Subscript indices must either be real positive integers or logicals?
1 次查看(过去 30 天)
显示 更早的评论
The below code is not able to execute due to the stated error in the line "U(i,:)=max(lb,U(i,:))" , but I really cant find any error.
tic
%%prob settings
lb=[-100 -100 ];
ub= [100 100 ];
%% parameters for differential evolution
np=10;
max=500;
%it=100;
cr=0.7;
F=0.8;
e=exp(1);
%x_ss=[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0];
%% START of DE
f=NaN(np,1);
fu=NaN(np,1);
d=length(lb);
U=NaN(np,d);
p= repmat(lb,np,1)+(repmat((ub-lb),np,1).*rand(np,d));
for i=1:np
x=p(i,1);
y=p(i,2);
f(i)= -(20*exp(-0.2*sqrt((1/2)*(x^2+y^2))))+20+e-exp((1/2)*(cos(2*pi*x)+cos(2*pi*y)));
end
%% iteration loop
for it=1:max
for i=1:np
candidates=[1:i-1 i+1:np];
idx= candidates(randperm(np-1,3));
X1= p(idx(1),:);
X2= p(idx(2),:);
X3= p(idx(3),:);
v= X1+F*(X2-X3);
for j=1:d
irand= randi(d,1);
if (rand<=cr || j==irand)
U(i,j)= v(j);
else
U(i,j)=p(i,j);
end
end
end
%% fitness function
for i=1:np
U(i,:)= min(ub,U(i,:));
U(i,:)= max(lb,U(i,:));
A=U(i,:);
x=A(1);
y=A(2);
fu(i)= -(20*exp(-0.2*sqrt((1/2)*(x^2+y^2))))+20+e-exp((1/2)*(cos(2*pi*x)+cos(2*pi*y)));
A=p(i,:);
x=A(1);
y=A(2);
f(i)= -(20*exp(-0.2*sqrt((1/2)*(x^2+y^2))))+20+e-exp((1/2)*(cos(2*pi*x)+cos(2*pi*y)));
if (fu(i)<f(i))
p(i,:)= U(i,:);
f(i)=fu(i);
else
p(i,:)=p(i,:);
end
end
end
toc
[bestfitness ind]= min(f)
bestsol=p(ind,:)
0 个评论
采纳的回答
Chunru
2021-5-29
You define max as a variable
max=500;
However you intend to use max as a function in:
U(i,:)= max(lb,U(i,:));
Since max is defined as a variable (overwriting as a function), the above statement is trying to index to max as a scalar variable. This causes the error.
To correnct the error, change max (as a variable) into other names, for example:
NMax = 500;
You may also need to change the following line accordingly:
for it=1:max % max==>NMax
0 个评论
更多回答(1 个)
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Startup and Shutdown 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!