I have prepared Matlab code (FFT and PSD) for frequency domain analysis of Quarter car passive suspension (Simulink Model Attached). The two Natural Frequencies of this 2DOF model are 3 Hz and 20 Hz.
The time-domain Acceleration vs Time data is in X.xlsx file, whereas Displacement vs Time data is in D.xlsx file. The respective Matlab files for FFT are 'accelfft.m' and 'dissfft.m'.
They seem to work, but I am not sure whether they are giving the right result or not. Also for displacement FFT gives a peak at 0 Hz, how to remove it?
Are these FFT nd PSD codes giving the right output? How can I find peak values and corresponding frequency by Matlab command?
Simulink Model, Matlab Code files are attached.
Your code seems to be correct. The peak frequency at zero corresponds to the DC component of the signal. You can type mean(Acc1) and see that the value at 0 Hz is the average signal component at DC in both cases, so nothing is wrong. You can easily find the peak frequency by searching for the max amplitude and find the corresponding frequency; e.g. for accelfft.m :
[maxValue, indexOfMaxValue] = max(yPower1);
fmax=f(indexOfMaxValue)
Ramy is correct. If you apply an FFT on your input signal, you are converting your time domain signal into frequency domain and you will get a spike for every change of information with time. If you take an FFT of a constant DC value, you will get only one spike irrespective of your input size and rest of contents value will be zero. You can you 'pwelch' command in matlab to observe your PSD.
Mr. Ramy and Mr. Md. Atiqul, Thanks a lot for your valuable responses. I have another query that when I m trying to obtain FFT of Unsprung Mass acceleration using file "musaccelfft.m" with excel file "E.xlsx" containing Unsprung Mass acceleration Vs Time data (Files attached), It gives two peaks one at 3 Hz and Other at 17 Hz.
The two natural frequency of quarter car model 'PassiveSuspensio.mdl' is 3 Hz and 20 HZ as per file "Naturalfreq.m".
My question is why the peak is coming at 17 Hz not at 20 Hz?
Why sprung mass acceleration FFT is showing only one peak, whereas Unsprung Mass acceleration FFT is showing Two Peaks?
One possible explanation could be the frequency resolution delta_f of your FFT. This is your sample frequency f_s multiplied by two and divided by your number of samples N. So maybe the 20Hz doesn`t fit in your grid and therefore the peak seems to be at 17Hz.
Thanks for your reply Mr. Marc, but your point is not very clear to me. Are you discussing about the line no 16 in the code? According to you what is the solution for this?
My other point was that why for unsprung mass acceleration I am getting two peaks, whereas for sprung mass acceleration (file shared earlier in this question) only one peak?
Pankaj Sharma the FFT shows the output signal frequency response. Natural frequency depends as well on the excitation signal. Your MATLAB code is correct but the excel file data may have been generated with different excitation.
By the way if you like any answers you can recommend them so that the system acknowledges the correct answer for others.
- Badges
- Science topic