08-25-2010 09:37 AM
I am using Matlab to calculate PSD-curves of random vibration data, especially from transport tests.
I will receive data from a shipping transport of 10 days, measured with 400hz ratio.
I think it is easy to see that this data is too huge to be analysed directly.
Furthermore, the sensor gives the data in packages of 30000 points (internal sensor reasons).
I'm able to make a PSD of this 30000-point-package.
But I need a PSD of the whole test.
Is there a possibility to "merge" PSD-curves?
Thank you for your help
truhti
08-25-2010 10:29 AM
Each of your PSD curves is over the same frequency range, right? You can use averaging to extend your analysis across all packages.
If your continuous time-domain data is logged, you can process the log with many different options: start, stop, block size, overlap, etc.
Here is a VI snippet showing how you can measure the averaged PSD in LabVIEW: