Font Size: a A A

The Coupling Algorithms Based On Glowworm Swarm Optimization And Bacterial Foraging Optimization

Posted on:2020-05-07Degree:MasterType:Thesis
Country:ChinaCandidate:Y C WangFull Text:PDF
GTID:2428330590956614Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Intelligent computing can solve many engineering problems effectively by simulating natural phenomena such as self-organization,self-adaptation and self-learning.However,any algorithm can only be applied to a certain type of problem according to no free lunch theorem.Therefore,the coupling between different algorithms will be an effective way to improve the performance of the algorithm furtherly.In this paper,Glowworm Swarm algorithm and Bacterial Foraging algorithm are choosen to construct efficient coupling methods for different problems such as single objective,multi-objective and many-objective.The main research contents are as follows:This paper taking the intelligent algorithm research of glowworm swarm optimization(GSO)and bacterial foraging optimization(BFO)algorithm as background.We design an adaptive step size adjustment strategy and random dimensional-by-dimension position renewal strategy against the disadvantages of fixed step size and directional position update.The two-point crossover replication operators are designed for the bacterial foraging algorithm.Finally,we designed the coupling rule based on the parallel evolution of individuals for the original algorithm,and replaced the first two individuals in the original population with the optimal individuals obtained by the two algorithms.Simulation results show that the algorithm proposed in this paper improves the performance of the algorithm furtherly.The multi-objective optimization problems are different from the single-objective optimization problems,and there is a Pareto relationship between its individuals.Therefore,a new Pareto rule is designed in the trend operator according to the Pareto relationship between individuals.In order to maintain the diversity of the population and improve the convergence speed of the algorithm,the replication operator based on GSO crossover and the dispersing operator based on crossover mutation are proposed.In addition,we also introduced fast non-dominant sorting and crowding distance methods to sort and select individuals.Through experimental analysis,the MGSO-BFO algorithm has more advantages than other algorithms.The Many-objective optimization problems are different from the multi-objective optimization problems.Therefore,on the basis of the multi-objective coupling algorithm,we add gaussian perturbation to the GSO position updating formula to solve the diversity problem of the many-objective optimization problems.Aiming at the dominating relation in the many-objective optimization problems,a coupling rule based on double external solution set is designed.What ‘s more,we introduce the BFE method for the internal update of the external solution set.Finally,we have carried out experiments on DTLZ and WFG functions,and the results show that the algorithm can improve the quality of the solution,increase the robustness of the algorithm,and can well solve the problem of many-objective optimization problems.
Keywords/Search Tags:Single objective optimization problems, Multi-objective optimization problems, Many-objective optimization problems, Coupling strategy, Glowworm Swarm Optimization, Bacterial Foraging Optimization
PDF Full Text Request
Related items