A Distributed Optimal Power Flow Algorithm that Incorporates A Priori Congestion Information through Bounded Dual Updates

Distributed optimal power flow (OPF) algorithms have been proposed as scalable optimization tools for the future grid; however, few of them incorporated a priori information about transmission congestion from operational experience, a prevalent technique to improve the computational performance when the OPF problem is solved in a centralized way. Centralized OPF algorithms typically include only a fraction of line flow limit constraints that are predicted to be binding. In this paper, a heuristic is proposed to incorporate a priori congestion information into the dual updates of a distributed OPF algorithm. The dual solution of the consensus constraints for phase angles is derived through duality theory, and it is shown to be correlated with the congestion status of nearby lines. Instead of updating the dual variables without restrictions, the proposed heuristic proposes to apply bounds determined from the congestion information in the dual update step. Numerical studies on a 3012-bus Polish system verified the effectiveness of the proposed method.