The Improvement Research Of Artificial Bee Colony Algorithm For Solving Global Optimization Problems

Artificial bee colony(ABC)is a very effective and efficient swarm-based heuristic algorithm,which is an intelligent optimization algorithm that simulates the collective foraging behavior of the honey bees.ABC has outstanding advantages such as easy implementation,good robustness and so on.However,it has strong exploration ability but poor exploitation ability because its solution search equation performs well in exploration but badly in exploitation when solving engineering application problems.Moreover,the distance and fitness information have the potential benefits by building the better effectively neighborhood structure to further improve its performance.Thus,ABC also has room for development and improvement.The research content of this paper is to deeply analyze the theoretical of ABC.According to the existing problems of some design defects,corresponding improvement methods are proposed to improve the optimization ability of ABC in solving global optimization problems.Thus,we give two improved versions of artificial bee colony algorithm(i.e.MPGABC and DFnABC).1).We put forward a new ABC variant,named MPGABC by combining the novel search strategy and new probability model with the basic framework of ABC.To be specific,in the novel search strategy,a parameter P is used to control which search equation to be used,i.e.the original search equation of ABC or the new proposed search equation.In the novel probability model,the selected probability of the good solution is absolutely significantly larger than that of the bad solution,which makes sure the good food sources can attract more onlooker bees to search.Through the comparison of MPGABC and some other state-of-theart ABC variants on two sets of test functions and one set of real world optimization problems,the experimental results show that MPGABC outperforms other competitors.2).We propose a new ABC variant with distance-fitness-based neighbor search mechanism(called DFnABC).To be specific,the employed bee exploits the information of a near-good-neighbor that not only has good fitness value but also is close to its own position to focus on the local exploitation around itself.Moreover,the selectable exploration scope of the employed bee decreases gradually with the process of the evolution and the search direction is guided by a randomly selected leader from the top Q solutions.In addition,each onlooker beefirstly selects a food source position that not only has high quality but also is far away from the current best position to search for the purpose of paying more attention to global exploration among the search space.Furthermore,the best neighbor's information of the selected food source position is used to generate the candidate solution.The simulation experiments were conducted on the benchmark functions,CEC2013 and real life optimization problems,the experimental results show that DFnABC outperforms other competitors.