Genetic Algorithm

Matlab code for simple genetic algorithm

function genealgo

syms x

n = input ('value of n:');

Q = randi([-20 20],n,n);

p = 2*n;

i=1;

while i<=p

    X(i,:)= randi([0 1],n,1);          % chromosome matrix

    i = i+1;

end

h=1;

u=6;

while u>1

i=1;

while i<=p

    E(i,1)= X(i,:)*Q*transpose(X(i,:)); % E= Evaluated matrix

    F(i,1)= exp(E(i,1));               % F = Fitness matrix

    i=i+1;

end

S=0;

i=1;

while i<=p

    S=S+E(i,1);                 % S = sum of all evaluations

    i=i+1;

end

i=1;

while i<=p

    P(i,1)= E(i,1)/S ;           % P = probability matrix

    i=i+1;

end

i=1;

c=0;

while i<=p;

    c=c+P(i,1);

    C(i,1)=c;      % C = Cummulative of probability matrix

    i=i+1;

end

R=rand(p,1);

i=1;

while i<= p

    j=1;

    while R(i,1)> C(j,1)    % roulette wheel check

        j=j+1;

    end

    X(i,1)=X(j,1);        % reprodution

    i=i+1;

end

N=rand(p,1);

j=1;

i=1;

while i<=n

    if (N(i,1)<0.25);

        A(j,1)=i;

        j=j+1;

        i=i+1;             % chromosomes for cross over

    else

        i=i+1;

    end

end

i=1;

B=X(A(1,1),:);

while i< j-2

    c = randi([1 n],1,1);

    X(A(i,1),c:end)= X(A(i+1,1),c:end);           % cross over

    i=i+1;

end

 c = randi([1 n],1,1);

 X(A(j-2,1),c:end)= X(B(1,1),c:end);          

i=1;

j=1;

while i<= p

    while j<=n

        c=rand(1,1);

        if c < 0.05

            if X(i,j)== 0;

              X(i,j)=1;                % mutation

            else

                X(i,j)=0;   

            end

            j=j+1;

        else

            j=j+1;

        end

    end

    i=i+1;

end

i=1;

while i<=p

  E(i,1)= X(i,:)*Q*transpose(X(i,:)); % E= Evaluated matrix                                                                                                                       after a generation

  i=i+1;

end

disp(E)

u=std2(E);

disp(u)

disp(Q)

disp(X)

disp(h)

h=h+1;

end

%Cross over probability taken as 0.8

%Mutation probability taken as 0.05

%Q is randomly generated with it’s all values between 20 and -20

% standard deviataion < 1 is taken as stopping criteria

1) Generate chromosome-chromosome number of the population, and the initialization value of the genes chromosome-chromosome with a random value.

2) Evaluation of fitness value of chromosomes by calculating objective function.

3) Chromosomes selection (roulette wheel-reproduction)

4) Crossover

5) Mutation

6) New Chromosomes (Offspring)

7) Iterate the process till desired offspring are obtained.

8) Solution (Best Chromosomes)