nens / servicedesk-issue-only

This repository will be used primarily by the service desk to report issues which will then be moved by the teamleads
0 stars 0 forks source link

[3Di]: bug als drain_level = NULL & surface_level != NULL #152

Open GeOdin opened 1 year ago

GeOdin commented 1 year ago

Topdesk URL (API)

https://nelen-schuurmans.topdesk.net/tas/secure/contained/incident?unid=00c63bb3-fb2a-4a0e-afdb-f35addd0f667

Incident Type

Error

What priority should we assign to this submission?

Prio1 Very Urgent

3Di component

API - v3 (Default)

What happened?

Ik heb een integraal model (van gemeente Assen), waarbij sommige connected putten alleen een surface_level hebben (dus drain_level is NULL), en sommige connected putten zowel geen surface_level als drain_level hebben (dus beiden NULL).

Als drain_level is NULL, dan zou hij de laagste pixel van de rekencel moeten gebruiken als uitwisselingshoogte. Dit is echter alleen het geval als surface_level ook NULL is. Zodra de surface_level wel is ingevuld pakt hij deze als uitwisselingshoogte (ook als drain_level is NULL).

Zie screenshot met de put waarbij ik deze bug als eerste opmerkte (de exchange-level van de 1D/2D-flowline = surface_level van manhole)

Stijn O. heeft het geverifieerd en zei dat het inderdaad een bug is.

Ter aanvulling, ik kwam er ook achter dat een surface-level verplicht is voor het gebruiken van de side-view tool.

Er werd altijd gezegd dat surface-level alleen voor eigen administratie is, maar dat klopt dus niet helemaal. Althans niet meer

How to Reproduce?

Ik heb een integraal model (van gemeente Assen), waarbij sommige connected putten alleen een surface_level hebben (dus drain_level is NULL), en sommige connected putten zowel geen surface_level als drain_level hebben (dus beiden NULL).

Als drain_level is NULL, dan zou hij de laagste pixel van de rekencel moeten gebruiken als uitwisselingshoogte. Dit is echter alleen het geval als surface_level ook NULL is. Zodra de surface_level wel is ingevuld pakt hij deze als uitwisselingshoogte (ook als drain_level is NULL).

Zie screenshot met de put waarbij ik deze bug als eerste opmerkte (de exchange-level van de 1D/2D-flowline = surface_level van manhole)

Stijn O. heeft het geverifieerd en zei dat het inderdaad een bug is.

Ter aanvulling, ik kwam er ook achter dat een surface-level verplicht is voor het gebruiken van de side-view tool.

Er werd altijd gezegd dat surface-level alleen voor eigen administratie is, maar dat klopt dus niet helemaal. Althans niet meer

What did you expect and what was the actual result?

No response

Add screenshots, video's etc

screenshot1676555055938

Relevant log output

No response

Additional info

Good luck! :)

GeOdin commented 1 year ago

Ps. Kasper said the bug was a prio 1.

GeOdin commented 1 year ago

More from the customer: Deze bug zorgt ervoor dat de drain level van putten (waarbij deze niet is ingevuld, en surface-level wel) 10-15 cm te hoog is en dus echt je systeemgedrag/uitkomsten negatief beïnvloedt. De uitwisseling tussen riolering en maaiveld wordt dan sterk onderschat.Het is daarmee wel echt een belangrijke bug om te fixen

Jonasvsl commented 1 year ago

can you please provide the link to the 3Di model on the management screens? And the simulation id?

Jonasvsl commented 1 year ago

@caspervdw is quite sure that the 3Di grid builder doesn't use the surface levels.

It is correct that surface levels are used of visualisation in the sideview tool

But there is a difference between what 3Di does and the assumption in this ticket. The drain level that we use in 3Di is NOT the lowest pixel in a cell, but the pixel value on the location of the manhole. If that is in the docs let me know with a link to the location and we will fix that in the docs.

I suspect that the surface levels have been derived from the DEM. And then it is logical that the surface level and drain level are the same.

If you need the lowest pixel in the cell to be used an additionial processing step by the user is required

GeOdin commented 1 year ago

I asked the questions to Kasper. I will give an update about it later.

GeOdin commented 1 year ago

I don’t have a model-id anymore, because I had to delete it. But the model was based on schematization with revision id 46234. The simulation id = 115120.

The confusion about what drain_level does when it is left empty has been cleared. Apparently, we had a different idea of what 3Di actually does. So for now I think we can close this ticket. There is a wrong comment in https://docs.3di.live/_downloads/84a660f75b3e8f891f1cd6c5f0b550f8/database-overview.pdf at the description of drain_level. This sentence is very confusing and possibly wrong, but not necessarily linked to this ticket.

I don’t know which additional steps I should take, but I’ve determined the drain_level differently.