LU507017B1 - A large-scale reservoir numerical simulation method and its numerical simulation system - Google Patents
A large-scale reservoir numerical simulation method and its numerical simulation system Download PDFInfo
- Publication number
- LU507017B1 LU507017B1 LU507017A LU507017A LU507017B1 LU 507017 B1 LU507017 B1 LU 507017B1 LU 507017 A LU507017 A LU 507017A LU 507017 A LU507017 A LU 507017A LU 507017 B1 LU507017 B1 LU 507017B1
- Authority
- LU
- Luxembourg
- Prior art keywords
- reservoir
- simulation
- model
- data
- scale
- Prior art date
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 139
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000004364 calculation method Methods 0.000 claims description 20
- 239000000284 extract Substances 0.000 claims description 11
- 230000003044 adaptive effect Effects 0.000 claims description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 9
- 230000008676 import Effects 0.000 claims description 8
- 238000012790 confirmation Methods 0.000 claims description 6
- 239000012530 fluid Substances 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 2
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 abstract description 3
- 238000012805 post-processing Methods 0.000 abstract description 2
- 238000007781 pre-processing Methods 0.000 abstract description 2
- 238000009877 rendering Methods 0.000 abstract description 2
- 238000011161 development Methods 0.000 description 4
- 230000010354 integration Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a large-scale reservoir numerical simulation method and its numerical simulation system, which relates to the field of reservoir numerical simulation technology. The invention divides the reservoir's three-dimensional area simulation model into grids to obtain computational tasks. According to the selected distributed parallel computing scheme, the computational area is divided into corresponding numbers of sub-regions, and then all sub-regions are executed in parallel. Finally, all simulation results obtained from executing the sub-regions are distributed outputted using a distributed input-output scheme, solving the problem of difficult pre- and post-processing of massive data involved in large-scale models. The invention also converts reservoir numerical simulation data into a three-dimensional model by rendering reservoir textures based on the analysis of numerical modeling situations. Additionally, the exported model format is compatible with mainstream formats, making it convenient for various platforms to access.
Description
DESCRIPTION LUS07017
A LARGE-SCALE RESERVOIR NUMERICAL SIMULATION METHOD AND
ITS NUMERICAL SIMULATION SYSTEM
This invention belongs to the field of reservoir numerical simulation, specifically involving a large-scale reservoir numerical simulation method and its numerical simulation system.
The development of large-scale reservoir numerical simulation is increasingly constrained. In 1998, Argonne National Laboratory and the University of Texas successfully attempted a super-large-scale reservoir numerical simulation problem with 4 million grid points and 32 million unknowns using IBM SP. Numerical simulation is a crucial analytical tool for oil field development evaluation and reservoir exploitation scheme design. The core of reservoir numerical simulation is to solve a set of nonlinear partial differential equations that reflect the flow of oil, water, and gas through numerical methods, aiming to simulate the history of oil field development, predict future reservoir dynamics, and adjust and formulate optimal oil recovery strategies.
In recent years, with the rapid development of reservoir geological modeling technology, reservoir geological models integrating disciplines such as seismic, well logging, geology, and reservoirs have reached a scale of several million or even tens of millions of grid nodes, reaching a sufficiently fine level. This fine reservoir simulation is of great significance for studying the distribution of remaining oil and promoting high and stable production in a considerable portion of China's old oil fields that have entered the high-water-cut late stage. The computational scale of this fine reservoir simulation needs to reach millions or tens of millions, with a typical history matching time exceeding 20 years. However, due to the lag in the application of high-performance computers and efficient parallel software by oil field departments LU507017 and the pursuit of timeliness in reservoir numerical simulation calculations for actual production, the current scale of reservoir numerical simulation is only at the level of 100,000 to 300,000 grid nodes, far from the required fine level. It is precisely because of the limitation of computer capabilities that high-precision geological models must be coarsened to be used in reservoir simulation, without considering the specificity of reservoirs, and unable to solve the problem of large-scale fine reservoir simulation.
Based on these factors, this invention is proposed.
The technical problem addressed by this invention is to overcome the shortcomings of existing technology and provide a large-scale reservoir numerical simulation method and its numerical simulation system, solving the problems outlined in the background technology.
To solve the above technical problem, the basic concept of the technical solution adopted in this invention is:
A large-scale reservoir numerical simulation method, comprising:
Step 1: Determine reservoir simulation data based on reservoir characteristics, core samples, well logging, and fluid test data, and establish a large-scale reservoir numerical simulation model;
Import reservoir simulation data into the modeling system and check for data matching. If the data does not match, prompt data corruption. If matching, proceed with establishing a numerical simulation model.
Step 2: Determine the reservoir boundary range and establish a three-dimensional reservoir region simulation model based on reservoir simulation data and large-scale mathematical simulation models;
Import reservoir simulation data and mathematical simulation models into the modeling system and check for data matching. If the data does not match, prompt data corruption. If matching, begin drawing the initial reservoir model. Further merge and process the initial model as needed, optimize the model, obtain attribute maps,
attach attribute maps to the model, and export the three-dimensional reservoir region LU507017 simulation model.
When drawing the initial reservoir model, extract XY values of reservoir network coordinates from imported data, merge and sort XY values according to a pattern.
Next, extract Z values of reservoir network coordinates, expand and sort Z values according to a pattern, then correspond XY values with Z values for each network point, extract XYZ data according to modeling rules, and finally draw the initial reservoir model based on XYZ data. Depending on application requirements, choose between slice models or layered models. For slice models, arrange initial model data longitudinally according to a pattern; for layered models, arrange initial model data horizontally according to a pattern.
Optimize the initial reservoir model by reducing the number of models and faces, extract attribute data from the optimized model, generate a color scale based on attribute data values, convert attribute data into attribute maps based on the color scale, attach UV coordinates of the optimized attribute maps to the model, and export the three-dimensional reservoir model after optimization.
Step 3: Mesh the three-dimensional region simulation model to obtain the calculation region;
Step 4: Divide the calculation region into corresponding numbers of subregions based on a predetermined computation scheme;
The predetermined computation scheme is a distributed parallel computation scheme, which includes domain decomposition of the meshed calculation region to obtain multiple subdomains. These subdomains are assigned to corresponding computing units for parallel computation, with interaction between computing units based on MPI communication.
Step 5: Execute all subregions using the predetermined adaptive implicit method to solve nonlinear equations. Utilize a third-party linear equation solver library and solve linear equations based on selected preconditioning methods. The predetermined methods include:
Using a predetermined fully implicit method and implicit pressure explicit LU507017 saturation method to solve nonlinear equations. Use a third-party linear equation solver library and solve linear equations based on selected preconditioning methods.
The adaptive implicit method involves checking if the saturation change of the target component in the target grid within one time step exceeds a predefined reference change value. If yes, use the fully implicit method to solve the nonlinear equation for the target grid. If not, use the implicit pressure explicit saturation method to solve the nonlinear equation for the target grid.
Preconditioning methods include selecting the CPR method when using the fully implicit method to solve linear equations, choosing domain decomposition method or
ILU(O) method when using the adaptive implicit method to solve linear equations. If the domain decomposition method fails, switch to the Jacob method.
Step 6: Distribute the results of all simulations obtained from executing subregions in a distributed manner.
A large-scale reservoir numerical simulation system includes:
Reservoir simulation model establishment module: Used to establish a numerical simulation model based on reservoir simulation data and create a three-dimensional reservoir simulation model based on the numerical simulation model.
Calculation region confirmation module: Used to mesh the three-dimensional region simulation model to obtain the calculation region.
Subregion confirmation module: Used to divide the calculation region into corresponding numbers of subregions based on a predetermined computation scheme.
Subregion execution module: Used to execute all subregions.
Simulation result output module: Used to distribute all simulation results obtained from executing subregions in a distributed manner.
Optionally, the reservoir simulation model establishment module comprises a reservoir numerical simulation model establishment module and a three-dimensional reservoir simulation model. The reservoir numerical simulation model establishment module is used to determine reservoir simulation data based on reservoir characteristics, core samples, well logging, and fluid test data, establish a large-scale LU507017 reservoir numerical simulation model. The three-dimensional reservoir simulation model is used to determine the reservoir boundary range based on reservoir simulation data and large-scale mathematical simulation models and establish a three-dimensional reservoir simulation model.
By adopting the above technical solution, this invention has the following beneficial effects compared to existing technology:
Converts reservoir numerical simulation data into a three-dimensional model format, renders reservoir texture maps based on numerical analysis, and exports models in mainstream compatible formats for easy integration across platforms.
Meshes the three-dimensional reservoir region simulation model to obtain calculation tasks, divides the calculation region into subregions based on a selected distributed parallel computation scheme, parallelizes the execution of all subregions, and distributes all simulation results obtained from executing subregions in a distributed manner, solving the difficulties in handling massive data processing associated with large-scale models.
The detailed description of specific embodiments of this invention will be further discussed in conjunction with the accompanying drawings.
The drawings in the following description are only some embodiments. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting creative efforts. In the attached picture:
Figure 1 is a schematic flowchart of the large-scale reservoir numerical simulation method;
Figure 2 is a schematic flow diagram of a large-scale reservoir numerical simulation system.
It should be noted that these drawings and text descriptions are not intended to limit the scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments.
The present invention will now be described in further detail with reference to the accompanying drawings.
In this embodiment, the method includes:
Step 1: Determine reservoir simulation data based on reservoir characteristics, core samples, well logging, and fluid test data, and establish a large-scale reservoir numerical simulation model.
Import reservoir simulation data into the modeling system, check for data matching. If there's a mismatch, prompt data corruption. If matching, proceed with establishing a numerical simulation model.
Step 2: Determine the reservoir boundary range and establish a three-dimensional reservoir region simulation model based on reservoir simulation data and large-scale mathematical simulation models.
Import reservoir simulation data and mathematical simulation models into the modeling system, check for data matching. If there's a mismatch, prompt data corruption. If matching, start drawing the initial reservoir model. Further merge and process the initial model as needed, optimize the model, obtain attribute maps, attach attribute maps to the model, and export the three-dimensional reservoir region simulation model.
Extract XY values of reservoir network coordinates from imported data, merge and sort XY values according to a pattern. Extract Z values of reservoir network coordinates, expand and sort Z values according to a pattern, then correspond XY values with Z values for each network point, and extract XYZ data according to modeling rules to draw the initial reservoir model. Choose between slice models or layered models based on application requirements.
Step 3: Mesh the three-dimensional reservoir region simulation model to obtain the calculation region.
Step 4: Divide the calculation region into corresponding numbers of subregions based on a predetermined computation scheme, which is a distributed parallel computation scheme involving domain decomposition and MPI communication for LU507017 parallel computation between multiple computing units.
Step 5: Execute all subregions using the predetermined adaptive implicit method to solve nonlinear equations, along with solving linear equations based on selected preconditioning methods.
In this embodiment, the method includes:
Step 5 (continued): Use the predetermined fully implicit method and implicit pressure explicit saturation method to solve nonlinear equations. The adaptive implicit method involves checking if the saturation change of the target component in the target grid within one time step exceeds a predefined reference change value. If yes, use the fully implicit method to solve the nonlinear equation for the target grid. If not, use the implicit pressure explicit saturation method to solve the nonlinear equation for the target grid. Preconditioning methods include selecting the CPR method when using the fully implicit method to solve linear equations, choosing domain decomposition method or ILU(0) method when using the adaptive implicit method to solve linear equations. If the domain decomposition method fails, switch to the Jacob method.
Step 6: Distribute the results of all simulations obtained from executing subregions in a distributed manner.
The large-scale reservoir numerical simulation system includes:
Reservoir simulation model establishment module: Used to establish a numerical simulation model based on reservoir simulation data and create a three-dimensional reservoir simulation model based on the numerical simulation model.
Calculation region confirmation module: Used to mesh the three-dimensional region simulation model to obtain the calculation region.
Subregion confirmation module: Used to divide the calculation region into corresponding numbers of subregions based on a predetermined computation scheme.
Subregion execution module: Used to execute all subregions.
Simulation result output module: Used to distribute all simulation results obtained from executing subregions in a distributed manner.
The reservoir simulation model establishment module comprises a reservoir LU507017 numerical simulation model establishment module and a three-dimensional reservoir simulation model. The reservoir numerical simulation model establishment module is used to determine reservoir simulation data based on reservoir characteristics, core samples, well logging, and fluid test data, establish a large-scale reservoir numerical simulation model. The three-dimensional reservoir simulation model is used to determine the reservoir boundary range based on reservoir simulation data and large-scale mathematical simulation models and establish a three-dimensional reservoir simulation model.
This invention's method of converting reservoir numerical simulation data into a three-dimensional model enables rendering reservoir texture maps based on numerical analysis and exports models in mainstream compatible formats for easy integration across platforms.
This invention grids a three-dimensional reservoir simulation model to obtain calculation tasks; divides the calculation area according to the selected distributed parallel computing scheme to obtain a corresponding number of sub-areas, and then executes all sub-areas in parallel ; Finally, all simulation results obtained through the execution sub-area will be distributed and output, using a distributed input and output scheme, which solves the problem of difficult pre- and post-processing of massive data involved in large-scale models.
The present invention is not limited to the above embodiments. Anyone should know that structural changes made under the inspiration of the present invention, and any technical solutions that are the same or similar to the present invention, fall within the protection scope of the present invention. The technology, shape, and structural parts not described in detail in the present invention are all known technologies.
Claims (10)
1. A large-scale reservoir numerical simulation method, characterized by: step 1: determine reservoir reservoir simulation data based on reservoir characteristics and testing data of cores, logging, and fluids, and establish a large-scale reservoir numerical simulation model; step 2: determine the reservoir boundary range based on the simulation data of the reservoir and the large-scale mathematical simulation model, and establish a three-dimensional area simulation model of the reservoir; step 3: divide the three-dimensional area simulation model into grids to obtain the computational area; step 4 divide the computational area into corresponding numbers of sub-regions according to the predetermined computing scheme; step 5: execute all sub-regions, use the predetermined adaptive implicit method to solve nonlinear equations, solve linear equations based on third-party linear equation solving libraries, and solve linear equations based on selected preconditioning methods; step 6: distribute output all simulation results obtained from executing the sub-regions.
2. A large-scale reservoir numerical simulation method according to claim 1, characterized in that in step 1, import reservoir simulation data, import reservoir simulation data into the modeling system, check if the data matches, prompt data damage if not matched, and perform operations to establish a numerical simulation model when matched.
3. A large-scale reservoir numerical simulation method according to claim 1, characterized in that in step 2, import reservoir simulation data and mathematical simulation models, import reservoir simulation data and mathematical simulation models into the modeling system, check if the data matches, prompt data damage if not matched, start drawing initial reservoir models when matched, further merge and process the initial models as required, optimize the models, obtain attribute textures,
attach attribute textures to the models, and export the three-dimensional area LU507017 simulation model of the reservoir.
4. A large-scale reservoir numerical simulation method according to claim 3, characterized in that when starting to draw the initial reservoir model, extract xy values of reservoir network coordinates from the imported data, merge and sort the xy values according to rules, then extract z values of reservoir network coordinates, expand and sort the z values according to rules, match the xy values of network points with z values, extract xyz data according to model drawing rules, and finally draw the initial reservoir model based on xyz data; additionally, based on application requirements, choose to slice the model or create a layered model; when choosing to slice the model, arrange the initial model obtained according to the longitudinal rules, and when choosing a layered model, arrange the initial model obtained according to the horizontal rules.
5. According to claim 4, a large-scale reservoir numerical simulation method is characterized by optimizing the initial reservoir model, which includes reducing the number of models and face counts; the optimized model extracts attribute data from the model and draws a color scale based on the numerical values of the attribute data; the attribute data is converted into attribute textures based on the color scale, and the uv coordinate points of the optimized attribute textures are attached to the model to obtain the three-dimensional reservoir model, which is then exported.
6. According to claim 1, a large-scale reservoir numerical simulation method is characterized by using a distributed parallel computing scheme as the predetermined computing scheme in step 4; the distributed parallel computing scheme involves domain decomposition of the computational area after grid partitioning, resulting in multiple subdomains; these subdomains are allocated to corresponding computing units for parallel computing, and the interaction between multiple computing units is implemented based on mpi communication.
7. According to claim 1, a large-scale reservoir numerical simulation method is LU507017 characterized by using a predetermined fully implicit method and an implicit pressure-explicit saturation method to solve nonlinear equations in step 5; linear equations are solved using third-party linear equation solving libraries and based on selected preconditioning methods; the adaptive implicit method includes determining whether the change in saturation of the target component in the target grid in one time step exceeds a predetermined reference change value; if yes, the fully implicit method is used to solve the nonlinear equation corresponding to the target grid; if no, the implicit pressure-explicit saturation method is used to solve the nonlinear equation corresponding to the target grid.
8. According to claim 7, a large-scale reservoir numerical simulation method is characterized by using preconditioning methods; when using the fully implicit method to solve linear equations, the cpr method is selected; when using the adaptive implicit method to solve linear equations, either the domain decomposition method or the ilu (0) method is selected; if the domain decomposition method fails, the jacob method is selected.
9. A large-scale reservoir numerical simulation system, characterized by: reservoir simulation model establishment module: used to establish a numerical simulation model based on reservoir simulation data and create a three-dimensional reservoir simulation model based on the numerical simulation model; calculation region confirmation module: used to mesh the three-dimensional region simulation model to obtain the calculation region; subregion confirmation module: used to divide the calculation region into corresponding numbers of subregions based on a predetermined calculation scheme; subregion execution module: used to execute all subregions; simulation result output module: used to distribute all simulation results obtained from executing subregions.
10. According to claim 9, a large-scale reservoir numerical simulation system, LU507017 characterized by the reservoir simulation model establishment module comprising a reservoir numerical simulation model establishment module and a three-dimensional reservoir simulation model; the reservoir numerical simulation model establishment module is used to determine reservoir simulation data based on reservoir characteristics, core samples, well logging, and fluid test data, establish a large-scale reservoir numerical simulation model; the three-dimensional reservoir simulation model is used to determine the reservoir boundary range based on reservoir simulation data and large-scale mathematical simulation models and establish a three-dimensional reservoir simulation model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| LU507017A LU507017B1 (en) | 2024-04-24 | 2024-04-24 | A large-scale reservoir numerical simulation method and its numerical simulation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| LU507017A LU507017B1 (en) | 2024-04-24 | 2024-04-24 | A large-scale reservoir numerical simulation method and its numerical simulation system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| LU507017B1 true LU507017B1 (en) | 2024-10-24 |
Family
ID=93155396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| LU507017A LU507017B1 (en) | 2024-04-24 | 2024-04-24 | A large-scale reservoir numerical simulation method and its numerical simulation system |
Country Status (1)
| Country | Link |
|---|---|
| LU (1) | LU507017B1 (en) |
-
2024
- 2024-04-24 LU LU507017A patent/LU507017B1/en active
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA3027332C (en) | Runtime parameter selection in simulations | |
| CA2986034A1 (en) | Parallel solution for fully-coupled fully-implicit wellbore modeling in reservoir simulation | |
| CN108573078B (en) | Fracturing effect prediction method based on data mining | |
| CN106055827A (en) | Oil deposit numerical value simulation parameter sensibility analysis device and method | |
| CN107908913B (en) | Earth power digital-analog method based on parallel computer | |
| CN107038302A (en) | A kind of semi-analytical calculation method in the Eddy Current Nondestructive Testing magnetic field containing columnar defects | |
| LU507017B1 (en) | A large-scale reservoir numerical simulation method and its numerical simulation system | |
| CN110863818B (en) | Description method and device for residual oil/gas distribution | |
| CA2865053C (en) | Input parsing and array manipulation in reservoir simulation | |
| CN115758911B (en) | Fusion point cloud residual error network and flow field and pressure field prediction method considering slippage | |
| Lian et al. | Parallel numerical simulation for a super large-scale compositional reservoir | |
| Imankulov et al. | Design of HPC system for analysis the gel-polymer flooding of oil fields | |
| CN114764546A (en) | Method for predicting residual oil of fault block oil reservoir based on generation of antagonistic neural network | |
| CN108090962B (en) | Method for quickly generating PLAAXIS geometric model based on ANSYS data file | |
| CN113761634A (en) | Building structure design method based on multi-objective optimization | |
| CN106022944B (en) | Data volume conversion method and device for conventional and streamline models in numerical reservoir simulation | |
| CN115481520A (en) | Oil deposit numerical simulation input data model framework | |
| CN117421939B (en) | Shale oil fracture system simulation agent method based on track piecewise linearization | |
| Hailong | Dynamic analysis of bottom hole pressure of a horizontal well in fracture-cavity reservoirs | |
| CN109241551A (en) | The persistence method of arch dam pouring construction simulation object | |
| CN112377185B (en) | Single-hole and multi-medium composite reservoir parameter sensitivity analysis method and device | |
| CN116629112B (en) | Oil reservoir attribute parameter inversion method and system based on scientific intelligence | |
| CN107526879B (en) | Three-dimensional model gradient finite element solving method based on ANSYS parametric design language | |
| Salazar et al. | External parallelization of reservoir simulators using a network of workstations and PVM | |
| Bao et al. | DTCN: Dynamic Temporal Convolution Network for Evaluating Dividing Coefficients of Water Well |