CRITICAL PSEUDO-MEASUREMENT SELECTION FOR UNREDUCED EXTERNAL SYSTEM MODELING

Attention is focused on state estimator based methods for unreduced ex ternal system representation. The crucial issues related to these meth ods are: how to prevent the contamination of internal system estimates by usually less accurate external data, and how to keep power mismatc hes at boundary buses within reasonable limits. The proposed strategy to solve both problems is based on the use of non-redundant pseudo-mea surements for the external system. In order to improve the quality of the external system representation, it is desirable to choose, from al l currently available external pseudo-measurements, those which are th e most reliable in some sense. This paper introduces a method to selec t the most reliable non-redundant pseudo-measurements for unreduced ex ternal system modeling. The proposed algorithm is combinatorial, and e xhibits as its main feature the flexibility to accommodate user-define d criteria for selecting the external data. The results obtained throu gh the proposed method are verified by performing contingency analysis studies in the IEEE 14-bus and 30-bus test systems, as well as in a 4 8-bus realistic network.