# matlab微粒群算法研究与仿真

matlab微粒群算法研究与仿真 摘要：随着现代科学技术的发展，优化问题是工业设计中经常遇到的问题,许多问题最后都可以归结为优化问题。 为了解决各种各样的优化问题,人们提出了许多优化算法。优化问题有两个主要问题：一是要求寻找全局最优点,二是要求有较高的收敛速度。粒子群算法是近年来发展起来的一种新的进化算法，这种属于进化算法的一种，和遗传算法相似，它也是从随机解出发，通过迭代寻找最优解，它也是通过适应度来评价解的品质，但它比遗传算法规则更为简单，它没有遗传算法的“交叉”和“变异”操作，它通过追随当前搜索到的最优值来寻找全局最优。这种算法以其实现容易、精度高、收敛快等优点引起了学术界的重视，并且在解决实际问题中展示了其优越性。
本文的围绕粒子群算法研究及仿真，并利用标准函数来逐一进行仿真，从而来完成对以上算法性能的比较。

Abstract：With the development of modern science and technology, optimize the industrial design problem is frequently encountered , many problems finally all boils down to optimization problem. In order to solve all kinds of optimization problem, people put forward a lot of optimization algorithms. Optimization problem has two main requirements:One is the requirement for the global optimization, second  is to a  higher convergence speed. In recent years particle swarm optimizaiton is a new evolutionary algorithm, which belongs to the evolutionary algorithm, PSO is similar to genetic algorithm to some degree,It is also a random solution based on the iterative search for optimal solution, it evaluates the quality of the solution according to its individual fitness , but it is better than genetic algorithm because its rules is more simpler, it has no genetic algorithm "cross" and "variation" operation, PSO carries on searching the optimal value according to the individuals past experience.
This algorithm is based on its implementation easy, high accuracy, fast convergence ,so it has caused the attention of the researchers, and its  solving practical problems shows its advantages.
Key words: particle swarm optimization    Compression particle   Linear decreasing  adaptive   random weight  simulated annealing

1.1微粒群算法的起源 4
1.1.1群智能算法的提出 4
1.1.2微粒群算法的提出 5
1.2微粒群算法的研究现状 5
1.2.1微粒群算法的发展 5
1.2.2微粒群算法的应用 6
1.3本文的组织 7

2.1微粒群算法概况 8
2.2基本粒子群算法 8
2.2.1算法原理 8
2.2.2算法步骤 9
2.2.3算法的实现 9
2.3带压缩因子的粒子群算法 11
2.3.1算法原理 11
2.4权重改进的粒子群算法 12
2.4.2自适应权重算法 15
2.4.3随机权重法 16
2.5变学习因子的粒子群算法 17
2.5.1同步变化的学习因子 17
2.5.2异步变化的学习因子 18
2.6二阶粒子群算法 20
2.6.1算法原理 20
2.6.2算法步骤 20
2.7二阶振荡粒子群算法 21
2.7.1算法原理 21
2.7.2算法步骤 21
2.8混沌粒子群算法 22
2.8.1算法原理 22
2.9混合粒子群算法 23
2.9.1基于自然选择的算法 23
2.9.2基于杂交的算法 24
2.9.3基于模拟退火的算法 26

3.1基本粒子群算法的出的结论 28
3.2同步变化的学习因子和异步变化的学习因子 30

4.1总结 33
4.2展望 34

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]  ... 下一页  >>

matlab微粒群算法研究与仿真下载如图片无法显示或论文不完整，请联系qq752018766