Cluster the data, separate the data
7 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Hi! I was measuring tilt for let's say 71 nozzles continiously. I got the inclinations. The results are shown in the attached pic. Now I have to split the data on 71 groups but the upper and low limits and time of measurements are different. When aplitting the data I just need the data with high density. Should I write a loop for defining the data points that are close in values until I reach some pick value and store it. Or there some solution to do. I would be appreciate. Thanks.
3 Kommentare
Adam Danz
am 23 Sep. 2019
This will require some trial-error. If the time is at a fixed sampling rate, I'd personally start with differentiating the inclinations vector using abs(diff(vector1)) and then I'd plot the results to see if I could use some kind of threshold to detect the large spikes. If there wasn't a clear threshold I'd move on to findpeaks() which will require you to read the documentation and play around with those options to find what works for you.
Akzeptierte Antwort
MS
am 23 Sep. 2019
As a quick answer (since you didn't upload the data), you may try to detect the peaks of the signal and then get the segments between them. For example, see the findpeaks method at: https://www.mathworks.com/help/signal/examples/peak-analysis.html#responsive_offcanvas.
4 Kommentare
Weitere Antworten (0)
Siehe auch
Kategorien
Mehr zu Time Series Events finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!