Distributed Multi-Area State Estimation for Power Systems With Switching Communication Graphs

We consider distributed multi-area state estimation algorithms for power systems with switching communication graphs. The power system is partitioned into multiple geographically non-overlapping areas and each area is assigned with an estimator to give a local estimate of the entire power system's state. The inter-area communication networks are assumed to switch among a finite set of digraphs. Each area runs a distributed estimation algorithm based on consensus+innovations strategies. By the binomial expansion of matrix products, time-varying system and algebraic graph theories, we prove that all area's local estimates converge to the global least square estimate with probability 1 if the measurement system is jointly observable and the communication graphs are balanced and jointly strongly connected. Finally, we demonstrate the theoretical results by an IEEE 118-bus system.