my_genetic.m

real genetic algorithm matlab code

clc;
clear all;
%%%%%%% initialization %%%%%%
N= 50;
Pc= 0.3;
Pm=0.05;
ITER=100;
m=2;
Lo=[-4 -1.5];
Hi=[2 1];
w=rand(N,m);
Population=[Lo(1)+(w(:,1).*(Hi(1)-Lo(1))) Lo(2)+(w(:,2).*(Hi(2)-Lo(2)))];
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
best_so_far=[];
average_fitness=[];
for it=1:ITER
    [selection_probability,fit,ave_fit,max_fit,opt_sol]=fit_eval(Population,N,m);
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    if it==1
        best_so_far(it)=max_fit;
        final_sol=opt_sol;
    elseif max_fit>best_so_far(it-1)
        best_so_far(it)=max_fit;
        final_sol=opt_sol;
    else
        best_so_far(it)=best_so_far(it-1);
    end
    average_fitness(it)=ave_fit;
    %%%%%%%%%%%%%%%%%%%%%%%%%%%% Reproduction %%%%%%%%%%%%%%%%%%%%%%%%%%%%
    
    %%%%%%%%%%%%%%%%%%%%%%%%%%%% Selection fot Cross over %%%%%%%%%%%%%%%%
    [mating_pool]=g_roulette_wheel(Population,N,selection_probability);
    
    %%%%%%%%%%%%%%%%%%%%%%%%%%%% Cross Over %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    [new_pop]=g_crossover(mating_pool,Pc,N);
    
    %%%%%%%%%% Check whether new population is in range or not %%%%%%%%%%%
    [new_pop]=rg_replacement(new_pop,N,m,Lo,Hi);
    
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%% Mutation %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    [Population]=g_mutation(new_pop,Pm,N,m);
    
    %%%%%%%%%%%%%%%%%%%%%%%%% End of reproduction %%%%%%%%%%%%%%%%%%%%%%%%
end
disp(' Final Solution');
disp(final_sol);
disp(' Optimum Fitness');
disp(best_so_far(end));
%%%%%%%%%%%%%%%%%%  Graph   %%%%%%%%%%%%%%%%%%%%%%
x=1:ITER;
figure,plot(x,best_so_far,'-r',x,average_fitness,'--b');
xlabel('Generation');
ylabel('Fitness Function');
legend('Best So Far','Average Fitness');
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%