public void Elite()
{
IChromosome<int>[] a = new IChromosome<int>[] { new DigitalChromosome().GenerateFromArray(new int[] { 9, 9, 9 }),
new DigitalChromosome().GenerateFromArray(new int[] { 7, 7, 7 }), new DigitalChromosome().GenerateFromArray(new int[] { 3, 3, 3 }),
new DigitalChromosome().GenerateFromArray(new int[] { 2, 2, 2 }), new DigitalChromosome().GenerateFromArray(new int[] { 7, 7, 7 }), new DigitalChromosome().GenerateFromArray(new int[] { 7, 7, 7 }),
new DigitalChromosome().GenerateFromArray(new int[] { 7, 7, 7 }) };
EliteSelection<int> selection = new EliteSelection<int>(2);
IChromosome<int>[] res = selection.Select(a, x => x.ToArray().Sum(), 6);
CollectionAssert.AreEqual(new IChromosome<int>[] { new DigitalChromosome().GenerateFromArray(new int[] { 9, 9, 9 }),
new DigitalChromosome().GenerateFromArray(new int[] { 7, 7, 7 }) }, res.Take(2).ToArray());
}
public static void CorretionCentroid()
{
//中心点数据
Stopwatch sw = new Stopwatch();
sw.Start();
List <Coordinate> centroidPoints = new List <Coordinate>(80);
StreamReader sr = new StreamReader(@"D:\MagicSong\OneDrive\2017研究生毕业设计\数据\项目用数据\中心点80.txt");
sr.ReadLine();//读取标题行
while (!sr.EndOfStream)
{
string[] line = sr.ReadLine().Split(',');
centroidPoints.Add(new Coordinate(double.Parse(line[1]), double.Parse(line[2])));
}
sr.Close();
//Bus数据,并且构造KD树
KdTree myKdtree = new KdTree(2);
IFeatureSet busFS = FeatureSet.Open(@"D:\MagicSong\OneDrive\2017研究生毕业设计\数据\项目用数据\BusStopGauss.shp");
List <Coordinate> busStopPoints = new List <Coordinate>(busFS.NumRows());
foreach (var item in busFS.Features)
{
var c = item.Coordinates[0];
busStopPoints.Add(c);
myKdtree.Insert(new double[] { c.X, c.Y }, item);
}
Console.WriteLine("数据读取完毕，开始构造遗传算法");
IFeatureSet newCentroid = new FeatureSet(FeatureType.Point);
newCentroid.Name = "优化过的中心点";
newCentroid.Projection = ProjectionInfo.FromEpsgCode(GAUSS_EPSG);
newCentroid.DataTable.Columns.Add("name", typeof(string));
//遗传算法，构造适应性函数
MyProblemChromosome.CandiateNumber = 5;
List <int[]> candinatesForEachControid = new List <int[]>(centroidPoints.Count);
foreach (var item in centroidPoints)
{
object[] nearest = myKdtree.Nearest(new double[] { item.X, item.Y }, MyProblemChromosome.CandiateNumber);
candinatesForEachControid.Add(nearest.Select((o) =>
{
var f = o as IFeature;
return(f.Fid);
}).ToArray());
}
MyProblemFitness fitness = new MyProblemFitness(centroidPoints, busStopPoints, candinatesForEachControid);
MyProblemChromosome mpc = new MyProblemChromosome(centroidPoints.Count);
//这边可以并行
MyProblemChromosome globalBest = null;
Console.WriteLine("遗传算法构造已经完成！");
sw.Stop();
Console.WriteLine("一共用时：{0}s", sw.Elapsed.TotalSeconds);
int GACount = 8;
Parallel.For(0, GACount, new Action <int>((index) =>
{
var selection = new EliteSelection();
var crossover = new TwoPointCrossover();
var mutation = new ReverseSequenceMutation();
var population = new Population(1000, 1200, mpc);
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
ga.Termination = new GenerationNumberTermination(1000);
Stopwatch sw1 = new Stopwatch();
sw1.Start();
Console.WriteLine("遗传算法任务{0}正在运行.......", index);
ga.Start();
var best = ga.BestChromosome as MyProblemChromosome;
if (globalBest == null || globalBest.Fitness < best.Fitness)
{
globalBest = best;
}
sw1.Stop();
Console.WriteLine("第{0}次遗传算法已经完成，耗费时间为：{1}s，最终的fitness为：{2},有效个数为：{3}", index, sw1.Elapsed.TotalSeconds, best.Fitness, best.Significance);
}));
Console.WriteLine("Final Choose!");
Console.WriteLine("最终的fitness为：{0},有效个数为：{1}", globalBest.Fitness, globalBest.Significance);
for (int i = 0; i < globalBest.Length; i++)
{
int index = candinatesForEachControid[i][(int)globalBest.GetGene(i).Value];
Coordinate c = busStopPoints[index];
var f = newCentroid.AddFeature(new Point(c));
f.DataRow.BeginEdit();
f.DataRow["name"] = busFS.GetFeature(index).DataRow["name"];
f.DataRow.EndEdit();
}
newCentroid.SaveAs("newCentroid.shp", true);
Console.ReadKey();
}
static void Main(string[] args)
{
#region Loading data from File
string location = Directory.GetCurrentDirectory() + "\\Data\\graf4_kod.txt";
int[,] data = LoadDataFromFile(location);
roads = new List <Road>();
for (int i = 0; i < data.GetLength(0); i++)
{
Road newRoad = new Road(data[i, 0], data[i, 1], data[i, 2]); // one direction cost
roads.Add(newRoad);
Road reverseRoad = new Road(data[i, 1], data[i, 0], data[i, 3]); // read reverse road
roads.Add(reverseRoad);
}
#endregion
var selection = new EliteSelection();
var crossover = new ThreeParentCrossover();
var mutation = new TworsMutation();
var fitness = new CPFitness();
var chromosome = new CPChromosome(3 * roads.Count);
var population = new Population(200, 400, chromosome);
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
ga.Termination = new GenerationNumberTermination(400);
Stopwatch timer = new Stopwatch();
timer.Start();
Console.WriteLine("GA running...");
ga.Start();
Console.WriteLine();
timer.Stop();
bool first = true;
int startCityIndex = Convert.ToInt32(ga.BestChromosome.GetGenes()[0].Value, CultureInfo.InvariantCulture);
int startCity = roads[startCityIndex].cityFrom;
int totalCost = 0;
foreach (Gene gene in ga.BestChromosome.GetGenes())
{
int ind = Convert.ToInt32(gene.Value, CultureInfo.InvariantCulture);
totalCost += roads[ind].cost;
if (CPFitness.everyRoadIsTraveled(roads) && roads[ind].cityTo == startCity)
{
Console.Write("-" + roads[ind].cityTo.ToString());
break;
}
else
{
if (first)
{
Console.Write(roads[ind].cityFrom.ToString() + "-" + roads[ind].cityTo.ToString());
first = false;
}
else
{
Console.Write("-" + roads[ind].cityTo.ToString());
}
roads[ind].isTravelled = true;
Road returnRoad = roads.Find(e => e.index.Equals(roads[ind].cityTo.ToString() + "-" + roads[ind].cityFrom.ToString()));
returnRoad.isTravelled = true;
}
}
TimeSpan ts = timer.Elapsed;
string elapsedTime = String.Format("{0:00}:{1:00}:{2:00}.{3:00}", ts.Hours, ts.Minutes, ts.Seconds, ts.Milliseconds / 10);
Console.WriteLine();
Console.WriteLine("Best solution found has {0} fitness.", ga.BestChromosome.Fitness);
Console.WriteLine("Best solution has total cost: {0}", totalCost);
Console.WriteLine("Time running: {0}", elapsedTime);
Console.ReadKey();
}
public OutputModel Process(InputModel input)
{
var toppingCount = Enum.GetValues(typeof(Topping)).Length;
var sliceMasks = GeneratePossibleSliceMasks(input.MinToppings * toppingCount, input.MaxCells).ToList();
var validSlices = new ConcurrentBag <Slice>();
Parallel.ForEach(sliceMasks, mask =>
{
for (var i = 0; i < input.Rows - mask.Width; i++)
{
for (var j = 0; j < input.Columns - mask.Height; j++)
{
var slice = new Slice(i, j, i + mask.Width, j + mask.Height);
if (IsValidSlice(input, slice))
{
validSlices.Add(slice);
}
}
}
});
var orderedSlices = validSlices.ToList();
var mutationRate = 1f;
var crossOverRate = 1f;
RandomizationProvider.Current = new FastRandomRandomization();
var fitness = new PizzaCutterFitness(input, orderedSlices);
var selection = new EliteSelection();
var crossOver = new EvolutionStrategyCrossOver();
var mutation = new FlipBitMutation();
var reinsertion = new EliteIncludeParentsReinsertion(fitness);
var currentBest = new PizzaCutterChromosome(orderedSlices.Count, (input.Rows * input.Columns) / (input.MinToppings * toppingCount));
currentBest.Fitness = fitness.Evaluate(currentBest);
var population = new Population(10, 20, currentBest);
var ga = new GeneticAlgorithm(population, fitness, selection, crossOver, mutation)
{
Reinsertion = reinsertion,
Termination = new FitnessStagnationTermination(),
CrossoverProbability = crossOverRate,
MutationProbability = mutationRate,
TaskExecutor = new SmartThreadPoolTaskExecutor()
{
MinThreads = Environment.ProcessorCount,
MaxThreads = Environment.ProcessorCount
}
};
ga.GenerationRan += (sender, args) =>
{
if (ga.BestChromosome.Fitness > currentBest.Fitness)
{
currentBest = ga.BestChromosome.Clone() as PizzaCutterChromosome;
}
};
ga.Start();
var output = new OutputModel()
{
Slices = currentBest.GetSlices(orderedSlices).ToList()
};
return(output);
}
private void btnStart_Click(object sender, EventArgs e)
{
Models.Instance i = new Models.Instance();
i.Days = 7;
i.Doctors = doctors;
int min = (int)numMin.Value;
int max = (int)numMax.Value;
var chromosome = new Models.Chromosome(21, i, r);
var population = new Population(min, max, chromosome);
var fitness = new Models.Fitness();
IMutation mutation = new TworsMutation();
ISelection selection = new RouletteWheelSelection();
ICrossover crossover = new OnePointCrossover(r.Next(20));
if (cbxMutation.SelectedItem.ToString() == "Insertion")
{
mutation = new InsertionMutation();
}
else if (cbxMutation.SelectedItem.ToString() == "Partial Shuffle")
{
mutation = new PartialShuffleMutation();
}
else if (cbxMutation.SelectedItem.ToString() == "Reverse Sequence")
{
mutation = new ReverseSequenceMutation();
}
if (cbxSelection.SelectedItem.ToString() == "Elitizam")
{
selection = new EliteSelection();
}
if (cbxCrossover.SelectedItem.ToString() == "Two-point")
{
int p1 = r.Next(19);
int p2 = r.Next(p1 + 1, 20);
crossover = new TwoPointCrossover(p1, p2);
}
else if (cbxCrossover.SelectedItem.ToString() == "Uniform")
{
crossover = new UniformCrossover();
}
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
ITermination termination = new FitnessStagnationTermination(50);
if (cbxTermination.SelectedItem.ToString() == "Generation number")
{
termination = new GenerationNumberTermination(200);
}
ga.Termination = termination;
ga.MutationProbability = (float)numProbability.Value;
ga.Start();
Gene[] g = ga.BestChromosome.GetGenes();
dataView.Rows.Clear();
for (int j = 0; j < 7; j++)
{
string[] row = new string[] { ((List <int>)g[j * 3].Value)[0].ToString() + " " + ((List <int>)g[j * 3].Value)[1].ToString(),
((List <int>)g[j * 3 + 1].Value)[0].ToString() + " " + ((List <int>)g[j * 3 + 1].Value)[1].ToString(),
((List <int>)g[j * 3 + 2].Value)[0].ToString() + " " + ((List <int>)g[j * 3 + 2].Value)[1].ToString() };
dataView.Rows.Add(row);
dataView.Rows[j].HeaderCell.Value = (j + 1).ToString();
}
lblFitness.Text = ga.BestChromosome.Fitness.ToString() + " , generacija broj " + ga.GenerationsNumber.ToString();
}
/// <summary>
/// Example:
///
/// http://diegogiacomelli.com.br/function-optimization-with-geneticsharp/
/// </summary>
static void Main(string[] args)
{
float maxWidth = 998;
float maxHeight = 680;
var chromosome = new FloatingPointChromosome(
new double[] { 0, 0, 0, 0, },
new double[] { maxWidth, maxHeight, maxWidth, maxHeight },
new int[] { 10, 10, 10, 10 },
new int[] { 0, 0, 0, 0 });
var population = new Population(50, 100, chromosome);
var fitness = new FuncFitness((c) =>
{
var fc = c as FloatingPointChromosome;
var values = fc.ToFloatingPoints();
var x1 = values[0];
var y1 = values[1];
var x2 = values[2];
var y2 = values[3];
return(Math.Sqrt(Math.Pow(x2 - x1, 2) + Math.Pow(y2 - y1, 2)));
});
var selection = new EliteSelection();
var crossover = new UniformCrossover(0.5f);
var mutation = new FlipBitMutation();
var termination = new FitnessStagnationTermination(100);
var ga = new GeneticAlgorithm(
population,
fitness,
selection,
crossover,
mutation);
ga.Termination = termination;
Console.WriteLine("Generation: (x1, y1), (x2, y2) = distance");
var latestFitness = 0.0;
ga.GenerationRan += (sender, e) =>
{
var bestChromosome = ga.BestChromosome as FloatingPointChromosome;
var bestFitness = bestChromosome.Fitness.Value;
if (bestFitness != latestFitness)
{
latestFitness = bestFitness;
var phenotype = bestChromosome.ToFloatingPoints();
Console.WriteLine(
"Generation {0,2}: ({1},{2}),({3},{4}) = {5}",
ga.GenerationsNumber,
phenotype[0],
phenotype[1],
phenotype[2],
phenotype[3],
bestFitness
);
}
};
ga.Start();
Console.ReadKey();
}
/// <summary>
/// Loads presetsProfile into service.
/// </summary>
/// <exception cref="T:System.InvalidOperationException">Thrown when an attempt to load presetsProfile twice is made.</exception>
public void AddGeneticAlgorithm(GADefinition definition, Guid Key)
{
var chromosome = new FloatingPointChromosome(
minValue: new double[] { 0, 0, 0, 0 },
maxValue: new double[] { _maxWidth, _maxHeight, _maxWidth, _maxHeight },
totalBits: new int[] { 20, 20, 20, 20 },
fractionDigits: new int[] { 0, 0, 0, 0 }, geneValues: _geneValues);
ICrossover gaCrossover = default;
switch (definition.Crossover)
{
case Crossovers.Uniform:
gaCrossover = new UniformCrossover(mixProbability: 0.5f);
break;
case Crossovers.OnePoint:
gaCrossover = new OnePointCrossover(swapPointIndex: 40);
break;
case Crossovers.ThreeParent:
gaCrossover = new ThreeParentCrossover();
break;
default:
throw new ArgumentOutOfRangeException(paramName: nameof(definition.Crossover),
actualValue: definition.Crossover, message: "Crossover has wrong value");
}
ISelection gaSelection = default;
switch (definition.Selection)
{
case Selections.Elite:
gaSelection = new EliteSelection();
break;
case Selections.Roulette:
gaSelection = new RouletteWheelSelection();
break;
case Selections.Tournament:
gaSelection = new TournamentSelection(
size: decimal.ToInt32(d: decimal.Multiply(definition.Population, new decimal(0.2f))));
break;
case Selections.StohasticUniversalSampling:
gaSelection = new StochasticUniversalSamplingSelection();
break;
default:
throw new ArgumentOutOfRangeException(paramName: nameof(definition.Selection),
actualValue: definition.Selection, message: "Selection has wrong value");
}
var gaMutation = new UniformMutation(true);
var gaPopulation = new Population(minSize: definition.Population,
maxSize: definition.Population, adamChromosome: chromosome);
var ga = new GeneticAlgorithm(population: gaPopulation,
fitness: new EuclideanDistanceFitness(), selection: gaSelection,
crossover: gaCrossover, mutation: gaMutation);
ga.MutationProbability = (float)definition.Mutation;
ga.GenerationRan += GeneticAlgorithmOnGenerationRan;
ga.Termination =
new FitnessStagnationTermination(expectedStagnantGenerationsNumber: 5);
_geneticAlgorithms.Add(key: Key, value: ga);
}
static void Main(string[] args)
{
#region Excel Input
FileInfo InputFile = new FileInfo(@"D:\InputPLASH.xlsx");
List <double> InputPrecipUp = new List <double>();
List <double> InputPrecipDown = new List <double>();
List <double> InputQObs = new List <double>();
List <double> InputEvap = new List <double>();
using (ExcelPackage package = new ExcelPackage(InputFile))
{
ExcelWorksheet worksheet = package.Workbook.Worksheets[1];
int ColCount = worksheet.Dimension.End.Column;
int RowCount = worksheet.Dimension.End.Row;
for (int row = 2; row <= RowCount; row++)
{
InputPrecipUp.Add(Convert.ToDouble(worksheet.Cells[row, 2].Value));
InputPrecipDown.Add(Convert.ToDouble(worksheet.Cells[row, 3].Value));
InputQObs.Add(Convert.ToDouble(worksheet.Cells[row, 4].Value));
InputEvap.Add(Convert.ToDouble(worksheet.Cells[row, 5].Value));
//Console.WriteLine("PrecipUp: {0}, PrecipDown: {1}, Evap: {2}, QObos: {3}", Math.Round(InputPrecipUp[row - 2],3), Math.Round(InputPrecipDown[row - 2], 3), Math.Round(InputQObs[row - 2],3), Math.Round(InputEvap[row - 2],3));
}
}
#endregion Excel Input
#region PLASH Simulation
#region Genetic Algorithm
int SimulationLength = InputPrecipUp.Count;
double Timestep = 24;
var chromosome = new FloatingPointChromosome(
new double[] { 0, 0, 2, 24, 0, //Param Upstream
0, 0, 0, 120, 6,
0, 0, 0, 0,
0, 0, 0, 0, //Initial Upstream
0, 0, 2, 24, 0, //Param Downstream
0, 0, 0, 120, 6,
0, 0, 0, 0,
0, 0, 0, 0, //Initial Downstream
12, 0.01 //Param Muskingum
},
new double[] {
10, 10, 10, 240, 300,
0.5, 0.5, 10, 3600, 120,
3, 500, 1, 1,
10, 10, 10, 10,
10, 10, 10, 240, 300,
0.5, 0.5, 10, 3600, 120,
20, 200, 1, 1,
10, 10, 10, 10,
180, 0.5
},
new int[] { 64, 64, 64, 64, 64,
64, 64, 64, 64, 64,
64, 64, 64, 64,
64, 64, 64, 64,
64, 64, 64, 64, 64,
64, 64, 64, 64, 64,
64, 64, 64, 64,
64, 64, 64, 64,
64, 64 },
new int[] { 3, 3, 3, 3, 3,
3, 3, 3, 3, 3,
3, 3, 3, 3,
3, 3, 3, 3,
3, 3, 3, 3, 3,
3, 3, 3, 3, 3,
3, 3, 3, 3,
3, 3, 3, 3,
3, 3 });
var population = new Population(50, 100, chromosome);
var fitness = new FuncFitness((c) =>
{
var fc = c as FloatingPointChromosome;
var values = fc.ToFloatingPoints();
if (values[12] < values[13] || values[30] < values[31])
{
return(double.NegativeInfinity);
}
DateTime[] TimeSeries = new DateTime[SimulationLength];
TimeSeries[0] = new DateTime(DateTime.Now.Year, DateTime.Now.Month, DateTime.Now.Day);
for (int i = 1; i < TimeSeries.Length; i++)
{
TimeSeries[i] = TimeSeries[0].AddHours(Timestep * i);
}
PLASHInput InputUp = new PLASHInput
{
DTE_Arr_TimeSeries = TimeSeries,
FLT_Arr_PrecipSeries = InputPrecipUp.ToArray(),
FLT_Arr_EPSeries = InputEvap.ToArray(),
FLT_Arr_QtObsSeries = InputQObs.ToArray(),
FLT_Arr_QtUpstream = new double[SimulationLength]
};
PLASHParameters ParamUp = new PLASHParameters
{
FLT_AD = 861.42252,
FLT_AI = 0.02,
FLT_AP = 0.95,
FLT_TimeStep = 24,
FLT_DI = values[0],
FLT_IP = values[1],
FLT_DP = values[2],
FLT_KSup = values[3],
FLT_CS = values[4],
FLT_CC = values[5],
FLT_CR = values[6],
FLT_PP = values[7],
FLT_KSub = values[8],
FLT_KCan = values[9],
FLT_CH = values[10],
FLT_FS = values[11],
FLT_PS = values[12],
FLT_UI = values[13]
};
PLASHInitialConditions Res0Up = new PLASHInitialConditions()
{
RImp0 = values[14],
RInt0 = values[15],
RSup0 = values[16],
RCan0 = values[17]
};
PLASHReservoir ReservoirUp = new PLASHReservoir();
PLASHOutput OutputUp = new PLASHOutput();
PLASH.Run(InputUp, ParamUp, Res0Up, ReservoirUp, OutputUp);
Muskingum Musk = new Muskingum()
{
FLT_K = values[36],
FLT_X = values[37],
FLT_Timestep = Timestep,
FLT_Arr_InputFlow = OutputUp.FLT_Arr_Qt_Calc
};
PLASHInput InputDown = new PLASHInput()
{
DTE_Arr_TimeSeries = TimeSeries,
FLT_Arr_PrecipSeries = InputPrecipDown.ToArray(),
FLT_Arr_EPSeries = InputEvap.ToArray(),
FLT_Arr_QtObsSeries = InputQObs.ToArray(),
FLT_Arr_QtUpstream = Muskingum.ProcessDamping(Musk)
};
PLASHParameters ParamDown = new PLASHParameters
{
FLT_AD = 727.8917, //Watershed Area (km2)
FLT_AI = 0.02, //Impervious Area Fraction (km2/km2)
FLT_AP = 0.95, //Pervious Area Fraction (km2/km2)
FLT_TimeStep = 24,
FLT_DI = values[18], //Maximum Impervious Detention (mm)
FLT_IP = values[19], //Maximum Interception (mm)
FLT_DP = values[20], //Maximum Pervious Detention (mm)
FLT_KSup = values[21], //Surface Reservoir Decay (h)
FLT_CS = values[22], //Soil Saturation Capacity (mm)
FLT_CC = values[23], //Field Capacity (%)
FLT_CR = values[24], //Recharge Capacity (%)
FLT_PP = values[25], //Deep Percolation (mm/h)
FLT_KSub = values[26], //Aquifer Reservoir Decay (d)
FLT_KCan = values[27], //Channel Reservoir Decay (h)
FLT_CH = values[28], //Hydraulic Conductivity (mm/h)
FLT_FS = values[29], //Soil Capilarity Factor (mm)
FLT_PS = values[30], //Soil Porosity (cm3/cm3)
FLT_UI = values[31] //Initial Moisture (cm3/cm3)
};
PLASHInitialConditions Res0Down = new PLASHInitialConditions()
{
RImp0 = values[32],
RInt0 = values[33],
RSup0 = values[34],
RCan0 = values[35]
};
PLASHReservoir ReservoirDown = new PLASHReservoir();
PLASHOutput OutputDown = new PLASHOutput();
PLASH.Run(InputDown, ParamDown, Res0Down, ReservoirDown, OutputDown);
if (ReservoirDown.FLT_Arr_ESSup.Sum() < 30 || ReservoirUp.FLT_Arr_ESSup.Sum() < 30)
{
return(double.NegativeInfinity);
}
//double objectiveNashSut = 1;
//double MeanSimulatedFlow = OutputDown.FLT_Arr_Qt_Calc.Average();
//double NashSutUpper = 0;
//double NashSutLower = 0;
//for(int i = 0; i < OutputDown.FLT_Arr_Qt_Calc.Length; i++)
//{
// NashSutUpper += Math.Pow(OutputDown.FLT_Arr_Qt_Calc[i] - InputDown.FLT_Arr_QtObsSeries[i], 2);
// NashSutLower += Math.Pow(InputDown.FLT_Arr_QtObsSeries[i] - MeanSimulatedFlow, 2);
//}
//objectiveNashSut -= (NashSutUpper / NashSutLower);
double objectiveSquareSum = 0;
for (int i = 0; i < OutputDown.FLT_Arr_Qt_Calc.Length; i++)
{
objectiveSquareSum += Math.Pow(OutputDown.FLT_Arr_Qt_Calc[i] - InputDown.FLT_Arr_QtObsSeries[i], 2);
}
//double objectiveAbsSum = 0;
//for(int i = 0; i < OutputDown.FLT_Arr_Qt_Calc.Length; i++)
//{
// objectiveAbsSum += Math.Abs(OutputDown.FLT_Arr_Qt_Calc[i] - InputDown.FLT_Arr_QtObsSeries[i]);
//}
//return objectiveAbsSum * -1;
return(objectiveSquareSum * -1);
//return objectiveNashSut;
});
var selection = new EliteSelection();
var crossover = new UniformCrossover(0.3f);
var mutation = new FlipBitMutation();
var termination = new FitnessStagnationTermination(250);
var ga = new GeneticAlgorithm(
population,
fitness,
selection,
crossover,
mutation);
ga.Termination = termination;
//Console.WriteLine("Generation: (x1, y1), (x2, y2) = distance");
Console.WriteLine("Genetic algorithm tests");
var latestFitness = 0.0;
ga.GenerationRan += (sender, e) =>
{
var bestChromosome = ga.BestChromosome as FloatingPointChromosome;
var bestFitness = bestChromosome.Fitness.Value;
if (bestFitness != latestFitness)
{
latestFitness = bestFitness;
var phenotype = bestChromosome.ToFloatingPoints();
Console.WriteLine("Generation {0}: {1}",
ga.GenerationsNumber,
bestFitness);
//Console.WriteLine(
// "Generation {0,2}: ({1},{2}),({3},{4}) = {5}",
// ga.GenerationsNumber,
// phenotype[0],
// phenotype[1],
// phenotype[2],
// phenotype[3],
// bestFitness
//);
}
};
ga.Start();
Console.WriteLine("GA Over!");
#endregion Genetic Algorithm
var bestChrom = ga.BestChromosome as FloatingPointChromosome;
var bestVal = bestChrom.ToFloatingPoints();
PLASHInput InputUpstream = new PLASHInput()
{
DTE_Arr_TimeSeries = new DateTime[SimulationLength],
FLT_Arr_PrecipSeries = InputPrecipUp.ToArray(),
FLT_Arr_EPSeries = InputEvap.ToArray(),
FLT_Arr_QtObsSeries = InputQObs.ToArray(),
FLT_Arr_QtUpstream = new double[SimulationLength]
};
PLASHParameters ParametersUpstream = new PLASHParameters()
{
FLT_AD = 861.42252, //Watershed Area (km2)
FLT_AI = 0.02, //Impervious Area Fraction (km2/km2)
FLT_AP = 0.95, //Pervious Area Fraction (km2/km2)
FLT_TimeStep = 24,
//Parameters
FLT_DI = bestVal[0], //Maximum Impervious Detention (mm)
FLT_IP = bestVal[1], //Maximum Interception (mm)
FLT_DP = bestVal[2], //Maximum Pervious Detention (mm)
FLT_KSup = bestVal[3], //Surface Reservoir Decay (h)
FLT_CS = bestVal[4], //Soil Saturation Capacity (mm)
FLT_CC = bestVal[5], //Field Capacity (%)
FLT_CR = bestVal[6], //Recharge Capacity (%)
FLT_PP = bestVal[7], //Deep Percolation (mm/h)
FLT_KSub = bestVal[8], //Aquifer Reservoir Decay (d)
FLT_KCan = bestVal[9], //Channel Reservoir Decay (h)
FLT_CH = bestVal[10], //Hydraulic Conductivity (mm/h)
FLT_FS = bestVal[11], //Soil Capilarity Factor (mm)
FLT_PS = bestVal[12], //Soil Porosity (cm3/cm3)
FLT_UI = bestVal[13] //Initial Moisture (cm3/cm3)
};
InputUpstream.DTE_Arr_TimeSeries[0] = new DateTime(DateTime.Now.Year, DateTime.Now.Month, DateTime.Now.Day);
for (int i = 1; i < InputUpstream.DTE_Arr_TimeSeries.Length; i++)
{
InputUpstream.DTE_Arr_TimeSeries[i] = InputUpstream.DTE_Arr_TimeSeries[0].AddHours(ParametersUpstream.FLT_TimeStep * i);
}
PLASHReservoir ReservoirUpstream = new PLASHReservoir();
PLASHInitialConditions InitialUpstream = new PLASHInitialConditions()
{
RImp0 = bestVal[14],
RInt0 = bestVal[15],
RSup0 = bestVal[16],
RCan0 = bestVal[17]
};
PLASHOutput OutputUpstream = new PLASHOutput();
PLASH.Run(InputUpstream, ParametersUpstream, InitialUpstream, ReservoirUpstream, OutputUpstream);
PLASHInput InputDownstream = new PLASHInput()
{
DTE_Arr_TimeSeries = new DateTime[SimulationLength],
FLT_Arr_PrecipSeries = InputPrecipDown.ToArray(),
FLT_Arr_EPSeries = InputEvap.ToArray(),
FLT_Arr_QtObsSeries = InputQObs.ToArray()
};
PLASHParameters ParametersDownstream = new PLASHParameters()
{
FLT_AD = 727.8917, //Watershed Area (km2)
FLT_AI = 0.02, //Impervious Area Fraction (km2/km2)
FLT_AP = 0.95, //Pervious Area Fraction (km2/km2)
FLT_TimeStep = 24,
//Parameters
FLT_DI = bestVal[18], //Maximum Impervious Detention (mm)
FLT_IP = bestVal[19], //Maximum Interception (mm)
FLT_DP = bestVal[20], //Maximum Pervious Detention (mm)
FLT_KSup = bestVal[21], //Surface Reservoir Decay (h)
FLT_CS = bestVal[22], //Soil Saturation Capacity (mm)
FLT_CC = bestVal[23], //Field Capacity (%)
FLT_CR = bestVal[24], //Recharge Capacity (%)
FLT_PP = bestVal[25], //Deep Percolation (mm/h)
FLT_KSub = bestVal[26], //Aquifer Reservoir Decay (d)
FLT_KCan = bestVal[27], //Channel Reservoir Decay (h)
FLT_CH = bestVal[28], //Hydraulic Conductivity (mm/h)
FLT_FS = bestVal[29], //Soil Capilarity Factor (mm)
FLT_PS = bestVal[30], //Soil Porosity (cm3/cm3)
FLT_UI = bestVal[31] //Initial Moisture (cm3/cm3)
};
InputDownstream.DTE_Arr_TimeSeries[0] = new DateTime(DateTime.Now.Year, DateTime.Now.Month, DateTime.Now.Day);
for (int i = 1; i < InputDownstream.DTE_Arr_TimeSeries.Length; i++)
{
InputDownstream.DTE_Arr_TimeSeries[i] = InputDownstream.DTE_Arr_TimeSeries[0].AddHours(ParametersDownstream.FLT_TimeStep * i);
}
PLASHReservoir ReservoirDownstream = new PLASHReservoir();
PLASHInitialConditions InitialDownstream = new PLASHInitialConditions()
{
RImp0 = bestVal[32],
RInt0 = bestVal[33],
RSup0 = bestVal[34],
RCan0 = bestVal[35]
};
PLASHOutput OutputDownstream = new PLASHOutput();
Muskingum DampenedUpstream = new Muskingum()
{
FLT_K = bestVal[36],
FLT_X = bestVal[37],
FLT_Timestep = 24,
FLT_Arr_InputFlow = OutputUpstream.FLT_Arr_Qt_Calc
};
DampenedUpstream.FLT_Arr_OutputFlow = Muskingum.ProcessDamping(DampenedUpstream);
InputDownstream.FLT_Arr_QtUpstream = DampenedUpstream.FLT_Arr_OutputFlow;
PLASH.Run(InputDownstream, ParametersDownstream, InitialDownstream, ReservoirDownstream, OutputDownstream);
Console.ReadKey();
//Console.WriteLine("");
//Console.ReadKey();
#endregion PLASH Simulation
#region Buwo Simulation
List <Buildup_Washoff> UsesUpstream = Buildup_Washoff.BuwoUpstreamList(Timestep, ReservoirUpstream.FLT_Arr_ESSup);
List <Buildup_Washoff> UsesDownstream = Buildup_Washoff.BuwoDownstreamList(Timestep, ReservoirDownstream.FLT_Arr_ESSup);
foreach (Buildup_Washoff Use in UsesUpstream)
{
Buildup_Washoff.fncBuildupWashoffProcess(Use);
}
Buildup_Washoff BuwoUpstream = Buildup_Washoff.AggregateUses(UsesUpstream, 863.178D);
foreach (Buildup_Washoff Use in UsesDownstream)
{
Buildup_Washoff.fncBuildupWashoffProcess(Use);
}
Buildup_Washoff BuwoDownstream = Buildup_Washoff.AggregateUses(UsesDownstream, 729.018D);
Buildup_Washoff Aggregate = Buildup_Washoff.Transport(BuwoUpstream, BuwoDownstream);
#endregion Buwo Simulation
//#region Excel Output
using (ExcelPackage excel = new ExcelPackage())
{
excel.Workbook.Worksheets.Add("Param_PLASHUp");
excel.Workbook.Worksheets.Add("Param_PLASHDown");
excel.Workbook.Worksheets.Add("PLASHReservoirUp");
excel.Workbook.Worksheets.Add("PLASHReservoirDown");
excel.Workbook.Worksheets.Add("PLASHInitialUp");
excel.Workbook.Worksheets.Add("PLASHInitialDown");
excel.Workbook.Worksheets.Add("Results_PLASHUp");
excel.Workbook.Worksheets.Add("Results_PLASHDown");
excel.Workbook.Worksheets.Add("Param_Muskingum");
excel.Workbook.Worksheets.Add("Results_Muskingum");
excel.Workbook.Worksheets.Add("Results_BuWo");
#region Parameters
var HeaderRowPLASHParam = new List <string[]>()
{
new string[]
{
"DI", "IP", "DP", "KSup", "CS", "CC", "CR", "PP", "Ksub", "KCan", "CH", "FS", "PS", "UI"
}
};
string headerRangePLASHParam = "A1:" + Char.ConvertFromUtf32(HeaderRowPLASHParam[0].Length + 64) + 1;
var worksheet = excel.Workbook.Worksheets["Param_PLASHUp"];
worksheet.Cells[headerRangePLASHParam].LoadFromArrays(HeaderRowPLASHParam);
List <object[]> cellDataPLASHParamUP = new List <object[]>();
cellDataPLASHParamUP.Add(new object[]
{
ParametersUpstream.FLT_DI, ParametersUpstream.FLT_IP, ParametersUpstream.FLT_DP, ParametersUpstream.FLT_KSup, ParametersUpstream.FLT_CS,
ParametersUpstream.FLT_CC, ParametersUpstream.FLT_CR, ParametersUpstream.FLT_PP, ParametersUpstream.FLT_KSub, ParametersUpstream.FLT_KCan,
ParametersUpstream.FLT_CH, ParametersUpstream.FLT_FS, ParametersUpstream.FLT_PS, ParametersUpstream.FLT_UI
});
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHParamUP);
worksheet = excel.Workbook.Worksheets["Param_PLASHDown"];
worksheet.Cells[headerRangePLASHParam].LoadFromArrays(HeaderRowPLASHParam);
List <object[]> cellDataPLASHParamDown = new List <object[]>();
cellDataPLASHParamDown.Add(new object[]
{
ParametersDownstream.FLT_DI, ParametersDownstream.FLT_IP, ParametersDownstream.FLT_DP, ParametersDownstream.FLT_KSup, ParametersDownstream.FLT_CS,
ParametersDownstream.FLT_CC, ParametersDownstream.FLT_CR, ParametersDownstream.FLT_PP, ParametersDownstream.FLT_KSub, ParametersDownstream.FLT_KCan,
ParametersDownstream.FLT_CH, ParametersDownstream.FLT_FS, ParametersDownstream.FLT_PS, ParametersDownstream.FLT_UI
});
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHParamDown);
#endregion Parameters
#region Reservoir
var HeaderRowPLASHreservoir = new List <string[]>()
{
new string[]
{
"ImpRes", "ImpEvap", "ImpFlow",
"IntRes", "IntEvap", "IntFlow",
"SurfRes", "SurfEvap", "SurfFlow",
"Inf", "InfCum", "IAE", "TP", "IAEAdim", "TPAdim",
"SoilRes", "SoilEvap", "SoilUpFlow", "SoilDownFlow",
"AquiRes", "AquiPerc", "AquiFlow",
"ChanRes", "ChanEvap", "ChanUpFlow", "ChanDownFlow"
}
};
string HeaderRangePLASHReservoir = "A1:" + Char.ConvertFromUtf32(HeaderRowPLASHreservoir[0].Length + 64) + 1;
worksheet = excel.Workbook.Worksheets["PLASHReservoirUp"];
worksheet.Cells[HeaderRangePLASHReservoir].LoadFromArrays(HeaderRowPLASHreservoir);
List <object[]> cellDataPLASHResUp = new List <object[]>();
for (int i = 0; i < SimulationLength; i++)
{
cellDataPLASHResUp.Add(new object[]
{
ReservoirUpstream.FLT_Arr_RImp[i], ReservoirUpstream.FLT_Arr_ERImp[i], ReservoirUpstream.FLT_Arr_ESImp[i],
ReservoirUpstream.FLT_Arr_RInt[i], ReservoirUpstream.FLT_Arr_ERInt[i], ReservoirUpstream.FLT_Arr_ESInt[i],
ReservoirUpstream.FLT_Arr_RSup[i], ReservoirUpstream.FLT_Arr_ERSup[i], ReservoirUpstream.FLT_Arr_ESSup[i],
ReservoirUpstream.FLT_Arr_Infiltration[i], ReservoirUpstream.FLT_Arr_Infiltration_Cumulative[i], ReservoirUpstream.FLT_Arr_IAE[i], ReservoirUpstream.FLT_Arr_TP[i], ReservoirUpstream.FLT_Arr_IAEAdim[i], ReservoirUpstream.FLT_Arr_TPAdim[i],
ReservoirUpstream.FLT_Arr_RSol[i], ReservoirUpstream.FLT_Arr_ERSol[i], ReservoirUpstream.FLT_Arr_EESol[i], ReservoirUpstream.FLT_Arr_ESSol[i],
ReservoirUpstream.FLT_Arr_RSub[i], ReservoirUpstream.FLT_Arr_PPSub[i], ReservoirUpstream.FLT_Arr_EESub[i],
ReservoirUpstream.FLT_Arr_RCan[i], ReservoirUpstream.FLT_Arr_ERCan[i], ReservoirUpstream.FLT_Arr_EECan[i], ReservoirUpstream.FLT_ARR_ESCan[i]
});
}
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHResUp);
worksheet = excel.Workbook.Worksheets["PLASHReservoirDown"];
worksheet.Cells[HeaderRangePLASHReservoir].LoadFromArrays(HeaderRowPLASHreservoir);
List <object[]> cellDataPLASHResDown = new List <object[]>();
for (int i = 0; i < SimulationLength; i++)
{
cellDataPLASHResDown.Add(new object[]
{
ReservoirDownstream.FLT_Arr_RImp[i], ReservoirDownstream.FLT_Arr_ERImp[i], ReservoirDownstream.FLT_Arr_ESImp[i],
ReservoirDownstream.FLT_Arr_RInt[i], ReservoirDownstream.FLT_Arr_ERInt[i], ReservoirDownstream.FLT_Arr_ESInt[i],
ReservoirDownstream.FLT_Arr_RSup[i], ReservoirDownstream.FLT_Arr_ERSup[i], ReservoirDownstream.FLT_Arr_ESSup[i],
ReservoirDownstream.FLT_Arr_Infiltration[i], ReservoirDownstream.FLT_Arr_Infiltration_Cumulative[i], ReservoirDownstream.FLT_Arr_IAE[i], ReservoirDownstream.FLT_Arr_TP[i], ReservoirDownstream.FLT_Arr_IAEAdim[i], ReservoirDownstream.FLT_Arr_TPAdim[i],
ReservoirDownstream.FLT_Arr_RSol[i], ReservoirDownstream.FLT_Arr_ERSol[i], ReservoirDownstream.FLT_Arr_EESol[i], ReservoirDownstream.FLT_Arr_ESSol[i],
ReservoirDownstream.FLT_Arr_RSub[i], ReservoirDownstream.FLT_Arr_PPSub[i], ReservoirDownstream.FLT_Arr_EESub[i],
ReservoirDownstream.FLT_Arr_RCan[i], ReservoirDownstream.FLT_Arr_ERCan[i], ReservoirDownstream.FLT_Arr_EECan[i], ReservoirDownstream.FLT_ARR_ESCan[i]
});
}
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHResDown);
#endregion Reservoir
#region Initial Conditions
var HeaderRowPLASHInitial = new List <string[]>()
{
new string[] { "RImp0", "RInt0", "RSup0", "RCan0" }
};
string headerRangePLASHInitial = "A1:" + Char.ConvertFromUtf32(HeaderRowPLASHInitial[0].Length + 64) + 1;
worksheet = excel.Workbook.Worksheets["PLASHInitialUp"];
worksheet.Cells[headerRangePLASHInitial].LoadFromArrays(HeaderRowPLASHInitial);
List <object[]> cellDataPLASHInitialUp = new List <object[]>();
cellDataPLASHInitialUp.Add(new object[] { InitialUpstream.RImp0, InitialUpstream.RInt0, InitialUpstream.RSup0, InitialUpstream.RCan0 });
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHInitialUp);
worksheet = excel.Workbook.Worksheets["PLASHInitialDown"];
worksheet.Cells[headerRangePLASHInitial].LoadFromArrays(HeaderRowPLASHInitial);
List <object[]> cellDataPLASHInitialDown = new List <object[]>();
cellDataPLASHInitialDown.Add(new object[] { InitialDownstream.RImp0, InitialDownstream.RInt0, InitialDownstream.RSup0, InitialDownstream.RCan0 });
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHInitialDown);
#endregion Initial Conditions
#region Results
//PLASH Upstream
var HeaderRowPLASH = new List <string[]>()
{
new string[] { "Precipitation", "Evapotranspiration", "Observed Flow",
"Impervious Reservoir", "Interception Reservoir", "Surface Reservoir", "Soil Reservoir", "Aquifer Reservoir", "Channel Reservoir",
"Calculated Basic Flow", "Calculated Surface Flow", "Calculated Total Flow" },
};
string headerRangePLASH = "A1:" + Char.ConvertFromUtf32(HeaderRowPLASH[0].Length + 64) + 1;
worksheet = excel.Workbook.Worksheets["Results_PLASHUp"];
worksheet.Cells[headerRangePLASH].LoadFromArrays(HeaderRowPLASH);
List <object[]> cellDataPLASHUp = new List <object[]>();
for (int i = 0; i < SimulationLength; i++)
{
cellDataPLASHUp.Add(new object[] { InputUpstream.FLT_Arr_PrecipSeries[i], InputUpstream.FLT_Arr_EPSeries[i], InputUpstream.FLT_Arr_QtObsSeries[i],
ReservoirUpstream.FLT_Arr_RImp[i], ReservoirUpstream.FLT_Arr_RInt[i], ReservoirUpstream.FLT_Arr_RSup[i], ReservoirUpstream.FLT_Arr_RSol[i], ReservoirUpstream.FLT_Arr_RSub[i], ReservoirUpstream.FLT_Arr_RCan[i],
OutputUpstream.FLT_Arr_QBas_Calc[i], OutputUpstream.FLT_Arr_QSup_Calc[i], OutputUpstream.FLT_Arr_Qt_Calc[i] });
}
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHUp);
//PLASH Downstream
worksheet = excel.Workbook.Worksheets["Results_PLASHDown"];
worksheet.Cells[headerRangePLASH].LoadFromArrays(HeaderRowPLASH);
List <object[]> cellDataPLASHDown = new List <object[]>();
for (int i = 0; i < SimulationLength; i++)
{
cellDataPLASHDown.Add(new object[] { InputDownstream.FLT_Arr_PrecipSeries[i], InputDownstream.FLT_Arr_EPSeries[i], InputDownstream.FLT_Arr_QtObsSeries[i],
ReservoirDownstream.FLT_Arr_RImp[i], ReservoirDownstream.FLT_Arr_RInt[i], ReservoirDownstream.FLT_Arr_RSup[i], ReservoirDownstream.FLT_Arr_RSol[i], ReservoirDownstream.FLT_Arr_RSub[i], ReservoirDownstream.FLT_Arr_RCan[i],
OutputDownstream.FLT_Arr_QBas_Calc[i], OutputDownstream.FLT_Arr_QSup_Calc[i], OutputDownstream.FLT_Arr_Qt_Calc[i] });
}
worksheet.Cells[2, 1].LoadFromArrays(cellDataPLASHDown);
#endregion Results
#region Muskingum Parameters
//Muskingum
var HeaderRowMuskingumParam = new List <string[]>()
{
new string[] { "K", "X" },
};
string headerRangeMuskingumParam = "A1:" + Char.ConvertFromUtf32(HeaderRowMuskingumParam[0].Length + 64) + 1;
worksheet = excel.Workbook.Worksheets["Param_Muskingum"];
worksheet.Cells[headerRangeMuskingumParam].LoadFromArrays(HeaderRowMuskingumParam);
List <object[]> cellDataMuskingumParam = new List <object[]>();
cellDataMuskingumParam.Add(new object[] { DampenedUpstream.FLT_K, DampenedUpstream.FLT_X });
worksheet.Cells[2, 1].LoadFromArrays(cellDataMuskingumParam);
#endregion Muskingum Parameters
#region Muskingum
var HeaderRowMuskingum = new List <string[]>()
{
new string[] { "Upstream Flow", "Downstream Flow" },
};
string headerRangeMuskingum = "A1:" + Char.ConvertFromUtf32(HeaderRowMuskingum[0].Length + 64) + 1;
worksheet = excel.Workbook.Worksheets["Results_Muskingum"];
worksheet.Cells[headerRangeMuskingum].LoadFromArrays(HeaderRowMuskingum);
List <object[]> cellDataMuskingum = new List <object[]>();
for (int i = 0; i < SimulationLength; i++)
{
cellDataMuskingum.Add(new object[] { DampenedUpstream.FLT_Arr_InputFlow[i], DampenedUpstream.FLT_Arr_OutputFlow[i] });
}
worksheet.Cells[2, 1].LoadFromArrays(cellDataMuskingum);
#endregion Muskingum
#region BuWo
//Buwo
var HeaderRow2 = new List <string[]>()
{
new string[] { "Precipitation", "Surface Flow", "Buildup", "Effective Washoff" },
};
string headerRange2 = "A1:" + Char.ConvertFromUtf32(HeaderRow2[0].Length + 64) + 1;
worksheet = excel.Workbook.Worksheets["Results_BuWo"];
worksheet.Cells[headerRange2].LoadFromArrays(HeaderRow2);
List <object[]> cellDataBuwo = new List <object[]>();
for (int i = 0; i < SimulationLength; i++)
{
cellDataBuwo.Add(new object[] { InputUpstream.FLT_Arr_PrecipSeries[i], ReservoirDownstream.FLT_Arr_ESSup[i], Aggregate.FLT_Arr_Buildup[i], Aggregate.FLT_Arr_EffectiveWashoff[i] });
}
worksheet.Cells[2, 1].LoadFromArrays(cellDataBuwo);
#endregion BuWo
FileInfo excelFile = new FileInfo(@"D:\dataGA.xlsx");
excel.SaveAs(excelFile);
}
////Console.WriteLine("Excel processed");
//#endregion Excel Output
//Console.ReadKey();
}
private static void Main(string[] args)
{
Console.WriteLine("Podaj nazwę plkiu z danymi(bez rozszerzenia)");
string fileName = Console.ReadLine();
string[] lines = File.ReadAllLines(Directory.GetCurrentDirectory() + "\\" + fileName + ".txt");
int edgesNumber = lines.Count();
List <Edge> edges = new List <Edge>();
int edgeIndex = 0;
string stringSeparator = " ";
foreach (string l in lines)
{
int[] values = l.Split(stringSeparator.ToCharArray(), StringSplitOptions.None).Select(s => int.Parse(s)).ToArray();
// Krawędź może być przechodzona w obu kierunkach
edges.Add(new Edge(values[0], values[1], values[2], edgeIndex));
// Ddodawana jest także w odwróconej wersji
edges.Add(new Edge(values[1], values[0], values[2], edgeIndex));
//Krawędź i jej odwrócona wersja mają ten sam indeks(dla łatwiejszego odnajdowania)
edgeIndex++;
}
EliteSelection selection = new EliteSelection();
ThreeParentCrossover crossover = new ThreeParentCrossover();
TworsMutation mutation = new TworsMutation();
FitnessFunction fitness = new FitnessFunction(edges);
Chromosome chromosome = new Chromosome(4 * edgesNumber, edges);
Population population = new Population(200, 400, chromosome);
GeneticAlgorithm ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation)
{
Termination = new GenerationNumberTermination(400)
};
Stopwatch timer = new Stopwatch();
timer.Start();
Console.WriteLine("START!");
ga.Start();
timer.Stop();
Chromosome bestChromosome = ga.BestChromosome as Chromosome;
int currentEdgeIndex = int.Parse(bestChromosome.GetGene(0).Value.ToString());
Edge currentEdge = edges[currentEdgeIndex];
int startVertex = currentEdge.VertexA;
int totalCost = currentEdge.Cost;
string verticesSequence = currentEdge.VertexA + "-" + currentEdge.VertexB;
Console.WriteLine("Funkcja dopasowania najlepszego rozwiązania wynosi: {0}", bestChromosome.Fitness);
for (int i = 1; i < bestChromosome.Length; i++)
{
currentEdgeIndex = int.Parse(bestChromosome.GetGene(i).Value.ToString());
currentEdge = edges[currentEdgeIndex];
currentEdge.Visited = true;
edges.SingleOrDefault(e => e.VertexA == currentEdge.VertexB && e.VertexB == currentEdge.VertexA).Visited = true;
totalCost += currentEdge.Cost;
verticesSequence += "-" + currentEdge.VertexB;
if (FitnessFunction.AllEdgesVisited(edges))
{
if (currentEdge.VertexB == startVertex)
{
break;
}
Edge possibleEdge = edges.SingleOrDefault(e => e.VertexA == currentEdge.VertexB && e.VertexB == startVertex);
if (possibleEdge != null)
{
totalCost += possibleEdge.Cost;
verticesSequence += "-" + possibleEdge.VertexB;
break;
}
}
}
Console.WriteLine("Ścieżka: {0}", verticesSequence);
Console.WriteLine("Koszt najlepszego rozwiązania: {0}", totalCost);
Console.WriteLine("Czas wykonania: {0}", timer.Elapsed.ToString(@"hh\:mm\:ss\:ff"));
Console.ReadKey();
}
public Core.Sudoku SolveGeneticSharp(Core.Sudoku s)
{
var populationSize = 10000;
IChromosome sudokuChromosome = new SudokuPermutationsChromosome(s);
//IChromosome sudokuChromosome = new SudokuCellsChromosome(s);
var fitnessThreshold = 0;
//var generationNb = 50;
var crossoverProbability = 0.75f;
var mutationProbability = 0.2f;
var fitness = new SudokuFitness(s);
var selection = new EliteSelection();
var crossover = new UniformCrossover();
var mutation = new UniformMutation();
IChromosome bestIndividual;
var solution = s;
int nbErrors = int.MaxValue;
do
{
var population = new Population(populationSize, populationSize, sudokuChromosome);
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation)
{
Termination = new OrTermination(new ITermination[]
{
new FitnessThresholdTermination(fitnessThreshold),
new FitnessStagnationTermination(10),
//new GenerationNumberTermination(generationNb)
//new TimeEvolvingTermination(TimeSpan.FromSeconds(10)),
}),
MutationProbability = mutationProbability,
CrossoverProbability = crossoverProbability,
OperatorsStrategy = new TplOperatorsStrategy(),
};
ga.GenerationRan += delegate(object sender, EventArgs args)
{
bestIndividual = (ga.Population.BestChromosome);
solution = ((ISudokuChromosome)bestIndividual).GetSudokus()[0];
nbErrors = solution.NbErrors(s);
Console.WriteLine($"Generation #{ga.GenerationsNumber}: best individual has {nbErrors} errors");
};
ga.Start();
//bestIndividual = (ga.Population.BestChromosome);
//solution = ((ISudokuChromosome)bestIndividual).GetSudokus()[0];
//nbErrors = solution.NbErrors(s);
if (nbErrors == 0)
{
break;
}
else
{
populationSize *= 2;
Console.WriteLine($"Genetic search failed with {nbErrors} resulting errors, doubling population to {populationSize}");
}
} while (true);
return(solution);
}
public GeneticOptimizer(GeneticOptimizerConfiguration configuration, Func <double[], double> objectiveFunction, Action <string> generationRanCallback = null)
{
//store configuration
_configuration = configuration;
_generationRanCallback = generationRanCallback;
//set min/max/precision of input variables
var minValues = new double[_configuration.Variables.Count];
var maxValues = new double[_configuration.Variables.Count];
var fractionDigits = new int[_configuration.Variables.Count];
for (int index = 0; index < _configuration.Variables.Count; index++)
{
minValues[index] = _configuration.Variables[index].MinimumValue;
maxValues[index] = _configuration.Variables[index].MaximumValue;
fractionDigits[index] = _configuration.Variables[index].NumberDigitsPrecision;
}
//total bits
var totalBits = new int[] { 64 };
//chromosome
var chromosome = new FloatingPointChromosome(minValues, maxValues, totalBits, fractionDigits);
//population
var population = new Population(MinimumNumberPopulation, MaximumNumberPopulation, chromosome);
//set fitness function
var fitnessFunction = new FuncFitness(c =>
{
var fc = c as FloatingPointChromosome;
var inputs = fc.ToFloatingPoints();
var result = objectiveFunction(inputs);
//add to results
if (!Double.IsNaN(result))
{
var list = inputs.ToList();
list.Add(result);
_result.IterationArray.Add(string.Join(",", list));
}
return(result);
});
//other variables
var selection = new EliteSelection();
var crossover = new UniformCrossover(0.5f);
var mutation = new FlipBitMutation();
var termination = new FitnessThresholdTermination();
_algorithm = new GeneticAlgorithm(population, fitnessFunction, selection, crossover, mutation)
{
Termination = termination,
};
//task parallelism
var taskExecutor = new ParallelTaskExecutor();
taskExecutor.MinThreads = 1;
taskExecutor.MaxThreads = _configuration.NumberThreadsToUse;
_algorithm.TaskExecutor = taskExecutor;
//if (_configuration.NumberThreadsToUse > 1)
//{
// var taskExecutor = new ParallelTaskExecutor();
// taskExecutor.MinThreads = 1;
// taskExecutor.MaxThreads = _configuration.NumberThreadsToUse;
// _algorithm.TaskExecutor = taskExecutor;
//}
//register generation ran callback
_algorithm.GenerationRan += AlgorithmOnGenerationRan;
}
public IHttpActionResult Generuj(DateTime poczatek, DateTime koniec, int naGodzine)
{
User = System.Web.HttpContext.Current.User;
int user;
int.TryParse(((ClaimsIdentity)User.Identity).Claims.First(c => c.Type == "Id").Value, out user);
((ClaimsIdentity)User.Identity).Claims.First(c => c.Type == "Admin");
if (((ClaimsIdentity)User.Identity).Claims.First(c => c.Type == "Admin").Value == "false")
{
return(Content(HttpStatusCode.Forbidden, "Brak uprawnień do wykonania zadania!"));
}
Models.Grafik grafik = null;
try
{
if (TRWA_GENEROWANIE)
{
return(Content(HttpStatusCode.Conflict, "Trwa Generowanie"));
}
TRWA_GENEROWANIE = true;
Models.DataBaseEntities db = new Models.DataBaseEntities();
List <Models.Pracownik> pracownicy = db.Pracownik.Where(p => p.GodzinWUmowie > 0 && p.StanowiskoPracownika.Count() > 0).ToList();
List <Models.Stanowisko> stanowiska = db.Stanowisko.Where(s => s.StanowiskoPracownika.Count() > 0).ToList();
// 1 bit dla czy w pracy
// pozostale numer stanowiska
int dlugoscKodonu = (int)Math.Round(Math.Log(stanowiska.Count(), 2), 0, MidpointRounding.AwayFromZero);
int iloscPrzedzialow = (int)Math.Round(koniec.Subtract(poczatek).TotalHours *naGodzine, 0, MidpointRounding.AwayFromZero) * naGodzine;
double przedzial = 60 / naGodzine;
List <double> genMin = new List <double>();
List <double> genMax = new List <double>();
List <int> genBits = new List <int>();
List <int> genDecimal = new List <int>();
for (int i = 0; i < iloscPrzedzialow * pracownicy.Count(); i++)
{
genMin.Add(0);
genMin.Add(0);
genMax.Add(1);
genMax.Add(stanowiska.Count() - 1);
genBits.Add(1);
genBits.Add(dlugoscKodonu);
genDecimal.Add(0);
genDecimal.Add(0);
}
var selection = new EliteSelection();
//var selection = new TournamentSelection(100, true);
var crossover = new UniformCrossover(0.5f);
//var crossover = new ThreeParentCrossover();
//var mutation = new ReverseSequenceMutation();
var mutation = new FlipBitMutation();
//var mutation = new UniformMutation(false);
var fitness = new OcenaGrafiku(pracownicy, stanowiska, naGodzine, poczatek, koniec);
var termination = new GeneticSharp.Domain.Terminations.OrTermination(
new FitnessStagnationTermination(1000), new FitnessThresholdTermination(0), new TimeEvolvingTermination(new TimeSpan(0, 5, 0)));
var chromosome = new FloatingPointChromosome(genMin.ToArray(), genMax.ToArray(), genBits.ToArray(), genDecimal.ToArray());
var population = new Population(1000, 10000, chromosome);
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
ga.Termination = termination;
ga.MutationProbability = 0.20f;
ga.Start();
var wynik = (ga.BestChromosome as FloatingPointChromosome).ToFloatingPoints();
double?x = (ga.BestChromosome as FloatingPointChromosome).Fitness;
DateTime czas = poczatek;
grafik = new Models.Grafik();
grafik.Poczatek = poczatek;
grafik.Koniec = koniec;
db.Grafik.Add(grafik);
int j = 0;
while (czas < koniec)
{
Models.Czas cz = new Models.Czas();
foreach (Models.Pracownik pr in pracownicy)
{
if (wynik[j] == 1)
{
Models.PracownikNaStanowisku pns = new Models.PracownikNaStanowisku();
pns.Pracownik = pr;
pns.Stanowisko = stanowiska[(int)wynik[j + 1] % stanowiska.Count()];
pns.Czas = cz;
db.PracownikNaStanowisku.Add(pns);
}
j += 2;
}
if (cz.PracownikNaStanowisku.Count() > 0)
{
cz.Grafik = grafik;
cz.Poczatek = czas;
cz.Koniec = czas.AddMinutes(przedzial);
db.Czas.Add(cz);
}
czas = czas.AddMinutes(przedzial);
}
db.SaveChanges();
}
catch (Exception ex)
{
return(Content(HttpStatusCode.InternalServerError, ex.ToString()));
}
TRWA_GENEROWANIE = false;
return(Ok(grafik.Id));
}
static void Main(string[] args)
{
Console.ForegroundColor = ConsoleColor.DarkGreen;
Console.WriteLine("GeneticSharp - ConsoleApp");
Console.ResetColor();
Console.WriteLine("Select the sample:");
Console.WriteLine("1) TSP (Travelling Salesman Problem)");
Console.WriteLine("2) Ghostwriter");
var sampleNumber = Console.ReadLine();
ISampleController sampleController = null;
switch (sampleNumber)
{
case "1":
sampleController = new TspSampleController(20);
break;
case "2":
sampleController = new GhostwriterSampleController();
break;
default:
return;
}
var selection = new EliteSelection();
var crossover = new UniformCrossover();
var mutation = new UniformMutation(true);
var fitness = sampleController.CreateFitness();
var population = new Population(50, 70, sampleController.CreateChromosome());
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
ga.MutationProbability = 0.4f;
ga.Termination = new FitnessStagnationTermination(100);
ga.TaskExecutor = new SmartThreadPoolTaskExecutor()
{
MinThreads = 25,
MaxThreads = 50
};
ga.GenerationRan += delegate
{
Console.CursorLeft = 0;
Console.CursorTop = 5;
var bestChromosome = ga.Population.BestChromosome;
Console.WriteLine("Generations: {0}", ga.Population.GenerationsNumber);
Console.WriteLine("Fitness: {0:n4}", bestChromosome.Fitness);
Console.WriteLine("Time: {0}", ga.TimeEvolving);
sampleController.Draw(bestChromosome);
};
try
{
ga.Start();
}
catch (Exception ex)
{
Console.ForegroundColor = ConsoleColor.DarkRed;
Console.WriteLine();
Console.WriteLine("Error: {0}", ex.Message);
Console.ResetColor();
Console.ReadKey();
return;
}
Console.ForegroundColor = ConsoleColor.DarkGreen;
Console.WriteLine();
Console.WriteLine("Evolved.");
Console.ResetColor();
Console.ReadKey();
}
/**
*
* Genetic Algorithm with Revit API
*
* This sample is using GeneticSharp Library (github.com/giacomelli/GeneticSharp)
* The MIT License (MIT)
* Copyright (c) 2013 Diego Giacomelli
*/
public void GridObjectPlacement(DesignAutomationData data)
{
m_doc = data.RevitDoc;
m_app = data.RevitApp;
InputData inputParameters = JsonConvert.DeserializeObject <InputData>(File.ReadAllText("params.json"));
// Family Symbol
Document familyProjectDoc = m_app.OpenDocumentFile("family.rvt");
string tempFamilyName = Path.GetFileNameWithoutExtension(inputParameters.FamilyFileName) + ".rfa";
ModelPath tempFamilyModelPath = ModelPathUtils.ConvertUserVisiblePathToModelPath(tempFamilyName);
FamilySymbol tempFamilySymbol = null;
FilteredElementCollector familyInstanceCollector
= new FilteredElementCollector(familyProjectDoc)
.WhereElementIsNotElementType()
.OfCategory(BuiltInCategory.OST_Furniture)
.OfClass(typeof(FamilyInstance));
foreach (Element familyInstanceElem in familyInstanceCollector)
{
FamilyInstance fi = familyInstanceElem as FamilyInstance;
Element superComponent = fi.SuperComponent;
if (superComponent == null)
{
tempFamilySymbol = fi.Symbol;
Family family = tempFamilySymbol.Family;
Document familyDoc = familyProjectDoc.EditFamily(family);
Family loadedFamily = familyDoc.LoadFamily(m_doc);
ISet <ElementId> familySymbolIds = loadedFamily.GetFamilySymbolIds();
foreach (ElementId familySymbolId in familySymbolIds)
{
FamilySymbol familySymbol = m_doc.GetElement(familySymbolId) as FamilySymbol;
m_familySymbol = familySymbol;
}
break;
}
}
if (!m_familySymbol.IsActive)
{
using (Transaction tx = new Transaction(m_doc))
{
tx.Start("Transaction Activate Family Symbol");
m_familySymbol.Activate();
tx.Commit();
}
}
// Room
m_room = m_doc.GetElement(inputParameters.RoomUniqueId) as Room;
// Level
m_level = m_doc.GetElement(m_room.LevelId) as Level;
// View
ElementId viewId = m_level.FindAssociatedPlanViewId();
if (viewId != null)
{
m_view = m_doc.GetElement(viewId) as View;
}
// Selected Placement Method
m_selectedPlacementMethod = int.Parse(inputParameters.GridTypeId);
// Construct Chromosomes with 3 params, m_objectDistanceX, m_objectDistanceY, m_selectedPlacementMethod
var chromosome = new FloatingPointChromosome(
new double[] { double.Parse(inputParameters.DistanceXMinParam), double.Parse(inputParameters.DistanceYMinParam), double.Parse(inputParameters.DistanceWallMinParam) },
new double[] { double.Parse(inputParameters.DistanceXMaxParam), double.Parse(inputParameters.DistanceYMaxParam), double.Parse(inputParameters.DistanceWallMaxParam) },
new int[] { 32, 32, 32 },
new int[] { 2, 2, 2 });
// Population Settings
//
// The population size needs to be 'large enough'.
// The question of when a population is large enough is difficult to answer.
// Generally, it depends on the project, the number of genes, and the gene value range.
// A good rule of thumb is to set the population size to at least 3x the number of inputs.
// If the results don't start to converge to an answer, you may need to increase the population size.
//
// by www.generativedesign.org/02-deeper-dive/02-04_genetic-algorithms/02-04-02_initialization-phase
var population = new Population(8, 12, chromosome);
// Fitness Function Settings
//
// Call CreateObjectPlacementPointList() and get count of points.
// This sample maximize a number of objects to place in a room.
//
// A fitness function is used to evaluate how close (or far off) a given design solution is from meeting the designer�fs goals.
//
// by www.generativedesign.org/02-deeper-dive/02-04_genetic-algorithms/02-04-03_evaluation-phase
var fitness = new FuncFitness((c) =>
{
var fc = c as FloatingPointChromosome;
var values = fc.ToFloatingPoints();
m_objectDistanceX = values[0];
m_objectDistanceY = values[1];
m_minimumDistanceFromObjectToWall = values[2];
List <XYZ> objectPlacementPointList = CreateObjectPlacementPointList();
return(objectPlacementPointList.Count);
});
var selection = new EliteSelection();
var crossover = new UniformCrossover(0.5f);
var mutation = new FlipBitMutation();
// Termination Condition Settings
//
// To finish the process in half an hour, this sample sets 2 conditions.
var termination = new OrTermination(
new GenerationNumberTermination(20),
new TimeEvolvingTermination(TimeSpan.FromMinutes(10)));
// Construct GeneticAlgorithm
var ga = new GeneticAlgorithm(
population,
fitness,
selection,
crossover,
mutation);
ga.Termination = termination;
Console.WriteLine("Generation: objectDistanceX, objectDistanceY, minimumDistanceFromObjectToWall = objectCount");
var latestFitness = 0.0;
// Callback Function of Generation Result
ga.GenerationRan += (sender, e) =>
{
var bestChromosome = ga.BestChromosome as FloatingPointChromosome;
var bestFitness = bestChromosome.Fitness.Value;
if (bestFitness != latestFitness)
{
latestFitness = bestFitness;
var phenotype = bestChromosome.ToFloatingPoints();
m_objectDistanceX = phenotype[0];
m_objectDistanceY = phenotype[1];
m_minimumDistanceFromObjectToWall = phenotype[2];
Console.WriteLine(
"Generation {0,2}: objectDistanceX: {1}, objectDistanceY: {2}, minimumDistanceFromObjectToWall: {3} = objectCount: {4}",
ga.GenerationsNumber,
m_objectDistanceX,
m_objectDistanceY,
m_minimumDistanceFromObjectToWall,
bestFitness
);
List <XYZ> objectPlacementPointList = CreateObjectPlacementPointList();
using (Transaction tx = new Transaction(m_doc))
{
tx.Start("Transaction Create Family Instance");
m_view.SetCategoryHidden(new ElementId(BuiltInCategory.OST_Furniture), false);
foreach (XYZ point in objectPlacementPointList)
{
FamilyInstance fi = m_doc.Create.NewFamilyInstance(point, m_familySymbol, m_level, Autodesk.Revit.DB.Structure.StructuralType.NonStructural);
m_doc.Regenerate();
BoundingBoxXYZ fiBB = fi.get_BoundingBox(m_view);
double xDiff = fiBB.Max.X - fiBB.Min.X;
double yDiff = fiBB.Max.Y - fiBB.Min.Y;
if (m_objectDistanceX / m_objectDistanceY >= 1.2 && yDiff / xDiff >= 1.2)
{
LocationPoint location = fi.Location as LocationPoint;
if (null != location)
{
XYZ fiLocationPoint = location.Point;
XYZ axisPoint = new XYZ(fiLocationPoint.X, fiLocationPoint.Y, fiLocationPoint.Z + 10);
Line axis = Line.CreateBound(fiLocationPoint, axisPoint);
location.Rotate(axis, Math.PI / 2.0);
}
}
else if (m_objectDistanceY / m_objectDistanceX >= 1.2 && xDiff / yDiff >= 1.2)
{
LocationPoint location = fi.Location as LocationPoint;
if (null != location)
{
XYZ fiLocationPoint = location.Point;
XYZ axisPoint = new XYZ(fiLocationPoint.X, fiLocationPoint.Y, fiLocationPoint.Z + 10);
Line axis = Line.CreateBound(fiLocationPoint, axisPoint);
location.Rotate(axis, Math.PI / 2.0);
}
}
}
DWGExportOptions dwgOptions = new DWGExportOptions();
ICollection <ElementId> views = new List <ElementId>();
views.Add(m_view.Id);
m_doc.Export(Directory.GetCurrentDirectory() + "\\exportedDwgs", m_level.Name + "_" + m_room.Name + "_Gen " + ga.GenerationsNumber + "_" + DateTime.Now.ToString("yyyyMMddHHmmss"), views, dwgOptions);
tx.RollBack();
}
}
};
ga.Start();
}
private void Go_Click(object sender, EventArgs e)
{
//initialize var
int cLength = Convert.ToInt32(chromoLength.Text);
EliteSelection<int> sel2 = new EliteSelection<int>(0);
RouletteSelection<int> sel1 = new RouletteSelection<int>();
GreedyCrossover cros = new GreedyCrossover(-1, -1, matrix);
GoldenMutation<int> mut = new GoldenMutation<int>(cLength);
pop = new Population<int>(mut, cros, sel1, sel2, x => 1 / cros.CalcFitness(x),
Convert.ToDouble(mProb.Text), Convert.ToDouble(cProb.Text));
maxF = new double[Convert.ToInt32(expCount.Text), Convert.ToInt32(iterCount.Text) + 1];
avgF = new double[Convert.ToInt32(expCount.Text), Convert.ToInt32(iterCount.Text) + 1];
double min = 100500; // Hi to Max ))
string bestChromo = null;
//experiments
for (int i = 0; i < Convert.ToInt32(expCount.Text); ++i)
{
//initial chromosomes
Trace.WriteLine("experiment #" + (i + 1).ToString());
Trace.Indent();
ChromosomesFromArray();
maxF[i, 0] = Math.Round(1 / pop.GetMaxFitness());
avgF[i, 0] = 1 / pop.GetPopulationFitness();
Trace.WriteLine("initial best fitness = " + Math.Round((1 / pop.GetMaxFitness())).ToString());
Trace.WriteLine("initial avg fitness = " + (1 / pop.GetPopulationFitness()).ToString());
Trace.WriteLine("initia;best fitness chromo = " + pop.GetMaxChromo());
// iterations
for (int j = 0; j < Convert.ToInt32(iterCount.Text); ++j)
{
Trace.WriteLine("iteration #" + (j + 1).ToString());
Trace.Indent();
pop.Iteration();
maxF[i, j + 1] = Math.Round(1 / pop.GetMaxFitness());
avgF[i, j + 1] = 1 / pop.GetPopulationFitness();
Trace.WriteLine(" best fitness = " + Math.Round((1 / pop.GetMaxFitness())).ToString());
Trace.WriteLine("avg fitness = " + (1 / pop.GetPopulationFitness()).ToString());
Trace.WriteLine("best fitness chromo = " + pop.GetMaxChromo());
Trace.Unindent();
}
if (Math.Round(1 / pop.GetMaxFitness()) < min)
{
min = Math.Round(1 / pop.GetMaxFitness());
bestChromo = pop.GetMaxChromo();
}
Trace.Unindent();
}
answer.Text = "Best Path: " + bestChromo + "Path Length" + min.ToString();
}
private void StartEvolution()
{
SetupSourceColorMatrix();
if (currentDrawing == null)
{
currentDrawing = GetNewInitializedDrawing();
}
lastSelected = 0;
var selection = new EliteSelection();
// var selection = new CustomSelection(20);
var crossover = new OrderedCrossover();
// var crossover = new CycleCrossover();
// var crossover = new CutAndSpliceCrossover();
// var crossover = new OnePointCrossover(2);
var mutation = new CustomMutation();
// var mutation = new UniformMutation(true);
var fitness = new DrawingFitness(sourceColors);
var chromosome = new PolygonChromosome();
var population = new Population(10, 100, chromosome);
_ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
_ga.Termination = new TimeEvolvingTermination(TimeSpan.FromHours(4));
//
_ga.GenerationRan += (sender, args) =>
{
generation++;
var best = _ga.Population.CurrentGeneration.Chromosomes.OrderBy(c => c.Fitness).First() as PolygonChromosome;
var drawing = new DnaDrawing(best);
if (best.Fitness.Value <= errorLevel)
{
errorLevel = best.Fitness.Value;
lock (currentDrawing)
{
currentDrawing = drawing;
}
selected++;
}
};
_ga.Stopped += (sender, args) =>
{
Stop();
};
_ga.Start();
// while (isRunning)
// {
// DnaDrawing newDrawing;
// lock (currentDrawing)
// {
// newDrawing = currentDrawing.Clone();
// }
// newDrawing.Mutate();
//
// if (newDrawing.IsDirty)
// {
// generation++;
//
// double newErrorLevel = FitnessCalculator.GetDrawingFitness(
// newDrawing,
// sourceColors);
//
// if (newErrorLevel <= errorLevel)
// {
// selected++;
// lock (currentDrawing)
// {
// currentDrawing = newDrawing;
// }
// errorLevel = newErrorLevel;
// }
// }
// //else, discard new drawing
// }
}
public void Solve(Maze maze)
{
const int NumberOfGenes = 5000;
const float MutationRate = 0.02f;
const float CrossOverRate = 0.6f;
var selection = new EliteSelection();
var crossover = new UniformCrossover();
var mutation = new UniformMutation(true);
var chromosome = new MazeSolverChromosome(NumberOfGenes);
var reinsertion = new ElitistReinsertion();
var population = new Population(50, 100, chromosome);
var fitness = new MazeWalkerFitness(maze);
IChromosome best = chromosome;
best.Fitness = fitness.Evaluate(best);
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
ga.MutationProbability = MutationRate;
ga.Reinsertion = reinsertion;
ga.Termination = new OrTermination(
new FitnessStagnationTermination(this.stagnationThreshold),
new TimeEvolvingTermination(TimeSpan.FromSeconds(this.maxRunTimeInSeconds)));
ga.CrossoverProbability = CrossOverRate;
ga.TaskExecutor = new SmartThreadPoolTaskExecutor()
{
MinThreads = Environment.ProcessorCount,
MaxThreads = Environment.ProcessorCount
};
ga.GenerationRan += (sender, eventargs) =>
{
if (this.generateUpdateImages)
{
if (ga.GenerationsNumber == 1 || ga.GenerationsNumber % this.updateImageFrequency == 0)
{
var winnerSteps = FillMazeStepsWalked(maze, ga.BestChromosome);
ImageDraw(winnerSteps, $"gen{ga.GenerationsNumber}.png");
}
}
if (ga.GenerationsNumber % 10 == 0)
{
Console.WriteLine(
$"{ga.GenerationsNumber} generations completed. Best fitness: {ga.BestChromosome.Fitness}. Best so far: {best.Fitness}. Time evolving: {ga.TimeEvolving.TotalMinutes} min.");
}
if (ga.BestChromosome.Fitness > best.Fitness)
{
best = ga.BestChromosome.Clone();
// ga.Population.CurrentGeneration.Chromosomes.Add(best);
}
};
ga.TerminationReached += (sender, eventargs) =>
{
Console.WriteLine($"Termination Reached");
var winnerSteps = FillMazeStepsWalked(maze, best);
ImageDraw(winnerSteps, "best.png");
var serializer = new XmlSerializer(typeof(string[]));
using (var sw = new StreamWriter("bestchromosome.xml"))
{
serializer.Serialize(sw, best.GetGenes().Select(g => g.Value as string).ToArray());
}
if (this.animate)
{
File.Delete("output.gif");
this.gif.Save("output.gif");
}
};
if (this.generateUpdateImages)
{
this.ImageDraw(maze, "gen0.png");
}
ga.Start();
}
