An ADMM-based fully distributed robust state estimation algorithm for mixed-integer nonlinear programming

Abstract

State estimation is the basis for the safe and reliable operation of power systems. On the one hand, the existence of bad data seriously affects the accuracy of the estimation results; on the other hand, the centralized state estimation calculation requires global grid information, which leaks information privacy and increases the communication burden. In this paper, a fully distributed robust state estimation algorithm based on the alternating direction multiplier method (ADMM) with mixed-integers nonlinear programming (MINLP) is proposed to detect the measurement error and obtain the state estimation solution. In this algorithm, first, a large-scale system is decomposed into multiple small systems by the node replication method, and a robust state estimation independent model with MINLP is constructed in parallel for each small system by introducing consistent variables. In this algorithm, the system only needs to exchange boundary coupling node information with adjacent systems to update its consistency variables and Lagrange multiplier information, thus realizing the protection of information privacy. The calculation example verifies that the algorithm proposed in this paper can accurately and effectively detect the bad data in each system, and obtain the same solution as the centralized one.