Segmenting datasets: Difference between revisions
From Atomix
No edit summary |
No edit summary |
||
| Line 8: | Line 8: | ||
* Determine how (or if at all) you should replace missing samples | * Determine how (or if at all) you should replace missing samples | ||
* [[Rotation of the velocity measurements|Rotate the measurements]] into the background (mean) flow's frame of reference | * [[Rotation of the velocity measurements|Rotate the measurements]] into the background (mean) flow's frame of reference | ||
* Use the concept of [[Reynolds decomposition]] to | * Use the concept of [[Reynolds decomposition]] to [[Decomposing velocity measurements|decompose the velocities]] into the mean, turbulence and surface waves component. | ||
Revision as of 23:39, 29 October 2021
Once the raw data has gone through QA/QC, then you must:
- Split the time series into shorter segments by considering:
- time and length scales of turbulence
- stationarity of the segment
- Taylor's frozen turbulence hypothesis, etc ...
- Determine how (or if at all) you should replace missing samples
- Rotate the measurements into the background (mean) flow's frame of reference
- Use the concept of Reynolds decomposition to decompose the velocities into the mean, turbulence and surface waves component.
