Improved Artificial Bee Colony Algorithm And Its Application

Swarm Intelligence Algorithm is a kind of stochastic optimization algorithm based onbehavior of biological swarm, which provides a new method to solve global optimizationproblem existed in the fields of computer science, management science, controlengineering and so on. So it becomes a focus among researchers in a long term. ArtificialBee Colony Algorithm is a meta-heuristic intelligence algorithm which is inspired by thehoney bees’ behavior in the nature. The algorithm has a unique role assignmentmechanism, and each agent in the algorithm is real bees, which quickly finds the solutionof optimization problems through the changing role of assistance, share information toquickly find the solution of optimization problems. And because of its few controlparameters, easy to implement and having a high convergence rate, there is the concern ofmany scholars at home and abroad. However, current research and application of artificialbee colony algorithm is still in its in fancy, and there are many issues to be resolved.Firstly, the basic artificial bee colony algorithm is described in detail and in-depthanalysis of strengths and advantages and disadvantages of the algorithm are described inthe paper. According to the question that the bee colony algorithm is easy to fall into localoptima, particle swarm optimization algorithm is introduced and it gives simulation inimproved artificial bee colony algorithm. The introduction of particle swarm optimizationfor the bee colony algorithm is easy to fall into local optima, and improved artificial beecolony algorithm simulation. The experimental results show that the improved artificialbee colony algorithm is better than the basic swarm algorithm, whether on the searchcapabilities or convergence speed and accuracy.Secondly, in order to maintain the diversity of the population during evolution,adaptive proportional selection strategy replaces the traditional proportional selectionstrategy, and experimental results show that the adaptive selection strategy in artificial beecolony algorithm has better search ability and convergence speed.Finally, the reactive power of power system optimization problem is introduced simply.According to nonlinear, discrete, uncertainty, and the characteristics of dynamic andmulti-goal and multi-variable control parameters, and more the combination of constraints,mixed nonlinear optimization in reactive power optimization problem, the correspondingobjective function is optimized. We make use of the improved artificial bee colonyalgorithm to simulate IEEE14node system. And experimental results show that theimproved artificial bee colony algorithm solving reactive power optimization problem hasmore applicability than the basic swarm algorithm.