Transformer tap position estimation and bad data detection using dynamic signal modelling

Abstract

Tracking estimation of a transformer tap position is considered. An estimation algorithm incorporating additional information about the dynamic behaviour of the measurements and tap positions is developed. It is based on the simultaneous state and tap position estimation method augmented by measurement prefiltering based bad data detection and incorporation of a tap position dynamics model into the estimation algorithm. Increased bad data identification reliability and the robustness of the estimates to the measurement noise is obtained. The performance of the algorithm has been examined using the IEEE 30 and 118-node systems. >