MATLAB遗传算法的性能仿真研究+文献综述
在一开始需要实现从表现型到基因型的映射即编码工作。但仿照基因编码是一种很复杂的工作，我们通常对其进行简化，如采用二进制编码。等到产生初代种群之后，便按照适者生存、优胜劣汰的生存机制，进行多次迭代演化逐渐趋于近似解。在每一代种群中，计算个体的适应度大小并挑选最优及次优个体，并依据遗传学的遗传算子按照一定的概率进行组合交叉和变异，生成代表新解集的种群。遗传算法的寻优思想导致种群像自然进化一样，新一代种群比前一代更适应于环境，经多次迭代寻优后，种群中最优个体经过解码，即为问题近似最优解。23206
本文在对进化计算及基本的寻优机制进行综述的基础上，给出了遗传算法及其改进算法的基本寻优思想。学习运用MATLAB相关指令及其编译方法，运用MATLAB 完成对遗传算法及其改进算法的设计与仿真，并利用标准测试函数来完成对以上算法性能的比较。
毕业论文关键词： 遗传算法；近似解；最优解；MATLAB；寻优；
The Performance Simulation of Genetic Algorithm
Abstract: Genetic algorithms imitate natural biological evolution based on Darwin's mechanism is a random intelligent mimic biological evolution optimization algorithm . The genetic algorithm is a problem from the target population may be the start of the potential solution set , and after a certain number of population by the gene coding for the inpidual , each with an inpidual chromosome actually physical characteristics . As the main carrier of genetic material , namely a collection of genes on chromosome , which is a combination of the internal representation of a gene , which determines the shape of the external representation of inpiduals .
Since the work is modeled on the gene encoding complexity , we tend to be simplified , such as the use of binary encoding. After the first generation of the population produces , according to the survival of the fittest , survival of the fittest mechanism of evolutionary iterations tended to approximate solutions .
In every generation the population , calculate the size of an inpidual 's fitness and the selection of the optimal and suboptimal inpiduals , and the help of genetics combined genetic operators of crossover and mutation based on certain probability , representative of the population to generate a new solution set.
Idea of genetic algorithm optimization leads to the same population as the natural evolution , new species more adapted to the environment than the previous generation , after several iterative optimization , the best inpidual populations decoded , the problem is the approximate optimal solution.
Based on the basic evolutionary computation and optimization mechanisms are reviewed on gives the basic idea of genetic optimization algorithm and its improved algorithm . Learning to use MATLAB commands and their associated methods of compiling , using MATLAB to complete the genetic algorithm and its improved algorithm design and simulation, and the use of standard test functions to accomplish the above comparison algorithm performance .
Keywords: genetic algorithm; approximate solution; optimal solution; MATLAB; optimization;
目录
摘要 i
Abstract i
目录 iii
1 绪论 1