基于多方法融合的进化算法研究-电路与系统专业论文.docx

优秀毕业论文精品参考文献资料ABSTRACTAs the performance rising, Evolutionary algorithm (EA) has become more and more important in optimization domain. Compared with classical optimization meth- ods, EAs require less parameter settings and less pre-knowledge. For the engineering applications with many constrains, EAs can work without complicated uniformization procedure. Besides, EAs have shown especially better performance on multi-modal and linked problems.Although EAs have achieved significant performance previously, several aspects of EAs still need to be improved. Generally speaking, there are three demerits of EAs as follows: (1) their universality, which means the ability of solving diverse kinds prob- lems, is still relatively poor; (2) their scalability, which is mainly the capability of solv- ing high dimensional problems, needs to be improved; (3) their application in engineer- ing optimization problems is still limited. For the first two drawbacks, we design a more robust and effective EA framework by using multi-method ensemble idea. Fur- thermore, the proposed EAs are applied to solve real world engineering applications and scientific problems, and the performance mostly surpass those of the classical EAs.This dissertation focuses on designing more effective and efficient multi-method ensemble based EAs, and then applying them to difficult engineering and scientific optimization problems. The important findings are as follows:In order to improve the scalability of Estimation of Distribution Algorithm (EDA), we design a new self-adaptive mixed distribution based sampling operator and then, propose a self-adaptive Mixed distribution based Uni-variate EDA (MUEDA). The advantages of MUEDA compared with the classical EDAs and state-of-the- art EAs are verified on the function optimization experiment scaling from 30 to 1500 dimension.In order to improve the universality of Particle Swarm Optimization (PSO), we propose a self-adaptive learning framework to extract strengthes