Equation solution is not found for all array element

1 次查看（过去 30 天）

Mustafa 2023-3-31

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1938634-equation-solution-is-not-found-for-all-array-element

评论： Mustafa 2023-4-3

Hi I have an equation that includes just one variable. With vpasolve command, it is just solved until 105th array value. But array length is 298x1. what is the wrong? and is there any suggestion? (my opinion is, when i equal to 106 and higher, program can't find real solution, but it should be)

clear all, clc, format shortG, close all
name="XXX";
S=load(name+".m");
F=S(:,1);
Vd=S(:,2);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
syms t_bulk Te n hw me t_p eps_inf eps_s w
% consts
e=1.602E-19;
h=6.626E-34;
h_p=1.05E-34;
eps_0=8.854E-12;
kb=1.38e-23;
T_lattice=300;
m0=9.1E-31;
thickness=5E-5;
% Parameters
n=(2.59E14)/thickness;
hw=0.034*e;
me=0.057*m0;
t_p=160E-15;
eps_inf=11.46*eps_0;
eps_s=13.94*eps_0;
w=(hw/h_p);
Nc=(2*((2*pi*me*kb*T_lattice)/(h^2))^1.5)*1E-6;
K0=(hw/(2*kb*Te));
t0=1/(((e^2)*(w)/(2*pi*h_p))*sqrt(me/(2*hw))*((1/(eps_inf))-(1/(eps_s))));
t_bulk=(t0+t_p*((n*kb*Te)/(2*Nc*hw))*(1-exp(-hw/(kb*Te))));
n0=(1/(exp(hw/(kb*T_lattice))-1));
Pexp=(e.*F*1000.*Vd);
Ptheory1=((sqrt(2/pi)*(hw/t_bulk)*((n0+1)*exp(-hw/(kb*Te))-n0)*(sqrt(hw/(2*kb*Te)))*exp(hw/(2*kb*Te))*besseli(0,K0)));
Ptheory1=simplify(Ptheory1);
for i=1:length(Pexp)
    T1(i,1)=double(vpasolve(Ptheory1==Pexp(i,1),Te, 1));
end

And error is below: Unable to perform assignment because the size of the left side is 1-by-1 and the size of the right side is 0-by-1.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

John D'Errico 2023-3-31

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1938634-equation-solution-is-not-found-for-all-array-element#answer_1205649

编辑：John D'Errico 2023-3-31

在 MATLAB Online 中打开

For SOME of these sub-problems, no solution was found.

I would suggest that when the result from vpasolve is empty, you could assign a NaN for those cases.

for i=1:length(Pexp)
  result = double(vpasolve(Ptheory1==Pexp(i,1),Te, 1));
  if isempty(result)
    T1(i,1) = NaN;
  else
    T1(i,1) = result;
  end
end

10 个评论
显示 8更早的评论隐藏 8更早的评论

Walter Roberson 2023-4-1

Your Ptheory1 does not depend upon F or Vd. Instead, the product of F and Vd combine with other constants to become Pexp, the value that Ptheory1 must equal.

If you fplot(Ptheory1) in the area near 700, such as 650 to 750, then you will see that it peaks slightly less than your Pexp values. The peak is at approximately Te = 714.80196643093804698707623194146 with a value of approximately 1.2109926630804e-09

Mustafa 2023-4-3