High Dimension Artificial Glowworm Swarm Optimization Algorithm Analysis Application And Research

Glowworm swarm optimization (GSO) is a new method of swarm intelligenceraised by K.N.Krishnanad and D.Ghose in2005. This algorithm has alreadyobtained certain success application in noisy text of sensor, multi-modal functionsoptimize, multi-supply oscillator track, noxious gas revelation and simulatingrobots. This algorithm becomes a new research hotspot of computationalintelligence draw our sights on it. Though GSO has strong universal, but there arealso premature, the search accuracy is not high, optimization of high-dimensionalis poor, post-iteration efficiency is not high defects. These shortcomings havegreatly limited the GSO’s application scope. Based on the above each kind ofreasons, this paper will conduct the deep research to the glowworm swarmoptimization algorithm, makes the improvement to the basic GSO algorithm,enhances the GSO algorithm in high dimensions space target function optimizationquestion solution ability, expands the application domain and the application scopeof the extand GSO algorithm.This paper mainly obtains following research results:(1)In the union algorithm own change process manifests the characteristic, proposes the max-min luciferin strategy in GSO. The experiment indicated thatproposed the improvement algorithm can effectively avoid precociousphenomenon, thus obviously improving the optimization global ability.(2)Raised a conception of definite updating search domains. Using this method tomake the position updating glowworm move closer to the best so that toimprove the accuracy and speeding up convergence. Through functions testing,experiment results show that the proposed algorithm has raised the convergencerate and the computational accuracy.(3)Raised a Parallel Glowworm Swarm Optimization with Master-Slave Structure.Through typical functions testing, experiment results show that the proposedalgorithm has better performance in view of decreasing computing time andavoid falling into local optimization.(4)Using optimized method to apply public transit vehicle dispatching and highdimension nonlinear equations system. Experiment results show that theimproved GSO can effectively solve the above problems.