Research On Improved Artificial Bee Colony Algorithm For Continuous Optimization Problem

Optimization problems can be divided into discrete optimization problems and continuous optimization problems based on the type of variables.In recent years,some scholars have proposed many intelligent algorithms to solve optimization problems.Aiming at the problem of continuous optimization,the algorithm of artificial swarm optimization is slow and easy to get into local optimal defect.OPIABC(Artificial bee colony with one-position inheritance)has made some improvements.In order to further improve human colony algorithm in the performance of the continuous optimization problems,this article on the OPIABC algorithm,based on the investigation of the guidance of the group evolution method of bees foraging bees in the investigation phase by introducing a parameter to adjust the search step length,and combined with group known some information about groups to adjust the search direction,artificial colony algorithm was put forward,in order to promote them as soon as possible find effective honey,corresponding to the algorithm is to ascend algorithm convergence speed.Then using two sets of benchmark test functions,will improve the new algorithm is compared with the original algorithm and related algorithm experiment,by comparing the experimental data analysis,proves that the improved algorithm in continuous argument function optimization problems,both in convergence speed and function,the optimal value shows superior performance.