上海交通大学学报 ›› 2021, Vol. 55 ›› Issue (S2): 51-59.doi: 10.16183/j.cnki.jsjtu.2021.S2.008

• • 上一篇    下一篇

计及风速分布与机组惯量转化不确定性的风电场可用惯量估计

巩伟峥1, 许凌1, 姚寅2()   

  1. 1.国家电网有限公司华东分部,上海 200120
    2.上海电力大学 电气工程学院,上海 200090
  • 收稿日期:2021-08-31 出版日期:2021-12-28 发布日期:2022-01-24
  • 通讯作者: 姚寅 E-mail:yin.yao@shiep.edu.cn
  • 作者简介:巩伟峥(1986-),女,山东省济南市人,硕士,高级工程师,从事电力系统安全稳定方面的研究.

Estimation of Wind Farm Available Inertia Considering Uncertainty of Wind Speed Distribution and Unit Inertia Transformation

GONG Weizheng1, XU Ling1, YAO Yin2()   

  1. 1. East Branch of State Grid Corporation of China, Shanghai 200120, China
    2. School of Electric Power Engineering, Shanghai University of Electric Power, Shanghai 200090, China
  • Received:2021-08-31 Online:2021-12-28 Published:2022-01-24
  • Contact: YAO Yin E-mail:yin.yao@shiep.edu.cn

摘要:

计及风电场风速分布及根据风速分布存在的相关性,首先根据风速分布对场内机组进行聚类分区,提出适用于实际应用的风电场风速分布建模方案;其次,根据欧式距离选择各区域的中心机组;最后,计及风速与可用惯量间的不确定性,采用极限学习机估计各机群的可用惯量分布.根据国内某风场的实际数据设置仿真算例,仿真结果表明与基于平均风速的等值机法相比,所提方法在低风速区间内的预测准确度有明显提升,且基于聚类后中心机组的算法计算效率较高.

关键词: 风速预测, Copula函数, 虚拟惯量, 数据驱动

Abstract:

Taking into account the wind speed distribution and correlation, the units in the field are first clustered according to the wind speed distribution, and the wind speed distribution modeling scheme is proposed for practical applications. Then, each area is selected according to the European distance. Finally, considering the uncertainty between wind speed and available inertia, neural network is used to predict the available inertia distribution of each cluster. A wind farm in China is selected as the simulation example. The simulation results show that compared with the equivalent machine method based on average wind speed, the prediction accuracy of the proposed method in the low wind speed range is significantly improved, and the calculation efficiency of the proposed algorithm based on the clustered central unit is higher.

Key words: wind speed forecast, Copula function, virtual inertia, data driven

中图分类号: