Research Of Artificial Bee Colony Algorithm Based On Multidimensional Search Optimization Strategy

The theory and method of optimization is an important branch of mathematics,which is applied in every aspect of our life.With the increasing complexity of optimization problems in every field of modern society,traditional optimization algorithms have been difficult to cope with.Some modern swarm intelligence optimization algorithms provide a new way for people to solve optimization problems.In general,these algorithms do not need to know the prior information of the objective function,and can well adapt to the optimization of the complex problem conditions.Since the proposed algorithm has been widely concerned and studied by scholars.Artificial bee colony algorithm is a new swarm intelligence optimization algorithm inspired by natural bee foraging behavior.Artificial bee colony algorithm has many advantages,such as few control parameters,easy implementation and strong robustness.It has been applied in more and more engineering fields.But the algorithm itself has some problems such as slow convergence speed,strong global search ability and weak local search ability.In this paper,a new artificial bee colony algorithm(MS-SABC)based on multi-dimensional search optimization strategy and adaptive step length is proposed in this paper.First,the uniform design method is introduced to the initialization stage of the population,and the initial population is evenly distributed in the search space,and the global optimization ability and convergence speed of the algorithm are improved.Secondly,aiming at the problem that the original algorithm searches for the new nectar source only with the blindness of searching in the random direction,the convergence speed is slow and the development ability is weak,a multi-dimensional search and optimization strategy is put forward to search each direction in the field of nectar source and select the best honey source,which can improve the calculation steadily and effectively.Local search and convergence accuracy of the method.Finally,with the unupdated frequency of honey source trial as the basis for the evolution of nectar source,an adaptive step size updating formula is proposed.When the trial is smaller,the nectar source is in the early stage of evolution with a larger step size.When the trial is close to the maximum unupdated number of unupdated limit,the nectar source is close to the convergence.Smaller step size.The local search ability of the algorithm is enhanced,and the search oscillations near the extremum are avoided effectively.Through the simulation of MS-SABC algorithm,standard artificial bee colony algorithm and swarm optimization based on global optimal guidance,the results of 3 multi peak test functions of 6 single peak test functions show that the MS-SABC algorithm is superior to the standard artificial swarm algorithm and the colony based on the global optimal guidance on the performance indexes such as convergence speed and convergence precision.Algorithm.It shows that the improved algorithm effectively improves the convergence speed and convergence accuracy of the algorithm,strengthens the local search ability of the algorithm,and has good stability at the same time.