CN112253097A - Oil-water high-permeability channel identification method based on big data analysis - Google Patents
Oil-water high-permeability channel identification method based on big data analysis Download PDFInfo
- Publication number
- CN112253097A CN112253097A CN202010974269.5A CN202010974269A CN112253097A CN 112253097 A CN112253097 A CN 112253097A CN 202010974269 A CN202010974269 A CN 202010974269A CN 112253097 A CN112253097 A CN 112253097A
- Authority
- CN
- China
- Prior art keywords
- well
- production
- water injection
- water
- oil
- 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
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B47/00—Survey of boreholes or wells
- E21B47/10—Locating fluid leaks, intrusions or movements
-
- 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
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B43/00—Methods or apparatus for obtaining oil, gas, water, soluble or meltable materials or a slurry of minerals from wells
- E21B43/30—Specific pattern of wells, e.g. optimizing the spacing of wells
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geology (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mining & Mineral Resources (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Fluid Mechanics (AREA)
- Environmental & Geological Engineering (AREA)
- Geochemistry & Mineralogy (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses an oil-water high-permeability channel identification method based on big data analysis, which comprises the following steps of collecting production data of a production well and a water injection well in a region; and step two, substituting the production data in the step one into a multiple linear regression model according to the multiple linear regression model of an injection and production system consisting of the water injection well and the production well to obtain the dynamic communication degree between the water injection well and the production well, representing the possible degree of high-permeability channels or microcracks generated between the water injection well and the production well, and providing support for subsequent deep profile control of the water injection well. Based on the production data of the water injection well and the production well, the high-permeability channel of the oil-water well for oil displacement and displacement in the deep part of the water injection well can be quickly judged, and a foundation is provided for subsequent oil displacement and displacement.
Description
Technical Field
The invention belongs to the field of petroleum industry, and particularly relates to an oil-water high-permeability channel identification method based on big data analysis
Background
Water well profile control and flooding have become the main measure for stabilizing the yield of low permeability oil fields. The annual workload reaches 800 wells. The optimization of profile control wells is one of the main ways to keep the effectiveness of low cost, high benefit measures. The corresponding relation of oil-water wells, namely the incoming water direction judgment of the high-water-content oil well, is the core of a well selection decision technology. At present, the identification of the corresponding relation of an oil-water well mainly takes dynamic data comparison and analysis as a main part, is greatly influenced by human factors, has low efficiency of whole well selection in a block and is difficult to adapt to production requirements.
At present, the corresponding relation of the oil-water well is quantitatively judged and identified at home and abroad mainly by solving by adopting the traditional mathematical methods such as a least square method, a genetic algorithm, a quasi-Newton method, Gaussian estimation and the like, and the implementation means mainly utilizes relatively professional tools such as SPSS software, MatLab, autonomous programming and the like. The technology threshold is high, and the method is not suitable for popularization and application.
Therefore, quantitative and rapid oil-water well corresponding relation identification methods need to be researched, and decision efficiency is improved.
Disclosure of Invention
The invention aims to provide an oil-water high-permeability channel identification method based on big data analysis, which is a quantitative and rapid oil-water well corresponding relation identification method by using a big data analysis algorithm and provides decision technology support for deep profile control and flooding of a water injection well.
An oil-water high-permeability channel identification method based on big data analysis,
collecting production data of a production well and a water injection well in a region;
and step two, substituting the production data in the step one into a multiple linear regression model according to the multiple linear regression model of an injection and production system consisting of the water injection well and the production well to obtain the dynamic communication degree between the water injection well and the production well, representing the possible degree of high-permeability channels or microcracks generated between the water injection well and the production well, and providing support for subsequent deep profile control of the water injection well.
In the second step, the multiple linear regression model is
In the formula:
is the estimated value of the liquid production of the jth production well in cm3S; t is the sampling time sequence number of the injection-production dynamic data; beta is a0jConstant terms for representing injection-production unbalance; n is the number of water injection wells; beta is aijDefining a multiple linear regression weight value between the ith water injection well and the jth production well as an interwell dynamic communication coefficient, and representing the dynamic communication degree of the injection well i and the production well j; i.e. ii(t) the water injection quantity of the water injection well at the ith port, cm3/s。
The production data comprises the monthly injection amount and the daily injection amount of the water injection well, and the monthly liquid recovery amount and the daily liquid recovery amount of the production well.
In the second step, the multivariate linear model obtains a constant term of injection-production unbalance and a multivariate linear regression weight value between the water injection well and the production well, wherein the multivariate linear regression weight value between the water injection well and the production well is the dynamic communication degree between the water injection well and the production well, and the higher the multivariate linear regression weight value between the water injection well and the production well is, the higher the dynamic communication degree between the water injection well and the production well is.
And the multivariate linear model obtains a constant term of injection-production unbalance and a multivariate linear regression weight value between the water injection well and the production well through a gradient descent method.
In the second step, when the gradient descent method is used for solving to obtain the injection-production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well, when the target function error is smaller than or equal to 2, the gradient descent is stopped, and the iterative injection-production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well are stable values.
And after the second step, sequencing the dynamic communication degrees between one water injection well and a plurality of production wells, or sequencing the dynamic communication degrees between one production well and a plurality of water injection wells, wherein a high-permeability channel or microcrack is considered to be generated between the first two wells from high to low, so that support is provided for subsequent deep profile control and flooding of the water injection wells.
The invention has the beneficial effects that: based on the production data of the water injection well and the production well, the high-permeability channel of the oil-water well for oil displacement and displacement in the deep part of the water injection well can be quickly judged, and a foundation is provided for subsequent oil displacement and displacement.
Drawings
FIG. 1 is a calculated hypertonic pathway map for West 32-29 wells;
FIG. 2 is a schematic representation of the hypertonic pathway for Western 32-29 well tracer tests;
FIG. 3 is a calculated hypertonic pathway map for West 28-35 wells;
FIG. 4 is a schematic representation of the hypertonic pathway for Western 28-35 well tracer tests;
FIG. 5 is a calculated hypertonic pathway map for West 31-20 wells;
FIG. 6 is a graph of the hypertonic channel of the West 31-20 well tracer test.
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Detailed Description
[ example 1 ]
An oil-water high-permeability channel identification method based on big data analysis,
collecting production data of a production well and a water injection well in a region;
and step two, substituting the production data in the step one into a multiple linear regression model according to the multiple linear regression model of an injection and production system consisting of the water injection well and the production well to obtain the dynamic communication degree between the water injection well and the production well, representing the possible degree of high-permeability channels or microcracks generated between the water injection well and the production well, and providing support for subsequent deep profile control of the water injection well.
The production data comprises the monthly injection amount and the daily injection amount of the water injection well, and the monthly liquid recovery amount and the daily liquid recovery amount of the production well.
The oil deposit is a dynamic balance system, the fluctuation of the oil well liquid production caused by the change of the water injection rate of the water well is the characteristic reflection of the communication in the oil-water well layer, and the fluctuation range of the oil well liquid production is related to the communication degree of the oil-water well. The better the connectivity between oil and water wells, the higher the correlation between water injection and fluid production data. Based on the concept of system analysis, the water injection wells, the production wells and the interwell reservoirs of the oil reservoir are regarded as a complete system, the input (excitation) of the system is the water injection amount of the water injection wells, and the output (response) of the system is the liquid production amount of the production wells.
In the second step, the multiple linear regression model is
In the formula:
is the estimated value of the liquid production of the jth production well in cm3S; t is the sampling time sequence number of the injection-production dynamic data; beta is a0jConstant terms for representing injection-production unbalance; n is the number of water injection wells; beta is aijDefining a multiple linear regression weight value between the ith water injection well and the jth production well as an interwell dynamic communication coefficient, and representing the dynamic communication degree of the injection well i and the production well j; i.e. ii(t) the water injection quantity of the water injection well at the ith port, cm3/s。
Representing the amount of liquid produced collected, ii(t) represents the collected injection amount, t represents the time corresponding to the injection amount and the liquid production amount, and the liquid production amount and the injection amount on the same day are put into a model, or the total liquid production amount of the production wells and the total injection amount of the injection wells i on the same month j are put into a model as data in month units.
Thus in oilfield production, the production variation of each well is linked to the co-operation of all the surrounding water injection wells with which it is in communication. According to the multiple linear regression concept, for an injection-production system consisting of a water injection well and a production well, the liquid production rate of the production well can be expressed as
By solving the formula, beta can be obtainedijThat is, the injection well i and the production well j are communicated, one water injection well usually corresponds to a plurality of production wells, and the production data of the other production wells are substituted into the solution, so that one water injection well and a plurality of production wells can be obtainedThe communication relationship between them. In actual use, according to historical data of well site and according to betaijThe value can be obtained to obtain the relation between a certain water injection well and which production wells, and whether a hypertonic channel or a micro-fracture exists is judged according to the correlation.
In the second step, the multivariate linear model obtains a constant term of injection-production unbalance and a multivariate linear regression weight value between the water injection well and the production well, wherein the multivariate linear regression weight value between the water injection well and the production well is the dynamic communication degree between the water injection well and the production well, and the higher the multivariate linear regression weight value between the water injection well and the production well is, the higher the dynamic communication degree between the water injection well and the production well is.
And the multivariate linear model obtains a constant term of injection-production unbalance and a multivariate linear regression weight value between the water injection well and the production well through a gradient descent method.
In the second step, when the gradient descent method is used for solving to obtain the injection-production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well, when the target function error is smaller than or equal to 2, the gradient descent is stopped, and the iterative injection-production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well are stable values.
Random numbers can be distributed by a system for initial values of the injection and production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well, new injection and production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well are obtained continuously in an iteration mode through a gradient descent method along with the introduction of production data until the injection and production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well tend to be stable, the injection and production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well at the moment are selected as final output values, and whether a high-permeability channel exists between the water injection well and the production well is judged through the final multiple linear regression weight value between the water injection well and the production well.
And when the error of the objective function is less than or equal to 2, considering that the constant term of injection-production unbalance and the multiple linear regression weight value between the water injection well and the production well tend to be stable, and stopping iteration of the regression descent method.
In the process of solving the stable value, the stable value is determined through an objective function, when the error range of the objective function is within 2, the stable state is considered to be achieved, and the iterated beta is selected0jAnd betaijAs a stable value.
In the formula, N is the number of water injection wells, and is the total amount of the water injection wells in the data, for example, the total data of 100 water injection wells in the imported big data, and then N is 100; and if 50 holes exist in the imported big data injection well data, taking 50N out of the N.
And after the second step, sequencing the dynamic communication degrees between one water injection well and a plurality of production wells, or sequencing the dynamic communication degrees between one production well and a plurality of water injection wells, wherein a high-permeability channel or microcrack is considered to be generated between the first two wells from high to low, so that support is provided for subsequent deep profile control and flooding of the water injection wells.
For example, the A water injection well is respectively communicated with B, C, D, E four production wells, and B, C, D three production wells are sequenced to be the first three, so that a hypertonic channel is considered to be formed between the A water injection well and B, C, D three production wells.
Wherein the solution in the multivariate linear model is as follows
The model is assumed to obey a multivariate linear relationship as follows:
qβ(I)＝β0+β1I1+…
at this time, q{i}＝βTI(i)+ε(i)Error epsilon(i)(1. ltoreq. i.ltoreq.m) are independent and identically distributed, subject to the central limit theorem that the mean is 0 and the variance is a specific value σ2A gaussian distribution of (a).
Likelihood function
Note:
an objective function:
gradient reduction:
and (3) solving a stagnation point, and obtaining an analytic expression of final parameters: beta ═ I (I)TI)-1ITq
Beta: refer to
Matrix array
I: refer to
Matrix array
q is to
Matrix array
[ example 2 ]
Based on example 1, 465 wells and 170 wells were used in the oil field. Wherein, 22860 dynamic month data and 685800 daily data of oil-water well, 21 tracer well groups are emphatically verified, and the accuracy is 75%. And rapid auxiliary identification can be carried out.
Typical well group analysis:
1. west 32-29 well software calculation and trace test comparison
As described in fig. 1, the software calculates the result: west 32-29 water injection wells and the oil wells with the highest correlation with the surrounding oil wells are West 31-27, West 31-29 and West 32-30 in sequence.
As depicted in fig. 2, trace test results: the injected water is mainly propelled in the west 32-30, west 33-31, west 31-29 directions, roughly north-east, where high permeability layers or microcracks are present. West 31-29 and West 32-30 are identical.
2. West 28-35 well software calculation and tracer trace test comparison
As shown in fig. 3, the software calculates the result: west 28-35 water injection wells and the oil wells with the highest correlation with the surrounding oil wells are 27-33 West, 27-34 West and 29-35 West in sequence.
As shown in fig. 4, the trace test results: the left side and the right side of the west 28-35 water injection well (west 27-33 well-west 29-38 well direction) are the main advancing direction of the injected water, and a high permeable layer or a micro-crack exists in the direction. The results of West 27-33, West 27-34 and West 29-35 are all consistent.
3. West 31-20 well software calculation and trace test comparison
As shown in fig. 5, the software calculates the result: west 31-20 water injection wells and the oil wells with the highest correlation with the surrounding oil wells are 31-19 West, 30-19 West and 32-22 West in sequence.
As shown in fig. 6, the trace test results: the West 31-20 water injection wells are communicated with the West 30-19 and the West 30-20 wells, and are the main advancing direction of injected water, and a high permeable layer or a micro-crack exists in the direction. West 31-19 gave identical results.
As shown in the embodiment, most of the calculated results and the actual tracing results are consistent, but the tracer test is high in cost and long in time consumption, the method is low in cost, and whether the hypertonic channel exists or not and the position where the hypertonic channel possibly exists indicate the large direction can be obtained by only using the existing data of a well site, so that the support is provided for the subsequent deep profile control of the water injection well.
Claims (7)
1. An oil-water high-permeability channel identification method based on big data analysis is characterized by comprising the following steps:
collecting production data of a production well and a water injection well in a region;
and step two, substituting the production data in the step one into a multiple linear regression model according to the multiple linear regression model of an injection and production system consisting of the water injection well and the production well to obtain the dynamic communication degree between the water injection well and the production well, representing the possible degree of high-permeability channels or microcracks generated between the water injection well and the production well, and providing support for subsequent deep profile control of the water injection well.
2. The oil-water hypertonic channel identification method based on big data analysis as claimed in claim 1, characterized in that: in the second step, the multiple linear regression model is
In the formula:
is the estimated value of the liquid production of the jth production well in cm3S; t is the sampling time sequence number of the injection-production dynamic data; beta is a0jConstant terms for representing injection-production unbalance; n is the number of water injection wells; beta is aijDefining a multiple linear regression weight value between the ith water injection well and the jth production well as an interwell dynamic communication coefficient, and representing the dynamic communication degree of the injection well i and the production well j; i.e. ii(t) the water injection quantity of the water injection well at the ith port, cm3/s。
3. The oil-water hypertonic channel identification method based on big data analysis as claimed in claim 1, characterized in that: the production data comprises the monthly injection amount and the daily injection amount of the water injection well, and the monthly liquid recovery amount and the daily liquid recovery amount of the production well.
4. The oil-water hypertonic channel identification method based on big data analysis as claimed in claim 1, characterized in that: in the second step, the multivariate linear model obtains a constant term of injection-production unbalance and a multivariate linear regression weight value between the water injection well and the production well, wherein the multivariate linear regression weight value between the water injection well and the production well is the dynamic communication degree between the water injection well and the production well, and the higher the multivariate linear regression weight value between the water injection well and the production well is, the higher the dynamic communication degree between the water injection well and the production well is.
5. The oil-water hypertonic channel identification method based on big data analysis as claimed in claim 4, characterized in that: and the multivariate linear model obtains a constant term of injection-production unbalance and a multivariate linear regression weight value between the water injection well and the production well through a gradient descent method.
6. The oil-water hypertonic channel identification method based on big data analysis as claimed in claim 5, characterized in that: in the second step, when the gradient descent method is used for solving to obtain the injection-production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well, when the target function error is smaller than or equal to 2, the gradient descent is stopped, and the iterative injection-production unbalanced constant term and the multiple linear regression weight value between the water injection well and the production well are stable values.
7. The oil-water hypertonic channel identification method based on big data analysis as claimed in claim 1, characterized in that: and after the second step, sequencing the dynamic communication degrees between one water injection well and a plurality of production wells, or sequencing the dynamic communication degrees between one production well and a plurality of water injection wells, wherein a high-permeability channel or microcrack is considered to be generated between the first two wells from high to low, so that support is provided for subsequent deep profile control and flooding of the water injection wells.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010974269.5A CN112253097A (en) | 2020-09-16 | 2020-09-16 | Oil-water high-permeability channel identification method based on big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010974269.5A CN112253097A (en) | 2020-09-16 | 2020-09-16 | Oil-water high-permeability channel identification method based on big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112253097A true CN112253097A (en) | 2021-01-22 |
Family
ID=74232472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010974269.5A Pending CN112253097A (en) | 2020-09-16 | 2020-09-16 | Oil-water high-permeability channel identification method based on big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112253097A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112901145A (en) * | 2021-03-19 | 2021-06-04 | 大庆油田有限责任公司 | Volume energy method for analyzing injection-production relation between oil-water wells |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130116998A1 (en) * | 2011-11-03 | 2013-05-09 | Bp Exploration Operating Company Limited | Statistical reservoir model based on detected flow events |
| CN110439515A (en) * | 2019-06-24 | 2019-11-12 | 中国石油化工股份有限公司 | Note adopts parameter optimization method and device |
-
2020
- 2020-09-16 CN CN202010974269.5A patent/CN112253097A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130116998A1 (en) * | 2011-11-03 | 2013-05-09 | Bp Exploration Operating Company Limited | Statistical reservoir model based on detected flow events |
| CN110439515A (en) * | 2019-06-24 | 2019-11-12 | 中国石油化工股份有限公司 | Note adopts parameter optimization method and device |
Non-Patent Citations (1)
| Title |
|---|
| 张明安: "油藏井间动态连通性反演方法研究", 油气地质与采收率, vol. 18, no. 3, pages 70 - 73 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112901145A (en) * | 2021-03-19 | 2021-06-04 | 大庆油田有限责任公司 | Volume energy method for analyzing injection-production relation between oil-water wells |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108240208B (en) | Oil field water-flooding classification well group development effect benchmarking method | |
| CN109441422B (en) | Shale gas well spacing optimization mining method | |
| CN107829718B (en) | Oil reservoir well pattern and injection-production program optimum design method based on balanced water drive theory | |
| CN112343587B (en) | Identification characterization method for dominant seepage channel of ultralow permeability oil reservoir | |
| CN110147561B (en) | Method for predicting volume fracture network of tight oil and gas reservoir containing natural fracture | |
| CN107387051B (en) | Repeated fracturing well selection method for multi-stage fractured horizontal well with low-permeability heterogeneous oil reservoir | |
| CN110130883A (en) | The determination method and device of formation parameters | |
| CN106547973A (en) | A kind of quantitative description of injection for heavy oil reservoir steam course channeling passage | |
| CN104091069A (en) | Method for determining oil driving efficiency and sweep coefficient of all layers and positions of heterogeneous reservoir stratum | |
| CN106611081B (en) | Judge automatically the integrated approach and system of fracture-pore reservoir production inter well connectivity | |
| CN108804382A (en) | A kind of automatic reverse method of parameter and device | |
| CN104060985A (en) | Method and system for testing entering depth of stratified oil deposit profile control water plugging agent | |
| CN112253097A (en) | Oil-water high-permeability channel identification method based on big data analysis | |
| CN106014365B (en) | A method of prediction waterflooding development Production Decline Prediction of Oilfield rate | |
| CN110751723B (en) | Same sub-basin river channel grid calculation order coding method | |
| CN111401716B (en) | Method for acquiring continuous parameters of aquifer based on superimposed water drainage test | |
| CN111553067B (en) | Method and device for testing and explaining non-stop well testing of shale oil reservoir fracturing horizontal well | |
| CN106484930A (en) | For determining the method and system of heterogeneous reservoir labyrinth well production | |
| CN109209361B (en) | Method for predicting stratum parameters of fractured ultra-low permeability reservoir | |
| CN110513089A (en) | A kind of Offshore Heavy Oil Field thermal recovery, which is handled up, develops production capacity multiple and determines method | |
| CN109653717A (en) | Combination, which is handled up, after individual well steam stimulation wells selects well decision-making technique | |
| RU2504654C1 (en) | Method of determination of oil extraction factor at nonlinear filtration | |
| CN109653730A (en) | One kind being used for drill-stem test operation downhole perforation well section depth calibration method | |
| CN110348154B (en) | Fracture-cavity type oil reservoir well group gas tracer interpretation and karst cave identification method | |
| CN109296358B (en) | Well bore quality evaluation method for well cementation |
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 |