基于特征聚类的区域风电短期功率统计升尺度预测

发布时间：2018-01-27 14:15

  本文关键词： 区域风电功率预测 EOF分解 层次聚类法 升尺度预测　出处：《电网技术》2017年05期 　论文类型：期刊论文

【摘要】：区域风电功率预测对于保障风电消纳及电网安全经济运行具有重要意义。由于新建风电场在并网初期尚未建立预测系统及各风电场预测精度参差不齐,经典的单场功率累加法预测精度并不高。提出一种基于风电功率数据特征聚类的区域风电功率统计升尺度预测方法,首先使用经验正交函数(empirical orthogonal function,EOF)法解析区域内风电出力特征,然后采用层次聚类法划分子区域,并利用风电场的相关系数和预测精度选取代表风电场,最后根据代表风电场的预测功率及权重系数完成区域风电功率的升尺度预测。应用冀北电网2015年的实际数据进行统计升尺度建模和方法验证。结果表明,相比累加法,文中提出的统计升尺度方法可改进区域风电功率预测精度,同时减少区域预测模型对单风电场数据完备性和预测精度的依赖。
[Abstract]:Regional wind power prediction is of great significance for ensuring wind power consumption and the safe and economic operation of power grid. Since the new wind farm has not yet established a prediction system in the initial stage of grid connection and the prediction accuracy of each wind farm is not uniform. The classical single-field power accumulation method is not accurate. A regional wind power statistical scaling prediction method based on wind power data clustering is proposed. Firstly, the empirical orthogonal orthogonal function EOF method is used to analyze the characteristics of wind power generation in the region. Then the hierarchical clustering method is used to divide the sub-area, and the correlation coefficient and prediction precision of wind farm are selected to represent the wind farm. Finally, according to the predicted power and weight coefficient of wind farm, the upscaling prediction of regional wind power is completed. The statistical scaling modeling and method verification are carried out by using the actual data of 2015 in Hebei power grid. The results show that. Compared with the cumulative method, the statistical scaling method proposed in this paper can improve the prediction accuracy of regional wind power and reduce the dependence of the regional prediction model on the data completeness and prediction accuracy of single wind farm.
【作者单位】： 电力系统及发电设备控制和仿真国家重点实验室(清华大学);国网冀北电力有限公司电力科学研究院;
【基金】：国家重点研发计划支持项目(2016YFB0900101) 国家自然科学基金项目(51677099)~~
【分类号】：TM614
【正文快照】： 0引言风电功率预测是新能源调度的关键基础,根据风电出力预测曲线优化日前机组组合、动态滚动调整常规机组出力,降低备用容量,从而降低系统运行成本[1-2]。风电功率预测通常针对单个风电场开展[3],但随着风资源富集地区的电源集中开发,逐步形成了大规模风电基地,区域风电出力，

本文编号：1468660