Improve Exploration of Arithmetic Optimization Algorithm by Opposition-based Learning

Abstract

An improved version of the arithmetic optimization algorithm (AOA) based on the opposition-based learning (OBL) strategy called OBLAOA is proposed in this paper. The proposed OBLAOA algorithm consists of two stages, and in the second stage we adds OBL to update the AOA population in each iteration. The improved AOA is compared with the original AOA by using 12 benchmark functions in different dimensions to validate the improvement on exploration with the OBL. Eventually ,we get a conclusion that the OBLAOA is committed to take both candidate solutions and their opposite solutions into consideration, which shows greater opportunity to reach the global optimal and faster convergence acceleration than AOA.