Optimization Algorithm Based On Chaotic Mixing

In recent year,with the deeper research of chaos theory,the application of chaostheory has become one of important issues and forefront project. In this dissertation,research works focus on chaos theory and its combined optimization,the primaryendeavor and contribution of this dissertation are as follows:First, combine with the research actuality and development tide of chaos theory,makes a systematic exposition for the development of the chaos theory, the definitionand characteristic of chaos. The advantage and disadvantages of traditionaloptimization algorithms are analyses. Using the properties of stochastic, ergodicityand regularity of chaos, the chaos optimization has became a new technology in thenonlinear systems.Second, The chaos system's outstanding character,ergodicity,is noticed. Chaoticvariables can approach to value in its defining range,so they can find the globaloptimal solution. Based on this feature,chaotic optimization algorithm is presented. Inthe simulation with chaotic optimization algorithm,the defect is found,and the natureof this method is discovered after carefully thinking. Chaotic variables have notuniform distribution. Therefore it is uncertain that the optimal solution can be found.Third, in order to solve the question,the author combined the simplex methodwith the chaotic optimization, a hybrid optimization strategy is proposed to solve theoptimization problems of complex functions. The hybrid optimization strategy utilizesthe chaos to find the global optimal solution, at the same time the initial points thatare needed in simplex method are formed, then use simplex method to perform localoptimization, which avoid the weakness of single method. This algorithm is applied totest functions optimization problem,and the simulation results illustrate that thismethod is simple and easy to implement,and is effective for the optimizationproblem.Forth, Genetic algorithm is a global optimization algorithm, and a lot ofachievements have been made in many fields. But in practice, premature convergenceoften happens, which greatly restrict the application of Genetic algorithm. Based onthe properties of simplex method and chaos optimization, a new chaos geneticsimplex optimization algorithm is presented. The algorithm has better stability ofnumerical value than genetic algorithm, and is an effective algorithm.As an implement, chaos has many applications. The chaos's history is short. Inorder to make the chaos had better performance and more abroad applications, a lot ofquestions should be discussed.