【优化求解】基于缎蓝园丁鸟优化算法 (SBO)求解单目标问题附matlab代码

1 简介

​

2 部分代码

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Satin Bowerbird Optimizer(SBO)



%%

clc;

clear;

close all;

tic



%% Problem Definition



Function_name='F1';



[lowerbound,upperbound,numbervar,costfcn]=cost_functions(Function_name);



% VarSize=[1 numbervar]; 



%% SBO Parameters



[MaxIt,nPop,alpha,pMutation,sigma]=SBO_parameters(lowerbound,upperbound);



%% Initialization



[pop,elite,BestCost]=Initialization(nPop,lowerbound,upperbound,numbervar,MaxIt,costfcn);



%% SBO Main Loop



for it=1:MaxIt

 newpop=pop;

 

 %Calculating the Fitness of each bower

 F=zeros(nPop,1);

 for i=1:nPop

•    if pop(i).Cost>=0

•      F(i)=1/(1+pop(i).Cost);

•    else

•      F(i)=1+abs(pop(i).Cost);

•    end

 end

 

 %Calculating the probability of each bower

 P=F/sum(F);

 

 %changes at any bower

 for i=1:nPop

•    for k=1:numbervar

•        

•        % Select target bower         

•        j=RouletteWheelSelection(P);

•        

•        % Calculating Step Size

•        lambda=alpha/(1+P(j));

•        

•        newpop(i).Position(k)=pop(i).Position(k) ...

•          +lambda*(((pop(j).Position(k)+elite(k))/2)-pop(i).Position(k));

•        

•        % Mutation

•      if rand<=pMutation



•        newpop(i).Position(k)=newpop(i).Position(k)+(sigma*randn);

•       

•      end

•        

•    end 

 

•    % Evaluation

•    newpop(i).Cost=costfcn(newpop(i).Position);

 

 end 

•     

  pop=[pop

•     newpop

•     ]; %#ok



 % Sort Population

[~, SortOrder]=sort([pop.Cost]);

 pop=pop(SortOrder);

 pop=pop(1:nPop);

 

 % Update Best Solution Ever Found

 BestSol=pop(1);

 elite=BestSol.Position;

 

 % Store Best Cost Ever Found

 BestCost(it)=BestSol.Cost;

 

 % Show Iteration Information

 

 disp(['SBO:: Iteration-> ' num2str(it) '<----->Best Cost = ' num2str(BestCost(it))]);

 

end

toc



%% Results



disp(['BestSol=' num2str(elite)]);

disp(['BestCost=' num2str(BestSol.Cost)]);

figure;

semilogy(BestCost,'LineWidth',2);

xlabel('Iteration');

ylabel('Best Cost');

img =gcf; %获取当前画图的句柄

print(img, '-dpng', '-r600', './img.png')     %即可得到对应格式和期望dpi的图像

3 仿真结果

4 参考文献

[1]王依柔、张达敏、樊英. "非均匀变异的互利自适应缎蓝园丁鸟优化算法." 计算机工程与科学 v.42;No.312.12(2020):135-143.

[2]鲁晓艺, 刘升, 韩斐斐,等. 基于自适应权重的缎蓝园丁鸟优化算法[J]. 智能计算机与应用, 2018, 8(6):7.

博主简介：擅长智能优化算法、神经网络预测、信号处理、元胞自动机、图像处理、路径规划、无人机等多种领域的Matlab仿真，相关matlab代码问题可私信交流。

部分理论引用网络文献，若有侵权联系博主删除。