：Wind power generation ，information granulation, support vector machine ，power prediction 1 引言 目前，国内外对于风电场功率预测的方法主要 [2] [3,4] [5-7]