The shuffled conic power flow equations: an improved angle-inclusive conic model

Abstract

This paper presents the Shuffled Conic Power Flow (PF) equations which enable a novel voltage angle substitution within a second-order cone (SOC) PF model. Computational experiments compare the performance of the novel voltage angle substitution against a conventional approximation used to model voltage angles within a state-of-the-art SOC PF model. Solving the PF problem for four radial distribution networks and the optimal power flow (OPF) problem for four demand scenarios in one of these networks, we show the following. First, voltage angles were improved in around 95% of cases when solving the PF problem and around 96% of cases when solving the OPF problem. Second, there is a particular improvement in voltage angles when voltage magnitudes divert from 1 p.u., and when there is demand growth. The results are also compared to benchmark software, Matpower.