public Population ExecuteGA(Population pop)
{
//criar a população
Population newPop = new Population();
List <Individual> popTemp = new List <Individual>();
//atribuir individuos a população temporaria
for (int i = 0; i < ConfigurationGA.sizePopulation; i++)
{
popTemp.Add(pop.GetPopulation()[i]);
}
//verificar se utilizara elitismo
Individual[] indElitsm = new Individual[ConfigurationGA.sizeElitism];
if (ConfigurationGA.elitism)
{
pop.Ordenate();
for (int i = 0; i < ConfigurationGA.sizeElitism; i++)
{
indElitsm[i] = pop.GetPopulation()[i];
}
}
for (int i = 0; i < (ConfigurationGA.sizePopulation / 2); i++)
{
//seleção dos pais
Individual father = selection(pop);
Individual mother = selection(pop);
//cruzamento dos pais
double sortCrossNumber = ConfigurationGA.random.NextDouble();
if (sortCrossNumber <= CrossOverRate)
{
Individual[] children = crossover(father, mother);
//mutação
if (ConfigurationGA.mutationType == GA.Mutation.NewIndividual)
{
children[0] = Mutation(children[0]);
children[1] = Mutation(children[1]);
}
//repopular os filhos no vetor
children[0].IndexOfArray = father.IndexOfArray;
children[1].IndexOfArray = mother.IndexOfArray;
popTemp[father.IndexOfArray] = children[0];
popTemp[mother.IndexOfArray] = children[1];
}
else
{
popTemp[father.IndexOfArray] = father;
popTemp[mother.IndexOfArray] = mother;
}
}
//apagar individuos velhos e inserir os individuos novos
for (int j = 0; j < ConfigurationGA.sizePopulation; j++)
{
newPop.SetIndividual(j, popTemp[j]);
}
popTemp = null;
//aplicação da mutação na nova população
if (ConfigurationGA.mutationType == GA.Mutation.InPopulation)
{
newPop = MutationPopulation(newPop);
}
//inserindo os individuos do elitismo na nova população
if (ConfigurationGA.elitism)
{
//avaliar a população
newPop.Evaluate();
//ordenar a população
newPop.Ordenate();
int startPoint = ConfigurationGA.sizePopulation - ConfigurationGA.sizeElitism;
int count = 0;
for (int a = startPoint; a < ConfigurationGA.sizePopulation; a++)
{
newPop.SetIndividual(a, indElitsm[count]);
count++;
}
}
//avaliação da população
newPop.Evaluate();
return(newPop);
}
public Individual[] CrossOverPMX(Individual father, Individual mother)
{
//variavel de retorno
Individual[] newInd = new Individual[2];
int[] parent1 = new int[ConfigurationGA.sizeChromossome];
int[] parent2 = new int[ConfigurationGA.sizeChromossome];
int[] offSpringArray1 = new int[ConfigurationGA.sizeChromossome];
int[] offSpringArray2 = new int[ConfigurationGA.sizeChromossome];
int[] replacement1 = new int[ConfigurationGA.sizeChromossome];
int[] replacement2 = new int[ConfigurationGA.sizeChromossome];
//Seleção dos pontos para cruzamento
int firstPoint = ConfigurationGA.random.Next(0, ConfigurationGA.sizeChromossome - 1);
int secondPoint = ConfigurationGA.random.Next(0, ConfigurationGA.sizeChromossome - 1);
if (firstPoint == secondPoint)
{
secondPoint = ConfigurationGA.random.Next(0, ConfigurationGA.sizeChromossome - 1);
}
if (secondPoint < firstPoint)
{
int temp = firstPoint;
firstPoint = secondPoint;
secondPoint = temp;
}
newInd[0] = new Individual();
newInd[1] = new Individual();
//transferir os genes para os pais
for (int i = 0; i < ConfigurationGA.sizeChromossome; i++)
{
parent1[i] = offSpringArray1[i] = father.GetGene(i);
parent2[i] = offSpringArray2[i] = mother.GetGene(i);
}
//popular o replacement para valores abaixo de 0
for (int i = 0; i < ConfigurationGA.sizeChromossome; i++)
{
replacement1[i] = replacement2[i] = -1;
}
//passo de cruzamento
for (int i = firstPoint; i <= secondPoint; i++)
{
offSpringArray1[i] = parent2[i];
offSpringArray2[i] = parent1[i];
replacement1[parent2[i]] = parent1[i];
replacement2[parent1[i]] = parent2[i];
}
//troca de genes repetidos
for (int i = 0; i < ConfigurationGA.sizeChromossome; i++)
{
if ((i >= firstPoint) && (i <= secondPoint))
{
continue;
}
int n1 = parent1[i];
int m1 = replacement1[n1];
int n2 = parent2[i];
int m2 = replacement2[n2];
while (m1 != -1)
{
n1 = m1;
m1 = replacement1[n1];
}
while (m2 != -1)
{
n2 = m2;
m2 = replacement2[n2];
}
offSpringArray1[i] = n1;
offSpringArray2[i] = n2;
}
for (int i = 0; i < ConfigurationGA.sizeChromossome; i++)
{
newInd[0].SetGene(i, offSpringArray1[i]);
newInd[1].SetGene(i, offSpringArray2[i]);
}
newInd[0].CalcFitness();
newInd[1].CalcFitness();
return(newInd);
}