Robust state estimator applied to the Ecuadorian electric power system

Abstract

This paper describes the selection process, design, and implementation of a robust state estimator on a realistic power system. The classical WLS estimator presents limitations when handling bad leverage points, and the masking effects that may arise in the presence of multiple bad data. On the other hand, robust estimators are able to handle measurement outliers and leverage points, remaining unbiased. In this paper we compare several robust state estimator methods and select the most promising one for development. We describe the implementation and interfacing aspects required to apply the robust estimator to a realistic system, and compare its performance against the WLS estimator.