Research And Application Of Swarm Intelligence Optimization Algorithm

There are various complex optimization problems in many fields,such as industrial engineering and science and technology,many of which are difficult to solve by traditional optimization algorithms.However,intelligent optimization algorithms have a strong adaptability and effectiveness to solve these complex optimization problems.It has no special requirements for the analytical properties of the objective function and the selection of the initial point of the algorithm,and the algorithm has a fast convergence speed,so it becomes an effective method to solve complex optimization problems.This paper mainly studies two kinds of swarm intelligence optimization algorithms.One is a particle swarm optimization algorithm that simulates the foraging behavior of bird swarms,and the other is an artificial bee colony algorithm that mimics the behavior of honey swarms.In the paper,the convergence effect of the algorithm is improved by introducing advanced evolutionary strategies,which are also applied to solve practical problems.The main content is summarized as follows:1.A dimensional evolutionary particle swarm optimization algorithm based on Gaussian and chaos enhancement strategies is proposed.By each particle is evolved in turn according to the fitness of the dimension to ensure that the entire particle is not degraded.To ensure population diversity,a Gaussian chaotic local contraction strategy is implemented near the optimal solution when the particles gather,helping the particles to jump out of the local optimal.The result of Benchmark function test shows that the optimization ability of the algorithm is significantly improved.The algorithm is introduced into the parameter tuning problem of PID control system,and the result indicates that the accuracy of the system control is improved and the control effect is fine.2.A chaotic particle swarm algorithm based on a variable-scale search strategy is proposed.The process of variable scale search is used to realize the rapid location of global optimal position and the fine search of local position.In order to avoid falling into local optimum,chaotic initialization and chaotic mutation are used to enhance population diversity.Through the time-varying acceleration coefficient and time-varying inertia weight balance the exploration and development ability.Comparative experiments show that the convergence speed and accuracy of the algorithm are significantly improved.Based on the algorithm,the location problem of logistics distribution center realizes the planning of the optimal distribution center and the optimal distribution of goods in the logistics network,and optimizes the logistics network structure.3.A hybrid optimization algorithm combining genetic algorithm with artificial bee colony algorithm is proposed.In the artificial bee colony algorithm,genetic algorithm selection,crossover,and mutation operations are introduced.The genetic operation performs a global and fast search,and bee colony algorithm is responsible for local accurate exploration,so that the exploration and development abilities of the algorithm are balanced.In order to verify the performance of the algorithm,the algorithm is applied to solve nonlinear optimization problem without constraint.The result shows that the proposed algorithm has obvious advantages over other methods.4.A quantum-derived artificial bee colony algorithm with adaptive grouping is proposed.The dynamic grouping of the population based on the average objective function value is realized to improve the local search ability,the quantum bit probability amplitude is used to encode the individual to improve the diversity of the population,and the bee colony mutation is implemented through the quantum non-gate and help it escape the local optimum.The comparative experiment of Benchmar function shows that the algorithm is feasible and effective.The two-dimensional maximum entropy image segmentation experiment based on the algorithm shows that the algorithm can better retain some details of the image and improve the segmentation effect of the image.