知识导航
资源分类

基于Adaboost算法的并网光伏发电系统的孤岛检测法

基于Adaboost算法的并网光伏发电系统的孤岛检测法

An Islanding Detection Method for Grid-Connected Photovoltaic Power System Based on Adaboost Algorithm

An Islanding Detection Method for Grid-Connected Photovoltaic Power System Based on Adaboost Algorithm

doi:
10.19595/j.cnki.1000-6753.tces.170016
摘要:
目前光伏并网发电系统的孤岛检测多采用主、被动法结合的方法,然而无法有效确定参考电气量的阈值,导致孤岛检测存在死区,同时也可能影响电能质量.基于此,提出了一种基于 Adaboost 算法的智能被动式孤岛检测方法,通过仿真建立光伏系统并网和孤岛运行的数据库,提取可判断并网和孤岛运行两种状态的特征量,利用Adaboost算法建立判断并网和孤岛运行状态的二分类模型,应用该模型根据某一运行状态下的特征量对其进行分类.该方法需要建立比较完备的数据库,选择合适的电气量作为判据.该方法避免了传统孤岛检测法各电气量阈值无法确定的问题,不引入扰动信号,不干扰电能质量.最后通过仿真验证了其检测孤岛的可靠性.
Abstract:
The universal islanding detection method is the combination of passive methods and active methods, however, the thresholds of the reference electrical quantities can hardly be accurately set, this will lead to none detection zone and negative effect to the power quality. Considering this, this paper proposes an intelligent passive islanding detection method based on Adaptive boosting algorithm. The data base of photovoltaic power systems' operation conditions are acquired through simulations. Six characteristic quantities which can distinguish islanding from grid connected operation are selected and Adaboost algorithm is employed to generate the binary classification model for judging grid-connected and islanding operation. This method needs to establish relatively complete database and choose appropriate electric quantities as the criterion. The proposed method set the thresholds automatically by intelligent algorithm, and it doesn't introduce disturbance signals and has no effect to the system's power quality, its validity is verified by simulation.
作者 贾科 [1] 宣振文 [1] 林瑶琦 [1] 魏宏升 [2] 李光辉 [3]
Author: Jia Ke[1] Xuan Zhenwen[1] Lin Yaoqi[1] Wei Hongsheng[2] Li Guanghui[3]
作者单位
  1. 新能源电力系统国家重点实验室(华北电力大学) 北京 102206
  2. 广州供电局有限公司 广州 510000
  3. 国家电网公司中国电力科学研究院 北京 100192
期 刊: 电工技术学报 ISTIC EI SCI PKU CSSCI
Journal: Transactions of China Electrotechnical Society
年,卷(期) 2018, 33(5)
分类号 TM615
关键词: 孤岛检测 知识脉络 Adaboost算法 知识脉络 智能被动式检测法 知识脉络 分类模型 知识脉络
Keywords: Islanding detection Adaptive boosting algorithm intelligent passive detection method classification model
机标分类号
基金项目 国家自然科学基金项目,中央高校基本科研业务费专项基金项目
参考文献和引证文献
返回顶部参考文献
返回顶部引证文献
返回顶部本文读者也读过
返回顶部相似文献
返回顶部相关博文

知识产权声明| 服务承诺| 联系我们| 人才招聘| 客户服务| 关于我们

京ICP证:010071  互联网出版许可证:新出网证(京)字042号  京公网安备11010802020237号
万方数据知识服务平台--国家科技支撑计划资助项目(编号:2006BAH03B01)©北京万方数据股份有限公司  万方数据电子出版社