New G&G Tool

Many of you know that today the most common tool for reservoir engineer is a Petrel.
It has a lot of prons but one big cons- it is very expensive.
What if someone will spend its time and resources to create the tool that will cover not all but main or vital functions of petrel. And will make this tool available for people for free.
If that will happen what you guys will contribute for this project in terms of your time and knowledge.
How many of you will help to test and tune functionalists of this tool?
So will be glad to hear your ideas

Petrel is a dedicated G & G tool, not a reservoir engineering tool...it just has Petrel-RE which is an Eclipse plug-in...

Other tools like Prisms' geographic and openworks...Jewel suite...are equally good tools

Sent from my HTC One

I would like to disagree with you Reservoir engineering is not something about production or placing the wells or running the cases, it is about knowledge and understanding physics of your reservoir.
Pass this thread my friend since you are on wrong direction.

All, i need more input from you guys.
Your ideas , your task are very welcome

I don't mean to drag an argument, but u mistake reservoir engineering for reservoir geoscience and reservoir geomechanics...if Petrel is a reservoir engineering tool what would you in effect call PvtSIM, sapphire, matbal...

I should assume you are a seasoned reservoir engineering, and I can understand what you mean. But your statement doesn't follow in that premise.

Sent from my HTC One

I'd say that Petrel is the defacto standard in static modeling right now, as Eclipse is the defacto standard for dynamic simulation. The RE functionality of Petrel is (IMHO) of some use as an adjunct to the reservoir simulation workflow, but poorly suited to being the center of the reservoir simulation workflow. I've been using Petrel RE for 7+ years now, and seen it come along in strides, but the more I use it (and the more complete it gets) strangely the less inclined I am to want to use it.

For creating your initial grids, creating custom FIPNUM etc arrays, generating complex well trajectory WELSPECS and COMPADAT's, adhoc simulation faults etc it works well and I find it useful. I generally though grab the data I need, and then work in textpad and third party post processor with Eclipse. Maybe I'm just showing my age, but at least I've been using Petrel RE for quite a few years so have some perspective on this. Most people who try to convince me I'm wrong have not tried for any significant period of time (ie 5 years +) to work without Petrel RE so really don't have the perspective to compare properly.

Anyway, not interested in getting into a heated debate on this - just wanted to express an opinion.

As to the topic at hand. Temr, are you talking about creating a G&G package, or a Reservoir Engineering package? The title says the former and the text says the latter..... could you provide some ideas of what you had in mind (sample workflows).

Like Bond, I understand you point on Petrel being useful for RE's, but while a good engineer certainly has to understand the geology and assumptions in the static model, they should not be expected to create the static model (any more than I'd trust a simulation model created by a geologist)

As someone who has had some exposure to software development - things always end up being far far more complex than you thought they were to begin with!

Just a quick sanity check though - most decent universities have access to commercial software, as do most oil & gas companies. As far as I can make out, the main target for such a custom package would be consultants who cant pass on costs to their clients? Am I wrong? As cool as the project sounds, I'd be unlikely to use it when I have Petrel & Eclipse easily to hand.... I say this not be sound mean, rather once you better understand the target 'market' likely to want to use it, then you will be better paced to decide what sort of workflows it should support.

I am very much agree with Temr. To become good simulation engineer you have to know Petrel G&G in details otherwise you are unable to use Petrel RE properly. There are many things in Petrel RE where it is linked with Petrel G&G. During upscaling of the model if you have no idea of propery modeling, your upscaled model will give different picture. With my little experience I can say that Petrel/Petrel RE is the ultimate tool for present day reservoir modeling.
Thanks

I am very much agree with Temr. To become good simulation engineer you have to know Petrel G&G in details otherwise you are unable to use Petrel RE properly. There are many things in Petrel RE where it is linked with Petrel G&G. During upscaling of the model if you have no idea of propery modeling, your upscaled model will give different picture. With my little experience I can say that Petrel/Petrel RE is the ultimate tool for present day reservoir modeling.
Thanks

What temr is asking is, if it is possible to develop an open source reservoir simulation tool.
In my opinion a very difficult task if not impossible and it doesn’t matter which route you go static or dynamic. Here is why: modeling the reservoir and PVT, as well as solving fluid flow equations is difficult and requires thousands line of code. It requires excellent knowledge of reservoir, PVT, geology and programming language. The number of reservoir engineers capable of doing that is very little and they are mostly specialized in one or another field, therefore such a project has to be a join effort of many specialists.
How big the chances are that ---h like this will come true?
Now let assume that such a program was build and offered for free to the users. Now, you have to convince them about the produced results not to mention the user friendly interface, support etc etc... very difficult task in my opinion.

[QUOTE=vinomarky;262716]I'd say that Petrel is the defacto standard in static modeling right now, as Eclipse is the defacto standard for dynamic simulation. The RE functionality of Petrel is (IMHO) of some use as an adjunct to the reservoir simulation workflow, but poorly suited to being the center of the reservoir simulation workflow. I've been using Petrel RE for 7+ years now, and seen it come along in strides, but the more I use it (and the more complete it gets) strangely the less inclined I am to want to use it.

For creating your initial grids, creating custom FIPNUM etc arrays, generating complex well trajectory WELSPECS and COMPADAT's, adhoc simulation faults etc it works well and I find it useful. I generally though grab the data I need, and then work in textpad and third party post processor with Eclipse. Maybe I'm just showing my age, but at least I've been using Petrel RE for quite a few years so have some perspective on this. Most people who try to convince me I'm wrong have not tried for any significant period of time (ie 5 years +) to work without Petrel RE so really don't have the perspective to compare properly.

Anyway, not interested in getting into a heated debate on this - just wanted to express an opinion.

As to the topic at hand. Temr, are you talking about creating a G&G package, or a Reservoir Engineering package? The title says the former and the text says the latter..... could you provide some ideas of what you had in mind (sample workflows).

Like Bond, I understand you point on Petrel being useful for RE's, but while a good engineer certainly has to understand the geology and assumptions in the static model, they should not be expected to create the static model (any more than I'd trust a simulation model created by a geologist)

As someone who has had some exposure to software development - things always end up being far far more complex than you thought they were to begin with![/QUOTE]
Vinomarky, for me G@G and Reservoir engineering is one synergy process. Since i am gathering all data from all departments in order to create geo model and then coarse geo model with fluid properties.
My plan is following, the suite is being coded now ( data base , functionalists, visualization) one the the sage will complete i will inform you and will wait for your feedback.
So far some of functionalists of this Suite will be available as a petrel plagins for you to test and rate it.
Now we almost prepare plagin for fault modeling that will 100 % be a part of new tool.
All plagins will be distributed for free. Current versions for petrel 2012 and 2013
So what i need from you is your input ideas what we as developers can bring on top of new suite

As for market , this tool will be available as free version and commercial version. I will post functionalists of each version later on . The main distinguished is commercial version will use open inventor library for visualization like petrel does. I will agree with you that you will not use something when you have Petrel me also, but if you cant effort license costing 300 K, if not you will need a tool , and that where we will come). Any way the project have been started. I need some help from you guys. As i posted earlier we will post information on product here then on dedicated site ( which now bieeng constructed) to give you a favor of future project we will produce petrel plagins with Aurora functions.

[QUOTE=Yengineer;262782]What temr is asking is, if it is possible to develop an open source reservoir simulation tool.
In my opinion a very difficult task if not impossible and it doesn’t matter which route you go static or dynamic. Here is why: modeling the reservoir and PVT, as well as solving fluid flow equations is difficult and requires thousands line of code. It requires excellent knowledge of reservoir, PVT, geology and programming language. The number of reservoir engineers capable of doing that is very little and they are mostly specialized in one or another field, therefore such a project has to be a join effort of many specialists.
How big the chances are that ---h like this will come true?
Now let assume that such a program was build and offered for free to the users. Now, you have to convince them about the produced results not to mention the user friendly interface, support etc etc... very difficult task in my opinion.[/QUOTE]
I will agree with you Yengineer, but currently on project 5 full time expert involved.
First version will let you create data base , make 3d gridding, property modeling, sw modelling, and upscaling. Also the will be support of eclipse binary files so you can import eclipse deck runs for visualization and analysis.
I am not saying that this will be easy i am sayng that this will be very interesting

I think it depends on the workflow. If it is MNCs with a troop of REs and Geologists/ Geophysicist, Pe--l would be a key. Ex-big blue people seems to be busy in the market, pushing Petrel sales.

One the other hand, ------- is still popular (atleast before it was tookover by IHS) among some small companies in the North Sea sector as well as other markets. DUG is also well known tool.

I had a privilege to work with some of the experienced REs. Some have excellent spreadsheets fit for their work. Some consultants simply didn't like/ want to use Eclipse or no-sense to use use Petrel simply because it was so expensive. Another reason is licensing strategy, it is a pain to share licenses. Technoguide did the brilliant job for creating Petrel but Schlumberger simply killed the product by integration and making profit, Microsoft.



[QUOTE=temr;262634]Many of you know that today the most common tool for reservoir engineer is a Petrel.
It has a lot of prons but one big cons- it is very expensive.
What if someone will spend its time and resources to create the tool that will cover not all but main or vital functions of petrel. And will make this tool available for people for free.
If that will happen what you guys will contribute for this project in terms of your time and knowledge.
How many of you will help to test and tune functionalists of this tool?
So will be glad to hear your ideas[/QUOTE]

[QUOTE=reservoirengineer;262999]I think it depends on the workflow. If it is MNCs with a troop of REs and Geologists/ Geophysicist, Pe--l would be a key. Ex-big blue people seems to be busy in the market, pushing Petrel sales.

One the other hand, ------- is still popular (atleast before it was tookover by IHS) among some small companies in the North Sea sector as well as other markets. DUG is also well known tool.

I had a privilege to work with some of the experienced REs. Some have excellent spreadsheets fit for their work. Some consultants simply didn't like/ want to use Eclipse or no-sense to use use Petrel simply because it was so expensive. Another reason is licensing strategy, it is a pain to share licenses. Technoguide did the brilliant job for creating Petrel but Schlumberger simply killed the product by integration and making profit, Microsoft.[/QUOTE]
In 2005 i meet with one of the founders of technoguide he said that the main reason they quite Roxar is the management did not want to make changes in RMS. i think the same is now happening with Petrel.....
Any way thanks you guys for you input. So far i think that having all skepticism in your mind still there is a seed that may grow, and i think the reason for that is the uniqueness of petrel with no other alternative...
I will keep you posted for updates on project.

I think what temr pointed out is quite interesting, I have been using petrel Re for the last 10 years and It s quit hard to work without it, however some small soft wares are really useful since they are cheap and too fast, I would highlight for example ..S3graph , it s fast, easy to work with and can do all the pre and post processing, there are still some issues with petrel that the user can do easily with another package (ex.. Rms) one example was what I experienced last week where I tried to update locally a structural model by just shifting 20m higher than it was from a designed ploygone... It was quit a lot of hassle to perform that, I also like the roxar rms in it s way on how you load data between different project... Petrel re has also some issues on simple reservoir engineering modeling ( advanced well model (ICD, ICV) is one of them, implementing a J function as formulas in intialization, modification of the CF for well connectivity if you apply a multiplier on permeability, the way how petrel export the VFP, and the way how petrel treat the WRFTPLT keyword..... In addition to the fact that the grids are left handed) I think one big scoop that petrel is lacking is the optimization on objective function, I have seen some plugins for history matching based on 4D seismic .... But I guess it still not mature and I think it s one of the reason why slb bought mepo
With regards to temr request, I think I would be really happy with a software like S3graph ..which is quite fast to visualize data, but I would rather need a simple workflow to modify a structural model, in addition to some thinking on petrophysical modelling and quick up scaling ... So the reservoir engineer can easily perform a modifications on his model based on dynamic data without going back to the geo model
Keep us on the loop
Gd luck

Miroine, thanks for you input. One of the advantages Aurora will have is that it will have a bridge between geomodel and simulation. What today is sold separately as plugins in petrel , already have been put to be implement (sw initialization and pvt module) in Aurora as standard set . I will defiantly put some information on data base and functionalists of Aurora later on. Again inputs from all you are very needed.

I think Petrel plug-in should be merged in Petrel package and to be sold as a single package by SLB. Every company it is not possible to purchase plug-in separately. Basically plug-in is the new updatation/addition of Petrel and should be merged in Petrel package.

Dipak_m, the problem is that petrel now is not flexible at all due to bureaucracy of SLB, and the only way to add some function is Ocean SDK that sold separetely. I bought one licence for 50 k , so i need my money back. This is the loop. Nevertheless i will issue plagins for free for the users of this forum to give the flavor of future Aurora functionality
For the moment two plaings are almost ready
1 is the fault modeling plagin ( this plagin can create faults using one or 3 polyline on fly)
2 reservoir summation plagin ( the plagin will make stat on pay intervals at the well )
On the roadmap in 2-3 month
3 Mesh Plaging ( making the 3d grid , horizons, zones, layering from scratch )
4 Petrophysical modeling ( plagin that will have functionality of data analysis(distribution, ranging) , petrophysical and facial modeling )
5 Upscaling plagin (this plagin will use unique algo for structure and property upscaling with ability of exporting all includes in auto mode)
6 Schedule plagin (the plagin that will help to create schedule section for eclipse)
this is only beginning

Again all this plagins wil have the same logic of Aurora modules

its great effort.........!
please try to incorporate every excellent and essential features from all available G&G software and make more user friendly and reliable. which will full fill the need of user .

Thanks