摘要
为解决现有大规模风电场集电线路保护存在误动风险的问题,分析了具有低电压穿越能力双馈风机在故障条件下输出电流与机端电压跌落深度的内在关联,根据风机故障电流的成分特性,提出一种基于聚类分析的风电场集电线路保护策略。通过采集不同运行条件下的故障电流的直流分量、工频分量、转速频率分量等故障信息组成历史特征样本集,利用聚类算法刻画故障历史数据特征和类别属性,并采用距离相似度判据判定实时特征样本属性,从而实现集电线路保护判断。在DIgSILENT中搭建了规模化风电场模型,模拟多种集电线路故障情况。仿真结果表明在故障条件下该方法无须设定整定值,且不会受到风机馈出电流的影响造成非故障线路保护误动,有效地解决了风电场集电线路保护选择性失配的问题。
To solve the problem of maloperation in the existing collector line protection for a large-scale wind farm,the intrinsic relationship between the output current and the voltage sags of a doubly-fed induction generator(DFIG)with low voltage ride through capability is analyzed under fault conditions. According to the composition characteristics of the DFIG fault current,a protection strategy for wind farm collector line is proposed based on clustering analysis. The protection is realized by collecting the historical feature sample set composed of DC component,AC component and rotating frequency component of fault current under different operation conditions,utilizing the proposed clustering algorithm to characterize the fault historical data features and category attributes,and using the distance similarity criterion to determine the real-time feature sample attribute. A large-scale wind farm model is built in DIgSILENT to simulate various fault conditions of collector lines. Simulation results show that the proposed protection method does not need to set the setting values under fault conditions. In addition,it does not cause maloperation of non-fault line protection under influences of the DFIG feeding current,thus effectively solving the problem of selective mismatch in wind farm collector line protection.
作者
赵国炫
鲁文军
喻锟
谢靖言
曾祥君
ZHAO Guoxuan;LU Wenjun;YU Kun;XIE Jingyan;ZENG Xiangjun(College of Electrical and Information Engineering,Hunan University,Changsha 410082,China;Key Laboratory of Renewable Energy Electric-Technology of Hunan Province(Changsha University of Science and Technology),Changsha 410114,China;Key Laboratory of Smart Grids Operation and Control of Hunan Provence,Changsha University of Science and Technology,Changsha 410114,China)
出处
《电力系统及其自动化学报》
CSCD
北大核心
2020年第7期39-46,共8页
Proceedings of the CSU-EPSA
基金
湖南省自然科学基金资助项目(2019JJ50645)
湖南省高校创新平台开放基金资助项目(18K047)
可再生能源电力技术湖南省重点实验室(长沙理工大学)开放基金资助项目(2018ZNDL006)。
关键词
风电场
集电线路
聚类算法
过电流保护
wind farm
collector line
clustering algorithm
overcurrent protection
