Hybrid Optimization Algorithm Based On Artificial Bee Colony And Particle Swarm Optimization

Many practical problems can be modeled as function problems in mathematics.In order to maximize the benefits or minimize the cost,it is especially important to solve the function extremum problem.Many traditional methods for solving the extremum of a function are subject to the conditions that can be differentiated and can be derived,while the swarm intelligence algorithm breaks through these limits and can give a satisfactory approximate solution if a certain termination condition is met.The idea of swarm intelligence optimization algorithm comes from bionic zoology.The artificial bee colony algorithm is a swarm intelligent algorithm for simulating honey bee collecting nectar process in recent years.The particle swarm optimization algorithm is a swarm intelligence algorithm that simulates the foraging behavior of birds.The artificial bee colony algorithm has relatively strong global exploration capability,while the particle swarm optimization algorithm has relatively strong local development capability.They all have their own advantages and characteristics,and the combination of different algorithms can complement each other.Therefore,hybrid optimization algorithms have become one of the research hotspots.The research content of this paper mainly includes the following aspects:1.Aiming at the problem that the population diversity of artificial bee colony algorithm is difficult to maintain and the evolution speed is slow,this paper proposed an improved artificial bee colony algorithm based on nonlinear decreasing selection strategy.In the employed bee phase,the algorithm adopts a nonlinear declining selection strategy to improve the diversity of the population and then improve the global exploration capability of the population;In the onlooker bee phase,the global optimal solution is used to guide the search for new solutions to improve the local development ability of the population;In the scout bee phase,a strategy close to the optimal solution is adopted to improve the quality of the generated new solution and accelerate the evolution of the population.2.The loss of population diversity for the standard particle swarm optimization algorithm is faster,which leads to the problem that population evolution is easy to fall into local optimality and evolutionary stagnation.Based on the distribution update rules,this paper introduces the unbalanced weight strategy and the local search method,and then forms a particle swarm optimization algorithm based on unbalanced weights and distribution update rules.The improved strategy can effectively solve the problem that the population is prone to premature convergence and evolutionary stagnation,and accelerate the convergence speed of the population.3.In order to make full use of the global exploration ability of artificial bee colony algorithm and the local development ability of particle swarm optimization algorithm,this paper combines the improved artificial bee colony algorithm based on nonlinear decrement selection strategy and particle swarm optimization algorithm based on unbalanced weight and distribution update rule.A new hybrid optimization algorithm is formed.Finally,the experimental results show that the algorithm has superior comprehensive performance.