CN113449417A - Method for predicting overflow layer section of water injection well - Google Patents
Method for predicting overflow layer section of water injection well Download PDFInfo
- Publication number
- CN113449417A CN113449417A CN202110674578.5A CN202110674578A CN113449417A CN 113449417 A CN113449417 A CN 113449417A CN 202110674578 A CN202110674578 A CN 202110674578A CN 113449417 A CN113449417 A CN 113449417A
- Authority
- CN
- China
- Prior art keywords
- water injection
- overflow
- well
- interval
- injection
- 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
- 238000002347 injection Methods 0.000 title claims abstract description 70
- 239000007924 injection Substances 0.000 title claims abstract description 70
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000010801 machine learning Methods 0.000 claims abstract description 11
- 238000004519 manufacturing process Methods 0.000 claims description 21
- 239000010410 layer Substances 0.000 claims description 12
- 239000004576 sand Substances 0.000 claims description 6
- 239000011229 interlayer Substances 0.000 claims description 5
- 238000007637 random forest analysis Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 4
- 239000012530 fluid Substances 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims description 2
- 238000011161 development Methods 0.000 description 5
- 239000000243 solution Substances 0.000 description 5
- 238000013145 classification model Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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 invention discloses a method for predicting an overflow interval of a water injection well, which is characterized in that a machine learning algorithm is integrated into the prediction process of the overflow interval of the water injection well, and the relationship among the geology, the oil deposit, the water injection characteristics and the overflow condition of the overflow interval is automatically learned by utilizing the strong self-learning capability of the machine learning algorithm, so that the accurate prediction of the overflow interval of the water injection well is realized.
Description
Technical Field
The invention belongs to the technical field of exploration and development of oil and gas reservoirs, and particularly relates to a method for predicting an overflow interval of a water injection well.
Background
In the process of oil field development, the phenomenon of ubiquitous phenomenon overflow in the water injection well production process in the hypotonic oil reservoir development process to overflow phenomenon is along with oil field development throughout, and the oil field stratum energy that arouses from this supplements inadequately, can cause the layer to use difference with, influences the development effect, consequently, carry out accurate prediction to water injection well overflow position, establish effective driving pressure difference, improve the displacement of reservoir oil effect, solve the problem that oil field injection can not advance to produce, just can effectively improve the production efficiency in oil field.
Disclosure of Invention
In order to solve the problems that the overflow position of a water injection well is difficult to predict and main factors influencing overflow are difficult to determine, the invention provides a method for predicting overflow intervals of the water injection well, so that the accuracy of prediction of the overflow position of the water injection well is improved.
The technical scheme of the invention is as follows:
a method for predicting an overflow interval of a water injection well comprises the following steps:
the method comprises the following steps: extracting water injection and overflow key parameter data to obtain injection and production connectivity, daily water injection quantity, water injection days, the number of surrounding oil production wells, interlayer heterogeneity, water injection pressure, water injection strength, average injection and production well spacing and overflow conditions of a water injection layer section;
step two: establishing an overflow model, namely acquiring an optimal prediction model of an overflow layer section of the water injection well by using key parameter data of water injection overflow through a machine learning algorithm;
step three: and (4) overflow horizon prediction, namely substituting the key overflow parameters of the interval to be predicted into a prediction model to obtain the prediction result of the overflow condition.
In the above technical solution, in the step one, the method for acquiring the injection-production connectivity is as follows: and comprehensively judging the injection-production connectivity through three indexes, namely the cross communication condition of different small-layer sand bodies, the communication condition of different gyratory sand bodies of the same small layer and the sand body contact condition.
In the above technical solution, in the first step, the method for acquiring the overflow condition is as follows: and (3) putting a casing pressure on the water injection well after normal water injection for a period of time, closing the well, opening the well after 24 hours of closing the well for testing, and if the pressure of the well mouth is more than 0Mpa and the fluid overflows from the well mouth, determining that the well is overflowed, otherwise, determining that the well is not overflowed.
In the above technical solution, in the second step, the method for obtaining the optimal prediction model is as follows: and selecting a model with the highest score as a water injection well overflow interval prediction model by using injection-production connectivity, daily water injection quantity, water injection days, the number of surrounding oil production wells, interlayer heterogeneity, water injection pressure, water injection strength and average injection-production well spacing in overflow key parameters as input characteristics and overflow conditions as output characteristics through the machine learning algorithm operation of a decision tree, a random forest, XGboost and Adaboost.
The beneficial effects of the invention are as follows:
the invention discloses a method for predicting an overflow interval of a water injection well, which is characterized in that a machine learning algorithm is integrated into the prediction process of the overflow interval of the water injection well, and the relationship among the geology, the oil deposit, the water injection characteristics and the overflow condition of the overflow interval is automatically learned by utilizing the strong self-learning capability of the machine learning algorithm, so that the accurate prediction of the overflow interval of the water injection well is realized.
Detailed Description
In order to make the technical solution of the present invention better understood, the technical solution of the present invention is further described below with reference to specific examples.
Example (b):
a method for predicting an overflow interval of a water injection well comprises the following steps:
according to the first step: extracting water injection and overflow key parameter data to obtain injection and production connectivity, daily water injection quantity, water injection days, the number of surrounding oil production wells, interlayer heterogeneity, water injection pressure, water injection strength, average injection and production well spacing and overflow conditions of a water injection layer section, wherein the results are shown in the following table 1;
TABLE 1
According to the second step: establishing an overflow model, namely calculating by using key parameter data of water injection overflow through machine learning algorithms such as a decision tree, a random forest, XGboost, Adaboost and the like, and selecting a random forest classification model with the highest score as a water injection well overflow layer section prediction model, wherein the scores of the prediction model and the verification accuracy are shown in the following table 2;
TABLE 2
Algorithm model | Model cross validation accuracy score |
XGboost classification model | 0.74 |
Decision tree classification model | 0.74 |
Random forest classification model | 0.83 |
AdaBoost classification model | 0.71 |
According to the third step: and (3) overflow horizon prediction, namely substituting the key overflow parameters of the interval to be predicted into a prediction model to obtain a prediction result of the overflow condition, wherein the prediction result is shown in the following table 3.
TABLE 3
The experimental results show that the machine learning algorithm is integrated into the prediction process of the overflow interval of the water injection well, and the relationship between the geology, the oil deposit, the water injection characteristics and the overflow condition of the overflow interval is automatically learned by utilizing the strong self-learning capability of the machine learning algorithm, so that the accurate prediction of the overflow interval of the water injection well is realized.
The invention has been described in an illustrative manner, and it is to be understood that any simple variations, modifications or other equivalent changes which can be made by one skilled in the art without departing from the spirit of the invention fall within the scope of the invention.
Claims (4)
1. A method for predicting an overflow interval of a water injection well is characterized by comprising the following steps: the method comprises the following steps:
the method comprises the following steps: extracting water injection and overflow key parameter data to obtain injection and production connectivity, daily water injection quantity, water injection days, the number of surrounding oil production wells, interlayer heterogeneity, water injection pressure, water injection strength, average injection and production well spacing and overflow conditions of a water injection layer section;
step two: establishing an overflow model, namely acquiring an optimal prediction model of an overflow layer section of the water injection well by using key parameter data of water injection overflow through a machine learning algorithm;
step three: and (4) overflow horizon prediction, namely substituting the key overflow parameters of the interval to be predicted into a prediction model to obtain the prediction result of the overflow condition.
2. The method for predicting the overflow interval of the water injection well according to claim 1, wherein: in step one, the method for acquiring the injection-production connectivity is as follows: and comprehensively judging the injection-production connectivity through three indexes, namely the cross communication condition of different small-layer sand bodies, the communication condition of different gyratory sand bodies of the same small layer and the sand body contact condition.
3. The method for predicting the overflow interval of the water injection well according to claim 1, wherein: in step one, the method for acquiring the overflow condition is as follows: and (3) putting a casing pressure on the water injection well after normal water injection for a period of time, closing the well, opening the well after 24 hours of closing the well for testing, and if the pressure of the well mouth is more than 0Mpa and the fluid overflows from the well mouth, determining that the well is overflowed, otherwise, determining that the well is not overflowed.
4. The method for predicting the overflow interval of the water injection well according to claim 1, wherein: in step two, the method for obtaining the optimal prediction model is as follows: and selecting a model with the highest score as a water injection well overflow interval prediction model by using injection-production connectivity, daily water injection quantity, water injection days, the number of surrounding oil production wells, interlayer heterogeneity, water injection pressure, water injection strength and average injection-production well spacing in overflow key parameters as input characteristics and overflow conditions as output characteristics through the machine learning algorithm operation of a decision tree, a random forest, XGboost and Adaboost.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110674578.5A CN113449417A (en) | 2021-06-17 | 2021-06-17 | Method for predicting overflow layer section of water injection well |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110674578.5A CN113449417A (en) | 2021-06-17 | 2021-06-17 | Method for predicting overflow layer section of water injection well |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113449417A true CN113449417A (en) | 2021-09-28 |
Family
ID=77811739
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110674578.5A Pending CN113449417A (en) | 2021-06-17 | 2021-06-17 | Method for predicting overflow layer section of water injection well |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113449417A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105408914A (en) * | 2013-10-22 | 2016-03-16 | 萨思学会有限公司 | Fluid flow back prediction |
CN108388921A (en) * | 2018-03-05 | 2018-08-10 | 中国石油集团工程技术研究院有限公司 | A kind of overflow leakage real-time identification method based on random forest |
CN109002574A (en) * | 2018-06-06 | 2018-12-14 | 西安石油大学 | A kind of stratified reservoir pulse period waterflooding extraction index prediction technique |
CN111414955A (en) * | 2020-03-17 | 2020-07-14 | 北京中油瑞飞信息技术有限责任公司 | Intelligent detection method and device for leakage and overflow of petroleum drilling well and electronic equipment |
-
2021
- 2021-06-17 CN CN202110674578.5A patent/CN113449417A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105408914A (en) * | 2013-10-22 | 2016-03-16 | 萨思学会有限公司 | Fluid flow back prediction |
CN108388921A (en) * | 2018-03-05 | 2018-08-10 | 中国石油集团工程技术研究院有限公司 | A kind of overflow leakage real-time identification method based on random forest |
CN109002574A (en) * | 2018-06-06 | 2018-12-14 | 西安石油大学 | A kind of stratified reservoir pulse period waterflooding extraction index prediction technique |
CN111414955A (en) * | 2020-03-17 | 2020-07-14 | 北京中油瑞飞信息技术有限责任公司 | Intelligent detection method and device for leakage and overflow of petroleum drilling well and electronic equipment |
Non-Patent Citations (1)
Title |
---|
MOHAMMAD ALJUBRAN 等: "Deep Learning and Time-Series Analysis for the Early Detection of Lost Circulation Incidents During Drilling Operations", 《IEEE ACCESS》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112343587A (en) | Ultra-low permeability reservoir dominant seepage channel identification and characterization method | |
CN112487582A (en) | Oil-gas drilling machinery drilling speed prediction and optimization method based on CART algorithm | |
CN115345378B (en) | Shale gas well yield evaluation method based on machine learning | |
CN113537585B (en) | Oil field production increasing measure recommendation method based on random forest and gradient lifting decision tree | |
CN107895092B (en) | Inter-well communication quantitative evaluation method based on complex nonlinear injection-production modeling | |
CN112360411A (en) | Local well pattern water injection development optimization method based on graph neural network | |
CN109063403B (en) | Optimal design method for slickwater fracturing | |
CN112211627B (en) | Selection method of low-permeability gas reservoir interference well testing test well | |
CN111461386A (en) | Shale gas sweet spot prediction method based on BP neural network | |
CN114781951A (en) | Shale oil reservoir carbon dioxide huff-puff development well selection decision method and system | |
CN111749688B (en) | Method for predicting development position and direction of dominant seepage channel | |
CN113449417A (en) | Method for predicting overflow layer section of water injection well | |
CN112257892B (en) | Optimization method for drainage gas production process system of complex gas reservoir | |
CN111749664B (en) | Oil field whole-course chemical flooding combined take-over opportunity optimization method and system | |
CN113052374B (en) | Data-driven intelligent prediction method for casing loss depth of oil well | |
CN115758868A (en) | Compact reservoir pressure maintaining shape maintaining coring sleeve material selection and model selection method based on CT scanning | |
CN112502677B (en) | Water injection development effect evaluation method based on multiple linear regression | |
CN112561279B (en) | Identification method and system for high-water-consumption zone | |
CN111101930B (en) | Single-well exploitation production-increasing potential evaluation method in gas reservoir development mode | |
CN116307670A (en) | Oil-gas well overflow risk monitoring method and system based on two-way long-short-term memory network | |
CN113988410A (en) | Cross-region tight oil reservoir oil well productivity prediction method based on KNN algorithm and polynomial regression algorithm combination | |
CN113153224A (en) | Diagnosis method and equipment suitable for shale gas well plunger process operation state | |
CN112709547A (en) | Method for judging water plugging time of water producing gas well | |
CN116066066B (en) | Method for screening nitrogen throughput measure well | |
Schiozer et al. | Integration of history matching and uncertainty analysis |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210928 |
|
RJ01 | Rejection of invention patent application after publication |