Real-Time State Estimation for the Nigerian Transmission Network Using Extended Kalman Filter Methods

This study develops and evaluates an extended Kalman filter based state estimation algorithm for real-time monitoring of the Nigerian 330kV and 132kV transmission network, addressing the inadequacy of the current TCN SCADA system's state estimation capabilities under high measurement redundancy loss conditions. Accurate real-time state estimation is fundamental to effective energy management system operation, enabling operators to verify network conditions, detect abnormal states, and support security-constrained economic dispatch. However, TCN's network experiences frequent SCADA measurement communication failures, creating bad data and missing measurement conditions that degrade conventional weighted least squares state estimator performance. The EKF-based estimator is formulated to exploit the dynamic relationships between successive network states through a linear prediction step that propagates the state estimate forward in time based on the power system's known dynamic behaviour, significantly reducing sensitivity to measurement dropouts compared to static WLS. The algorithm is implemented in MATLAB and evaluated on a model of the TCN 330kV network constructed from published network data, with artificial measurement noise and communication failure scenarios applied based on analysis of TCN SCADA failure statistics. Performance metrics include state estimation accuracy under varying measurement availability fractions, convergence speed, and computational requirement. Results demonstrate that the EKF estimator maintains acceptable state estimation accuracy with measurement availability as low as 55 percent, compared to WLS divergence below 70 percent availability. Computational time is within the one-second EMS refresh cycle requirement on standard server hardware. Keywords: state estimation, extended Kalman filter, transmission network, SCADA, Nigeria.