How to create a multisine signal input using idinput given duration and peak value?
26 views (last 30 days)
Show older comments
Hello everyone,
I am setting up a system identification experiment, and I came across the following example for an industrial arm robot with flexible joints:
I am trying to replicate the same input information. However, for my application, the maximum amplitude is 12, and I would like to simulate for 10s. The signals are:
- Multisine signals with a flat amplitude spectrum in the frequency interval 1-40 Hz with a peak value of 16 rad/s.
- The multisine signal is superimposed on a filtered square wave with amplitude 20 rad/s and cut-off frequency 1 Hz.
- Multisine signal with frequencies 0.1, 0.3, and 0.5 Hz, with peak value 40 rad/s.
clear all
T = 1;
Ts = 1e-3; % Sample time
Tsinput = 0.2; %
Period = T/Ts; % Period = T/Ts*Tsinput
%Generate a Sum-of-Sinusoids Signal.
% Specify that the signal has N samples in each period and P periods.
NumPeriod = 1;
Range = [-12 12];
%Specify the frequency range of the signal between 0 and Nyquist [0 1]
Band = [0 .1];
%Now modify the number, frequency, and phase of the sinusoids
% 2*pi*[1:GridSkip:fix(Period/2)]/Period and the passband pi*Band.
NumSinusoids = 10;
NumTrials = 10;
GridSkip = 2;
SineData = [NumSinusoids,NumTrials,GridSkip];
%Generate the signal using default characteristics for the sine waves.
[u,freq] = idinput([Period 1 NumPeriod],'sine',Band,Range,SineData);
%% Verification
ufft = fft(u);
Fs = 1/Ts;
L = length(u);
w = (0:L-1)*Fs/L;
figure;
subplot(121); stem(w(1:L/2),abs(ufft(1:L/2))) % Plot until Nyquist frequency
subplot(122); stem(w(1:L/2),abs(ufft(1:L/2))) % Just a little zoom
xlim([0 50]);
title('Single-Sided Amplitude Spectrum of u(t)')
xlabel('Frequency (rad/s)')
ylabel('Amplitude')
%Find Peak-Magnitude-to-RMS Ratio of Sinusoid
disp(['Peak-Magnitude-to-RMS Ratio: ' num2str(peak2rms(u))])
%Returns the sum-of-sinusoids signal in u and the frequencies of the sinusoids in freq.
freq = freq/Ts;
%% Final Signal - For hardware
u = iddata([],u,Tsinput,'TimeUnit','seconds');
figure;
plot(u)
Could someone point out a direction to accomplish this task?
Thank you in advance
0 Comments
Answers (0)
See Also
Categories
Find more on Linear Model Identification in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!