while loop won't run?

Hello,

I am programming the simulation of a thermoelectric device coupled to PV cells and for some reason the while loop I have coded won't seem to run. When I run my script it skips it and goes directly to end (which I can tell by putting breakpoints). If anyone has an idea as to how I can fix this, that would be of great help! Thank you :)

   %%Task 5
   clc;close all; clear all;

   %%1. Guess Tpv:
   Tpv=293; %K

   %%2. Run Task 3 to compute the power output Pl and waste heat Qdotpv
      Rlpv=.007; %Ohm
      Rs=0; %Ohm
      Apv=(.1)^2; %m^2
      rc=12; %nounit
      Id=1350; %J.s^-1.m^-2
      Pin=rc*Id; %J.s^-1.m^-2

   PL=powdev(Tpv, Rlpv);
   display ('in W=J/s');

   Qdotpv=Pin*Apv-PL;
   display('in W=J/s');

   %%3. Set TH and run Task 4 to get TC and QHdot
      % Parameters
   Sigcont=300; %W/K
   Tsat=90; %K
   Rlte=.10; %ohm
   n=12; %nounit
   rhoA=.002; %ohm.cm
   rhoB=.003; %ohm.cm
   lamA=.032; %W/(cm.K)
   lamB=.021; %W/(cm.K)
   alpha=.0017; %V/K
   kweff=6.5; %W.m^-1.K^-1
   tw=4.5*10^-3; %m
   TCguess=90; %K
      % To get TC
   TH=Tpv-Qdotpv/Sigcont 
    %
  [TC,Err2]=optTC(TCguess,TH,Tsat,Rlte,n,rhoA,rhoB,lamA,lamB,alpha,kweff,tw,Apv);
   display(TC);
   display('both in K');
   Th=TH-273
   Tc=TC-273
   display('both in deg C');
   %
      % To get QHdot :
          %a
          RSigmin=((lamA*rhoA)^0.5+(lamB*rhoB)^0.5)^2; %W.Ohm.K^-1
          %b
          Z=alpha^2/RSigmin; %K^-1
          %c
          mmaxeff=(1+0.5.*(TH+TCguess)*Z).^0.5; %nounit
          %d
          R=Rlte./(n*mmaxeff); %ohm
          %iii : efficiency
          eta=((TH-TC)/TH).*((1+mmaxeff).^2./mmaxeff.*(Z.*TH)^-1+1+1./(2.*mmaxeff).*(1+TC./TH)).^-1; %nounit
          %
          %iv : power output
          Rbatt=n.*R; %Ohm
          Wdot=n.^2.*alpha.^2.*(TH-TC).^2.*Rlte./(Rlte+Rbatt).^2; %W.m^2
          %
          %v : heat input and rejection for the TE device
          QHdot=Wdot./eta %W
          display('in W=J/s');

   %%4. Compute the error and optimize Tpv

   Error=Qdotpv-QHdot; %W
   i=10;

while Error>=10^-5 

      if sign(Error)==-1
           Tpv=Tpv+i
           PL=powdev(Tpv, Rlpv);% W
           Qdotpv=Pin*Apv-PL; % W
           TH=Tpv-Qdotpv/Sigcont; %K
           TCguess=TC; %K
  %           [TC,Err2]=optTC(TCguess,TH,Tsat,Rlte,n,rhoA,rhoB,lamA,lamB,alpha,kweff,tw,Apv);
           RSigmin=((lamA*rhoA)^0.5+(lamB*rhoB)^0.5)^2;%W.Ohm.K^-1
           Z=alpha^2/RSigmin; %K^-1
           mmaxeff=(1+0.5.*(TH+TCguess)*Z).^0.5; %nounit
           R=Rlte./(n*mmaxeff); %ohm
           eta=((TH-TC)/TH).*((1+mmaxeff).^2./mmaxeff.*(Z.*TH)^-1+1+1./(2.*mmaxeff).*(1+TC./TH)).^-1; %nounit
           Rbatt=n.*R; %Ohm
           Wdot=n.^2.*alpha.^2.*(TH-TC).^2.*Rlte./(Rlte+Rbatt).^2; %W.m^2
           QHdot=Wdot./eta; %W
           Error=abs(Qdotpv-QHdot) %W
      else
          Tpv=Tpv-i;
           i=i*0.5;
           Tpv=Tpv+i
           PL=powdev(Tpv, Rlpv);% W
           Qdotpv=Pin*Apv-PL; % W
           TH=Tpv-Qdotpv/Sigcont; %K
           TCguess=TC; %K
  %           [TC,Err2]=optTC(TCguess,TH,Tsat,Rlte,n,rhoA,rhoB,lamA,lamB,alpha,kweff,tw,Apv);
           RSigmin=((lamA*rhoA)^0.5+(lamB*rhoB)^0.5)^2;%W.Ohm.K^-1
           Z=alpha^2/RSigmin; %K^-1
           mmaxeff=(1+0.5.*(TH+TCguess)*Z).^0.5; %nounit
           R=Rlte./(n*mmaxeff); %ohm
           eta=((TH-TC)/TH).*((1+mmaxeff).^2./mmaxeff.*(Z.*TH)^-1+1+1./(2.*mmaxeff).*(1+TC./TH)).^-1; %nounit
           Rbatt=n.*R; %Ohm
           Wdot=n.^2.*alpha.^2.*(TH-TC).^2.*Rlte./(Rlte+Rbatt).^2; %W.m^2
           QHdot=Wdot./eta; %W
           Error=abs(Qdotpv-QHdot) %W
      end
  end
  display (Tpv);

[per isakson]

"my script it skips it and goes directly to end" &nbsp Because Error is zero or less - I guess

What's powdev , optTC , etc. ?

[maud]

I think the problem is coming from my function optTC. here's a copy of it, if you see anything wrong let me know! I think it may because of syms or the fact that there is two outputs, not sure...

%%Optimizing TC
function [TC,Err2]=optTC(TCguess,TH,Tsat,Rlte,n,rhoA,rhoB,lamA,lamB,alpha,kweff,tw,Apv)
%a
RSigmin=((lamA*rhoA)^0.5+(lamB*rhoB)^0.5)^2; %W.Ohm.K^-1
%b
Z=alpha^2/RSigmin; %K^-1
%c
mmaxeff=(1+0.5.*(TH+TCguess)*Z).^0.5; %nounit
%d
R=Rlte./(n*mmaxeff); %ohm
%e
Sigma=RSigmin./R; %W.K^-1
%ii : Qcboil
QCboil=Apv*kweff.*(TCguess-Tsat)/tw; %W.m^2
%iii : efficiency
eta=((TH-TCguess)/TH).*((1+mmaxeff).^2./mmaxeff.*(Z.*TH)^-1+1+1./(2.*mmaxeff).*(1+TCguess./TH)).^-1; %nounit
%iv : power output
Rbatt=n.*R; %Ohm
Wdot=n.^2.*alpha.^2.*(TH-TCguess).^2.*Rlte./(Rlte+Rbatt).^2; %W.m^2
%v : heat input and rejection for the TE device
QHdot=Wdot./eta; %W
QCdotTE=QHdot-Wdot; %W
%vi : error
Err1=double(vpa(QCdotTE-QCboil,10)); %W
Err2=1;
i=10;
TC=TCguess;
while abs(Err2)>=10^-5
    if sign(Err1)==sign(Err2)
        TC=TC+i;
        %c
        mmaxeff=(1+0.5.*(TH+TC)*Z).^0.5; %nounit
        %d
        R=Rlte./(n*mmaxeff); %ohm
        %e
        Sigma=RSigmin./R; %W.K^-1
        %ii : Qcboil
        QCboil=Apv*kweff.*(TC-Tsat)/tw; %W.m^2
        %iii : efficiency
        eta=((TH-TC)/TH).*((1+mmaxeff).^2./mmaxeff.*(Z.*TH)^-1+1+1./(2.*mmaxeff).*(1+TC./TH)).^-1; %nounit
        %iv : power output
        Rbatt=n.*R; %Ohm
        Wdot=n.^2.*alpha.^2.*(TH-TC).^2.*Rlte./(Rlte+Rbatt).^2; %W.m^2
        %v : heat input and rejection for the TE device
        QHdot=Wdot./eta; %W
        QCdotTE=QHdot-Wdot; %W
        %vi : error
        Err2=double(vpa(QCdotTE-QCboil,10)); %W
        %
    else
        TC=TC-i;
        i=0.5*i;
        TC=TC+i;
        %c
        mmaxeff=(1+0.5.*(TH+TC)*Z).^0.5; %nounit
        %d
        R=Rlte./(n*mmaxeff); %ohm
        %e
        Sigma=RSigmin./R; %W.K^-1
        %ii : Qcboil
        QCboil=Apv*kweff.*(TC-Tsat)/tw; %W.m^2
        %iii : efficiency
        eta=((TH-TC)/TH).*((1+mmaxeff).^2./mmaxeff.*(Z.*TH)^-1+1+1./(2.*mmaxeff).*(1+TC./TH)).^-1; %nounit
        %iv : power output
        Rbatt=n.*R; %Ohm
        Wdot=n.^2.*alpha.^2.*(TH-TC).^2.*Rlte./(Rlte+Rbatt).^2; %W.m^2
        %v : heat input and rejection for the TE device
        QHdot=Wdot./eta; %W
        QCdotTE=QHdot-Wdot; %W
        %vi : error
        Err2=double(vpa(QCdotTE-QCboil,10)); %W
    end
end
end

[maud]

here it the other function powder if you need it:

function [Pl,efficiency]=powdev(Tpv, Rlpv)

qe=1.6022*10^-19; %J
kB=1.38*10^-23; %J.K^-1
Tsource=6000; %K temperature of the sun
Apv=(10*10^-2)^2; %m^2

rc=12;
Id=1350; %J.s^-1.m^-2
Pin=rc*Id; %J.s^-1.m^-2
Phi=Pin*2.404*15/(pi^4*kB*Tsource); %m^-2.s^-1
Phig=0.416*1.3495429*Phi; %m^-2.s^-1
Inu=Phig*qe*Apv; %C.s^-1=A
Io=Inu*(3*10^-8.*Tpv.^3-3*10^-5.*Tpv.^2+0.0098.*Tpv-2.083); 
%A (see excel for details)

syms x;
solve (Rlpv*x==-Tpv*kB/qe*log(1+(Inu-x)/Io),x);
Il=eval(ans); %a=C.s^-1

Vl=Rlpv*Il; %V
Pl=Vl.*Inu-Vl.*Io.*(exp(qe.*Vl/(kB*Tsource))-1); %W=J/s
efficiency=Pl/(Pin*Apv);

end

Answer 1

[Roger Stafford]

I am guessing that you need to change the initial ‘Error’ to:

   Error = abs(Qdotpv-QHdot); %W

As you have it, if the error is large but negative with Qdotpv<QHdot, the ‘while’ loop will never execute.

[maud]

thanks!! I have added the abs but still isn't working because of my optTC function apparently, I have posted it on the previous comment.