FFT for Lift coefficient to find the frequency

Hello, everyone!

I have different data files(.txt) includes simulation time, lift coefficient and drag coefficient.

I want to find the some frequencies which make periodic motion of lift.

The code is follwed :

load CdCl5.txt; a=CdCl5;    Time5=a(1:end,2); Lift5=a(1:end,4);
load CdCl7.txt; a=CdCl7;    Time7=a(1:end,2); Lift7=a(1:end,4);

figure(1)   % Graph of the life coeffi.
subplot(211)
plot(Time5, Lift5)
title('Scheme 1')
xlabel('Time')
ylabel('C_L')
subplot(212)
plot(Time7, Lift7)
title('Scheme 2')
xlabel('Time')
ylabel('C_L')

% Parameter
N = length(Time5);
T = Time5(end);  % Time length
dt = T/N;
df = 1/T;   % frequency interval
f = df*(0:N-1); % frequency sequence

X5 = fft(Lift5)/N;
X7 = fft(Lift7)/N;

figure(2)
subplot(211)
plot(f,abs(X5))
title('Scheme 1')
ylabel('|X5(f)|')
xlabel('Frequncy[hz]')
subplot(212)
plot(f,abs(X7))
title('Scheme 2')
ylabel('|X7(f)|')
xlabel('Frequncy[hz]')

Do I analyze the figure(2) graphs to find the frequencies?

Could someone help me?

Thank you so much!

Sopo.

Answer 1

[Mathieu NOE]

hello

I prefer to put the repetitive fft lines in a subfunction , so the code can be cleaner and more compact

for the fft result , we are interested only in the positive half frequency points

you can use max or findpeaks to get the peaks value

a = readmatrix('CdCl5.txt');
Time5=a(1:end,2); Lift5=a(1:end,4);

a = readmatrix('CdCl7.txt');
Time7=a(1:end,2); Lift7=a(1:end,4);

figure(1)   % Graph of the life coeffi.
subplot(211)
plot(Time5, Lift5)
title('Scheme 1')
xlabel('Time')
ylabel('C_L')
subplot(212)
plot(Time7, Lift7)
title('Scheme 2')
xlabel('Time')
ylabel('C_L')

% FFT plot 
[f1,fft_spectrum1] = do_fft(Time5,Lift5);
[f2,fft_spectrum2] = do_fft(Time7,Lift7);

figure(2)
[PKS1,LOCS1] = max(fft_spectrum1);
f1_peak = f1(LOCS1)
subplot(211)
semilogx(f1,fft_spectrum1,f1_peak,PKS1,'dr')
text(f1_peak*1.25,PKS1 , (['f1 = ' num2str(f1_peak) ' Hz']));
title('Scheme 1')
ylabel('|X5(f)|')
xlabel('Frequency[hz]')
    
subplot(212)
[PKS2,LOCS2] = max(fft_spectrum2);
f2_peak = f2(LOCS2)
semilogx(f2,fft_spectrum2,f2_peak,PKS2,'dr')
text(f2_peak*1.25,PKS2 , (['f2 = ' num2str(f2_peak) ' Hz']));
title('Scheme 2')
ylabel('|X7(f)|')
xlabel('Frequency[hz]')


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [freq_vector,fft_spectrum] = do_fft(time,data)

dt = mean(diff(time));
Fs = 1/dt;
nfft = length(data); % maximise freq resolution => nfft equals signal length

fft_spectrum = abs(fft(data))/nfft;

% one sidded fft spectrum  % Select first half 
    if rem(nfft,2)    % nfft odd
        select = (1:(nfft+1)/2)';
    else
        select = (1:nfft/2+1)';
    end
fft_spectrum = fft_spectrum(select,:);
freq_vector = (select - 1)*Fs/nfft;

end

[Sopo Yoon]

Thank you, appreciate it.

It looks good and easy to understand.

Best

Sopo

[Mathieu NOE]

My pleasure !