A new AC optimal power flow formulation and solution using Genetic Algorithm based on P-Q decomposition

This paper presents a new P-Q decomposition approach to solve the AC optimal power flow (OPF) problem. An alternative power flow model based on line flow equations (LFB) is used. In traditional power flow, the voltage magnitude and phase angle at each bus are treated as the independent variables. In the LFB model, the real and reactive power flows through the line along with the bus voltage magnitudes constitute the independent variables. The formulation is done by decoupling the OPF problem into active power and reactive power subproblems using line flow equations. Unlike traditional decomposition methods, the technique adopted in this approach does not ignore the weak coupling between the subjacobians. Taking a pragmatic point of view and eschewing complicated mathematical algorithms, the paper presents a solution procedure using Genetic Algorithm methodology. Following a detailed description of the steps in the new approach, test results from an implementation of the proposed algorithm on a bench mark test system are presented. The results demonstrate the usefulness of this approach for achieving a practically acceptable solution to the AC OPF problem.