基于相似日和人工神经网络的风电功率短期预测.pdfVIP

第 34 卷 第 12 期 电 网 技 术 Vol. 34 No. 12 2010 年 12 月 Power System Technology Dec. 2010 文章编号：1000-3673 （2010 ）12-0163-05 中图分类号：TM 72 文献标志码：A 学科代码：470·40 基于相似日和人工神经网络的风电功率短期预测 1 1 2 1 1 1 孟洋洋 ，卢继平 ，孙华利 ，潘雪 ，高道春 ，廖勇 （1．输配电装备及系统安全与新技术国家重点实验室(重庆大学) ，重庆市 沙坪坝区 400044 ； 2．云南电力调度中心，云南省 昆明市 650000 ） Short-Term Wind Power Forecasting Based on Similar Days and Artificial Neural Network 1 1 2 1 1 1 MENG Yangyang , LU Jiping , SUN Huali , PAN Xue , GAO Daochun , LIAO Yong (1. State Key Laboratory of Power Transmission Equipment System Security and New Technology(Chongqing University), Shapingba District, Chongqing 400044, China; 2. Yunnan Electric Power Dispatch Center, Kunming 650000, Yunnan Province, China) ABSTRACT: It is significant to forecast short-term generated 划，则可有效地减轻风电场对接入电网的影响，减 power of wind farm for the security and stability of 少电力系统运行成本和旋转备用，提高风电穿透功 grid-connected wind farm. According to the factors impacting 率极限，同时为风电场参与发电竞价奠定基础。另 wind power output a method to select similar days of wind 外，准确的风力发电功率预测便于安排机组维护和 farm’s generated power is proposed and a short-term wind 检修，从而提高风电场容量系数。 power forecasting model based on similar days and Chebyshev neural network is designed. The designed model is verified by 目前，根据预测的物理量不同，风电功率预测 the data of a certain wind farm located in Yunnan province, and 方法分可以分为2 类[5] ：第 1类是先对风速进行预测， the forecasting error and its probability distribution are 然后根据风电机组或风电场的功率曲线得到风电场 analyzed. Analysis results show that the results forecasted by 功率输出；第 2 类为直接预测风电场的