14. Preprocessing Data
When the data have been collected from the identification experiment, they are not likely to be in shape for immediate use in identification algorithms. There are several possible deficiencies in the data that should be attended to:
1. High-frequency disturbances in the data record, above the frequencies of interest to the system dynamics
2. Occasional bursts and outliers, missing data, non-continuous data records
3. Drift and offset, low-frequency disturbances, possibly of periodic character
It must be stressed that in off-line applications, one should always first plot the data in order to inspect them for these deficiencies. In this section we shall discuss how to preprocess the data so as to avoid problems in the identification ...
Get System Identification: Theory for the User, 2nd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
