基于电流极性和相对熵的广域继电保护新算法
发布时间:2018-06-09 13:31
本文选题:广域继电保护 + 边界电流极性 ; 参考:《西南交通大学学报》2017年03期
【摘要】:为了提高广域继电保护的容错性和灵敏度,结合母线采样值保护和线路故障分量差动保护原理,提出一种基于电流极性和能量相对熵的广域继电保护新算法.首先,对线路IED保护关联域进行划分,针对不同关联域,提取正序工频故障分量电流,并给出关联域边界电流、综合计算电流和综合虚拟电流概念;其次,定义综合电流采样值能量相对熵,量化综合电流的差异程度,同时,通过求取关联域边界电流之间的夹角大小来表征边界电流的极性关系,并利用电流采样值能量相对熵和极性关系识别关联域内外故障;最后,利用PSCAD/EMTDC软件搭建IEEE-3机9节点系统模型,分别在不同故障点设置不同故障类型进行仿真检验算法性能.研究结果表明:在关联域区内故障时,综合电流几乎一样,相对熵值很小,接近为0;边界电流间夹角小于π/2,极性近似相同;区外故障时,综合电流波形及采样值能量差异显著,相对熵值大于1;边界电流间夹角接近于π,极性近似相反,二者差异明显;域内单相接地故障时,在800Ω的高阻条件下,相对熵最大为0.428,夹角最大为0.388,仍能可靠满足判据,具有较高的灵敏度.实验结果验证了本文算法性能基本不受过渡电阻的影响,能较好克服TA饱和、数据畸变等不利因素的影响.
[Abstract]:In order to improve the fault tolerance and sensitivity of wide area relay protection, a new algorithm for wide area relay protection based on current polarity and energy relative entropy is proposed, which combines bus sampling value protection with line fault component differential protection principle. Firstly, the correlation domain of line IED protection is divided, the positive sequence power frequency fault component current is extracted according to different correlation domain, and the boundary current of correlation domain is given, and the concepts of integrated calculation current and synthetic virtual current are given. The relative entropy of integrated current sampling value is defined to quantify the difference of integrated current. At the same time, the polarity relationship of boundary current is represented by calculating the angle between the boundary currents in the correlation domain. The relative entropy and polarity relation of current sampling values are used to identify the faults in and out of the correlation domain. Finally, a 9-bus system model of IEEE-3 computer is built by using PSCAD / EMTDC software, and different fault types are set up at different fault points to verify the performance of the algorithm. The results show that the integrated current is almost the same, the relative entropy is very small, close to zero, the angle between boundary currents is less than 蟺 / 2, and the polarity is approximately the same. The relative entropy is greater than 1, the angle between boundary currents is close to 蟺, the polarity is close to opposite, and the difference between them is obvious. Under the condition of high resistance of 800 惟, the maximum relative entropy is 0.428 and the angle of inclusion is 0.388. the criterion can be satisfied reliably. It has high sensitivity. The experimental results show that the proposed algorithm can overcome the influence of TA saturation, data distortion and other unfavorable factors.
【作者单位】: 西南交通大学电气工程学院;四川理工学院自动化与电子信息学院;
【基金】:人工智能四川省重点实验室项目(2013RYY01,2012RYY06,2014RYY05,2015RYY01)的资助
【分类号】:TM77
【相似文献】
相关期刊论文 前10条
1 张彦峰;;基于灰色关联分析的相对熵集结模型[J];西北工业大学学报;2013年02期
2 曾青松;贺卫国;;基于相对熵度量的行为识别方法[J];河南科技大学学报(自然科学版);2009年06期
3 李小鹏;何正友;武骁;;基于S变换能量相对熵的高压输电线路极性比较式纵联保护[J];电网技术;2014年08期
4 张震;郭崇慧;;基于相对熵的多粒度不确定语言型群决策方法[J];大连理工大学学报;2012年06期
5 白凤兰;徐s,
本文编号:1999891
本文链接:https://www.wllwen.com/kejilunwen/dianlidianqilunwen/1999891.html