Improved And Applied Research Of The Glowworm Swarm Optimization Algorithm

Glowworm Swarm Optimization algorithm (GSO) is a novel swarmoptimization algorithm which is proposed by India scholars K.N.Krishnanad andD.Ghose in2005. So far, the glowworm swarm optimization algorithm has beensuccessfully application in the multi-modal function optimization, multi-sourceissue tracking, multi-source location problem, the problem of harmful gas leaklocation and combinatorial optimization, and show good performance. However,the glowworm swarm optimization algorithm also has its own shortcomings,there are many places need to improve and perfect. For example: into a localoptimum, slow convergence in the late iteration, the solution accuracy isn’t highand lack of mathematics theory foundation etc. In the application, the glowwormswarm optimization algorithm compared to the particle swarm optimization(PSO) algorithm, ant colony optimization (ACO) algorithm, the glowworm swarmoptimization algorithm is still relatively narrow range of applications, need to befurther widened.Based on this, in this paper, aim at the disadvantage of the basic glowwormswarm optimization algorithm has bad effect in high dimensional functionoptimization, we put forward use the complex method guidance the swarmoptimization algorithm for solving high dimensional function optimizationproblem, the experimental results show that the glowworm swarm optimizationalgorithm based on the complex method guidance is effective to solvethe shortcoming which the basic glowworm swarm optimization algorithm haspoor quality in the high dimension space optimization, make the optimizeperformance in the high dimensional function optimization is greatly improved. Inaddition, in according to the glowworm swarm optimization algorithm has lowsearch accuracy and low optimal efficiency in the latter part of iterative, aglowworm swarm optimization algorithm with local search operator is proposed.This algorithm introduce the local search operator in the basic glowworm swarmoptimization algorithm, the result of simulation shows that the proposed algorithmhas improved greatly in the search precision and the efficiency of optimization. Finally, we use the complex method guidance the swarm optimization algorithmapplied to the problem of the solving nonlinear equations, through the numericalexperiments show that the improved algorithm in solving nonlinear equations has avery good solving effect.