Hybrid Method Based on Metaheuristics and Interior Point for Optimal Power Flow

Abstract

In this paper we present a hybrid algorithm based on metaheuristics and the interior point (IP) method from MATPOWER to solve the optimal power flow problem. The control variables optimized are the real power and voltage of the generators, the transformer tap ratios and angles and the settings of the static volt-ampere reactive compensators (SVARs). The metaheuristic is used to optimize the discrete variables while MATPOWER is used at every evaluation of the fitness function to compute optimized values for the continuous variables. Compared to methods relying only on metaheuristics, our proposed approach is able to optimize the control settings for networks that are much larger. Compared to using MATPOWER alone, our proposed approach is able to optimize the transformer and the SVAR settings. To select the metaheuristic that is best suited for this application, five metaheuristics were implemented and compared. The software was implemented in MATLAB and parallelized to run on a computer cluster. The proposed algorithm was tested on networks up to 2383 buses.