search

CRAVA / crava

6 stars 13 forks source link

Problem with compaction correlation structure #227

Closed alfbr closed 7 years ago

alfbr commented 8 years ago

Seems like a bug, with frequent crashes. Harald can provide input and data set for reproduction.

hflesche commented 8 years ago

background_base_conform.pptx This files illustrates the sharp lateral changes that can come when base or top conform correlation structures are used to create background models. A proportional model would be preferrable, but as discussed there can be limitations in the FFT.

hgolsen commented 8 years ago

https://jira.nr.no/browse/CRA-837

alfbr commented 8 years ago

What is the status here?

hgolsen commented 8 years ago

Ragnar has started looking at this issue. There was some missing data. We received those today.

alfbr commented 8 years ago

Any update on this?

RagnarHauge commented 8 years ago

Problems were related to resampling issues and trying to write SegY files with very thin cells in the original grid. A fix for this has been implemented, and waits for code review by Håvard in a pull request.

hgolsen commented 8 years ago

The fix is checked in now.

RagnarHauge commented 8 years ago

The undef problems observed in this case does not relate to the compaction structure use, but is a separate bug due to lack of well data when trying to scale wavelet locally in a zone. Will create issue for it and fix it after next week.

alfbr commented 7 years ago

Great, thanks!