Perturbation Particle Swarm Optimization Algorithm Based On Local Far-neighbor Differential Enhancement

Due to the lack of interaction among individuals in the search space,the particle swarm algorithm gradually loses the genetic diversity during the iteration process,which causes the algorithm to converge prematurely,falls into a local extremum and cannot escape,and proposes a local distant relative Differential Enhanced Disturbance Particle Swarm Optimization Algorithm.First,in order to guarantee the randomness of initialization population and to avoid the situation of local aggregation caused by random initialization,a semi-homogeneous initialization strategy is proposed to distribute the initialization population in the overall solution space in an overall uniform and locally random manner;In order to expand the population search space and increase the diversity of feasibility solutions,we propose disturbance factors that will be perturbed in the course of speed and location renewal,so that the inertia weights and learning factors will remain within small-scale fluctuations in the overall trend of change;In order to apply the optimization information of the current population to the next generation of populations,use the fitness value of the particle to propose the reconstruction probability,and select the individuals with low fitness values based on the reconstruction probability to rebuild the intermediate population;finally,to further enrich the diversity of the population,Retain the individual genes that have a greater impact on the population gene bank,propose particle irrelevance,and perform differential enhancement operations on distant relatives that have a large difference between the particle irrelevance selection and the excellent individuals.In order to verify the effectiveness of the proposed algorithm,a large number of simulation experiments were conducted with basic PSO and similar PSO improved algorithms.The results show that this algorithm uses a floating change search of population position and velocity,which significantly expands the exploration space for understanding,and through the individual with distant relatives.The differential enhancement operation enables the individuals with high fitness values in the middle population to remain,effectively increasing the diversity of the population,and thus has a strong ability to escape the local extremum,accelerating the particle to approach the global optimum.