1081 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Joakim Hove	8cf963ba3f	Using TableManager class to get input tables.	2015-09-01 12:21:11 +02:00
Andreas Lauser	8c612009ac	Merge pull request #46 from andlaus/implement_slgof_support ... ECL satfuncs: implement support for the SLGOF keyword	2015-08-25 21:55:29 +02:00
Andreas Lauser	a58531c3de	fix the satfunc family detection code in ECL scaling points class ... thanks to [at]totto82 for finding this.	2015-08-25 11:38:31 +02:00
Andreas Lauser	5e447cb83a	ECL satfuncs: implement support for the SLGOF keyword ... if my understanding of this keyword is correct, it is identical to SGOF except that the saturation column is specified in terms of Swco + So instead of as Sg	2015-08-24 21:51:46 +02:00
Tor Harald Sandve	6c403d457e	Merge pull request #45 from andlaus/shift_gas_saturations_again ... shift the gas saturations of the oil relperm tables again	2015-08-24 15:30:12 +02:00
Andreas Lauser	c6cafe3a7a	shift the gas saturations of the oil relperm tables again ... it seems like this was accidentally undone while resolving conflicts in #42.	2015-08-24 13:52:09 +02:00
Andreas Lauser	086b129773	Merge pull request #43 from andlaus/fix_compile ... fix compilation in debug mode	2015-08-19 16:07:59 +02:00
Andreas Lauser	6e27d1ab1d	fix compilation in debug mode	2015-08-19 16:07:07 +02:00
Andreas Lauser	5490017b3f	Merge pull request #42 from totto82/family2 ... Support for Family II saturation functions input	2015-08-19 13:06:06 +02:00
Andreas Lauser	220fe9cffa	Merge pull request #38 from andlaus/SplineLaw_implement_twophase_inversion_api ... SplineTwoPhaseMaterial: implement the two-phase inversion functions	2015-08-19 11:02:56 +02:00
Tor Harald Sandve	3c7a087d3d	Merge pull request #41 from andlaus/ecl_satfunc_improvements ... Ecl satfunc improvements	2015-08-19 10:43:26 +02:00
Tor Harald Sandve	4431ff6d7d	Add function to check saturation function family ... Adapt to change in the EclipseState i.e. the method to get the saturation function family is no longer available and a similar method is implemented in EclMaterialLawManager	2015-08-19 09:29:20 +02:00
Andreas Lauser	ef28c31368	EclHysteresisTwoPhaseLaw: use '<=' instead of '<' to decide whether to use the imbibition or the drainage curve ... this does not change the results in any way, but for the Norne deck, it improves performance (i.e., It Must Be Right (TM))...	2015-08-18 17:35:35 +02:00
Andreas Lauser	57948012a5	PiecewiseLinearTwoPhaseMaterial: remove unnecessary calls to min() and max() ... it is nicer to read like this and it may improve performance a tiny bit.	2015-08-18 17:35:22 +02:00
Andreas Lauser	0edc7b5f70	EclMaterialLawManager: shift the saturations of the oil phase the in gas-oil twophase system by the connate water saturation ... for the oil phase relperm, the gas saturation column of the SGOF keyword actually is Sg - Swco. why? it's ECL! don't ask why!	2015-08-18 17:24:28 +02:00
Tor Harald Sandve	3cffd067e3	PiecewiseLinearTwoPhaseMaterial: make the saturation columns of the quantities independent	2015-08-18 17:24:28 +02:00
Andreas Lauser	3693edb044	Merge pull request #40 from andlaus/revert_family_2_support ... Revert "Add support for saturation family 2 keywords"	2015-08-18 17:23:10 +02:00
Andreas Lauser	7e83191892	Revert "Add support for saturation family 2 keywords" ... This reverts commit a041e3ccff. it does not compile with the current opm-parser.	2015-08-18 17:20:29 +02:00
Andreas Lauser	4a7e099246	Merge pull request #37 from totto82/family2 ... Add support for saturation family II keywords	2015-08-18 17:02:54 +02:00
Tor Harald Sandve	a041e3ccff	Add support for saturation family 2 keywords ... Saturation functions can no be given using the family 2 keywords SWFN, SGFN and SOF3.	2015-08-17 07:08:13 +02:00
Andreas Lauser	8e11a8254e	Merge pull request #39 from andlaus/change_connate_water_treatment ... EclDefaultMaterial: do not add the connate water saturation to Sw_ow	2015-08-14 17:42:50 +02:00
Andreas Lauser	14e4692386	EclDefaultMaterial: do not add the connate water saturation to Sw_ow	2015-08-14 17:40:05 +02:00
Andreas Lauser	8e51155f24	Merge pull request #36 from andlaus/inconsistent_hysteresis_updates ... EclDefaultMaterial: use inconsistent saturations for the hysteresis update	2015-08-14 15:32:03 +02:00
Andreas Lauser	fb6eb1279a	SplineTwoPhaseMaterial: implement the two-phase inversion functions ... these are required for ECL hysteresis.	2015-08-14 14:04:20 +02:00
Andreas Lauser	a8aaa168e2	EclDefaultMaterial: allow the parameters to decide if consistent or inconsistent hysteresis are used	2015-08-14 09:33:28 +02:00
Andreas Lauser	893538c9f3	EclDefaultMaterial: use inconsistent saturations for the hysteresis update ... with this, I got slightly better performance than the opm-core master version if flow is tasked on simulating the full Norne deck. Be aware that from the physical POV, this is wrong.	2015-08-14 09:26:28 +02:00
Andreas Lauser	83de8aceb1	Merge pull request #35 from andlaus/connate_water_fixes ... Connate water fixes	2015-08-14 09:26:10 +02:00
Andreas Lauser	a52c279fc8	Merge pull request #34 from andlaus/fix_two-point_krw_scaling ... fix saturation scaling points for two-point krw scaling	2015-08-13 17:39:24 +02:00
Andreas Lauser	0d1c711d8e	EclDefaultMaterial: increase the epsilon for division-by-zero-avoidance to 1e-20 ... also, treat saturations smaller than 0 as 0. this also makes clamping the kro values to range [0, 1] superfluous.	2015-08-13 17:37:19 +02:00
Andreas Lauser	797012a8ba	EclDefaultMaterial: add the connate water saturation to the wetting-phase saturation in the gas-oil system	2015-08-13 17:34:21 +02:00
Andreas Lauser	be4593841f	EclDefaultMaterial: Make sure that the water saturation is at least the connate water saturation ... I thought this would not make a difference if the tables are extrapolated by constants, but apparantly it makes a difference for endpoint scaling.	2015-08-13 17:31:56 +02:00
Andreas Lauser	ca2a4bbb4e	fix saturation scaling points for two-point krw scaling ... kudos to [at]osae for finding this!	2015-08-12 18:07:05 +02:00
Andreas Lauser	10dc683c75	Merge pull request #33 from andlaus/only_non-wetting_phase_ecl_hysteresis ... limit ecl relperm hystersis to the non-wetting phase	2015-08-12 13:21:35 +02:00
Andreas Lauser	edf345219f	limit ecl relperm hystersis to the non-wetting phase ... I have trouble seeing why this can possibly physically justified (because oil is once the wetting and once the non-wetting phase), but things seem to be like this as the documentation says that in case of chosing the relperm hysteresis model 0 via the EHYSTR keyword, hysteresis is only used for the non-wetting phase.	2015-08-11 17:23:02 +02:00
Andreas Lauser	f48891ffb0	Merge pull request #30 from andlaus/no_limits_in_krn ... EclDefaultMaterial: do not limit the saturations in krn()	2015-08-10 12:31:38 +02:00
Andreas Lauser	89cbb8aab7	EclMaterialLawManager: slightly clean up the materialLawParams() methods	2015-08-07 18:23:32 +02:00
Andreas Lauser	a82648a002	embed the scaled endpoints directly into the relevant parameter objects ... this hopefully improves performance slightly.	2015-08-07 18:23:32 +02:00
Andreas Lauser	8b9cfa0a65	EclDefaultMaterial: do not limit the saturations in krn() ... this is required to get a match with the current opm-core satfuncs for Norne.	2015-08-07 16:47:23 +02:00
Andreas Lauser	53309405c2	EclMaterialLawManager: make it use compressed instead of logically Cartesian elements ... for grids with many inactive cells, this can change quite a bit of memory and might even lead to slightly better performance because of the memory prefetching potentially works better.	2015-08-06 16:44:00 +02:00
Andreas Lauser	5d2dadc6e1	replace custom conditional type selector by std::conditional ... I did not know about std::conditional until today...	2015-07-29 13:56:53 +02:00
Andreas Lauser	fc2e6c9918	implement a EclMaterialLawManager class ... this class internalizes and manages the material parameters from the ECL deck because this turns out to be a non-trivial task.	2015-07-28 17:24:31 +02:00
Andreas Lauser	7cbffc22c6	introduce SimpleModularFluidState and make ModularFluidState not depend on the FluidSystem anymore ... it is called "simple" because the usage is somewhat simpler: The quantities are stored and those which are not can be specified using simple boolean template arguments instead of having to pass the class names of the storage modules. The definition of the class is quite a bit more involved than the non-"simple" variant, though.	2015-07-28 17:24:27 +02:00
Andreas Lauser	a25cee980b	implement a multiplexer three-phase fluid-matrixinteraction for ECL problems ... this allows to switch between the Default three-phase implementation, Stone1 and Stone2 at runtime.	2015-07-28 17:24:25 +02:00
Andreas Lauser	11ccc40e44	implement Stone2 three-phase fluid-matrixinteractions for ECL problems	2015-07-28 17:24:22 +02:00
Andreas Lauser	d46f4bafe3	implement Stone1 three-phase fluid-matrixinteractions for ECL problems	2015-07-28 17:24:20 +02:00
Andreas Lauser	c7c4784746	some updates for EclDefaultMaterial ... it now uses shared pointers to the underlying twophase material law parameter objects and it can be used with hysteresis.	2015-07-28 17:24:18 +02:00
Andreas Lauser	014a8faa04	implement ECL hysteresis ... so far, this is only for relperm hysteresis. the Eclipse docu seems to be incorrect (or at least vague) for capillary pressure hysteresis...	2015-07-28 17:24:16 +02:00
Andreas Lauser	4dea8c0642	add invertion methods to some of the two-phase fluid-matrix interactions ... these are required for hysteresis	2015-07-28 17:24:14 +02:00
Andreas Lauser	01aaaea95c	Brooks-Corey: fix some comments ... they were misleading or wrong.	2015-07-28 17:24:12 +02:00
Andreas Lauser	bb100ced1c	implement a two-phase fluid-matrix interaction for ECL end-point scaling	2015-07-28 17:24:09 +02:00