# Why I cannot exit the for cycle with the "z" counter?

1 view (last 30 days)
Emilio Pulli on 15 Nov 2021
Edited: Emilio Pulli on 15 Nov 2021
I want that if a certain condition is met the iterations inside the cycle where the condition is met ends and the external for proceeds to the next case. The if condition is the "Cp(j,z)<=0" one and once it is met, I want that the code exits from the "z" for by going to the "j" for and increases it. The counter "z" is set on the vector omega_rpm correctly defined out of the concatenated for cycles.
Anyone can help?
for k=1:length(TSR)
k
for t=1:length(r_rif)
t
for j=1:length(v)
j
for z=1:length(omega_rpm)
z
omega(z)=omega_rpm(z)*(2*pi)/60;
for i=1:length(r)
i
a_new(j,:)=a_glauert(k,:);
a1_new(j,:)=a1_glauert(k,:);
a_old(j,:)=zeros(1,length(r));
a1_old(j,:)=zeros(1,length(r));
sigma(i)=chord_dim(k,t)*Nb/(2*pi*r(i));
counter=0;
while (abs(a_new(j,i)-a_old(j,i))>toll || abs(a1_new(j,i)-a1_old(j,i))>toll) && counter<10000
if abs(a_new(j,i)-a_old(j,i))>toll
a_old(j,i) = eps*a_new(j,i) + (1-eps)*a_old(j,i);
end
if abs(a1_new(j,i)-a1_old(j,i))>toll
a1_old(j,i) = eps*a1_new(j,i) + (1-eps)*a1_old(j,i);
end
fi(j,i)=atan(((1-a_old(j,i))*v(j))/((1+a1_old(j,i))*omega(z)*r(i)));
W(j,i)=((1-a_old(j,i))*v(j))/sin(fi(j,i));
f(j,i)=Nb*(R-r(i))/(2*r(i)*sin(fi(j,i)));
F(j,i)=2/pi*acos(exp(-f(j,i)));
Re(j,i)=ro_air*chord_dim(k,t)*W(j,i)/mu_air;
B(j,i) = fi(j,i) - Bc(k,t);
[CL_vect, CD_vect, B_vect]=polars_offdesign(Re(j,i),r(i),root,primary,tip,profile_rt,profile_p,profile_t);
[CL(j,i),CD(j,i),error(j,i)]=beta_interp(CL_vect,CD_vect,B_vect,B(j,i));
if error(j,i)==1
check=error(j,i);
lambda_error=TSR(k)
r_rif_error=r_rif(t)
v_error=v(j)
omega_error=omega(z)*60/(2*pi)
r_error=r(i)
return
end
if a_old(j,i)<0.4
a_new(j,i)=sigma(i)*W(j,i)*(CL(j,i)/tan(fi(j,i))+CD(j,i))/(4*v(j)*F(j,i));
a1_new(j,i)=sigma(i)*W(j,i)*(CL(j,i)-CD(j,i)/tan(fi(j,i)))/(4*omega(z)*r(i)*F(j,i));
elseif a_old(j,i)>0.4
Cx0=2;
a_new(j,i)=(0.5*Nb*W(j,i)^2*chord_dim(k,t)*(CL(j,i)*cos(fi(j,i))+CD(j,i)*sin(fi(j,i)))/(pi*r(i)*F(j,i)*v(j)^2)-Cx0+4*(sqrt(Cx0)-1))/(4*(sqrt(Cx0)-1));
a1_new(j,i)=(0.5*Nb*W(j,i)^2*chord_dim(k,t)*(CL(j,i)*cos(fi(j,i))-CD(j,i)*sin(fi(j,i))))/((1-a_new(j,i))*4*pi*r(i)^2*omega(z)*v(j)*F(j,i));
end
counter=counter+1;
end
a(j,i)=a_new(j,i);
a1(j,i)=a1_new(j,i);
dFt(j,i)=0.5*ro_air*W(j,i)^2.*chord_dim(k,t)*(CL(j,i)*sin(fi(j,i))-CD(j,i)*cos(fi(j,i)));
dFx(j,i)=0.5*ro_air*W(j,i)^2*chord_dim(k,t)*(CL(j,i)*cos(fi(j,i))+CD(j,i)*sin(fi(j,i)));
dM(j,i)=0.5*ro_air*W(j,i)^2*chord_dim(k,t)*(CL(j,i)*sin(fi(j,i))-CD(j,i)*cos(fi(j,i)))*r(i);
dCx(j,i)=sigma(i)*W(j,i)^2*(CL(j,i)*cos(fi(j,i))+CD(j,i)*sin(fi(j,i)))/v(j)^2;
dCp(j,i)=Nb*omega(z)/(pi*R^2*v(j)^3)*(chord_dim(k,t)*W(j,i)^2*(CL(j,i)*sin(fi(j,i))-CD(j,i)*cos(fi(j,i)))*r(i));
dPw(j,i)=ro_air*Nb*omega(z)/2*(chord_dim(k,t)*W(j,i)^2*(CL(j,i)*sin(fi(j,i))-CD(j,i).*cos(fi(j,i)))*r(i));
end
Cp(j,z)=trapz(r,dCp(j,:));
Pw(j,z)=trapz(r,dPw(j,:));
Cm(j,z)=Cp(j,z)/(omega(z)*R/v(j));
M(j,z)=Pw(j,z)/(omega(z));
Fx(j,z)=trapz(r,dFx(j,:));
Cx(j,z)=trapz(r,dCx(j,:));
if Cp(j,z)<=0
break;
end
end
end
end
end

the cyclist on 15 Nov 2021
It looks your syntax is correct for what you are trying to do, and the if statement is correctly placed.
My best guess is that your condition
Cp(j,z)<=0
is not met when you expect it to be. You could put a debugging breakpoint at the line where you try to break out of the loop, and see if the code ever gets there.
Emilio Pulli on 15 Nov 2021
Figured it out! I do not know how but I simply restart Matlab and now it works…I have never experienced something similar, it looked like a bug…