遺傳算法試驗報告

1、實驗五計算智能(2).實驗目的理解遺傳算法的求解思想，掌握遺傳算法的原理，通過運用 MATLAB編程(或相關編程語言)實現(xiàn)遺傳算法，并求解實際問題，對求解結(jié)果進行分析。通過分析結(jié)果了解遺傳算法在求解實際問題的特點和優(yōu)勢。.實驗內(nèi)容編寫一個遺傳算法及其實際應用的程序，能運用遺傳算法求解實際問題。.實驗報告要求(1)簡述實驗原理及方法，并請給出程序設計流程圖。遺傳算法本質(zhì)上是對染色體模式所進行的一系列運算，即通過選擇算子將當前種群中的優(yōu)良模式遺傳到下一代種群中，利用交叉算子進行模式重組，利用變異算子進行模式突變。通過這些遺傳操作，模式逐步向較好的方向進化，最終得到問題的最優(yōu)解基本組成為：a.編碼(

2、產(chǎn)生初始種群)b.適應度函數(shù)c.遺傳算子(selection, crossover, mutation)d.運行參數(shù)求下述二元函數(shù)的最大值：f(x1,x2)=x1八2+x2八2,x1,x2 取值 1-7 的自然數(shù)開始*產(chǎn)生初始種群-計算適應度.ir適亞度是否達到期望值或迭代次數(shù)是否達到最大值結(jié)束適應度是否達到期望值或送代次數(shù)是否達到最大值.否變異(2)源程序清單：#include #include #include /The definiton of user data/#define Cmax 100 /certain maximal value#define Cmin 0 /certain

3、 minimum value#define LENGHT1 3 /the chromosome length of 1st variable#define LENGHT2 3 /the chromosome length of 2nd variable/總?cè)倔w長度#define CHROMLENGTH LENGHT1+LENGHT2const int MaxGeneration = 100; / 最大代數(shù)const int PopSize = 10; /樣本大小const double Pc = 0.6; /交叉概率const double Pm = 0.001; /變異概率/數(shù)據(jù)結(jié)構(gòu)定義 /

4、struct Individualchar chromCHROMLENGTH + 1; /一個個體的染色體double value; / 目標值double fitness; / 適應度;int generation ; / int bestIndex; / int worstIndex; /進化次數(shù)最好個體的下標最壞個體的游標Individual bestIndividual ; / Individual worstIndividual ; / / best individual by now Individual currentBest ;/ Individual population P

5、opSize當前一代中的最好個體 當前一代中的壞個體到目前為止的最好個體;/樣本/void generateInitialPopulation();void generateNextPopulation();void evalutePopulation();long decomdeChromosome(char*, int, int);void calculateObjectValue();void calculateFitnessValue();void findBestAndWorstIndividual();void performEvolution();void selectionOp

6、erator();void crossoverOperator();void mutationOperator();void outputTextReport();/int main()generation = 0;generateInitialPopulation();evalutePopulation();while (generation MaxGeneration) generation+;generateNextPopulation();evalutePopulation();performEvolution();outputTextReport();return 0;/產(chǎn)生第一代樣

7、本/void generateInitialPopulation() int i, j;srand(unsigned)time(NULL);for (i = 0; i PopSize; i+) for (j = 0; j CHROMLENGTH; j+) populationi.chromj = (rand() % 10) 5) ? 0 : 1;populationi.chromCHROMLENGTH = /0;/產(chǎn)生下一代樣本/void generateNextPopulation() selectionOperator();crossoverOperator(); mutationOper

8、ator();/變異算子/void mutationOperator() int i, j;double p;/ bit mutationfor (i = 0; i PopSize; i+) for (j = 0; j CHROMLENGTH; j+) p = rand() % 1000 / 1000.0;if (p Pm) populationi.chromj= (populationi.chromj0) ?1: 0; /交叉算子/void crossoverOperator() int i, j;int indexPopSize;int point, temp;double p;char

9、ch;for (i = 0; i PopSize; i+) indexi = i;for (i = 0; i PopSize; i+) point = rand() %(PopSize - i);temp = indexi;indexi = indexpoint + i;indexpoint + i = temp;for (i = 0; i PopSize - 1; i+=2) p = rand() % 1000 / 1000.0;if (p Pc) point = rand()% (CHROMLENGTH - 1) + 1;for (j = point; j CHROMLENGTH; j+)

10、 ch = populationindexi.chromj;populationindexi.chromj = populationindexi + 1.chromj;populationindexi + 1.chromj = ch; / 選擇算子/void selectionOperator() int i, index;double p, sum = 0.0;double cfitnessPopSize;Individual newpopulationPopSize;for (i = 0; i PopSize; i+) sum += populationi.fitness;for (i =

11、 0; i PopSize; i+) cfitnessi = populationi.fitness / sum;/ calculate cumulative fitnessfor (i = 1; i PopSize; i+) cfitnessi = cfitnessi + cfitnessi - 1;for (i = 0; i cfitnessindex) index+;newpopulationi = populationindex;for (i = 0; i PopSize; i+) populationi = newpopulationi;/依據(jù)某些公式對樣本進行評價/void eva

12、lutePopulation() calculateObjectValue();calculateFitnessValue();findBestAndWorstIndividual();找出到目前為止最好的個體/void findBestAndWorstIndividual() int i;double sum = 0.0;bestIndividual = population0;worstIndividual = population0;for (i = 0; i bestIndividual.fitness) bestindividual = populationi;bestindex =

13、 i; else if (populationi.fitness currentBest.fitness) currentBest = bestindividual;/計算適應度/void calculateFitnessValue() int i;long tempi, temp2;double x1, x2;for (i = 0; i PopSize; i+) tempi = decomdeChromosome(populationi.chrom, 0, LENGHT1);temp2 = decomdeChromosome(populationi.chrom, LENGHT1, LENGH

14、T2);x1 = tempi * tempi;x2 = temp2 * temp2;populationi.fitness = x1+x2;/計算目標值/ 目標函數(shù)為 f(x) = x1* x1 + x2*x2void calculateObjectValue() int i;long temp1, temp2;double x1, x2;for (i = 0; i PopSize; i+) temp1 = decomdeChromosome(populationi.chrom, 0, LENGHT1);temp2 = decomdeChromosome(populationi.chrom,

15、LENGHT1, LENGHT2);x1 = tempi * tempi;x2 = temp2 * temp2;populationi.value = x1 + x2;)/把二進制轉(zhuǎn)化為十進制long decomdeChromosome(char* string, int point, int length) int i;long decimal = 0L;char * pointer;for(i = 0, pointer=string+point; i length;i+,pointer+) decimal += (*pointer 0) currentBest.fitness) curre

16、ntBest = populationbestIndex; else populationworstIndex = currentBest;/打印當前樣本信息/void outputTextReport() int i;double sum;double average;sum = 0.0;for (i = 0; i PopSize; i+) sum += populationi.value;average = sum / PopSize;printf(gen=%d,avg=%f,best=%f,generation,average,currentBest.value);printf( chr

17、omosome=);for( i = 0; i 111111ffen=7.au98.00003U chronDSome-111111Ersn-16,avg=98.0Q000H. bestw38,00000H chromosome=111111ffen=17.auf=98.000000, best =?8.000000 chromosome =111111Efcn=18.aug=98 .000000 best =98 .000001 chromosome =111111Sjen=19,avg=98 ,009000,. best =98 .00B0O0 cJiromosome =111111gen

18、-20,.avgSB rQdBQdOx best-98.000000 chronosone-111111EVsn-21,avSf=98 .000000, hest=?8 .000000 chromDsone=llllllE(en=22aug=98.000000 best=98.09QQQQ chromosome=111111E(en=23,aug=98 .00B000, best=98.000000 cliionosome =111111gen=24,avg=98.003000, best-98.00000U chronosome=111111gfen25avg=98.0000HF best

19、-98 ,00000(1 cht*onosome=llllllgen=26.av$=98,000000, best =9S . 000000 chi*QniiDsone =111111gen=27,aug=98.000000 best=98.000000 chromosome=111111E(en =28 ,aug=98 .00B00B,. best =98 .000000 clii*oriiosoie=llllligren-29,avg=98.00B000 best-98.00003U chromosome-111111E(en=30avSfS .000000 best =98 -00090H chronosone=llllilgen=31,auf=98.000000 best=?S .QQQQQQ chpomosoie=llllllgen=32.avg=98.00 0000best =98 .000000 chromo some =111111g(en=33,au-98.000000,. best =98.00003(4 chrono some =111111gen34avg=98.000000, hest-9