MATLAB: Spectrum analysis of wind

wind spectrum

Hello every one.

I am having Wind hourly data. Now I want to know spectral analysis of wind for cycles/day..

could any one help me out.. It would be better appreciation.

Thanks in advance.. phanindra

Best Answer

If you have the Signal Processing Toolbox, you can use periodogram() to get a power spectral density estimate of the wind data in cycles/day. periodogram() is nice because it will output an appropriate frequency vector for you and confidence intervals if you desire.

I'll show you how to do that with your sampling frequency, 1 sample/hour. I'll have to simulate some data because you did not attach yours.

30 days of data:

   t = 0:1:(30*24)-1;
     x = 1.5*cos(2*pi*(1/24)*t)+randn(size(t));  % 1 cycle per days plus noise
     [Pxx,F] = periodogram(x,rectwin(length(x)),length(x),1);
     plot(F*24,10*log10(Pxx))
     xlabel('Cycles/Day'); ylabel('dB');

Otherwise, you can use fft() in base MATLAB, to the same in base MATLAB, see

PSD using FFT

Related Solutions

MATLAB: Evaluate the variation in the power of a known periodicity

You'll need to use spectrogram() for that. Typically, to get a nice (not blocky looking) output from spectrogram, you overlap your moving windows, but you don't have to do that. You can set the NOVERLAP to 0.

I'm not sure what you intend with your example. It seems like your example has a time vector sampled in increments of 1/24, but you really want your time vector in hours

   t = 1:(365*24);

The problem with just using a 24-hour window when your sampling is once an hour is that you only have 24 data points.

Are you sure you can't use a bit longer window, say 72 hours?

 t = 1:(365*24);
 x = cos((2*pi)/12*t).*(t<72)+1/2*sin((2*pi)/12*t).*(t>72 & t<=96)+1/4*cos((2*pi)/12*t).*(t>96 & t<=144)+randn(size(t));
% compute spectrogram -- moving window of 72 samples, no overlap
[y,f,t,p] = spectrogram(x,72,0,128,1);

The power estimates for the moving windows are contained in the columns of the matrix p

   plot(f,p(:,1))

MATLAB: Is it possible to find the frequency specture from a signal imported from data points (excel)

hello

for your help see my little demo below for audio wav file analysis. Just adapt the load to your data file format

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


% signal
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% [signal,Fs]=wavread('audio_in_noise1.wav');  %(older matlab)
% or 
[data,Fs]=audioread('audio_in_noise1.wav'); %(newer matlab)
channel = 1;
signal = data(:,channel);
samples = length(signal);
dt = 1/Fs;
t = (0:dt:(samples-1)*dt);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% FFT parameters
samples = length(signal);
NFFT = 512*4;    % 
NOVERLAP = round(0.75*NFFT);
w = hanning(NFFT);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%



% display 1 : averaged FFT spectrum
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
[sensor_spectrum, freq] = pwelch(signal,w,NOVERLAP,NFFT,Fs);
figure(1),semilogx(freq,20*log10(sensor_spectrum));grid
title(['Averaged FFT Spectrum  / Fs = ' num2str(Fs) ' Hz / Delta f = ' num2str(freq(2)-freq(1)) ' Hz ']);
xlabel('Frequency (Hz)');ylabel('Amplitude (dB)');
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% display 2 : time / frequency analysis : spectrogram demo
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
saturation_dB = 80;  % dB range scale (means , the lowest displayed level is 80 dB below the max level)
[sg,fsg,tsg] = specgram(signal,NFFT,Fs,w,NOVERLAP);  
% FFT normalisation and conversion amplitude from linear to dB peak
sg_dBpeak = 20*log10(abs(sg))+20*log10(2/length(fsg));     % NB : X=fft(x.*hanning(N))*4/N; % hanning only
% saturation to given dB range
mini_dB = round(max(max(sg_dBpeak))) - saturation_dB;
sg_dBpeak(sg_dBpeak<mini_dB) = mini_dB;
% plots spectrogram
figure(2);
imagesc(tsg,fsg,sg_dBpeak);axis('xy');colorbar('vert');grid
title(['Spectrogram  / Fs = ' num2str(Fs) ' Hz / Delta f = ' num2str(fsg(2)-fsg(1)) ' Hz ']);
xlabel('Time (s)');ylabel('Frequency (Hz)');

Best Answer

Related Solutions

MATLAB: Evaluate the variation in the power of a known periodicity

MATLAB: Is it possible to find the frequency specture from a signal imported from data points (excel)

Related Question