A compensation approach for conic optimal power flow in weakly meshed networks -

Abstract

This research presents an efficient approach for solving the optimal power flow (OPF) problem in weakly meshed distribution networks. The aim of OPF is to minimize the total active power loss in the system without violating the network’s technical constraints. The proposed approach builds on techniques for convex relaxation in optimal power flow, and complex power compensation in power flow. The iterative forward-backward sweep method is commonly used to solve the power flow problem in radial networks; extension to meshed networks is achieved via a compensation technique applied after a loop breaking process. This research shows that a similar approach could be applied to solve weakly meshed OPF problems, by exchanging the forward-backward sweep power flow method with a radial OPF solver. For the radial OPF, the proposed approach employs convex relaxation to attain a second order conic programming problem whose global solution could be efficiently computed. In the case of weakly meshed topology, multi-port complex power compensation is iterated with the solution of the convex network OPF to effectively model meshes in the radial network OPF solver. The proposed approach is tested on distribution networks with up to 3000 nodes, and the results are validated by comparison with a benchmark OPF solver.