Enhanced Heuristic Dragonfly Optimization Algorithm And Its Application

Dragonfly algorithm(DA)is a new swarm intelligence optimization algorithm proposed by Mirjalili in 2016.It is inspired by the static and dynamic swarm intelligence behaviors of dragonflies in nature.It mainly simulates the collision avoidance,clustering,gathering,foraging and enemy avoidance of dragonflies to conduct global and local search.DA has the characteristics of simple and clear structure and strong searching ability.With the deepening of the research,the researchers found that the algorithm has some shortcomings such as low precision,slow convergence speed and easy to fall into local minimum.This thesis analyzes and improves the shortcomings of dragonfly algorithm,proposes three improved dragonfly algorithms,and applies the improved algorithms to practical optimization problems,in order to further improve the theoretical basis of DA algorithm and broaden its application scope.(1)To improve the overall optimization ability of the algorithm,improve the natural mechanism of the algorithm and avoid falling into the local optimal,a kind of Dragonfly algorithm(Wind driven Dragonfly algorithm,WDDA)based on the wind driven optimization algorithm was proposed by introducing the position updating formula of the wind driven optimization algorithm.WDDA is applied to 23 different types of benchmark function testing and engineering optimization problems,and compared with 5 meta-heuristic algorithms,experimental results show that WDDA has strong competitiveness in function optimization and engineering examples.(2)To balance the exploration capability and mining capability of the algorithm,a discrete Dragonfly algorithm(Enhanced Discrete Dragonfly Algorithm,EDDA)based on enhancement is proposed by introducing global discrete and local discrete methods.EDDA is used to solve the difficult coloring problem of four-color map.The experimental results show that EDDA is feasible and provides a new method to solve planar graph coloring problem by optimizing the coloring of 8 real maps and comparing with other 4 intelligent algorithms.(3)to expand the application scope of basic DA algorithm,a hybrid Dragonfly algorithm(Hybrid Dragonfly algorithm,HDA)was proposed to solve the location problem of multiple distribution centers by integrating the cooling process of simulated annealing in the middle and late iterations of Dragonfly algorithm.Through a practical case analysis and comparison with other 5algorithms,the simulation results show that the algorithm gives an optimal solution to solve the location of multiple distribution centers.