【优化算法】社会群体优化算法（SGO）【含Matlab源码 1449期】

一、获取代码方式

获取代码方式1： 通过订阅紫极神光博客付费专栏，凭支付凭证，私信博主，可获得此代码。

获取代码方式2： 通过紫极神光博客主页开通CSDN会员，凭支付凭证，私信博主，可获得此代码。

获取代码方式3： 完整代码已上传我的资源：【优化算法】社会群体优化算法（SGO）【含Matlab源码 1449期】

备注：开通CSDN会员，仅只能免费获得1份代码（有效期为开通日起，三天内有效）； 订阅紫极神光博客付费专栏，可免费获得2份代码（有效期为订阅日起，三天内有效）；

二、部分源代码

%--------------------------------------------------------------------------
% % Developed in MATLAB R2016a
 

clc;
clear all;
close all;

'MSGO'

pn=30; %population size
gn=250; %maximum number of iteration
% dim=30; %dimension of the fuction
% lb=-30; %lower bound
% ub=30;  %upper bound

func_no=21;
[lb,ub,dim]=Test_Functions_Range(func_no);

[fitsgo,objvalue]=msgofunction(pn,dim,gn,lb,ub,func_no);
 
objvalue
fitsgo;

if objvalue>0
     semilogy(fitsgo,'r','LineWidth',2);
 else
     plot(fitsgo,'r','LineWidth',2);
 end
 xlabel('Iterations');
 ylabel('Fitness');
 title('Convergence characteristics')
 
% for i=1:5
%     point(i)=i;
% end

%plot(point,fitsgo(point),'-hb');

% %% 
% title('Convergence characteristics')
% %% 
% semilogy(fitsgo)
% %plot(fitsgo)
% xlabel('Number of Iterations');
% ylabel('Fitness value');
% %axis tight
 grid on
% legend('MSGO','SouthEast');
% 
%--------------------------------------------------------------------------
% % Developed in MATLAB R2016a
% The code is based on the following papers.
% A. Naik, A. Abraham and S.C. Satapathy(2020), Modified Social Group Optimization-a meta-heuristic 
% algorithm to solve Short-term Hydrothermal Scheduling, Applied softcomputing,doi:10.1016/j.asoc.2020.106524 
%--------------------------------------------------------------------------

function[fit,objvalue]=msgofunction(pn,dim,gn,lb,ub,func_no)
obj=@Test_Functions;
%Test_Functions(pop(i,:),func_no,dim);
pop=lb+rand(pn,dim)*(ub-lb);
c=0.2;
for i=1:pn
   f(i)=obj(pop(i,:),func_no,dim);
end
[value ibest]=min(f);
% from here iteration start
fit=zeros(gn,1);
for g=1:gn
% improvement due to best person of the group
[value ibest]=min(f);
value;
guru=pop(ibest,:);
for i=1:pn
     for j=1:dim
     newpop(i,j)=c*pop(i,j)+rand*(guru(j)-pop(i,j));
          if((newpop(i,j)<lb))
              newpop(i,j)=lb;
          end
          if((newpop(i,j)>ub))
              newpop(i,j)=ub;
          end
     end

f1=obj(newpop(i,:),func_no,dim);
       
     if(f1<f(i))
         pop(i,:)=newpop(i,:);
         f(i)=f1;
     end  
end
 [a b]=min(f);
  gpop=pop(b,:);
  SAP=0.7;
             
for i=1:pn
        r1=floor(1+rand*pn);
        while(r1==i)
            r1=floor(1+rand*pn);
        end
        if(f(i)<f(r1))
            if rand>SAP
                 for j=1:dim
                     popnew(i,j)=pop(i,j)+rand*(pop(i,j)-pop(r1,j))+rand*(gpop(j)-pop(i,j));
                     if(popnew(i,j)<lb)
                         popnew(i,j)=lb;
                     end
                     if(popnew(i,j)>ub)
                         popnew(i,j)=ub;
                     end
                 end
            
            else
                 for j=1:dim
                      popnew(i,:)=lb+rand*(ub-lb); 
                 end
            end
        else
            for j=1:dim
                popnew(i,j)=pop(i,j)+rand*(pop(r1,j)-pop(i,j))+rand*(gpop(j)-pop(i,j));
                if(popnew(i,j)<lb)
                     popnew(i,j)=lb;
                end
                if(popnew(i,j)>ub)
                     popnew(i,j)=ub;
                end
            end
        end
end
               
for i=1:pn

f2(i)=obj(popnew(i,:),func_no,dim);   
    if(f2(i)<f(i))
       pop(i,:)=popnew(i,:);
       f(i)=f2(i);
    end 
end
%end of learner phase
[minvalue position]=min(f);
fit(g)=minvalue;
objvalue=fit(g)
end

end
 
 

三、运行结果

四、matlab版本及参考文献

1 matlab版本 2014a

2 参考文献 [1] 包子阳,余继周,杨杉.智能优化算法及其MATLAB实例（第2版）[M].电子工业出版社，2016. [2]张岩,吴水根.MATLAB优化算法源代码[M].清华大学出版社，2017.