interp2 using iterated values in a loop - NaN values returned for subsequent iterations after defined initial condition

Question

George Verrier 2023-1-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1892300-interp2-using-iterated-values-in-a-loop-nan-values-returned-for-subsequent-iterations-after-define

评论： Walter Roberson 2023-3-30

For my script I am tyrying to get interp2 to read values from previous lines completed in an iterative loop. Please see the code:

for i=1:n
    m(i+1) = m(i) + dt*mdot;
    T(i+1) = T(i) + dt*mdot*(T(i)*V/mmol+he-h(i))/(m(i)*cp(i)-m(i)*V/mmol);
    P(i+1) = P(i) + dt*(mmol*m(i)*(T(i+1)-T(i))/dt+T(i)*mmol*m(i)); 
    h(i+1) = interp2(Ph,Th,Vhn,P(i+1)/10^6,T(i+1),"linear"); %h(i+1) values calculated using (P(i+1) and T(i+1)
    cp(i+1) = interp2(Pcp,Tcp,Vcpn,P(i+1)/10^6,T(i+1),"linear"); %see note above for cp(i+1)
end

The code, with the rest of the scrip runs without error, but the loop is not able to calculate values for h(i+1) and cp(i+1). Instead it returns NaN values for subsequent values of h and cp when opening the array in the workspace. I understand that NaN values could indicate that the values for Xp or Yq are outside of the limits of X or Y respectively, however this is not the case. I think the issue lies in the use of interp2 in an iterative loop. I am hoping that this can be solved so h(i+1) and cp(i+1) can be solved for at each iteration loop from the first to the nth run. Please can anyone offer a way to resolve this issue.

2 个评论
显示无隐藏无

Torsten 2023-1-11

There cannot be another reason for h(i+1) and/or cp(i+1) to be NaN values as

a) the inputs contain NaN/Inf values or

b) Xq and/or Yq are out of range.

Walter Roberson 2023-1-11

Note that when you request linear, which is the default, then extrapolation is not automatically turned on.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Aditya Srikar 2023-3-30

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1892300-interp2-using-iterated-values-in-a-loop-nan-values-returned-for-subsequent-iterations-after-define#answer_1204579

编辑：Aditya Srikar 2023-3-30

Hi George

The function interp2 returns interpolated values of a function at specific query points using linear interpolation. It might return NaN values if the input contains NaN or Inf values.

Consider these statements

h(i+1) = interp2(Ph,Th,Vhn,P(i+1)/10^6,T(i+1),"linear");

cp(i+1) = interp2(Pcp,Tcp,Vcpn,P(i+1)/10^6,T(i+1),"linear");

Suppose if any of the values P(i+1) or T(i+1) become NaN or INF in current iteration, then the output returned will be NaN and it will effect all the subsequent iterations.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Walter Roberson 2023-3-30

Much more common of a problem, though, is requesting to interpolate at a point that is out of the defined range when you have not turned extrapolation on.

请先登录，再进行评论。