CORC  > 清华大学
最小信息损失状态估计在拓扑错误辨识中的应用
孙宏斌 ; 高峰 ; 张伯明 ; 杨滢 ; SUN Hong-bin ; GAO Feng ; ZHANG Bo-ming ; YANG Ying
2010-06-10 ; 2010-06-10
关键词电力系统 状态估计 拓扑错误辨识 数据融合 应用 信息理论 能量管理系统 Power system State estimation Topology error identification Data fusion Application Information theory Energy management system TM73
其他题名APPLICATION OF MINIMUM INFORMATION LOSS BASED STATE ESTIMATION TO TOPOLOGY ERROR IDENTIFICATION
中文摘要网络拓扑的错误导致状态估计结果不可用,一直是电力系统状态估计应用中的难题。该文将最小信息损失(MIL)状态估计理论应用到拓扑错误辨识中,提出了具有严格数学基础的MIL拓扑检错新方法。该方法综合利用了蕴涵在遥测和遥信中的信息量,在总体信息损失最小的估计目标下,对遥信和遥测实施统一最优估计,并给出拓扑检错结果。文中还分析了MIL法与传统方法的内在联系,指出了传统残差法是MIL法的特例,讨论了新方法的在线实现问题。给出了小算例系统和实际系统的对比试验研究,结果验证了MIL法比传统残差法具有更低的错误率和更强的可扩展性。; Network topology error,which results in failure of state estimation,is one of the most difficult problems in application of state estimation to power systems.In the paper,a novel minimum information loss(MIL) based method with rigorous mathematic foundation for topology error identification is presented.In the proposed MIL method,information contained in hybrid data including remote-analogue and remote-digital can be used synthetically,and power flow and topology can be estimated synchronously,hence topology errors can be identified.The inner relationship between the MIL and traditional method are discussed.It is pointed out that the traditional residual based method is a special case of the MIL method.Discussions about online implementation of the MIL method are done.Numerical tests on a small case and a real-life power network are reported,the results show that,in comparison with the traditional residual based method,the MIL method is of low error ratio and strong expansibility.; 国家自然科学基金项目(50107005)。~~
语种中文 ; 中文
内容类型期刊论文
源URL[http://hdl.handle.net/123456789/62997]  
专题清华大学
推荐引用方式
GB/T 7714
孙宏斌,高峰,张伯明,等. 最小信息损失状态估计在拓扑错误辨识中的应用[J],2010, 2010.
APA 孙宏斌.,高峰.,张伯明.,杨滢.,SUN Hong-bin.,...&YANG Ying.(2010).最小信息损失状态估计在拓扑错误辨识中的应用..
MLA 孙宏斌,et al."最小信息损失状态估计在拓扑错误辨识中的应用".(2010).
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace