Ebcro: Hybrid Chemical Reaction With Employed Bee Operator

Zhiyong Li,Tingkun Yuan,Bo Yang,Shilong Jiang,Yanmei Xie

DOI: https://doi.org/10.1109/FSKD.2017.8393061

2017-01-01

Abstract:his article proposes a new hybrid algorithm EBCRO, which combines the local search operator of chemical reaction optimization algorithm (CRO) with modified employed bee operator (EB) main part of artificial bee colony algorithm (ABC), and the introduction of probability-based selection of molecular method. This algorithm creates new solutions not only by local search operation of CRO but also mechanisms of EB. CRO is a newly, effectively swarm intelligence optimization algorithm. However, owing to random characteristics of CRO and not make use of other solutions information, its search ability and convergence speed are inefficient. EB as a part of ABC has simplicity and efficiency global search ability, but the solution is updated only by other solutions information which leading to lack of local search ability. Therefore, this paper presents EBCRO, which can balance the exploitation and exploration of algorithm of numerical function optimization. In this study, EBCRO is tested on 23 benchmark numerical functions, and the simulation results of EBCRO, CRO, orthogonal chemical reaction optimization (OCRO), hybrid particle swarm and chemical reaction optimization (HP-CRO2), and ABC showed that EBCRO outperforms other algorithms in convergence performance.