Distributed agent-based state estimation considering controlled coordination layer

• A coordination strategy considering controlled coordination layer is presented.
• A distributed agent-based SE defined three types of agents is proposed.
• Observability analysis of DSE is investigated.
• Dependence measurements are used to identify boundary bad data.
• The proposed method is tested on the IEEE 14-bus and 118-bus system.

The buses involved in re-estimation is fixed in the traditional distributed state estimation. On the one hand, if only the boundary buses were re-estimated, will lead to a high power mismatch in a short or low impedance transmission line. On the other hand, if all the buses were re-estimated again, will lead to unnecessary computational burden. To solve this problem, this paper presents a new distributed agent-based state estimation that considers controlled coordination layer. Each local agent of a subsystem is in charge of the state estimation of its own system. The global sensitivity agent is responsible for the sensitivity analysis of the entire system. The boundary bus aggregation agent determines the range of coordination layer to be re-estimated according to the state estimation deviation of each boundary bus defined in this paper. Furthermore, the observability analysis judgment theorem of the overlapped DSE and the topology analysis suitable for the bad data of boundary region are investigated. The proposed method is illustrated with the IEEE 14-bus system, and the test results for the IEEE 118-bus system are provided.