Self Adaptive Artifcial Bee Colony For Global Numerical Optimization

Optimization problems play an important role in both industrial application fieldsand the scientific research world. During the past decade, we have viewed differentkinds of meta-heuristic algorithms advanced to handle optimization problems. Amongthem, Meta-heuristic based methods,such as simulated annealing (SA), geneticalgorithm (GA), estimation of distribution algorithms (EDA),particle swarmoptimization algorithm (PSO), ant colony optimization (ACO), biogeography basedoptimization (BBO), artificial bee colony (ABC), and differential evolution (DE) maybe one of the most popular methods.Among them, artificial bee colony algorithm, ABC for short,(Karaboga,&Basturk,2008) is a population-based heuristic evolutionary algorithm inspired bythe intelligent foraging behaviour of the honeybee swarm. The artificial bee colony(ABC) algorithm has been used in many practical cases and has demonstrated goodconvergence rate. This algorithm had admitted rapidly in the internationaloptimization field,and widely used in production and life,because they aresimple,convenient, good convergence rate, robust and strong.In this paper, we propose a self adaptive artificial bee colony, called self adaptiveABC, for the global numerical optimization. Specifically, in order to improve theconvergence rates, a new self adaptive perturbation is introduced in the basic ABCalgorithm. In this way, we can balance the exploration and the exploitation of ABC.Additionally, in order to verify the performance of self adaptive ABC,23benchmarkfunctions are employed. Experimental results indicate our approach is effective andefficient. Compared with other algorithms, self adaptive ABC performs better than, orat least comparable to the basic ABC algorithm and other state-of-the-art approachesfrom literature when considering the quality of the solution obtained.Theexperimental results show that the improved ABC algorithm is correct and effectivefor solving optimization problems.