自适应多目标演化优化算法研究

自适应多目标演化优化算法研究(论文12000字)
摘要：近年来，演化算法在多目标优化问题中得到了广泛的应用，但是多目标演化算法在使用过程中会遇到各种参数设置问题，本文旨在对多目标演化算法中的参数进行自适应改进。主要工作为：（1）详细介绍了演化算法中最典型的遗传算法，对多目标优化的基本概念运用数学公式进行了描述。（2）针对多目标优化算法中使用固定参数（交叉概率、变异概率）的缺点，提出了一种将自适应机制建立在非支配排序算法（Nondominated Sorting Genetic Algorithm II,NSGA-II）基础上的自适应NSGA-II算法（ANSGA-II），利用经典测试函数比较ANSGA-II和NSGA-II的最优解分布情况。（3）对ANSGA-II算法用于车间工作生产优化的案例进行了分析，体现了该算法的实际应用性很强。
关键词：多目标优化；遗传算法；NSGA-II；自适应机制

Multi-objective differential evolution algorithm with elist archive
and crowding entropy-based Problems and its application
Abstract：In recent years, evolutionary algorithms have been widely used in multi-objective optimization problems, but the multi-objective evolutionary algorithm will encounter a variety of parameter setting problems in the use of the process, this paper aims to improve the parameters of the multi-objective evolutionary algorithm adaptive.The main work of this paper is as follows: (1) The most typical genetic algorithm in evolutionary algorithm is introduced in detail. The basic concept of multi-objective optimization is described by mathematical formula. (2) To overcome the disadvantages of using fixed parameters (crossover probability and mutation probability) in the multi-objective optimization algorithm, they proposed an adaptive NSGA-II Algorithm (ANSGA-II) that based on Non-dominated Sorting Algorithm II (Nondominated Sorting Genetic Algorithm II, NSGA-II) and dominated the adaptive mechanism, and compared the distribution of the optimal solution between ANSGA-II Algorithm and NSGA-II Algorithm using the classical test function. (3) The case of ANSGA-II algorithm used in workshop production optimization is analyzed, which shows that the algorithm has strong practical application.
Keywords：Multi-objective optimization；Genetic algorithm；NSGA-II；Adaptive mechanism
 

目录
1绪论    1
1.1研究背景与研究意义    1
1.1.1研究背景    1
1.1.2研究意义    1
1.2国内外研究现状    1
1.2.1遗传算法的发展    1
1.2.2多目标优化遗传算法的发展    2
1.2.3自适应技术的发展    2
1.3论文研究内容    3
1.4论文结构    3
2遗传算法    3
2.1遗传算法基本思想    3
2.2遗传算法流程    4
2.2.1编码    5
2.2.2适应度函数计算    5
2.2.3遗传算法的基本操作    5
2.3单目标函数优化分析    6
3多目标优化的遗传算法    8
3.1多目标优化问题    8
3.2非支配排序遗传算法    8
3.2.1快速非支配排序    9
3.2.2拥挤度    9
3.2.3精英策略    10
3.2.4算法流程    11
3.3自适应机制    12
3.4函数验证与分析    13
3.4.1测试函数    13
3.4.2算法分布性分析    14
3.4.3测试结果分析    15
4自适应多目标遗传算法在车间生产中的应用    17
4.1车间生产加工数据    17
4.2仿真实验    17
5结论与展望    19
5.1结论    19
5.2展望    19
参考文献    20
致谢    22