Topology Identification, Bad Data Processing, and State Estimation Using Fuzzy Pattern Matching

Abstract

A new method for topology identification, bad data detection and state estimation (SE) is proposed in this paper. The method utilizes fuzzy clustering and pattern matching for this purpose. The proposed approach is very efficient and accurate for simple SE and to aid in improving the accuracy and time performance for conventional weighted least squares state estimator (WLSE). In this method, a fuzzy pattern vector is generated based on the analog measurement vector (telemetry data) received. The topology identification and gross errors are detected from difference between the fuzzy pattern vector and the analog measurement data. The gross errors and topology errors in the measurement data are detected and corrected by using the fuzzy pattern vector which can directly be used as good measurement data for SE. The effectiveness of the proposed method is demonstrated on IEEE 14-bus, IEEE 57-bus, practical 75-bus Indian system and IEEE 118-bus systems. The SE results obtained by proposed method are also compared with ANN models reported in the literatures.