一个简单实用的遗传算法c程序(转载)

      这是一个非常简单的遗传算法源代码,是由Denis Cormier (North Carolina State University)开发的,Sita S.Raghavan (University of North Carolina at Charlotte)修正。代码保证尽可能少,实际上也不必查错。对一特定的应用修正此代码,用户只需改变常数的定义并且定义“评价函数”即可。注意代码的设计是求最大值,其中的目标函数只能取正值;且函数值和个体的适应值之间没有区别。该系统使用比率选择、精华模型、单点杂交和均匀变异。如果用Gaussian变异替换均匀变异,可能得到更好的效果。代码没有任何图形,甚至也没有屏幕输出,主要是保证在平台之间的高可移植性。读者可以从ftp.uncc.edu,目录 coe/evol中的文件prog.c中获得。要求输入的文件应该命名为‘gadata.txt’;系统产生的输出文件为‘galog.txt’。输入的文件由几行组成:数目对应于变量数。且每一行提供次序——对应于变量的上下界。如第一行为第一个变量提供上下界,第二行为第二个变量提供上下界,等等。

/**************************************************************************/
/* This is a simple genetic algorithm implementation where the */
/* evaluation function takes positive values only and the      */
/* fitness of an individual is the same as the value of the    */
/* objective function                                          */
/**************************************************************************/

#include <stdio.h>
#include <stdlib.h>
#include <math.h>

/* Change any of these parameters to match your needs */

#define POPSIZE 50               /* population size */
#define MAXGENS 1000             /* max. number of generations */
#define NVARS 3                  /* no. of problem variables */
#define PXOVER 0.8               /* probability of crossover */
#define PMUTATION 0.15           /* probability of mutation */
#define TRUE 1
#define FALSE 0

int generation;                  /* current generation no. */
int cur_best;                    /* best individual */
FILE *galog;                     /* an output file */

struct genotype /* genotype (GT), a member of the population */
{
  double gene[NVARS];        /* a string of variables */
  double fitness;            /* GT’s fitness */
  double upper[NVARS];       /* GT’s variables upper bound */
  double lower[NVARS];       /* GT’s variables lower bound */
  double rfitness;           /* relative fitness */
  double cfitness;           /* cumulative fitness */
};

struct genotype population[POPSIZE+1];    /* population */
struct genotype newpopulation[POPSIZE+1]; /* new population; */
                                          /* replaces the */
                                          /* old generation */

/* Declaration of procedures used by this genetic algorithm */

void initialize(void);
double randval(double, double);
void evaluate(void);
void keep_the_best(void);
void elitist(void);
void select(void);
void crossover(void);
void Xover(int,int);
void swap(double *, double *);
void mutate(void);
void report(void);

/***************************************************************/
/* Initialization function: Initializes the values of genes    */
/* within the variables bounds. It also initializes (to zero)  */
/* all fitness values for each member of the population. It    */
/* reads upper and lower bounds of each variable from the      */
/* input file `gadata.txt’. It randomly generates values       */
/* between these bounds for each gene of each genotype in the  */
/* population. The format of the input file `gadata.txt’ is    */
/* var1_lower_bound var1_upper bound                           */
/* var2_lower_bound var2_upper bound …                       */
/***************************************************************/

void initialize(void)
{
FILE *infile;
int i, j;
double lbound, ubound;

if ((infile = fopen(“gadata.txt”,”r”))==NULL)
      {
      fprintf(galog,”/nCannot open input file!/n”);
      exit(1);
      }

/* initialize variables within the bounds */

for (i = 0; i < NVARS; i++)
      {
      fscanf(infile, “%lf”,&lbound);
      fscanf(infile, “%lf”,&ubound);

      for (j = 0; j < POPSIZE; j++)
           {
           population[j].fitness = 0;
           population[j].rfitness = 0;
           population[j].cfitness = 0;
           population[j].lower[i] = lbound;
           population[j].upper[i]= ubound;
           population[j].gene[i] = randval(population[j].lower[i],
                                   population[j].upper[i]);
           }
      }

fclose(infile);
}

/***********************************************************/
/* Random value generator: Generates a value within bounds */
/***********************************************************/

double randval(double low, double high)
{
double val;
val = ((double)(rand()%1000)/1000.0)*(high – low) + low;
return(val);
}

/*************************************************************/
/* Evaluation function: This takes a user defined function.  */
/* Each time this is changed, the code has to be recompiled. */
/* The current function is:  x[1]^2-x[1]*x[2]+x[3]           */
/*************************************************************/

void evaluate(void)
{
int mem;
int i;
double x[NVARS+1];

for (mem = 0; mem < POPSIZE; mem++)
      {
      for (i = 0; i < NVARS; i++)
            x[i+1] = population[mem].gene[i];
     
      population[mem].fitness = (x[1]*x[1]) – (x[1]*x[2]) + x[3];
      }
}

/***************************************************************/
/* Keep_the_best function: This function keeps track of the    */
/* best member of the population. Note that the last entry in  */
/* the array Population holds a copy of the best individual    */
/***************************************************************/

void keep_the_best()
{
int mem;
int i;
cur_best = 0; /* stores the index of the best individual */

for (mem = 0; mem < POPSIZE; mem++)
      {
      if (population[mem].fitness > population[POPSIZE].fitness)
            {
            cur_best = mem;
            population[POPSIZE].fitness = population[mem].fitness;
            }
      }
/* once the best member in the population is found, copy the genes */
for (i = 0; i < NVARS; i++)
      population[POPSIZE].gene[i] = population[cur_best].gene[i];
}

/****************************************************************/
/* Elitist function: The best member of the previous generation */
/* is stored as the last in the array. If the best member of    */
/* the current generation is worse then the best member of the  */
/* previous generation, the latter one would replace the worst  */
/* member of the current population                             */
/****************************************************************/

void elitist()
{
int i;
double best, worst;             /* best and worst fitness values */
int best_mem, worst_mem; /* indexes of the best and worst member */

best = population[0].fitness;
worst = population[0].fitness;
for (i = 0; i < POPSIZE – 1; ++i)
      {
      if(population[i].fitness > population[i+1].fitness)
            {     
            if (population[i].fitness >= best)
                  {
                  best = population[i].fitness;
                  best_mem = i;
                  }
            if (population[i+1].fitness <= worst)
                  {
                  worst = population[i+1].fitness;
                  worst_mem = i + 1;
                  }
            }
      else
            {
            if (population[i].fitness <= worst)
                  {
                  worst = population[i].fitness;
                  worst_mem = i;
                  }
            if (population[i+1].fitness >= best)
                  {
                  best = population[i+1].fitness;
                  best_mem = i + 1;
                  }
            }
      }
/* if best individual from the new population is better than */
/* the best individual from the previous population, then    */
/* copy the best from the new population; else replace the   */
/* worst individual from the current population with the     */
/* best one from the previous generation                     */

if (best >= population[POPSIZE].fitness)
    {
    for (i = 0; i < NVARS; i++)
       population[POPSIZE].gene[i] = population[best_mem].gene[i];
    population[POPSIZE].fitness = population[best_mem].fitness;
    }
else
    {
    for (i = 0; i < NVARS; i++)
       population[worst_mem].gene[i] = population[POPSIZE].gene[i];
    population[worst_mem].fitness = population[POPSIZE].fitness;
    }
}
/**************************************************************/
/* Selection function: Standard proportional selection for    */
/* maximization problems incorporating elitist model – makes  */
/* sure that the best member survives                         */
/**************************************************************/

void select(void)
{
int mem, i, j, k;
double sum = 0;
double p;

/* find total fitness of the population */
for (mem = 0; mem < POPSIZE; mem++)
      {
      sum += population[mem].fitness;
      }

/* calculate relative fitness */
for (mem = 0; mem < POPSIZE; mem++)
      {
      population[mem].rfitness =  population[mem].fitness/sum;
      }
population[0].cfitness = population[0].rfitness;

/* calculate cumulative fitness */
for (mem = 1; mem < POPSIZE; mem++)
      {
      population[mem].cfitness =  population[mem-1].cfitness +      
                          population[mem].rfitness;
      }

/* finally select survivors using cumulative fitness. */

for (i = 0; i < POPSIZE; i++)
      {
      p = rand()%1000/1000.0;
      if (p < population[0].cfitness)
            newpopulation[i] = population[0];     
      else
            {
            for (j = 0; j < POPSIZE;j++)     
                  if (p >= population[j].cfitness &&
                              p<population[j+1].cfitness)
                        newpopulation[i] = population[j+1];
            }
      }
/* once a new population is created, copy it back */

for (i = 0; i < POPSIZE; i++)
      population[i] = newpopulation[i];     
}

/***************************************************************/
/* Crossover selection: selects two parents that take part in  */
/* the crossover. Implements a single point crossover          */
/***************************************************************/

void crossover(void)
{
int i, mem, one;
int first  =  0; /* count of the number of members chosen */
double x;

for (mem = 0; mem < POPSIZE; ++mem)
      {
      x = rand()%1000/1000.0;
      if (x < PXOVER)
            {
            ++first;
            if (first % 2 == 0)
                  Xover(one, mem);
            else
                  one = mem;
            }
      }
}
/**************************************************************/
/* Crossover: performs crossover of the two selected parents. */
/**************************************************************/

void Xover(int one, int two)
{
int i;
int point; /* crossover point */

/* select crossover point */
if(NVARS > 1)
   {
   if(NVARS == 2)
         point = 1;
   else
         point = (rand() % (NVARS – 1)) + 1;

   for (i = 0; i < point; i++)
        swap(&population[one].gene[i], &population[two].gene[i]);

   }
}

/*************************************************************/
/* Swap: A swap procedure that helps in swapping 2 variables */
/*************************************************************/

void swap(double *x, double *y)
{
double temp;

temp = *x;
*x = *y;
*y = temp;

}

/**************************************************************/
/* Mutation: Random uniform mutation. A variable selected for */
/* mutation is replaced by a random value between lower and   */
/* upper bounds of this variable                              */
/**************************************************************/

void mutate(void)
{
int i, j;
double lbound, hbound;
double x;

for (i = 0; i < POPSIZE; i++)
      for (j = 0; j < NVARS; j++)
            {
            x = rand()%1000/1000.0;
            if (x < PMUTATION)
                  {
                  /* find the bounds on the variable to be mutated */
                  lbound = population[i].lower[j];
                  hbound = population[i].upper[j]; 
                  population[i].gene[j] = randval(lbound, hbound);
                  }
            }
}

/***************************************************************/
/* Report function: Reports progress of the simulation. Data   */
/* dumped into the  output file are separated by commas        */
/***************************************************************/

void report(void)
{
int i;
double best_val;            /* best population fitness */
double avg;                 /* avg population fitness */
double stddev;              /* std. deviation of population fitness */
double sum_square;          /* sum of square for std. calc */
double square_sum;          /* square of sum for std. calc */
double sum;                 /* total population fitness */

sum = 0.0;
sum_square = 0.0;

for (i = 0; i < POPSIZE; i++)
      {
      sum += population[i].fitness;
      sum_square += population[i].fitness * population[i].fitness;
      }

avg = sum/(double)POPSIZE;
square_sum = avg * avg * POPSIZE;
stddev = sqrt((sum_square – square_sum)/(POPSIZE – 1));
best_val = population[POPSIZE].fitness;

fprintf(galog, “/n%5d,      %6.3f, %6.3f, %6.3f /n/n”, generation,
                                      best_val, avg, stddev);
}

/**************************************************************/
/* Main function: Each generation involves selecting the best */
/* members, performing crossover & mutation and then          */
/* evaluating the resulting population, until the terminating */
/* condition is satisfied                                     */
/**************************************************************/

void main(void)
{
int i;

if ((galog = fopen(“galog.txt”,”w”))==NULL)
      {
      exit(1);
      }
generation = 0;

fprintf(galog, “/n generation  best  average  standard /n”);
fprintf(galog, ” number      value fitness  deviation /n”);

initialize();
evaluate();
keep_the_best();
while(generation<MAXGENS)
      {
      generation++;
      select();
      crossover();
      mutate();
      report();
      evaluate();
      elitist();
      }
fprintf(galog,”/n/n Simulation completed/n”);
fprintf(galog,”/n Best member: /n”);

for (i = 0; i < NVARS; i++)
   {
   fprintf (galog,”/n var(%d) = %3.3f”,i,population[POPSIZE].gene[i]);
   }
fprintf(galog,”/n/n Best fitness = %3.3f”,population[POPSIZE].fitness);
fclose(galog);
printf(“Success/n”);
}
/***************************************************************/

    原文作者:遗传算法
    原文地址: https://blog.csdn.net/AmiRural/article/details/416333
    本文转自网络文章,转载此文章仅为分享知识,如有侵权,请联系博主进行删除。
点赞