Estimación de Estado Generalizada (GSE) con aplicación al Sistema Nacional Interconectado del Ecuador

Abstract

One of the main foundations of conventional estimation algorithm (SE) is to assume that the network topology information is error free, which often is not true, causing serious distortions in the results. Numerous algorithms that attempt to solve this problem by including the network topology within the state estimation process have been proposed. These algorithms have been grouped into what is called Generalized State Estimation (GSE). In this paper, based on a proposed methodology, several GSE algorithms were compared and three of them were selected and implemented in MATLAB. The algorithm that presented the best features called State and Topology Coestimation (CNET by Spanish acronyms) was selected and applied to the National Electrical System of Ecuador (SNI by Spanish acronyms). CNET algorithm uses information of SNI by means an interface called GSE-EMS to get the detailed model of the power system (bus section-breaker model). Finally, CNET and SE algorithms were compared and the conditions under CNET algorithm is more reliable that SE algorithm were determined. Index Terms-- State Estimation, Generalized State Estimator, State and Topology Coestimation, Energy Management System, Power Systems.