【优化求解】基于蜻蜓算法求解多目标优化问题matlab代码

骑猪看日落 2022-09-17 05:18 190阅读 0赞

1 模型

![1e04c4123a75b439c7aa4d125c3b87f9.png][]

![01be15b3763c326086dc2204011c14f6.png][]

![85e52e22a494cd7ee6ea5a3e9533bc6b.png][]

![7cb07ad9824497acf3b57b85cd2e4192.png][]

## 2 部分代码 ##

%-------------------------------------------------------------------------%

%\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_%

% Multi-Objective Dragonfly Algorithm (MODA) source codes demo %

% version 1.0 %

clc;

clear;

close all;

% Change these details with respect to your problem%%%%%%%%%%%%%%

ObjectiveFunction=@ZDT1;

dim=5;

lb=0;

ub=1;

obj\_no=2;

if size(ub,2)==1

ub=ones(1,dim)\*ub;

lb=ones(1,dim)\*lb;

end

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

% Initial parameters of the MODA algorithm

max\_iter=100;

N=100;

ArchiveMaxSize=100;

Archive\_X=zeros(100,dim);

Archive\_F=ones(100,obj\_no)\*inf;

Archive\_member\_no=0;

r=(ub-lb)/2;

V\_max=(ub(1)-lb(1))/10;

Food\_fitness=inf\*ones(1,obj\_no);

Food\_pos=zeros(dim,1);

Enemy\_fitness=-inf\*ones(1,obj\_no);

Enemy\_pos=zeros(dim,1);

X=initialization(N,dim,ub,lb);

fitness=zeros(N,2);

DeltaX=initialization(N,dim,ub,lb);

iter=0;

position\_history=zeros(N,max\_iter,dim);

for iter=1:max\_iter

r=(ub-lb)/4+((ub-lb)\*(iter/max\_iter)\*2);

w=0.9-iter\*((0.9-0.2)/max\_iter);

my\_c=0.1-iter\*((0.1-0)/(max\_iter/2));

if my\_c<0

my\_c=0;

end

if iter<(3\*max\_iter/4)

s=my\_c; % Seperation weight

a=my\_c; % Alignment weight

c=my\_c; % Cohesion weight

f=2\*rand; % Food attraction weight

e=my\_c; % Enemy distraction weight

else

s=my\_c/iter; % Seperation weight

a=my\_c/iter; % Alignment weight

c=my\_c/iter; % Cohesion weight

f=2\*rand; % Food attraction weight

e=my\_c/iter; % Enemy distraction weight

end

for i=1:N %Calculate all the objective values first

Particles\_F(i,:)=ObjectiveFunction(X(:,i)');

if dominates(Particles\_F(i,:),Food\_fitness)

Food\_fitness=Particles\_F(i,:);

Food\_pos=X(:,i);

end

if dominates(Enemy\_fitness,Particles\_F(i,:))

if all(X(:,i)<ub') && all( X(:,i)>lb')

Enemy\_fitness=Particles\_F(i,:);

Enemy\_pos=X(:,i);

end

\[Archive\_X, Archive\_F, Archive\_member\_no\]=UpdateArchive(Archive\_X, Archive\_F, X, Particles\_F, Archive\_member\_no);

if Archive\_member\_no>ArchiveMaxSize

Archive\_mem\_ranks=RankingProcess(Archive\_F, ArchiveMaxSize, obj\_no);

\[Archive\_X, Archive\_F, Archive\_mem\_ranks, Archive\_member\_no\]=HandleFullArchive(Archive\_X, Archive\_F, Archive\_member\_no, Archive\_mem\_ranks, ArchiveMaxSize);

else

Archive\_mem\_ranks=RankingProcess(Archive\_F, ArchiveMaxSize, obj\_no);

end

Archive\_mem\_ranks=RankingProcess(Archive\_F, ArchiveMaxSize, obj\_no);

% Chose the archive member in the least population area as foods

% to improve coverage

index=RouletteWheelSelection(1./Archive\_mem\_ranks);

if index==-1

index=1;

end

Food\_fitness=Archive\_F(index,:);

Food\_pos=Archive\_X(index,:)';

% Chose the archive member in the most population area as enemies

% to improve coverage

index=RouletteWheelSelection(Archive\_mem\_ranks);

if index==-1

index=1;

end

Enemy\_fitness=Archive\_F(index,:);

Enemy\_pos=Archive\_X(index,:)';

for i=1:N

index=0;

neighbours\_no=0;

clear Neighbours\_V

clear Neighbours\_X

% Find the neighbouring solutions

for j=1:N

Dist=distance(X(:,i),X(:,j));

if (all(Dist<=r) && all(Dist~=0))

index=index+1;

neighbours\_no=neighbours\_no+1;

Neighbours\_V(:,index)=DeltaX(:,j);

Neighbours\_X(:,index)=X(:,j);

end

% Seperation%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Eq. (3.1)

S=zeros(dim,1);

if neighbours\_no>1

for k=1:neighbours\_no

S=S+(Neighbours\_X(:,k)-X(:,i));

end

S=-S;

else

S=zeros(dim,1);

end

% Alignment%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Eq. (3.2)

if neighbours\_no>1

A=(sum(Neighbours\_V')')/neighbours\_no;

else

A=DeltaX(:,i);

end

% Cohesion%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Eq. (3.3)

if neighbours\_no>1

C\_temp=(sum(Neighbours\_X')')/neighbours\_no;

else

C\_temp=X(:,i);

end

C=C\_temp-X(:,i);

% Attraction to food%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Eq. (3.4)

Dist2Attraction=distance(X(:,i),Food\_pos(:,1));

if all(Dist2Attraction<=r)

F=Food\_pos-X(:,i);

iter;

else

F=0;

end

% Distraction from enemy%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Eq. (3.5)

Dist=distance(X(:,i),Enemy\_pos(:,1));

if all(Dist<=r)

E=Enemy\_pos+X(:,i);

else

E=zeros(dim,1);

end

for tt=1:dim

if X(tt,i)>ub(tt)

X(tt,i)=lb(tt);

DeltaX(tt,i)=rand;

end

if X(tt,i)<lb(tt)

X(tt,i)=ub(tt);

DeltaX(tt,i)=rand;

end

if any(Dist2Attraction>r)

if neighbours\_no>1

for j=1:dim

DeltaX(j,i)=w\*DeltaX(j,i)+rand\*A(j,1)+rand\*C(j,1)+rand\*S(j,1);

if DeltaX(j,i)>V\_max

DeltaX(j,i)=V\_max;

end

if DeltaX(j,i)<-V\_max

DeltaX(j,i)=-V\_max;

end

X(j,i)=X(j,i)+DeltaX(j,i);

end

else

X(:,i)=X(:,i)+Levy(dim)'.\*X(:,i);

DeltaX(:,i)=0;

end

else

for j=1:dim

DeltaX(j,i)=s\*S(j,1)+a\*A(j,1)+c\*C(j,1)+f\*F(j,1)+e\*E(j,1) + w\*DeltaX(j,i);

if DeltaX(j,i)>V\_max

DeltaX(j,i)=V\_max;

end

if DeltaX(j,i)<-V\_max

DeltaX(j,i)=-V\_max;

end

X(j,i)=X(j,i)+DeltaX(j,i);

end

Flag4ub=X(:,i)>ub';

Flag4lb=X(:,i)<lb';

X(:,i)=(X(:,i).\*(~(Flag4ub+Flag4lb)))+ub'.\*Flag4ub+lb'.\*Flag4lb;

end

display(\['At the iteration ', num2str(iter), ' there are ', num2str(Archive\_member\_no), ' non-dominated solutions in the archive'\]);

end

figure

Draw\_ZDT1();

hold on

if obj\_no==2

plot(Archive\_F(:,1),Archive\_F(:,2),'ko','MarkerSize',8,'markerfacecolor','k');

else

plot3(Archive\_F(:,1),Archive\_F(:,2),Archive\_F(:,3),'ko','MarkerSize',8,'markerfacecolor','k');

end

legend('True PF','Obtained PF');

title('MODA');

## 3 仿真结果 ##

![72cab00e9434bd634f05bfa980ae9f3e.png][]

## 4 参考文献 ##

\[1\]董海, 徐德珉. 基于蜻蜓算法和最小二乘向量机的小批量生产质量预测\[J\]. 科技管理研究, 2019, 000(022):256-260.

![62636861e5c0f0131e4e3e44f72747b8.png][]

[1e04c4123a75b439c7aa4d125c3b87f9.png]: /images/20220828/631d5c21a3544dfe9588b7018c42298a.png
[01be15b3763c326086dc2204011c14f6.png]: /images/20220828/061d1cfe7233453db1524d7a9ba13898.png
[85e52e22a494cd7ee6ea5a3e9533bc6b.png]: /images/20220828/c3405b83907746b9a4d1fa46502c5aa8.png
[7cb07ad9824497acf3b57b85cd2e4192.png]: /images/20220828/69901c919aee4d31a2e6c0e446b8250b.png
[72cab00e9434bd634f05bfa980ae9f3e.png]: /images/20220828/f0c968bcf11d49cc8dbb411b9d734c20.png
[62636861e5c0f0131e4e3e44f72747b8.png]: /images/20220828/6c4f37d909294edd8a50d3a38c7f75b9.png