Simple workflow to readjust dynamic models their permeability parameters values for the uncored depth intervals
Due to the importance of the permeability parameters influence on dynamic models history match i.e. (PERMX,Y&Z) and in order to accelerate obtaining better knowledge about porosity-permeability relationship trends, a simple workflow can be implemented as part of Assisting History Match technology that would help to better defining porosity-permeability relationship with the objective to re-adjustment the predicted permeability values for the uncored depth intervals.
In the following case study example, a reservoir has 3 main permeability layers trend in general between (1) high layers permeability intervals, (2) low layers permeability intervals and, (3) buffer layers type intervals, which is not fully dense. Then by using simple formula “PERM=C1xEXP(POROxC2)” where C1&C2 are representing constant parameters that would drive the shape of the porosity-permeability final relationship in few alterations runs prior to proceed with complex history match process later.
By automated modifying the values of the constants of C1&C2 per each layer scheme through activating Python function in the tNavigator simulator, this process will allow generating multi permeability models quickly within the simulator model that will be evaluated through “Objective Function” to achieve optimum case. This workflow shall accelerate achieving the most possible cloud shape of the porosity-permeability relationship, which would be achieved in a very short time period overall.
The plot is representing the cloud area ranges of the generated porosity-permeability relationship based on the measure cored data.
Cross-section from the dynamic model, which shows the 3 main layers schemes that are representing (1) high, (2) low, and (3) buffers permeability layers.
The regulation lines per each layer scheme i.e. (high, low, and buffer permeability zones) that are required to be adjusted through "Objective Function" and within min and max values ranges of C1 & C2.
MULTNUM parameter in tNavigator is representing layers schemes, GXi is representing C1 constant, and GRi is representing C2 constant, the tNavigator simulator is going to change the values of GXi and GRi vie each iteration while trying to reduce the errors between the actual and simulated profile values.
The Assisting-History-Match tool of tNavigator is going to change GXi and GRi to find the possible optimum values that were showing the best solution before going into complex history match processes later.
--
4 年Assalamu Alikum Wahrahmat Allah hope your all females save from corona virus and your long life
Principal Reservoir Engineer (Simulation)
4 年@All, the ultimate goal is to have cloud relationship between poro-perm, which would assist to achieve acceptable history match. Once the regulation lines per zones were created then it would require to general cloud around the regulation lines by using permeability multipliers per well regions and/or layers and/or zone ..etc The following plot is comparing between two cases where each one has different poro-perm cloud distribution while both cases were showed simile FIP's and identical Sw distribution i.e. (the RRT's grouping was made based on 3D static Sw model using ADNOC Fast-Track approach that is discussed in other articular)
Principal Reservoir Engineer (Simulation)
4 年@Merouane, we are trying to enhance simulation model history match of GOR and WCut profiles per well, of course those parameters are more controlled by Kr’s curves mainly. As you observed that I’m discussing how to redesign obsolete permeability models i.e. (PERMX,Y&Z) within simulation model as part of complex workflow that has over 1000s variables analyzed in parallel. What I had highlighted to you here is representing only 5% of the total workflow function that designed to modify tons of variables in smart way, such as well region, zone and high permeability streak layers, connectivity and others.
Manager Concept Development & valuations | Subsurface project leader | Lead Reservoir Engineer | concept architect @ Equinor
4 年Interesting approach... what kind of history matching objective are u trying to achieve ? (Mass balance .. breakthrough timing? Drainage area ? Well productivity? ) .... I guess when u mention permeability parameters ... u refer to effective permeability .... or does this affect rel-perms (as they are binned to water saturation which usually is a pure function of per-poro relationship) ... for relperm shape .. u should check pyscal module in python (under pypi) that makes generating shapes very simple and efficient...