A Methodology for Transient State Estimation Based on Numerical Derivatives, Optimal Monitoring, and Filtered Measurements

Abstract

This paper proposes a methodology for transient state estimation in power systems. The proposed methodology is formulated using approximation methods for derivatives to relate the state variables to measurements. It does not require knowledge of the steady state to establish the predisturbance operation conditions. The method uses an optimal monitoring system based on topological analysis to obtain full observability. A saving index is introduced to analyze the effectiveness of the instrumentation used. The adverse effect of noisy measurements in the estimation process is mitigated by using an infinite impulse response filter. A transient index is introduced to estimate the fault location. The transient state estimation is assessed using two test systems. The results are validated through direct comparison against those obtained by simulation using SimPowerSystems toolbox of Simulink. With the proposed methodology, the transient state estimation can be obtained with an important saving in the implementation of the measuring system and with considerably less computational effort.