Convex Relaxation Method of Optimal Solution in Power Distribution Networks

Abstract

For operation and scheduling of power systems, AC Optimal power flow is needed. To obtain
an optimal solution to optimal power flow (OPF) problem, convex relaxation methods are
used. It is essential to recover to feasible solution when exactness of convex relaxation is not
guarantee. In this paper, second order cone programming SOCP which will be alternative
solution for convex optimization method. A penalty convex procedure used for first
formulated for Difference-of-convex programming problem. Linearization of concave parts of
power flow constraints gives solution for convex approximation of DCP problem. Global
optimal solution can be obtained by using a solution of tightened SCOP OPF model as initial
point. By the conduction of numerical tests, this method outperforms same as semi definite
based algorithms.