Research On Artificial Bee Colony Algorithm With Orthogonal Experimental Design

Nowadays,optimization plays an important role in various fields such as production scheduling and logistics distribution,but traditional optimization algorithms cannot effectively solve complex optimization problems.In order to solve complex optimization problems,the meta-heuristic algorithm using stochastic optimization technology to search for solutions has been widely concerned and applied by researchers and technicians.Meta-heuristic algorithm includes evolutionary algorithm(EA),meme computation(MC),swarm intelligence(SI),etc.As a new swarm intelligence algorithm,artificial bee colony algorithm has been widely concerned by many researchers and technicians,and has been widely used in many fields such as complex optimization problem,social production scheduling and image processing.Artificial bee colony algorithm(ABC)is a population-based global optimization technique first proposed by Karaboga and Basturk in 2005.ABC was inspired by the behavior of bees in gathering honey.ABC algorithm has been proved to perform better than other evolutionary algorithms through a series of experiments.Artificial bee colony algorithm simulates the behavior of bee collecting nectar,compares the advantages and disadvantages of the problem through the local optimal behavior of individuals in the population,and finally makes the global optimal solution emerge in the population,which has the advantages of less control parameters,easy to realize and high search efficiency.However,constrained by the search pattern and search equation of artificial bee colony algorithm,the artificial bee colony algorithm has strong exploration ability but weak development ability.Therefore,it has become an important issue in the field of artificial bee colony algorithm research how to maintain the characteristics of strong algorithm exploration ability while improving the development ability of the algorithm,so as to improve the performance of the algorithm.In view of the above problems existing in artificial bee colony algorithm,this paper studies the search mode and search equation of the algorithm respectively.The main work is as follows:Firstly,according to the characteristics of OL(orthogonal learning)and on the premise of maximizing the number of function evaluations,this paper optimizes the dimension of each participation in orthogonal learning,and proposes a QOL(quarter orthogonal learning)method.On this basis,the method of random selection and elite guidance is combined for dimensional selection to maintain the exploration ability and accelerate the convergence speed.Secondly,this article comparative study in terms of search equation of the existing search equation,in the observation of bee phase two new search equation and a new equation method,put forward two search equation are more likely to develop,but is responsible for the search of different areas,combined with the new equation method,can effectively enhance development ability of the algorithm.The experimental results show that QOL method can significantly improve the performance of artificial bee colony algorithm.The proposed artificial bee colony algorithm improves significantly in solution accuracy,convergence speed and success rate.