Fourier transform of dynamic mode coefficients for impinging jet along a line from nozzle exit to plate at a distance 0.6 times jet diameter

Hi everyone. I want to calculate the Fourier transform of dynamic mode coefficients for impinging jet along a line from nozzle exit to plate at a distance 0.6d. Then I want to plot the normalized wavenumber(K) spectra (Amplitude vs Kd). I have shared the code here.

Phib=reshape(Phi(:,2),96,1280);
for i=29
    for j=945:-1:618
        Phibb=Phib(i,945:-1:618);
    end
end
Phibb=Phibb-mean(Phibb);
Phibn=Phibb./max(Phibb);
figure(88);
xl=0:4/327:4;
plot(xl,Phibn);
Px=Phibb./max(Phibb);
ff=fft(Px);
ffk=fftshift(ff);
ffn=sqrt((real(ffk).^2)+(imag(ffk).^2));
% Compute the corresponding frequencies
Ln = length(Px);
fre = linspace(-Ln/2, Ln/2, Ln);
kk=2*pi.*fre;
kkn=kk*0.01; %nondimensionalizing wrt jet dia
% xlim=3.5:-3.5/18135:0;
% subplot(r,1,i);
figure(900)
plot(kkn,ffn);
xlabel('kD')
ylabel('|Amplitude|')

I am not getting proper results. Is there any mistake I am doing in the plot? Please check my code.

Answer 1

[Paul]

Hi DEBI,

It will be easier for people to help if yor provide all of the information needed to run your code. You should save the variable Phi in a .mat file and then add it to your question using the paperclip icon.

Also, it would be helpful if you show, or at least explain, why the result is not what you expect and what the expected result is.

Also, just from looking at the code, the calculation for fre is incorrect. It should be

fre = linspace(-Ln/2, Ln/2-1, Ln);     % if Ln is even
fre = linspace(-(Ln-1)/2,(Ln-1)/2,Ln)  % if Ln is odd

But the next line is:

kk=2*pi.*fre;

which kind of suggests that fre is supposed to be in units of Hz, either cycles/sample or cycles/sec. But fre is just a vector of integers, and probably needs to be scaled by 1/Ln or some other factor to get the correct units.

[DEBI SARANGI]

Hi...

Thank you for the response. I have attached the data of Phibb matrix and attached the plot I am getting from my code as well as the plot which I am expecting.Plz check.

[Paul]

Not working. Why not save phibb in a .mat file?

data = readtable('phibb.xlsx');

Warning: Column headers from the file were modified to make them valid MATLAB identifiers before creating variable names for the table. The original column headers are saved in the VariableDescriptions property.
Set 'VariableNamingRule' to 'preserve' to use the original column headers as table variable names.

who

Your variables are:

ans     cmdout  data    

data

data =

  0×328 empty table

[Image Analyst]

All your data seems to be complex numbers in Excel and readmatrix reads them all in as NANs. Please give us code for reading in the data. Otherwise attach the variables in a .mat file.

y = readmatrix('phibb.xlsx')

y = 1×328

NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN

whos y

  Name      Size             Bytes  Class     Attributes

  y         1x328             2624  double              

[DEBI SARANGI]

Hi...

Thank you for the response. Please find the data of phibb in the matlab file.

[Walter Roberson]

Note that Excel does not support complex data at all without an extension, and the extension codes complex data similar to =COMPLEX(real,imaginary) . Therefore the only way to handle that xlsx file would be to use a reading function that treated all of the columns as text and then converted the text to numeric.

[DEBI SARANGI]

Okay sir. Thank you for the response. I have attached the .mat file of phib. Please check the data and my code shared above.

[DEBI SARANGI]

The pixel value corresponds to the distance from nozzle exit to plate surface which is 0 to 4 cm. And the pixel size 29 corresponds to 0.6 times jet dia where complete pixel value for the nozzle dia is from 0 to 96. The jet dia is 1 cm. I want to plot the spatial frequency vs amplitude plot. The data related to this I have shared above. Please check my code.

Thank you.