Genetic algorithms and simple implementation of Java code

See Address:

https://www.jianshu.com/p/ae5157c26af9

 

Code:

GA class {public 

	Private ChrNum int = 10; // number of chromosomes 
	private String [] ipop = new String [ChrNum]; // Number of chromosomes in a population 
	private int generation = 0; // chromosome code 
	public static final int GENE = 46 ; // number of genes 
	private double bestfitness = Double.MAX_VALUE; // function optimal solution 
	private int bestgenerations; // All the best parent and offspring chromosomes 
	private String beststr; // chromosome optimum binary code 
	
	/ ** 
	 * initialize a chromosome (represented by binary strings) 
	 * / 
	Private initChr string () { 
		string RES = ""; 
		for (int I = 0; I <the GENE; I ++) { 
			IF (Math.random ()> 0.5) { 
				RES + = "0"; 
			} the else { 
				RES = + ". 1"; 
			} 
		} 
		return RES; 
	}

	/ ** 
	 * initialize a population (10 chromosomes) 
	 * / 
	Private String [] initPop () { 
		String [] iPOP = new new String [ChrNum]; 
		for (int I = 0; I <ChrNum; I ++) { 
			iPOP [I ] = initChr (); 
		} 
		return iPOP; 
	} 

	/ ** 
	 * converted into a chromosome values of x, y variables 
	 * / 
	Private Double [] calculatefitnessvalue (string STR) { 

		// binary 23 before the binary string x after 23 to binary string y 
		int a = the Integer.parseInt (str.substring (0, 23 is), 2);       
		int B = the Integer.parseInt (str.substring (23 is, 46 is), 2); 

		Double X = a * (6.0 - 0) / (Math.pow (2, 23) - 1); // gene of x 
		double y = b * (6.0 - 0) / (Math.pow (2, 23) - 1) ; // Y gene 

		// function to be optimized 
		double fitness = 3 - Math.sin (2 * x) * Math.sin (2 * x)
				- Math.sin, (Y * 2) * Math.sin, (2 * Y); 
		
		Double [] Returns = {X, Y, Fitness}; 
		; return Returns 

	} 

	/ ** 
	 * roulette select 
	 each of the individual groups calculated * fitness value; 
	 * according to a rule determined by the selected individual fitness value of the individual entering the next generation; 
	 * / 
	Private void sELECT () { 
		Double evals [] = new new Double [ChrNum]; // all chromosomes adaptation value 
		double p [] = new [ChrNum ] double; // the probability of selecting each chromosome 
		double q [] = new double [ ChrNum]; // cumulative probability 
		double F = 0; // integrated adaptation sum value 
		for (int i = 0 ; I <ChrNum; I ++) { 
			evals [I] = calculatefitnessvalue (iPOP [I]) [2]; 
			IF (evals [I] <bestfitness) {// record the minimum value in the population, i.e. the optimal solution 
			} 
				bestfitness = evals [i];
				bestgenerations = Generation; 
				beststr = iPOP [I]; 

			F. = F. + evals [I]; // all chromosomes adaptation sum value 
		} 

		for (int I = 0; I <ChrNum; I ++) { 
			P [I] = evals [I ] / F.; 
			IF (I == 0) 
				Q [I] = P [I]; 
			the else { 
				Q [I] = Q [I -. 1] + P [I]; 
			} 
		} 
		for (int I = 0; I <ChrNum; I ++) { 
			Double = R & lt Math.random (); 
			IF (R & lt <= Q [0]) { 
				iPOP [I] = iPOP [0]; 
			} the else { 
				for (int. 1 = J; J <ChrNum; ++ J) { 
					IF (R & lt <Q [J]) { 
						iPOP [I] = iPOP [J]; 
					} 
				} 
			} 
		} 
	} 

	/ ** 
	 * crossover intersection was 60%, with an average of 60% of the cross-chromosome
	 * / 
	private void cross() {
		String temp1, temp2;
		For (int I = 0; I <ChrNum; I ++) { 
			IF (Math.random () <0.60) { 
				int POS = (int) (Math.random () * the GENE) + 1'd; // POS site binary string longitudinal cross 
				temp1 = iPOP [I] .substring (0, POS) + iPOP [(I +. 1)% ChrNum] .substring (POS); 
				temp2 of iPOP = [(I +. 1)% ChrNum] .substring (0 , POS) + iPOP [I] .substring (POS); 
				iPOP [I] = temp1; 
				iPOP = temp2 of [(I +. 1) / ChrNum]; 
			} 
		} 
	} 

	/ ** 
	 * mutation genetic variation operation 1% 
	 * / 
	void mutation Private () { 
		for (int I = 0; I <. 4; I ++) { 
			int NUM = (int) (Math.random () * * ChrNum the GENE +. 1); 
			int chromosomeNum = (int) (NUM / the GENE ) + 1; // chromosome number
 
			int mutationNum = num - (chromosomeNum - 1) * gENE; // number gene
			IF (mutationNum == 0) 
				mutationNum =. 1; 
			chromosomeNum chromosomeNum = -. 1; 
			IF (chromosomeNum> = ChrNum) 
				chromosomeNum =. 9; 
			String TEMP; 
			String A; // record the encoded mutation site variants 
			if ([chromosomeNum] ipop .charAt (mutationNum - 1) == ' 0') {// when the mutation site is 0, 
                A = ". 1"; 
			} the else {    
				A = "0"; 
			} 
			// if mutation site in the first, middle and mutation at the tail 
			IF (mutationNum ==. 1) { 
				TEMP = a + iPOP [chromosomeNum] .substring (mutationNum); 
			} {the else 
				IF (= mutationNum the GENE!) { 
				} else { 
					TEMP = iPOP [chromosomeNum] .substring (0, mutationNum -1) + A
							IPOP + [chromosomeNum] .substring (mutationNum); 
					TEMP = iPOP [chromosomeNum] .substring (0, mutationNum -. 1) + A; 
				} 
			} 
			// record Chromosomal Variation		 
			iPOP [chromosomeNum] = TEMP; 
		} 
	} 

	public static main void (String args []) { 

		GA GA tryer new new = (); 
		Tryer.ipop Tryer.initPop = (); // initial population 
		String STR = ""; 
		
		// number of iterations 
		for (int i = 0; i <1000000; I ++) { 
			Tryer.select (); 
			Tryer.cross (); 
			Tryer.mutation (); 
			Tryer.generation = I; 
		} 
		
		Double [] X = Tryer.calculatefitnessvalue (Tryer.beststr); 

		STR = "Min "+ Tryer.bestfitness + '\ n' +" No. "
		        + Tryer.bestgenerations + "chromosomes: <" + + Tryer.beststr ">" + '\ n-'  
				+ "= X" + X [0] + '\ n-'+ "y=" + x[1];

		System.out.println (STR); 

	} 
}