抗差最小二乘法状态估计

研究电力系统状态估计具有重要的理论价值和实际意义 ,在现代调度控制中心 ,状态估计是实现EMS/DMS的许多功能 :调度员潮流、事故预想分析和调度员模拟培训等的基础。在实际运行的电网自动化系统中 ,要想实时得到正确的网络状态估计结果是很困难的 ,因为在进行状态估计的过程中 ,必须考虑多个坏数据和多个拓扑错误同时存在的情况。抗差估计理论主要研究抗拒少量粗差 (泛指离群的误差 )对估值的影响。拓扑错误和坏数据可以分别看作带有粗差的网络参数和量测数据 ,因此可以将抗差最小二乘法用于存在拓扑错误和坏数据时的状态估计。算例结果表明 ,抗差最小二乘法具有良好的抗粗差能力和收敛可靠性 ,收敛速度快 ,并能够将抗粗差和状态估计在计算过程中一次完成 ,不需要象普通最小二乘法一样进行多次的状态估计计算。

State estimation with the least robustness square method

The study of state estimation in power systems is very important both in theoretical and practical meaning.In modern dispatch control center,state estimation is fundamental to many functions of EMS/DMS,such as tide flow for dispatchers,expected events analysis,simulation training for dispatchers,etc..However it is difficult to get the correct real time network states in practical Electric Network Automation,for in state estimations,the situation that a lot of bad data and topology errors exist simultaneously must be taken into account.Robustness square estimation theory pays much attention to the influence of little outlier(refer to the errors of those not in the mass)resistance to estimation.Since topology errors and bad data can be considered as network parameters with outlier and measured data separately,the least robustness square method can be used in state estimations with topology errors and bad data.As shown in the results of calculation examples,the least robustness square method has favorable outlier resistance,convergence reliability and high convergence speed.Furthermore,unlike ordinary least square methods,the least robustness square method can combine outlier resistance and state estimation in one calculation process rather than perform repeated calculations.