Engineering Research Initiation Grant: An Efficient Hierarchical State Estimator for Large Power Networks (REU Supplement)

本课题的目的是研究一种适用于大型电力系统的有效静态估计器。所提出的估计器包括执行两级计算：在较低（第一）层次上，对所有子系统同时执行常规状态估计；然后在上层协调这些局部估计。该方法利用电力系统的物理特性，将局部状态估计问题表述为两个线性规划子问题，并依次求解。它还建议研究Dantzig和Wolf分解原理的一种形式，以减小问题的规模。将使用IEEE标准和其他现有的电力系统进行测试和比较。本文的研究为解决大规模电力系统的状态估计问题提供了创新的方法。它肯定会增加对线性规划状态估计器的性质的理解，并对它们可能存在的问题提出补救措施。同时，本文还提出了一种适合于线性规划状态估计器的状态估计问题分解的新方法。

The objective of this project is to investigate an efficient static state estimator for very large power systems. The proposed estimator consists of performing a two-level calculation: in the lower (first) hierarchical level, a conventional state estimation is performed simultaneously for all subsystems; these local estimations are then coordinated at the upper level. The approach exploites the physical properties of power systems and formulates the local state estimation problem as two linear programming sub-problems which would be sequentially solved by a proposed method. It also suggests investigating a form of the decomposition principle of Dantzig and Wolf to reduce the size of the problem. Standard IEEE and other existing power systems will be used for testing and comparison purposes. The proposed research suggests innovative techniques to solve the problem of state estimation on large scale power systems. It will certainly increase the understanding of the properties of linear programming state estimators and would suggest remedies to the problems which they might have. It is also expected to result in a new approach for decomposing the state estimation problem that is suitable for application with linear programming state estimators.