Improvement And Application Of Selfish Herd Optimizer

Selfish herd optimizer is a new swarm intelligence optimization algorithm which simulates the behavior of prey-predator relationship in nature.The algorithm realizes the search process by simulating the behavior of prey avoiding predation risk and hunter hunting behavior in nature,and balances global search and local search by controlling the number of two groups of individuals.The algorithm has the characteristics of high accuracy and strong robustness.However,with the deepening of research,it is found that selfish swarm optimization algorithm has some shortcomings,such as slow optimization speed,low precision and easy to fall into local optimum.In view of the shortcomings of selfish herd optimizer,three improved versions of selfish herd optimizer are proposed and applied to some practical engineering problems.The purpose is to improve the overall optimization performance of the algorithm and expand its application fields.This paper mainly includes three aspects:(1)(1)A Elite opposition-based selfish herd optimizer is proposed.The elite opposition-based learning strategy is introduced into the selfish beast swarm optimization algorithm.According to the characteristics of the selfish beast swarm optimization algorithm,elite reverse learning is carried out for individuals in the prey group,which enlarges the search space of the population,enhances the diversity of the population and avoids the algorithm falling into local optimum.The experimental results show that the Elite opposition-based Selfish herd optimizer has better convergence speed and optimization accuracy in solving function optimization and engineering case problems.(2)A discrete version of selfish herd optimizer algorithm is proposed.According to the attributes function of hunter and prey individuals,a discrete location updating mechanism is introduced,and a differential selection strategy is introduced to select the prey individuals,which improves the optimization speed of the algorithm.The coloring problem of six planar graphs and two example graphs given randomly is tested.The experimental results show that the discrete version of the selfish swarm optimization algorithm can obtain effective solutions.(3)A new selfish herd optimizer algorithm is proposed,which introduces the strategy of eliminating the leader mechanism to prevent individuals from falling into local optimum and improve the global search ability of the algorithm.The algorithm is applied to the deployment of monitoring area nodes in wireless sensor networks.The experimental results show that compared with other intelligent algorithms,the convergence speed is faster.It has high precision and good stability.