[Opm] User-configurable Summary support

[Jørgen Kvalsvik]

Hi,

We're almost ready to merge a massive change set to opm-output, and a highlight from that change set is that the new Summary support is switched on. As with all new development we, unfortunately so, have to prepare for bugs and quirks. Sorry about that.

With the apologies out of the way, here's a quick list of known issues or uncertainties.

# Nonsensical values
We're still a bit in limbo when it comes to handling obviously nonsensical values. An example:
Well W1 is a producer, but WWIR water injection rate for all wells is requested. Should we:
* Print 0 (i.e. no injection)
* Print the negative injection (i.e. production) for the well? If the well produces at the rate 2.3, then -2.3 would be printed
* Print a default "undefined" constant?

# Approximately zero
How do we handle values that are essentially zero? The current reference summary file does not print 1.29967e-18 for the water injection rate a timestep for Norne, but the simulator reports this. This can be numerical noise and it can be a tiny rate. What about 1.56839e-09? For reference, these numbers are given as-is by the simulator and read from the simulator's own state, so it's the numbers the simulator works with.

# Known issues and limitations
* Accumulation resets don't work properly yet.
* Only well- and group keywords are supported, and the list is still fairly short

There's likely to be other bugs as well. That being said, the data now written is configurable and plentiful compared to what we had, so I still consider this an improvement. It might be rough in the transition period, but hopefully with more users we'll see rapid improvements.

Happy squashing!


-------------------------------------------------------------------
The information contained in this message may be CONFIDENTIAL and is
intended for the addressee only. Any unauthorised use, dissemination of the
information or copying of this message is prohibited. If you are not the
addressee, please notify the sender immediately by return e-mail and delete
this message.
Thank you
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <//opm-project.org/pipermail/opm/attachments/20160606/6ae9b36b/attachment.html>