gppso126c.m

PSO算法实现的图分割问题的matlab源码。原创

%PSO >> function for the PSO ALGORITHM
%
% USAGES:   1.) [fxmin, xmin, Swarm, history] = PSO(psoOptions);
%           2.) [fxmin, xmin, Swarm, history] = PSO;
%           3.) fxmin = PSO(psoOptions);
%           3.) PSO 
%           etc.
%
% Arguments     : psoOptions--> A Matlab stucture containing all PSO related options. (see also: get_psoOptions)
% Return Values : [fxmin, xmin, Swarm, history]
%                     |     |       |       |_The history of the algorithm. Depicts how the function value of GBest changes over the run.
%                     |     |       |_The final Swarm. (A matrix containing co-ordinates of all particles)
%                     |     |_The co-ordinates of Best (ever) particle found during the PSO's run.
%                     |__The objective value of Best (^xmin) particle.
%
%  History        :   Author      :   NB (Niu Ben)
%                     Created on  :   02122004 (Friday. 2nd DEC, 2004)
%                     Comments    :   The basic PSO algorithm.
%                     Modified on :   1012004 (Thursday. 10th DEC, 2004)
%                     Comments    :   It uses psoOptions structure now. More organized.
%  
%                     see also: get_psoOptions


%Globals
%ave=0;
%for i=1:2
%fidSPSO=fopen('SF1Rosenbrock.txt','w');

SwarmSize = 20;
 Dim=6;
 AA= [1,0,0,0,1,0
     0,1,1,1,1,1
     0,1,1,1,1,1
     0,1,1,1,0,0
     1,1,1,0,1,0
     0,1,1,0,0,1];
 
S = [3,4,3,7,5,2];


for pppp=1:2
Vmax=4;
GM = 0; % Global minimum (used in the stopping criterion)
ErrGoal = 1e-10; % Desired accuracy
Iterations=9;
% initializing variables
success=0;
iter = 0;   % Iterations' counter

%maxstep=10;
%cross=0;
fevals = 0; % Function evaluations' counter

%y=zeros(2,(Iterations+1));



% Using params---
% Determine the value of weight change
w = 0.729;   %Initial inertia weight's value
c1=1.494;
c2=1.494;

% Initialize graph
%A = [0,0,0,0,1,0
 %    0,0,1,1,1,1
  %   0,1,0,1,1,1
   %  0,1,1,0,0,0
    % 1,1,1,0,0,0
     %0,1,1,0,0,0];
     



%while (Dim>1)% Sensor still left for scheduling
% Initialize parometers in the cost function

    
% Initialize Swarm and Velocity

    
Swarm = round(rand(SwarmSize, Dim));
VStep = rand(SwarmSize, Dim)*Vmax*2 - Vmax;


fSwarm= fit(Swarm,AA,S,Dim,SwarmSize);
fevals = fevals + SwarmSize;

% Initializing the Best positions matrix and
% the corresponding function values
PBest = Swarm;
fPBest = fSwarm;

% Finding best particle in initial population
[fGBest, g] = max(fSwarm);
lastbpf = fGBest;
Best = Swarm(g,:); %Used to keep track of the Best particle ever
fBest = fGBest;

% Output
disp(sprintf('Clique\t\tfGBest\t\t\tfSwarm'));
    


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%                  THE  PSO  LOOP                          %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

iiter=0;

while( iter <= Iterations)

iter = iter+1; 

    %%%%%%%%%%%%%%%%%
    % The PLAIN PSO %
    
    %Set GBest
    A= repmat(Swarm(g,:), SwarmSize, 1); %A = GBest. repmat(X, m, n) repeats the matrix X in m rows by n columns.
     
    % Generate Random Numbers
    R1 = rand(SwarmSize, Dim);
    R2 = rand(SwarmSize, Dim);
    
    % Calculate Velocity
    VStep = w*VStep + c1*R1.*(PBest-Swarm) + c2*R2.*(A-Swarm);

  
    % Apply Vmax Operator for v > Vmax
    changeRows = VStep > Vmax;
    VStep(find(changeRows)) =Vmax;
    % Apply Vmax Operator for v < -Vmax
    changeRows = VStep < -Vmax;
   VStep(find(changeRows)) = -Vmax;
    
  
    
    % ::UPDATE POSITIONS OF PARTICLES::
    PP=rand(SwarmSize,Dim);
    
    for i= 1:SwarmSize%evaluate new Swarm
        for j = 1:Dim
            if PP(i,j)<(1/(1+exp(-VStep(i,j))))
                Swarm(i,j)=1;
            else
                Swarm(i,j)=0;
            end
        end
    end
    


      
  

% Find initial function values



    fSwarm= fit(Swarm,AA,S,Dim,SwarmSize);
    fevals = fevals + SwarmSize;
   
  
    % Updating the best position for each particle
    changeRows = fSwarm > fPBest;
    fPBest(find(changeRows)) = fSwarm(find(changeRows));%适应值跟换
    PBest(find(changeRows), :) = Swarm(find(changeRows), :);%粒子位置跟换
    
    lastbpart = PBest(g, :);%保存最好位置
    
    % Updating index g
   [fGBest, g] = max(fPBest);  % 保存最好适应值
    
    %Update Best. Only if fitness has improved.
    if fGBest > lastbpf    %如果本次适应值好于上次则保存
        [fBest, b] = max(fPBest);%最好适应值为fBest
        Best = PBest(b,:);%最好位置为Best
    end
    
    lastbpf = fGBest; %To be used to find Best保留本次适应值准备与下一次比较
          
    
         
        x(iter)=iter;
       % y(pppp,iter)=fGBest   
       
    
    %%OUTPUT%%   
    
        %disp(sprintf('%4d\t\t\t%.5g\t\t\t%5d', iter, fGBest, fevals));
        
        
end 
%fclose all;
  
  %thebest(i)=fGBest;
  %ave=ave+thebest(i);


%axis([0 1000 0 200]);
%xlabel('iterations');
%ylabel( 'Best-fitness=fGBest(i)')
%title('Multi-pattern Evolutionary process')

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%                  END  OF PSO  LOOP                       %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%[fxmax, b] = max(fPBest)
%xmax = PBest(b, :)
%disp(sprintf('%4d\t\t\t%.5g\t\t\t%5d', Best,iter,fSwarm ));
[AA,S,Dim]=NewAAS(Dim,Best,AA,S);

end


%plot(x,log10(y(2,:)))