The Studyabout Swarm Intelligence Optimizationalgorithm Based On Chaotic System

In this thesis, we focus on the key technology about swarm intelligence algorithm which basedon chaotic image, and its’ application in optimization problems.When solving complexoptimization problems, swarm intelligence algorithm has been appearing poor convergence, toolong search time phenomenon, and it is easy to stagnate in local optimal solution. With thecomprehensive consideration of the logistic, tent, henon chaotic image features, the improvementmeasures are put forward to increase their performance. Firstly, a new2d hyperchaos system isproposed, whose kinetic equation contains non-linear terms, strengthening hyper-chaotic systemcomplexity. Its’ phase diagram, lyapunov exponents and bifurcation diagram are presented andstudied, and the binarization algorithm of chaotic image is given.The analyses shows that the newhyperchaos system has better pseudo-random sequence performance.Secondly, a new chaotic antcolony algorithm (CACO) is put forward, and the formula which can map chaotic system intointelligent algorithm is presented and studied, the new algorithm solves vehicle routing problem(VRP) which is multi-objective optimization problem. Thirdly, a new chaotic particle swarmoptimization algorithm which based on improved2d henon image is proposed and studied, and itswork mechanism is given. The benchmark standard test problems are selected as simulationexperiment. Finally, by computer simulation, we make an overall illustration about the above newmethods. It effectively enhanced the swarm intelligence algorithm performance, increased theirability in solving complex optimization problems.