CN106919756A - A kind of steam soak injection parameter optimization method based on approximate model - Google Patents
A kind of steam soak injection parameter optimization method based on approximate model Download PDFInfo
- Publication number
- CN106919756A CN106919756A CN201710117719.7A CN201710117719A CN106919756A CN 106919756 A CN106919756 A CN 106919756A CN 201710117719 A CN201710117719 A CN 201710117719A CN 106919756 A CN106919756 A CN 106919756A
- Authority
- CN
- China
- Prior art keywords
- steam
- steam soak
- injection parameter
- injection
- approximate model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of steam soak injection parameter optimization method based on approximate model, comprise the following steps：1) steam soak three-dimensional oil reservoir numerical simulator is set up；2) steam injection mass dryness fraction, steam injecting temperature, four injection parameters of steam injection rate and cyclic steam injection volume are chosen as design optimization variable；3) initial span is chosen to injection parameter；4) sample point is chosen for the injection parameter of steam soak using test design method；5) numerical simulation calculation is carried out to sample point using steam soak three-dimensional oil reservoir numerical simulator；6) object function is chosen, the approximate model for characterizing steam soak injection parameter is built using the injection parameter and its object function response of each sample point；7) precision of pairing approximation model is verified, if meeting required precision, carries out step 8), if being unsatisfactory for required precision, return to step 3)；8) genetic algorithm is combined to injection parameter global optimizing, so as to obtain optimal steam soak injection parameter.
Description
Technical field
The present invention relates to a kind of steam soak injection parameter optimization method based on approximate model, belong to oil field steam soak
Development technique field.
Background technology
Steam soak is a kind of effective method for improving recovery ratio of heavy crude reservoir, and it is main by injecting with certain
The steam of mass dryness fraction, the then stewing well of closing well, the mode that the discharge opeing that driven a well after a few days is recovered the oil.Reduction crude oil can be reached using steam soak
Viscosity, the purpose for improving viscous crude mobility and production capacity.Steam soak injection parameter affects Simulation on whole pay zones effect, optimal to obtain
Simulation on whole pay zones effect, Simulation on whole pay zones conceptual design need to carry out the optimization of injection parameter.Current steam soak injection parameter optimization
Local sensitivity analysis are mainly limited to, i.e., using fixed other specification, sensitivity analysis are carried out just for a certain parameter is changed
Method to find " flex point " is optimized.This method wastes time and energy, and can not take into full account the friendship between injection parameter
Interaction, Optimal Parameters be often before design load, it is difficult to obtain optimal injection parameter assembled scheme.Orthogonal experiment design method
Test number (TN) can be reduced, but injection parameter best of breed, optimum results cannot be obtained in whole injection parameter excursion
Without continuity, reliability is poor.
The content of the invention
Regarding to the issue above, it is an object of the invention to provide a kind of amount of calculation is small and optimum results reliability is high based on
The steam soak injection parameter optimization method of approximate model.
To achieve the above object, the present invention uses following technical scheme：A kind of steam soak injection based on approximate model
Parameter optimization method, comprises the following steps：1) the geological reservoir feature according to subject oil field sets up steam soak three-dimensional oil reservoir number
Value simulation model；2) choose steam injection mass dryness fraction, steam injecting temperature, steam injection rate and cyclic steam injection volume etc. four and effect is developed to steam soak
Fruit influences larger injection parameter as design optimization variable；3) initial span is chosen to aforementioned four injection parameter；
4) sample point is chosen for the injection parameter of steam soak using test design method；5) using step 1) steam set up
Three-dimensional oil reservoir numerical simulator of handling up carries out numerical simulation calculation to sample point, obtains the steam soak corresponding to each sample point
Oil production, gas oil ratio, net present value (NPV), it is cold adopt on the basis of steam soak yield amplification, recovery ratio amplification and it is cold adopt on the basis of steaming
Vapour is handled up net present value (NPV) amplification；6) with steam soak oil production, gas oil ratio, net present value (NPV), it is cold adopt on the basis of steam soak yield increase
Width, recovery ratio amplification and it is cold adopt on the basis of one of steam soak net present value (NPV) amplification as object function, using each sample point
Injection parameter and its object function response characterize the approximate model of steam soak injection parameter to build；7) pairing approximation model
Precision is verified, if meeting required precision, carries out step 8), if being unsatisfactory for required precision, return to step 3)；8) basis
The steam soak injection parameter of structure and the approximate model of object function, with reference to genetic algorithm to injection parameter global optimizing, from
And obtain optimal steam soak injection parameter.
The step 1) in, steam soak three-dimensional oil reservoir numerical simulator is to simulate business software by thermal recovery to set up
's.
The step 4) employed in test design method centered on composite design method, Box-Behnken design sides
Method or Latin hypercube body method for designing.
The step 6) described in approximate model be using polynomial response surface approximate model, wherein, Quadratic response
Approximate model mathematic(al) representation is as follows：
In formula, y is response, and k is the number of design variable, and ε is random error, xiIt is i-th point of k dimension independents variable x
Amount, β0, βiAnd βijIt is undetermined coefficient, it is constituted into column vector β according to certain sequential arrangement, it is to obtain above-mentioned near to solve vector β
Like model.
Column vector β is solved by minimum secondary method.
The step 7) in, tested using the precision of multiple correlation coefficient index pairing approximation model, wherein, phase relation
Number is to meet required precision more than 0.98.
Also include step 9) it is as follows：Optimum results are carried out with analysis of uncertainty using Monte Carlo model, quantify risk,
Verify the reliability of optimal case.
Due to taking above technical scheme, it has advantages below to the present invention：1st, the invention provides a set of based on approximate
The steam soak injection parameter optimization method of modelling technique and genetic algorithm, using approximate model technology, builds steam soak note
Enter the approximate model between parameter and its object function response, finally the true implicit function of description steam soak challenge
Explicit approximate function expression formula is converted into, can be substituted with less amount of calculation a large amount of on the premise of computational accuracy is ensured
Numerical simulation works.Global optimizing can be carried out in injection parameter excursion in combination with genetic algorithm, it is to avoid injection ginseng
The discontinuity of number optimum results, so as to improve optimum results reliability.2nd, The present invention gives quantification, exercisable technology
Method and implementation steps.3rd, the present invention is also suitable for land oil field suitable for the determination of offshore oilfield steam soak injection parameter
The determination of steam soak injection parameter.
Brief description of the drawings
Fig. 1 is schematic flow sheet of the invention.
Specific embodiment
The present invention is described in detail with reference to the accompanying drawings and examples.
A kind of steam soak injection parameter optimization method based on approximate model of the present invention, comprises the following steps：
1) the geological reservoir feature according to subject oil field sets up steam soak three-dimensional oil reservoir numerical simulator, general to use
Ripe thermal recovery simulation business software, such as CMG-STARS, ECLIPSE-E300 simulator, the three-dimensional oil of the steam soak set up
Hide numerical simulator and should try one's best and embody the actual characteristics of reservoirs such as Reservoir Heterogeneity.
2) steam injection mass dryness fraction, steam injecting temperature, steam injection rate and cyclic steam injection volume etc. four is chosen to steam soak development effectiveness
The larger injection parameter of influence is used as design optimization variable.
3) initial span is chosen to aforementioned four injection parameter, such as with reference to the actual heat injection equipment energy of Bohai Bay Oil
Power requirement, the span of aforementioned four injection parameter can distinguish 0.0~0.4,300~340 DEG C of value, 200~300m3/ d and
3000~5000m3(steam water equivalent).
4) sample point is chosen for the injection parameter of steam soak using test design method, it is less in order to carry out
Experiment can just obtain enough information.Composite design method, Box-Behnken design sides centered on above-mentioned test design method
One of method, Latin hypercube body method for designing.
In a preferred embodiment, from Box-Behnken methods for designing, the method for designing is that one kind meets rotation
Property or the almost spherical design of rotatability, extensively should due to its rational Sampling Strategy and good result performance
With.
5) using step 1) the steam soak three-dimensional oil reservoir numerical simulator set up carries out numerical simulation to sample point
Calculate, obtain the steam soak oil production corresponding to each sample point, gas oil ratio, net present value (NPV), it is cold adopt on the basis of steam soak produce
Amount amplification, recovery ratio amplification and it is cold adopt on the basis of steam soak net present value (NPV) amplification.
6) with steam soak oil production, gas oil ratio, net present value (NPV), it is cold adopt on the basis of steam soak yield amplification, recovery ratio
Amplification and it is cold adopt on the basis of one of steam soak net present value (NPV) amplification as object function, using each sample point injection parameter and
Its object function response characterizes the approximate model of steam soak injection parameter to build, and thus description steam soak is injected and is joined
Several true implicit functions for influenceing this challenge on target function value are converted into the approximate function expression formula of display.
Approximate model is divided into Local approximation, global approximate and medium range according to the big I of the design space that can be described
Approximate three class.The present invention uses polynomial response surface approximate model, and the approximate model belongs to global approximate, convenient with setting up
Simply, fast convergence rate, can be expressed using explicit function, the advantage such as the transparency is good.Specifically can be approximate using Quadratic response
Model, Quadratic response approximate model mathematic(al) representation is as follows：
In formula, y is response, and k is the number of design variable, and ε is random error, xiIt is i-th point of k dimension independents variable x
Amount, β0, βiAnd βijIt is undetermined coefficient, it is constituted into column vector β according to certain sequential arrangement, solves vector β as setting up secondary
The key of response surface model.
The data of each sample point are substituted into above formula respectively, during substitution, y values are the object function of current sample point, k values
It is 4, i.e. independent variable x is 4 dimension independents variable, its four components are respectively steam injection mass dryness fraction, steam injecting temperature, steam injection rate and cycle steam injection
Amount.
Then vectorial β can be tried to achieve using least square method, so as to try to achieve each undetermined coefficient, and then it is near to obtain Quadratic response
Like model.
7) precision of pairing approximation model is verified, if meeting required precision, carries out step 8), will if being unsatisfactory for precision
Ask, then return to step 3).
For the checking of approximate model accuracy, multiple correlation coefficient (R is generally available2) index tests.Complex phase relation
Number has reacted the accuracy of approximate model, its numerical value 0<R2≤ 1, its value is closer to 1, and regression equation is more accurate.For viscous crude oil
Steam soak exploitation is hidden, application claims coefficient correlation is more than 0.98.
8) according to the steam soak injection parameter for building and the approximate model of object function, injection is joined with reference to genetic algorithm
Number global optimizing, so as to obtain optimal steam soak injection parameter.
The present invention with the quadratic response surface model of steam soak object function as fitness function, using floating-point code, just
Beginning population is produced by random fashion, and selection operation is carried out based on geometry distribution, and crossover operation is carried out using arithmetic crossover method, is adopted
Mutation operation is carried out with non-uniform mutation method.Global optimizing is realized in the region for meet constraints, is obtained and is met convergence bar
The injection parameter optimum combination of part.
In completion step 8) after, implementation steps 9 can be selected)：
9) optimum results are carried out with analysis of uncertainty using Monte Carlo model, quantifies risk, checking optimal case
Reliability.
The various embodiments described above are merely to illustrate the present invention, and wherein implementation steps of method etc. all can be what is be varied from,
Every equivalents carried out on the basis of technical solution of the present invention and improvement, should not exclude in protection scope of the present invention
Outside.
Claims (7)
1. a kind of steam soak injection parameter optimization method based on approximate model, comprises the following steps：
1) the geological reservoir feature according to subject oil field sets up steam soak three-dimensional oil reservoir numerical simulator；
2) steam injection mass dryness fraction, steam injecting temperature, steam injection rate and cyclic steam injection volume etc. four is chosen to steam soak development effect influence
Larger injection parameter is used as design optimization variable；
3) initial span is chosen to aforementioned four injection parameter；
4) sample point is chosen for the injection parameter of steam soak using test design method；
5) using step 1) the steam soak three-dimensional oil reservoir numerical simulator set up carries out numerical simulation calculation to sample point,
Obtain the steam soak oil production corresponding to each sample point, gas oil ratio, net present value (NPV), it is cold adopt on the basis of steam soak yield increase
Width, recovery ratio amplification and it is cold adopt on the basis of steam soak net present value (NPV) amplification；
6) with steam soak oil production, gas oil ratio, net present value (NPV), it is cold adopt on the basis of steam soak yield amplification, recovery ratio amplification
With one of the steam soak net present value (NPV) amplification on the basis of cold adopting as object function, using the injection parameter and its mesh of each sample point
Scalar functions response characterizes the approximate model of steam soak injection parameter to build；
7) precision of pairing approximation model is verified, if meeting required precision, carries out step 8), if being unsatisfactory for required precision,
Then return to step 3)；
8) it is complete to injection parameter with reference to genetic algorithm according to the steam soak injection parameter and the approximate model of object function for building
Office's optimizing, so as to obtain optimal steam soak injection parameter.
2. a kind of steam soak injection parameter optimization method based on approximate model as claimed in claim 1, it is characterised in that：
The step 1) in, steam soak three-dimensional oil reservoir numerical simulator is to simulate business software by thermal recovery to set up.
3. a kind of steam soak injection parameter optimization method based on approximate model as claimed in claim 1, it is characterised in that：
The step 4) employed in test design method centered on composite design method, Box-Behnken methods for designing or Latin
Hypercube method for designing.
4. a kind of steam soak injection parameter optimization method based on approximate model as claimed in claim 1, it is characterised in that：
The step 6) described in approximate model be using polynomial response surface approximate model, wherein, Quadratic response approximate model
Mathematic(al) representation is as follows：
In formula, y is response, and k is the number of design variable, and ε is random error, xiIt is i-th component of k dimension independents variable x, β0,
βiAnd βijIt is undetermined coefficient, it is constituted into column vector β according to certain sequential arrangement, it is to obtain above-mentioned approximate model to solve vector β.
5. a kind of steam soak injection parameter optimization method based on approximate model as claimed in claim 4, it is characterised in that：
Column vector β is solved by minimum secondary method.
6. a kind of steam soak injection parameter optimization method based on approximate model as claimed in claim 1, it is characterised in that：
The step 7) in, tested using the precision of multiple correlation coefficient index pairing approximation model, wherein, coefficient correlation is more than
0.98 is to meet required precision.
7. a kind of steam soak injection parameter optimization method based on approximate model as claimed in claim 1, it is characterised in that：
Also include step 9) it is as follows：Optimum results are carried out with analysis of uncertainty using Monte Carlo model, quantifies risk, verified optimal
The reliability of scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710117719.7A CN106919756A (en) | 2017-03-01 | 2017-03-01 | A kind of steam soak injection parameter optimization method based on approximate model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710117719.7A CN106919756A (en) | 2017-03-01 | 2017-03-01 | A kind of steam soak injection parameter optimization method based on approximate model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106919756A true CN106919756A (en) | 2017-07-04 |
Family
ID=59454316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710117719.7A Pending CN106919756A (en) | 2017-03-01 | 2017-03-01 | A kind of steam soak injection parameter optimization method based on approximate model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106919756A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106886649A (en) * | 2017-03-01 | 2017-06-23 | 中国海洋石油总公司 | A kind of multielement hot fluid is handled up injection parameter optimization method |
| CN107515978A (en) * | 2017-08-17 | 2017-12-26 | 广东工业大学 | The method of response surface model is built based on genetic algorithm and applies its system |
| CN109359332A (en) * | 2018-09-07 | 2019-02-19 | 中国石油化工股份有限公司 | A kind of shallow-thin layer reservoir numerical simulation method for establishing model and the method for turning steam drive |
| CN110046754A (en) * | 2019-03-29 | 2019-07-23 | 中国海洋石油集团有限公司 | Latin Hypercube Sampling oil field liquid production structure optimization method, storage medium and terminal |
| CN110513090A (en) * | 2019-09-06 | 2019-11-29 | 中海石油(中国)有限公司 | A kind of Offshore Heavy Oil Field steam soak is with producing the method for determination |
| CN112508727A (en) * | 2019-08-26 | 2021-03-16 | 中国石油天然气股份有限公司 | Method and system for determining recovery ratio of steam huff-puff heavy oil reservoir |
| CN114548568A (en) * | 2022-02-24 | 2022-05-27 | 中国石油大学(华东) | Gas injection huff-puff parameter collaborative optimization method for tight shale oil reservoir |
| CN116561907A (en) * | 2023-04-12 | 2023-08-08 | 中国石油大学(华东) | Flue gas reinjection parameter optimization method based on differential evolution algorithm |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106246152A (en) * | 2016-08-18 | 2016-12-21 | 中国石油天然气股份有限公司 | Method for optimizing huff-puff steam injection parameters of complex fault block heavy oil reservoir |
-
2017
- 2017-03-01 CN CN201710117719.7A patent/CN106919756A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106246152A (en) * | 2016-08-18 | 2016-12-21 | 中国石油天然气股份有限公司 | Method for optimizing huff-puff steam injection parameters of complex fault block heavy oil reservoir |
Non-Patent Citations (2)
| Title |
|---|
| 师俊峰等: "遗传算法用于蒸汽吞吐井注采参数的优化设计", 《石油天然气学报》 * |
| 李俊键等: "基于代理模型与遗传算法的海上油田聚合物驱注入参数优化控制方法", 《中国海上油气》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106886649A (en) * | 2017-03-01 | 2017-06-23 | 中国海洋石油总公司 | A kind of multielement hot fluid is handled up injection parameter optimization method |
| CN107515978A (en) * | 2017-08-17 | 2017-12-26 | 广东工业大学 | The method of response surface model is built based on genetic algorithm and applies its system |
| CN109359332A (en) * | 2018-09-07 | 2019-02-19 | 中国石油化工股份有限公司 | A kind of shallow-thin layer reservoir numerical simulation method for establishing model and the method for turning steam drive |
| CN110046754A (en) * | 2019-03-29 | 2019-07-23 | 中国海洋石油集团有限公司 | Latin Hypercube Sampling oil field liquid production structure optimization method, storage medium and terminal |
| CN110046754B (en) * | 2019-03-29 | 2021-12-10 | 中国海洋石油集团有限公司 | Latin hypercube sampling oil field liquid production structure optimization method, storage medium and terminal |
| CN112508727A (en) * | 2019-08-26 | 2021-03-16 | 中国石油天然气股份有限公司 | Method and system for determining recovery ratio of steam huff-puff heavy oil reservoir |
| CN110513090A (en) * | 2019-09-06 | 2019-11-29 | 中海石油(中国)有限公司 | A kind of Offshore Heavy Oil Field steam soak is with producing the method for determination |
| CN114548568A (en) * | 2022-02-24 | 2022-05-27 | 中国石油大学(华东) | Gas injection huff-puff parameter collaborative optimization method for tight shale oil reservoir |
| CN116561907A (en) * | 2023-04-12 | 2023-08-08 | 中国石油大学(华东) | Flue gas reinjection parameter optimization method based on differential evolution algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106919756A (en) | A kind of steam soak injection parameter optimization method based on approximate model | |
| Dachanuwattana et al. | Application of proxy-based MCMC and EDFM to history match a Vaca Muerta shale oil well | |
| CN106886649A (en) | A kind of multielement hot fluid is handled up injection parameter optimization method | |
| Wu et al. | Integrated characterization of the fracture network in fractured shale gas reservoirs—stochastic fracture modeling, simulation and assisted history matching | |
| Xu et al. | A hybrid embedded discrete fracture model for simulating tight porous media with complex fracture systems | |
| CN104750896B (en) | A kind of fractured-cavernous carbonate reservoir method for numerical simulation | |
| Akın et al. | Optimization of well placement geothermal reservoirs using artificial intelligence | |
| CN104899380A (en) | Side slope stable reliability sensitivity analysis method based on Monte Carlo simulation | |
| Zhai et al. | Optimisation of hydraulic fracturing parameters based on cohesive zone method in oil shale reservoir with random distribution of weak planes | |
| Zhai et al. | Production optimization in waterfloods with a new approach of inter-well connectivity modeling | |
| Shiqian et al. | A history matching framework to characterize fracture network and reservoir properties in tight oil | |
| Jiang et al. | Inverse analysis for geomaterial parameter identification using Pareto multiobjective optimization | |
| Zhang et al. | Thermal-hydraulic-mechanical-chemical modeling and simulation of an enhanced geothermal system based on the framework of extended finite element methods-Embedded discrete fracture model | |
| Ren et al. | Propagation simulation and structural characterization of multiple hydraulic fractures in naturally fractured unconventional hydrocarbon reservoirs | |
| Hanea et al. | Robust multi-objective field development optimization for the mariner asset | |
| Zhou et al. | Artificial Neural Network‐(ANN‐) Based Proxy Model for Fast Performances’ Forecast and Inverse Schedule Design of Steam‐Flooding Reservoirs | |
| Huang et al. | Simulation of the production performance of fractured horizontal wells in shale gas reservoirs considering the complex fracture shape | |
| Zhao et al. | A flownet-based method for history matching and production prediction of shale or tight reservoirs with fracturing treatment | |
| MoradiDowlatabad et al. | New approach for predicting multiple fractured horizontal wells performance in tight reservoirs | |
| Liu et al. | A general method for simulating reactive dissolution in carbonate rocks with arbitrary geometry | |
| Xu et al. | Efficient production optimization for naturally fractured reservoir using EDFM | |
| CN117517061A (en) | Fine intelligent sensing method and system for strength characteristics of all-domain rock of tunnel face | |
| Wang et al. | Fracture network mapping using integrated micro-seismic events inverse with rate-transient analysis | |
| US20240232459A9 (en) | ENSEMBLE KALMAN FILTER (EnKF)-BASED METHOD AND SYSTEM FOR REALTIME INVERSION OF HYDRAULIC FRACTURE | |
| Shahin et al. | A new model based on evolutionary computing for predicting ultimate pure bending of steel circular tubes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100010 Beijing, Chaoyangmen, North Street, No. 25, No. Applicant after: China Offshore Oil Group Co., Ltd. Applicant after: CNOOC research institute limited liability company Address before: 100010 Beijing, Chaoyangmen, North Street, No. 25, No. Applicant before: China National Offshore Oil Corporation Applicant before: CNOOC Research Institute |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170704 |