基于DE和NSGA-Ⅱ的进化多目标优化算法及应用-电路与系统专业论文.docxVIP

摘 要 摘 要 I 摘 要 从上世纪 60 年代开始，多目标问题吸引了越来越多不同背景研究人员的注意 力。多目标优化问题在现实世界中的普遍存在性以及求解的困难性，促使人们付 出了很多精力来寻求解决的方法。 进化多目标优化(EMO)算法是近年来发展起来的一种利用进化计算方法求解 多目标优化问题的新的优化算法，而随着进化算法的提出和研究的不断深入，使 得其作为解决多目标优化问题的新方法受到了相当的关注，并逐步发展成为有效 解决多目标优化问题的重要技术。 本文首先对国内外进化多目标领域的研究现状和研究方法进行了总结，接着 详细介绍了 DE 算法和 NSGA-II 算法的原理和具体流程，并据此提出了一种基于 DE 和 NSGA-II 的进化多目标优化算法。该算法在 DE 算法和 NSGA-II 算法的理论 基础上，保留了 NSGA-II 的快速非支配排序和种群多样性保持策略，利用差分方 式取代 NSGA-II 中的模拟二进制交叉和多项式变异操作，从而产生子代个体。在 第三章和第四章中，我们把本文提出的算法应用到多目标聚类及图像分割任务上， 通过仿真实验验证了本文算法的有效性。 关键词：多目标优化 进化算法 聚类 聚类分析 图像分割 Abstr Abstract III Abstract Since the 1960s, multi-objective optimization problems have attracted more and more attention of many researchers with different background.In the real world,with the prevalence of solving the difficulty for Multi-objective optimization problem,it allows people to pay a lot of effort to seek a solution. Evolutionary muti-objective optimization algorithem (EMOA) is a new method developed recently, which employs eolutionary computation to solve multi-objective optimization problem. With the development of Nature-inspired computation research, as a new algorithm used to solve the multio-bjective problems, Nature-inspired computation algorithm has been attracting more attention, and is becoming one of the most important methods to solve the multiobjective problems. At first,the research status and methods of evlutionary multi-objective optimization were summarized. Then , it give a detail introduction on a evolutionary multi-objective optimization algorithm based on DE and NSGA-II, and accordingly proposed a evolutionary multi-objective optimization algorithm based on DE and NSGA-II. The algorithm is based on the theory of the DE and NSGA-II.The proposed algorithm retained the fast non-dominated sorting approach and the diversity preservation method of NSGA-II ,and it made use of the differential way for obtaining the offspring in stead of the simulated binary crossover and polynomial mutation. In the chapter 3 and 4, the algorithm