Improved Artificial Bee Colony Based on Orthognal Learning for Optimal Power Flow

Abstract

Optimal power flow (OPF) problem is to optimize an objective function (usually total cost of generation), while satisfying system constraints. The OPF is a non-linear and non-convex problem, and an artificial bee colony (ABC) algorithm is utilized to handle the problem. Heuristic methods are credited for their simplicity to solve complex non-linear optimization problem without simplifying approximation of the system. However, the original ABC has poor efficiency on exploitation search, thus in order to find better global optimum, this paper proposes an improved ABC (IABC) based on orthogonal learning. The IABC implements the idea of orthogonal experiment design (OED) based on the orthogonal learning. The validity and effectiveness of the method are tested in the IEEE-30 bus system.