Highly distributed state estimation for a DC spacecraft power system

To enable the development of an autonomous power control system comprised of highly-distributed software agents, a distributed implementation of a power system state estimator must be developed. The estimator uses agents located at each bus to estimate the local state and boundary conditions that are then communicated to neighboring agents. Using an iterative approach, the agents can compute a near-optimal state estimate using only local information and the communicated boundary conditions. The algorithm is implemented and tested against a subset of the International Space Station's power system. The results are compared and validated with those of a centralized state estimator.