Segmenting datasets: Difference between revisions
From Atomix
No edit summary |
No edit summary |
||
| Line 1: | Line 1: | ||
Once the raw data has gone through QA/QC, then you must: | Once the raw data has gone through QA/QC, then you must: | ||
* Split the time series into shorter segments by considering stationarity | * Split the time series into shorter segments by considering: | ||
** stationarity | |||
** the time and length scales of turbulence | |||
* determine how (or if at all) missing samples should be replaced | * determine how (or if at all) missing samples should be replaced | ||
* Determine the contribution of the [[Reynolds decomposition|mean flow versus turbulence]] to the velocity field | * Determine the contribution of the [[Reynolds decomposition|mean flow versus turbulence]] to the velocity field | ||
* Rotate the measurements into the frame of reference of the mean flow | * Rotate the measurements into the frame of reference of the mean flow | ||
* | * Taylor's hypothesis, etc ... | ||
* Estimate surface wave velocities | * Estimate surface wave velocities | ||
* Calculate the mean flow past the sensor | * Calculate the mean flow past the sensor | ||
Revision as of 19:02, 29 October 2021
Once the raw data has gone through QA/QC, then you must:
- Split the time series into shorter segments by considering:
- stationarity
- the time and length scales of turbulence
- determine how (or if at all) missing samples should be replaced
- Determine the contribution of the mean flow versus turbulence to the velocity field
- Rotate the measurements into the frame of reference of the mean flow
- Taylor's hypothesis, etc ...
- Estimate surface wave velocities
- Calculate the mean flow past the sensor
