Plotting the the acceleration data of the degree of freedom versus the time.

I am given the data for acceleration (almost 3 pages) and the question says that the sampling rate for all of the experiments was 50 Hz (50 samples per second, or delta_t= 0.02 sec. I am supposed to plot the acceleration data vs time and also find the peaks of acceleration that accurs during the free vibration. I loaded the acceleration data file and extracted data from the column but I am not sure how to specify the time data when I am only given the time step and how to mark the peaks. I appreciate the help! I get the following error:

Code:

clc; clear all; close all;
load VibrationSDOF_Data.txt
acc = VibrationSDOF_Data(:,1);
time =[0: 0.02: 110]
plot (time,acc,'r-');

Error:

error using plot

vectors must be the same length

Answer 1

[Walter Roberson]

Change

time =[0: 0.02: 110]

to

dt = 0.02;
time = (0:numel(acc)-1) * dt;

[Obsidian]

Thank you it worked!!!

Do you have any suggestion on how to find and mark the peaks of acceleration?

It is supposed to look like this:

[Walter Roberson]

See findpeaks()

[Obsidian]

I tried findpeaks(acc,time) but it says I need to install a toolbox and everytime I try to install the toolbox my matlab crashes. Is there any other way to find the peaks without having to install the toolbox?

[Walter Roberson]

Finding peaks of noisy data is a bit difficult to reliably tell the difference between noise and signal.

Especially with acceleration data (which is notorious for being noisy), it is common to first apply a low-pass filter or moving-median filter. After that, there are a number of different approaches.

One of the approaches is to use max() to find the maximum; that is one peak. Then set that location and a distance on each side of it to NaN, and find the max() again. Keep repeating that until the max value you found is less than some threshold you set .

[Obsidian]

I'll try that. Thank you so much!

Answer 2

[Chunru]

Make sure time and acc have the same size

time =[0: length(acc)-1]*0.02;
plot (time,acc,'r-');