Trapz error (ORDER contains an invalid permutation index)

Question

Alvin 2017-11-3

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/364829-trapz-error-order-contains-an-invalid-permutation-index

评论： Alvin 2017-11-4

clear
C1 = linspace(0,1e+04,3000);
k1 = 1;
k2 = 0;
n1 = 0;
n2 = 0;
nm = 0;
D = 0;
O = 0;
gam = 1e-04.*k1;
alph = 0;
 w = linspace(-5,5,3000);
 for ii=1:length(C1)
    for jj=1:length(w)
        TA = (w(jj)+D)./(k1./2);
        TB = (w(jj)-O)./(gam/2);
        xa = 1-1i*TA;
        xb = 1-1i*TB;
        C = (2.*1i.*sqrt(C1(ii)))./(sqrt(gam).*((xa.*xb)+C1(ii)));
        D = (2.*xa)./(sqrt(gam).*((xa.*xb)+C1(ii)));
        MC = (abs(C)).^2;
        MD = (abs(D)).^2;
        Sbb = 2.*pi*(MC.*(n1+0.5)+MD.*(nm+0.5));
        A = trapz(w,Sbb)./(2.*pi);
        AA(ii)=A;Cum
        figure(1)
        plot(w,Sbb)
        set(gca,'FontSize',13)
        xlabel('\omega')
        ylabel('S_{bb}')
        %ylim([0,200])
        drawnow
    end
end

As stated in my title, the error:ORDER contains an invalid permutation index was returned upon compiling the code. I've searched this up and found that if I put my trapz to

trapz(w,Sbb,1)
or
trapz(w,Sbb,2)

I could remedy the problem. But it appears to not work. Can anyone assist me?

Much thanks!

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。

Answer 1

Walter Roberson 2017-11-3

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/364829-trapz-error-order-contains-an-invalid-permutation-index#answer_289180

在 MATLAB Online 中打开

You have trapz(w,Sbb) with Sbb being a scalar.

There are two trapz() syntaxes with two arguments:

trapz(X, Y)
trapz(X, dim)

The way that MATLAB determines which one is being used is it looks to see whether the second argument is a vector or not: if it is then it assumes the first syntax, but if it is a scalar (as is the case for you) then it assumes the second syntax is being used.

Your w values form a line. You should not be trying to trapz() a scalar Y against a vector of X.

I think if you investigate, you will find that you can vectorize the Sbb calculation over all w values instead of using a loop. It does make sense to trapz with a vector of Sbb.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Alvin 2017-11-4

Thanks! I vectorized w and kept only one loop (over C1) and this seems to work. That way my Sbb is kept as a vector so trapz(w,Sbb) functions normally!

请先登录，再进行评论。