Modified Re-Iterated Kalman Filter for Handling Delayed and Lost Measurements in Power System State Estimation

Abstract

State estimator (SE) is a critical tool in the power system control center (CC) and is used by the system operators for real-time control and monitoring. Traditional SE techniques (static and dynamic) work under the assumption that there is no delay in communicating the sensor measurements to the CC. But in actual scenario, measurements are often randomly delayed and are not immediately available for processing. When the observations arrive with a delay, the current estimate of the system state may not be evaluated correctly, since the missing measurement may be carrying relevant information for estimating the states. With the increased deployment of phasor measurement units (PMUs) as sensors in the grid, time-stamped synchronized measurements are readily available. Time-stamped measurements enable the delay to be exactly known at the estimator. In this paper, a modified re-iterated Kalman filter that can handle both lost and delayed measurements without having any knowledge of the delay statistics, is proposed. The proposed algorithm is tested and validated in real-time using Real Time Digital Power System Simulator (RTDS).