static void Main(string[] args)
{
FitFunc fitFunc = new FitFunc();
GeneticAlgorithm ga = new GeneticAlgorithm(fitFunc.Fitness);
ga.Run();
Console.ReadLine();
}
//To generate new individual Run sequence
internal static Individual Random_Sequence()
{
FitFunc newtest = new FitFunc();
Individual I = new Individual();
string Indiv_s;
byte[][] Indiv_b = new byte[15][];
for (int i = 0; i < 9; i++)// elements 0-8 reperesent parameter levels
{
bool[] L = new bool[5];
int lvl = rnd.Next(0, 5); // generate random parameter level
L[lvl] = true; // set level to next a parameter and convert to byte array
byte[] L_b = Array.ConvertAll(L, b => b ? (byte)1 : (byte)0);
Indiv_b[i] = L_b;
}
for (int i = 9; i < 14; i++) // elements 9-13 reperent device presence
{
int D_presence = rnd.Next(0, 2); // Randomly decide if device during run
bool[] D = new bool[1];
if (D_presence == 1)
{
D[0] = true;
}
else
{
D[0] = false;
}
byte[] D_b = Array.ConvertAll(D, b => b ? (byte)1 : (byte)0);
Indiv_b[(i)] = D_b;
}
bool[] T = new bool[3];
int T_Level = rnd.Next(0, 3); //Randomly choose a Timer level
T[T_Level] = true;
byte[] T_b = Array.ConvertAll(T, b => b ? (byte)1 : (byte)0);
Indiv_b[14] = T_b; // Element 14 Represents Timer Lels
Indiv_s = ByteSeq_To_String(Indiv_b);
//Set Properties Of new Individual
I.Byte_Sequence = Indiv_b;
I.String_Sequence = Indiv_s;
I.Fitness = newtest.evalFunc(Indiv_s);
return(I);
}
public static Tuple <Individual, Individual> Crossover_and_Mutate(Individual Parent1, Individual Parent2, int Site, int mutationrate)
{
Individual child1 = new Individual();
Individual child2 = new Individual();
FitFunc newtest = new FitFunc();
// Initialise new Jagged Array for Children
child1.Byte_Sequence = new byte[15][];
child2.Byte_Sequence = new byte[15][];
//First Crossover
for (int i = 0; i < Parent1.Byte_Sequence.Length; i++)
{
if (i < Site)
{
child1.Byte_Sequence[i] = Parent1.Byte_Sequence[i];
child2.Byte_Sequence[i] = Parent2.Byte_Sequence[i];
}
else
{
child1.Byte_Sequence[i] = Parent2.Byte_Sequence[i];
child2.Byte_Sequence[i] = Parent1.Byte_Sequence[i];
}
}
//Then Perform muation function (mutation will happen depending on mutation rate)
Mutate(child1, mutationrate);
Mutate(child2, mutationrate);
//Updates Child's String_Sequence and fitness property
child1.String_Sequence = ByteSeq_To_String(child1.Byte_Sequence);
child1.Fitness = newtest.evalFunc(child1.String_Sequence);
child2.String_Sequence = ByteSeq_To_String(child2.Byte_Sequence);
child2.Fitness = newtest.evalFunc(child2.String_Sequence);
//Returns two new children
return(new Tuple <Individual, Individual>(child1, child2));
}
public void Selection(int parentPoolSize)
{
FitFunc selection = new FitFunc();
int fitness;
parentIndex = ((totalPopulation * parentPoolSize) / 100);
for (int i = 0; i < chroms.Count; i++)
{
fitness = selection.evalFunc(chroms[i].Genes);
chroms[i].Grade = fitness;
}
chroms.Sort((a, b) => b.CompareTo(a));
genBest = chroms[0].Grade;
chroms.RemoveRange(parentIndex, totalPopulation - parentIndex);
if (genBest > overallBest)
{
overallBest = genBest;
bestChrom = chroms[0].Genes;
}
}
public void Initialise(int populationSize)
{
chroms.Clear();
Random random = new Random();
FitFunc grading = new FitFunc();
totalPopulation = populationSize;
int fitness;
string createChrom;
for (int i = 0; i < populationSize; i++)
{
createChrom = alleleFive[random.Next(5)] + alleleFive[random.Next(5)] + alleleFive[random.Next(5)] + alleleFive[random.Next(5)] +
alleleFive[random.Next(5)] + alleleFive[random.Next(5)] + alleleFive[random.Next(5)] + alleleFive[random.Next(5)] +
alleleFive[random.Next(5)] + random.Next(0, 2) + random.Next(0, 2) + random.Next(0, 2) + random.Next(0, 2) + random.Next(0, 2)
+ alleleThree[random.Next(3)];
fitness = grading.evalFunc(createChrom);
chroms.Add(new Chromosome {
Genes = createChrom, Grade = fitness
});
}
}
//********************************************************************************************************************************
static void Main(string[] args)
{
int POPULATION_SIZE = 1000;
List <string> chromosomeList = new List <string>();
var rand = new Random();
for (int n = 0; n < POPULATION_SIZE; n++) // here 10 is the population size
{
Random random = new Random(); //create a random object
List <int> list1 = new List <int>(); //create a list for parameter 1
List <int> list2 = new List <int>(); //create a list for parameter 2
List <int> list3 = new List <int>(); //create a list for parameter 3
List <int> list4 = new List <int>(); //create a list for parameter 4
List <int> list5 = new List <int>(); //create a list for parameter 5
List <int> list6 = new List <int>(); //create a list for parameter 6
List <int> list7 = new List <int>(); //create a list for parameter 7
List <int> list8 = new List <int>(); //create a list for parameter 8
List <int> list9 = new List <int>(); //create a list for parameter 9
List <int> list10 = new List <int>(); //created a list10 correct for 10 - 14 parameters (5 elements)
List <int> list11 = new List <int>(); //create a list 11 for 15th parameter (without only 1 rest 0)
// Adding elements to List
list1.Add(0);
list1.Add(0);
list1.Add(0);
list1.Add(0);
list1.Add(0);
list1[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list2.Add(0);
list2.Add(0);
list2.Add(0);
list2.Add(0);
list2.Add(0);
list2[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list3.Add(0);
list3.Add(0);
list3.Add(0);
list3.Add(0);
list3.Add(0);
list3[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list4.Add(0);
list4.Add(0);
list4.Add(0);
list4.Add(0);
list4.Add(0);
list4[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list5.Add(0);
list5.Add(0);
list5.Add(0);
list5.Add(0);
list5.Add(0);
list5[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list6.Add(0);
list6.Add(0);
list6.Add(0);
list6.Add(0);
list6.Add(0);
list6[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list7.Add(0);
list7.Add(0);
list7.Add(0);
list7.Add(0);
list7.Add(0);
list7[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list8.Add(0);
list8.Add(0);
list8.Add(0);
list8.Add(0);
list8.Add(0);
list8[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list9.Add(0);
list9.Add(0);
list9.Add(0);
list9.Add(0);
list9.Add(0);
list9[random.Next() % 5] = 1; // changing one 0 to 1 at random position
for (int i = 0; i < 5; i++) //loop 5 times
{
int r = random.Next(0, 2); //genearte random number 0 or 1
list10.Add(r); //add the random number to list10
}
list11.Add(0);
list11.Add(0);
list11.Add(0);
list11[random.Next() % 3] = 1; // changing one 0 to 1 at random position
List <int> Individualslist = list1.Concat(list2).Concat(list3).Concat(list4).Concat(list5).Concat(list6).Concat(list7).Concat(list8).Concat(list9).Concat(list10).Concat(list11).ToList();
var chromosome = string.Join("", Individualslist); // concatenated chromosomeList is in List<int> data type and needs to be converted to string and this string needs to be stored in "chromosome"
chromosomeList.Add(chromosome);
} // of "for loop" of initialising Population
// printing the population
for (int i = 0; i < POPULATION_SIZE; i++)
{
Console.WriteLine(chromosomeList[i]);
}
List <string> testPopulationlist = new List <string>();
testPopulationlist.AddRange(chromosomeList);
List <int> fitnesseslist = new List <int>(testPopulationlist.Count); // 100 is the constant population size
// calculate fitness for test population.
for (int i = 0; i < 1; ++i) // this loop evaluates the fitness grade of the engine configurations chromosome string
{
FitFunc generationtest = new FitFunc();
foreach (string chromosomeString in testPopulationlist)
{
int gradeValue = generationtest.evalFunc(chromosomeString);
fitnesseslist.Add(gradeValue);
}
}
List <string> topScorerList = Selection1(testPopulationlist, fitnesseslist);
for (int i = 0; i < fitnesseslist.Count; i++)
{
if (testPopulationlist[i] == topScorerList[0])
{
Console.WriteLine(topScorerList[0] + " scored the highest fitness grade of " + fitnesseslist[i]);
break;
}
}
testPopulationlist = topScorerList; //now contain only the top 10% student names
//Clear the contents of the gradeslist completely and keep only
fitnesseslist.Clear(); // clear this list
List <string> runPopulationlist = new List <string>();
List <string> OptimumPopulationlist = new List <string>();
List <string> MutatePopulationlist = new List <string>();
List <string> Generationlist = new List <string>();
Generationlist.AddRange(testPopulationlist);
int generation = 10;
// continuously generate populations until number of iterations is met.
for (int iter = 1; iter < generation; ++iter)
{
if (iter == generation)
{
break; // condition for breaking out of generation loop.
}
while (Generationlist.Count <= POPULATION_SIZE)
{
// crossover
// Instantiating random number generator
var random1 = new Random();
//OptimumPopulationlist which will store randomly selected parent strings as per calculations for crossover
for (int c = 0; c < 10; c++) // random 10 chromosome strings (n = 10)
{
//generating a random index
int index1 = random1.Next(Generationlist.Count);
// adding the chromosome string present at randomly generated index in runPopulationlist to OptimumPopulationlist
OptimumPopulationlist.Add(Generationlist[index1]);
}
//Console.Write("\n random 10 crossover parent strings: \n");
// loop to print the 9 random chromosome strings in OptimumPopulationlist
//for (int i = 0; i < 10; i++)
// {
// Console.WriteLine(OptimumPopulationlist[i]);
//}
int[] array = { 5, 10, 15, 20, 25, 30, 35, 40, 45, 50 }; // array of crossover index points
string[] list = OptimumPopulationlist.ToArray();
//Console.Write("\n crossover : \n");
string[] result = crossover(list, array); // send to crossover function
//for (int i = 0; i < result.Length; i++)
// Console.WriteLine(result[i] + " --> offspring " + (i + 1));
List <string> crossoverResultlist = new List <string>(result);
runPopulationlist.AddRange(crossoverResultlist); //crossoverResultlist has 900 chromosomees for 10 randomly selected chromosome strings from OptimumPopulationlist (from top 5% of test Populationlist of 1000 population size)
// Mutation
Program p = new Program();
List <string> resultlist = p.Mutation(Generationlist); // strings in resultlist from 1 randomly selected string // resultlist has 50 strings
//function to remove chromosomes strings in resultlist which are identical to OptimumPopulationlist
RemoveIdenticalString(resultlist, Generationlist); //string is reference type, so resultList is updated on ReturnIdenticalString
//MutatePopulationlist which will store randomly selected 50 mutated chromosome strings for 1000 population
for (int m = 0; m < 50; m++) // random 1 string from runPopulationlist so 1 x 50 combinations is 50 for mutation
{
//generating a random index
int index1 = random1.Next(resultlist.Count);
// adding the chromosome string present at randomly generated index in runPopulationlist to MutatePopulationlist
MutatePopulationlist.Add(resultlist[index1]);
// 50 strings generated from one randomly selectedstring
}
// loop to print the 50 random mutated chromosome strings in MutatePopulationlist
//for (int i = 0; i < 50; i++)
//{
// Console.WriteLine(MutatePopulationlist[i]);
// }
resultlist.Clear();
//********************************************************************************************************************
Generationlist.AddRange(MutatePopulationlist);
Generationlist.AddRange(runPopulationlist);
}
List <int> fitnesseslist1 = new List <int>(Generationlist.Count);
// calculate fitness for test population.
for (int i = 0; i < 1; ++i) // this loop evaluates the fitness grade of the engine configurations chromosome string
{
FitFunc generationtest1 = new FitFunc();
foreach (string chromosomeString in Generationlist)
{
int gradeValue1 = generationtest1.evalFunc(chromosomeString);
fitnesseslist1.Add(gradeValue1);
}
}
List <string> topScorerList1 = Selection2(Generationlist, fitnesseslist1);
for (int i = 0; i < fitnesseslist1.Count; i++)
{
if (Generationlist[i] == topScorerList1[0])
{
Console.WriteLine(topScorerList1[0] + " scored the highest fitness grade of " + fitnesseslist1[i] + " in generation:" + iter);
break;
}
}
Generationlist = topScorerList1;//now contain only the top N% chromosome strings.
fitnesseslist1.Clear(); //Clear the contents of the fitnesseslist1 completely.
runPopulationlist.Clear();
MutatePopulationlist.Clear();
}
Console.WriteLine("Hello World!");
}
private void BruteWorker_DoWork(object sender, DoWorkEventArgs e)
{
string
bruteExTemp = "",
bruteInTemp = "",
bruteCyPre = "",
bruteVOP = "",
bruteLoTor = "",
bruteNOx = "",
bruteCO = "",
bruteHC = "",
brutePM = "",
bruteUCLan = "",
bruteString = "",
optimalSettings = "";
int
gradeValue = 0,
newGradeValue,
progress = 0,
Iterations = 0;
//Extern temp
for (int i = 0; i < 5; i++)
{
bruteExTemp = toBinary(2 ^ i);
//Intern temp
for (int j = 0; j < 5; j++)
{
bruteInTemp = bruteExTemp + toBinary(2 ^ j);
//Cylinder Pressure
for (int k = 0; k < 5; k++)
{
bruteCyPre = bruteExTemp + bruteInTemp + toBinary(2 ^ k);
//Valve Opening Pressure
for (int l = 0; l < 5; l++)
{
bruteVOP = bruteExTemp + bruteInTemp + bruteCyPre + toBinary(2 ^ l);
//Load torque
for (int m = 0; m < 5; m++)
{
bruteLoTor = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + toBinary(2 ^ m);
//NOx Emissions
for (int n = 0; n < 5; n++)
{
bruteNOx = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + toBinary(2 ^ n);
//CO Emissions
for (int o = 0; o < 5; o++)
{
bruteCO = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + bruteNOx + toBinary(2 ^ o);
//HC Emissions
for (int p = 0; p < 5; p++)
{
bruteHC = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + bruteNOx + bruteCO + toBinary(2 ^ p);
//PM Emissions
for (int q = 0; q < 5; q++)
{
brutePM = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + bruteNOx + bruteCO + bruteHC + toBinary(2 ^ q);
//UCLan Stuff
for (int r = 0; r < 32; r++)
{
bruteUCLan = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + bruteNOx + bruteCO + bruteHC + brutePM + toBinary(r);
//
for (int s = 0; s < 3; s++)
{
//temp
if (s == 0)
{
//high
bruteString = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + bruteNOx + bruteCO + bruteHC + brutePM + bruteUCLan + "001";
}
else if (s == 1)
{
//Middle
bruteString = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + bruteNOx + bruteCO + bruteHC + brutePM + bruteUCLan + "010";
}
else
{
//low
bruteString = bruteExTemp + bruteInTemp + bruteCyPre + bruteVOP + bruteLoTor + bruteNOx + bruteCO + bruteHC + brutePM + bruteUCLan + "100";
}
//shows the progress in progress bar
progress = (Iterations + 1 / 187500000);
BruteWorker.ReportProgress(progress, BruteProgress);
System.Threading.Thread.Sleep(1000);
if (BruteWorker.CancellationPending)
{
e.Cancel = true;
}
FitFunc newtest = new FitFunc();
newGradeValue = newtest.evalFunc(bruteString);
if (newGradeValue > gradeValue)
{
gradeValue = newGradeValue;
optimalSettings = bruteString;
}
}
}
}
}
}
}
}
}
}
}
}
//fin
MessageBox.Show("Optimal Settings found: " + optimalSettings + Environment.NewLine + "With a grade of: " + gradeValue);
Result_Box.Text = "Optimal Settings found: " + optimalSettings + Environment.NewLine + "With a grade of: " + gradeValue;
}
public int[] GeneticAlgorythm(int PoolSize, int Cycles, int GeneNumber)
{
int[][]
GeneInt = new int[PoolSize][]; //to store the config parts as integer array array
int
mutate;
string[]
GeneString = new string[PoolSize]; //to store the configurations
FitFunc
EvaluateGene = new FitFunc();
Random randNumGen = new Random();
//GENERATES POOL MEMBERS
GeneInt = poolGen(PoolSize);
//Genetic Algorythm code to follow:
//EVALUATE FIRST BATCH
for (int i = 0; i < PoolSize; i++)
{
GeneString[i] = ToBinary(GeneInt[i]);
GeneInt[i][11] = EvaluateGene.evalFunc(GeneString[i]);
}
for (int i = 0; i < Cycles; i++)
{
//SORT ARRAY
GeneInt = sort(GeneInt, PoolSize);
Evaluation.Series["Genetic Algorythm"].Points.AddXY(i, GeneInt[0][11]);
System.Threading.Thread.Sleep(1); //Thread sleep to give UI time to updatew with fresh graph
//REPOPULATE BOTTOM 2/3
GeneInt = childmaker(GeneInt, PoolSize);
if (i < 100)
{
mutate = randNumGen.Next(0, 101 - i); //non-Uniform mutation prevents later pools from being too random allowing better solutions pools to be kept
}
else
{
mutate = 0;
}
//MUTATE
for (int m = PoolSize / GeneNumber; (m < mutate) && (m < PoolSize); m++) //random number of mutations
{
int r = randNumGen.Next(0, 11); //mutate random parameter
GeneInt[m][r] = GeneInt[m][r] + randNumGen.Next(0, PoolSize);
}
//EVALUATE CHILDREN
for (int j = 0; j < PoolSize; j++)
{
GeneString[j] = ToBinary(GeneInt[j]);
GeneInt[j][11] = EvaluateGene.evalFunc(GeneString[j]);
}
}
GeneInt = sort(GeneInt, PoolSize);
Result_Box.Text = Result_Box.Text + "The Genetic Algorithm solution is: " + ToBinary(GeneInt[0]) +
Environment.NewLine + "With a grade of: " + GeneInt[0][11] + Environment.NewLine;
return(GeneInt[0]);
}
public int[] ParticleSwarm(int PoolSize, int Cycles)
{
foreach (var series in ParticleVisual.Series)
{
series.Points.Clear();
}
int[][][]
Generation = new int[Cycles][][];
int[][]
ParticleInt = new int[PoolSize][],
ParticlePlot = new int[PoolSize][],
ParticleLocalBest = new int[PoolSize][],
temp = new int[PoolSize][],
GlobalBest = new int[Cycles][];
double[][]
velocity = new double[PoolSize][];
double
random;
string[]
ParticleString = new string[PoolSize];
string
particle; //unused was part of attampt to graph particle score over time(cycles)
FitFunc
EvaluateParticle = new FitFunc();
Random randNumGen = new Random();
//initial velocity
for (int a = 0; a < PoolSize; a++)
{
velocity[a] = new double[12];
for (int b = 0; b < 12; b++)
{
velocity[a][b] = 0.1;
}
}
//generate random pool of particles
ParticleInt = poolGen(PoolSize);
Generation[0] = ParticleInt;
for (int a = 1; a < Cycles; a++)
{
Generation[a] = ParticleInt;
}
for (int i = 0; i < PoolSize; i++)
{
ParticleString[i] = ToBinary(Generation[0][i]);
ParticleInt[i][11] = EvaluateParticle.evalFunc(ParticleString[i]);
}
ParticlePlot = sort(ParticleInt, PoolSize);
GlobalBest[0] = ParticleInt[0];
for (int gen = 0; gen < Cycles; gen++)
{
GlobalBest[gen] = Generation[gen][0];
for (int i = 0; i < PoolSize; i++)
{
ParticleString[i] = ToBinary(Generation[gen][i]);
Generation[gen][i][11] = EvaluateParticle.evalFunc(ParticleString[i]);
}
ParticleLocalBest = Generation[gen];
// Step 2 – for cycles many times do this
for (int i = 0; i < Cycles; i++)
{
//System.Threading.Thread.Sleep(1); //Thread sleep to give UI time to updatew with fresh graph
//PlotConvergence(i, Generation[gen][i][11]);
// Step 3 - For each particle do this
for (int j = 0; j < PoolSize; j++)
{
//temp = (sort(Generation[gen], PoolSize));
//Step 4 – Evaluate the particle’s position.
ParticleString[j] = ToBinary(Generation[gen][j]);
Generation[gen][j][11] = EvaluateParticle.evalFunc(ParticleString[j]);
//Step 5 – Compare particle’s current position to particle’s best local position, if current position is better replace it.
if (Generation[gen][j][11] > ParticleLocalBest[j][11])
{
int[] temp2 = new int[12];
temp2 = Generation[gen][j];
for (int k = 0; k > 12; k++)
{
ParticleLocalBest[j][k] = temp2[k];
}
}
//Step 6 – Compare particle’s current position to global best position of the previous cycle, if better, replace
if (gen > 0)
{
if (Generation[gen][j][11] > GlobalBest[gen - 1][11])
{
for (int k = 0; k > 12; k++)
{
GlobalBest[gen][k] = Generation[gen][j][k];
}
}
else if (Generation[gen][j][11] < GlobalBest[gen - 1][11])
{
for (int k = 0; k > 12; k++)
{
GlobalBest[gen][k] = Generation[gen - 1][j][k];
}
}
}
// Step 7 - Calculate velocity vector for particles
for (int k = 0; k < 11; k++)
{
random = randNumGen.Next(1, 101) / 100.00;
if (gen < 1)
{
velocity[j][k] = Convert.ToDouble(velocity[j][k] + 2 + random * (ParticleLocalBest[j][k] - Generation[gen][j][k]) + 2 * random * (GlobalBest[gen][k] - Generation[gen][j][k]));
}
else
{
velocity[j][k] = Convert.ToDouble(velocity[j][k] + 2 + random * (ParticleLocalBest[j][k] - Generation[gen][j][k]) + 2 * random * (GlobalBest[gen - 1][k] - Generation[gen][j][k]));
}
// Step 8 - Move particle according to velocity vector.
if (gen + 1 >= Cycles)
{
//
}
else if (k < 9)
{
if ((velocity[j][k] < 0) && (Generation[gen][j][k] != 1))
{
Generation[gen + 1][j][k] = Generation[gen][j][k]--;
}
else if ((velocity[j][k] > 0) && (Generation[gen][j][k] != 5))
{
Generation[gen + 1][j][k] = Generation[gen][j][k]++;
}
else if (gen + 1 >= Cycles)
{
//
}
else
{
}
}
else if (k == 9)
{
if ((velocity[j][k] < 0) && (Generation[gen][j][k] != 1))
{
Generation[gen + 1][j][k] = Generation[gen][j][k]--;
}
else if ((velocity[j][k] > 0) && (Generation[gen][j][k] != 32))
{
Generation[gen + 1][j][k] = Generation[gen][j][k]++;
}
else if (gen + 1 >= Cycles)
{
//
}
else
{
}
}
else if (k == 10)
{
if ((velocity[j][k] < 0) && (Generation[gen][j][k] != 1))
{
Generation[gen + 1][j][k] = Generation[gen][j][k]--;
}
else if ((velocity[j][k] > 0) && (Generation[gen][j][k] != 3))
{
Generation[gen + 1][j][k] = Generation[gen][j][k]++;
}
else if (gen + 1 >= Cycles)
{
//
}
else
{
}
}
//ParticleVisual.Series["Velocity per cycle"].Points.AddXY(k, velocity[j][k]);
}
}
//Step 9 – Return to Step 2.
//ParticleGraph.RunWorkerAsync();
}
if (gen < 1)
{
Evaluation.Series["Particle Swarm"].Points.AddXY(gen, GlobalBest[gen][11]);
}
else
{
Evaluation.Series["Particle Swarm"].Points.AddXY(gen, GlobalBest[gen - 1][11]);
}
}
Result_Box.Text = Result_Box.Text + "The Particle Swarm solition is: " + ToBinary(GlobalBest[Cycles - 1]) +
Environment.NewLine + "With a grade of: " + GlobalBest[Cycles - 1][11] + Environment.NewLine;
return(GlobalBest[0]);
}
} // end of sorting function
static void Main(string[] args)
{
int POPULATION_SIZE = 100;
//List<List<int>> initialPopulationlist = new List<List<int>>();
List <string> chromosomeList = new List <string>();
var rand = new Random();
for (int n = 0; n < POPULATION_SIZE; n++) // here 10 is the population size
{
Random random = new Random(); //create a random object
List <int> list1 = new List <int>(); //create a list for parameter 1
List <int> list2 = new List <int>(); //create a list for parameter 2
List <int> list3 = new List <int>(); //create a list for parameter 3
List <int> list4 = new List <int>(); //create a list for parameter 4
List <int> list5 = new List <int>(); //create a list for parameter 5
List <int> list6 = new List <int>(); //create a list for parameter 6
List <int> list7 = new List <int>(); //create a list for parameter 7
List <int> list8 = new List <int>(); //create a list for parameter 8
List <int> list9 = new List <int>(); //create a list for parameter 9
List <int> list10 = new List <int>(); //created a list10 correct for 10 - 14 parameters (5 elements)
List <int> list11 = new List <int>(); //create a list 11 for 15th parameter (without only 1 rest 0)
// Adding elements to List
list1.Add(0);
list1.Add(0);
list1.Add(0);
list1.Add(0);
list1.Add(0);
list1[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list2.Add(0);
list2.Add(0);
list2.Add(0);
list2.Add(0);
list2.Add(0);
list2[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list3.Add(0);
list3.Add(0);
list3.Add(0);
list3.Add(0);
list3.Add(0);
list3[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list4.Add(0);
list4.Add(0);
list4.Add(0);
list4.Add(0);
list4.Add(0);
list4[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list5.Add(0);
list5.Add(0);
list5.Add(0);
list5.Add(0);
list5.Add(0);
list5[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list6.Add(0);
list6.Add(0);
list6.Add(0);
list6.Add(0);
list6.Add(0);
list6[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list7.Add(0);
list7.Add(0);
list7.Add(0);
list7.Add(0);
list7.Add(0);
list7[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list8.Add(0);
list8.Add(0);
list8.Add(0);
list8.Add(0);
list8.Add(0);
list8[random.Next() % 5] = 1; // changing one 0 to 1 at random position
list9.Add(0);
list9.Add(0);
list9.Add(0);
list9.Add(0);
list9.Add(0);
list9[random.Next() % 5] = 1; // changing one 0 to 1 at random position
for (int i = 0; i < 5; i++) //loop 5 times
{
int r = random.Next(0, 2); //genearte random number 0 or 1
list10.Add(r); //add the random number to list10
}
list11.Add(0);
list11.Add(0);
list11.Add(0);
list11[random.Next() % 3] = 1; // changing one 0 to 1 at random position
List <int> Individualslist = list1.Concat(list2).Concat(list3).Concat(list4).Concat(list5).Concat(list6).Concat(list7).Concat(list8).Concat(list9).Concat(list10).Concat(list11).ToList();
var chromosome = string.Join("", Individualslist); // concatenated chromosomeList is in List<int> data type and needs to be converted to string and this string needs to be stored in "chromosome"
chromosomeList.Add(chromosome);
} // of "for loop" of initialising Population
for (int i = 0; i < 100; i++)
{
Console.WriteLine(chromosomeList[i]);
}
List <string> runPopulationlist = new List <string>();
runPopulationlist.AddRange(chromosomeList);
// Create fitnesses list with as many elements as the chromosome strings in the OptimumPopulationlist;
List <int> fitnesseslist = new List <int>(POPULATION_SIZE); // 100 is the constant population size
// calculate fitness for initliased population.
for (int i = 0; i < 1; ++i) // this loop evaluates the fitness grade of the engine configurations chromosome string
{
//fitnesseslist.Clear();
FitFunc generationtest = new FitFunc();
foreach (string chromosomeString in runPopulationlist)
{
int gradeValue = generationtest.evalFunc(chromosomeString);
fitnesseslist.Add(gradeValue);
}
}
List <string> topScorerList = Selection(runPopulationlist, fitnesseslist);
for (int i = 0; i < fitnesseslist.Count; i++)
{
if (runPopulationlist[i] == topScorerList[0])
{
Console.WriteLine(topScorerList[0] + " scored the highest fitness grade of " + fitnesseslist[i]);
break;
}
}
runPopulationlist = topScorerList; //now contain only the top 10% student names
//Clear the contents of the gradeslist completely and keep only
fitnesseslist.Clear(); // clear this list after each generation loop.
Console.WriteLine("Hello World!");
}
