Exact Distribution Optimal Power Flow (D-OPF)Model using Convex Iteration Technique

TL;DR

This paper introduces a convex iteration method to solve the distribution optimal power flow (D-OPF) problem, ensuring feasible and optimal solutions for PV hosting capacity without relaxation inaccuracies.

Contribution

It proposes a novel convex iteration technique that guarantees feasible solutions for D-OPF problems where traditional SOCP relaxations fail.

Findings

Successfully applied to IEEE test systems

Achieves feasible solutions with maximum PV capacity

Improves solution accuracy over existing relaxations

Abstract

The distribution optimal power flow (D-OPF) models have gained attention in recent years to optimally operate acentrally-managed distribution grid. On account of nonconvex formulation that is difficult to solve, several relaxation methods have been proposed; the exactness of the solutions obtained from the relaxed models, however, remain a concern. In this paper, we identify one such problem related to radial distribution feeder where second-order cone program (SOCP) relaxation does not yield a solution that is feasible with respect to the original nonlinear OPF model. Specifically, we formulate an OPF model for PV hosting capacity problem to obtain maximum PV capacity that a feeder can integrate without violating the operating constraints. The SOCP relaxation for this problem yields infeasible solutions. To address this concern, we propose a convex iteration technique to simultaneously achieve optimal and feasible OPF solution for the PV hosting (maximization) problem. The proposed approach minimizes the feasibility gap with respect to original nonlinear constraints at each SOCP iteration. The effectiveness of the proposed approach is validated using IEEE-13 node and IEEE-123 node test systems.