Distributed Cubature Kalman Filter with Performance Comparison for Large-scale Power Systems

This paper studies a distributed cubature Kalman filter algorithm for the dynamic state estimation of a large-scale power system, which is interconnected and consists of some sub-areas. Using the hybrid measurements, each subsystem runs the proposed algorithm independently and in parallel to forecast and estimate its local state, which helps in reducing the whole execution time and the communication load without a central coordinator. The edge measurements and the exchanged information with the neighboring subsystems are utilised to correct the local state estimation of each subsystem, which makes this algorithm improve some known methods in literature. Simulation results of the IEEE 118-bus system are provided to demonstrate the performance of the proposed algorithm by comparing it with the distributed methods based on the extended Kalman filter and unscented Kalman filter.