Is it possible that ploting bode diagram without transfer fuction?

Hi everyone,

Is it possible that ploting bode diagram without transfer fuction?

I have some values of the Power and the Speed as inputs and some output values in excel format. However the transfer fuction of the system has not been defined.

Are there any one who knows any method or code to plotting bode diagram without transfer function?

Answer 1

[Star Strider]

Yes. Use the fft function.

input = randn(1, 10000);
output = randn(1, 10000);
time = linspace(0, 1000, 10000);

Fs = 100;                           % Sampling Frequency
Fn = Fs/2;                          % Nyquist Frequency
L = numel(time);
FTinp = fft(input)/L;
FTout = fft(output)/L;
TF = FTout ./ FTinp;                % Transfer Function
Fv = linspace(0, 1, fix(L/2)+1)*Fn; % Frequency Vector
Iv = 1:numel(Fv);                   % Index Vector

figure
subplot(2,1,1)
plot(Fv, 20*log10(abs(TF(Iv))))
title('Amplitude')
ylabel('dB')
subplot(2,1,2)
plot(Fv, angle(TF(Iv))*180/pi)
title('Phase')
ylabel('°')

Provide your own vectors for ‘input’, ‘output’, and ‘time’.

[Heng Jiang]

Hello, could you please tell me that, if the input is a sine wave with changing frequencies, like: input = 10*sin(2*pi*w.*time), w = linspace(0, 100, 10000), how should I modify the Frequency vector? Should Fv be replaced with w?

[Star Strider]

That would appear to be a linear chirp sort of input. I doubt that the code would change much (if at all), since the chirp input woud likely be similar to an impulse input in that it covers a wide band of frequencies. It might be wortth exploring the input and output separately with one of the spectrogram plots (specttrogram or pspectrum). Otherwise, do tthat experiment yourself and see what the result is.