CN112541571A - Injection-production connectivity determination method based on machine learning of double parallel neural networks - Google Patents
Injection-production connectivity determination method based on machine learning of double parallel neural networks Download PDFInfo
- Publication number
- CN112541571A CN112541571A CN202011339272.6A CN202011339272A CN112541571A CN 112541571 A CN112541571 A CN 112541571A CN 202011339272 A CN202011339272 A CN 202011339272A CN 112541571 A CN112541571 A CN 112541571A
- Authority
- CN
- China
- Prior art keywords
- production
- connectivity
- injection
- neural network
- double parallel
- 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
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 71
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000010801 machine learning Methods 0.000 title claims abstract description 13
- 238000002347 injection Methods 0.000 claims abstract description 33
- 239000007924 injection Substances 0.000 claims abstract description 33
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 15
- 238000012360 testing method Methods 0.000 claims abstract description 13
- 238000003062 neural network model Methods 0.000 claims abstract description 7
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 7
- 230000009977 dual effect Effects 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 239000013598 vector Substances 0.000 claims description 2
- 238000004891 communication Methods 0.000 abstract description 6
- 238000004364 calculation method Methods 0.000 abstract description 4
- 230000008901 benefit Effects 0.000 abstract description 3
- 230000003068 static effect Effects 0.000 abstract description 2
- 238000011161 development Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 230000035699 permeability Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012407 engineering method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000003129 oil well Substances 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
- 230000005514 two-phase flow Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK 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
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK 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
- E21B43/16—Enhanced recovery methods for obtaining hydrocarbons
- E21B43/20—Displacing by water
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Mining & Mineral Resources (AREA)
- Geology (AREA)
- Economics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Molecular Biology (AREA)
- Geochemistry & Mineralogy (AREA)
- Marketing (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Business, Economics & Management (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- Biomedical Technology (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
Abstract
The invention relates to a method for determining injection-production connectivity based on machine learning of a double parallel neural network, which specifically comprises the following steps: the method comprises the following steps: acquiring input and output data of the double parallel neural network model, and performing data preprocessing, wherein the step two is as follows: constructing a double parallel neural network, and performing model training, wherein the third step is as follows: based on the trained double parallel neural network, performing connectivity analysis, and performing the fourth step: the generalization ability of the model is checked using the test set data. The method has the advantages of high calculation speed, low economic cost, high representation precision, no need of considering geological static parameters and the like. The network can also be used for production prediction, and has important guiding significance for adjusting injection-production relation on site of an oil field, adopting water shutoff and profile control measures, optimizing a well pattern and the like; through historical injection and production data, the evolution process of the injection and production dynamic connectivity and the total communication condition in the production history can be accurately calculated and evaluated, two kinds of connectivity analysis are integrated into a whole, and the practical application requirements can be met.
Description
Technical Field
The invention belongs to the field of petroleum engineering, and particularly relates to an injection-production connectivity determination method based on machine learning of a double parallel neural network.
Background
With the continuous development of oil and gas field development technology, how to utilize lower cost to carry out oil and gas exploitation, the economic benefit maximization is realized, and the method becomes a research field which is more and more emphasized. The method comprises the following steps of performing injection-production relation optimization, water shutoff and profile control and the like, wherein accurate evaluation of connectivity among injection-production wells cannot be performed.
The conventional engineering method for determining injection-production connectivity usually needs to consume a large amount of time cost, has high dependency on the experience of oil field experts, and is difficult to popularize. Therefore, how to utilize the oil field big data and accurately and quickly determine the injection-production connectivity by a machine learning method has important significance for oil-gas field development.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides an injection-production connectivity analysis method based on machine learning of a double parallel neural network. In the training process of the neural network, the importance degree of different variables is reflected through the weight, so that the data of the injection and production wells are corresponding to the data of the injection and production wells, and the injection and production communication relation is determined. Firstly, collecting production data information of each water injection well and each oil production well in an oil field, and then performing normalization processing on original injection and production data to obtain a training and testing data set of a double-parallel neural network; building a double parallel neural network, and using a back propagation algorithm as a learning algorithm to realize the rapid updating optimization of neural network parameters; and quantitatively representing injection-production connectivity based on the trained neural network weight parameters.
In order to achieve the purpose, the invention adopts the following technical scheme:
the injection-production connectivity determination method based on machine learning of the double parallel neural networks specifically comprises the following steps:
the method comprises the following steps: acquiring input and output data of the double parallel neural network model, and performing data preprocessing
Step two: constructing double parallel neural networks and carrying out model training
Step three: performing connectivity analysis based on the trained dual parallel neural network
Step four: the generalization ability of the model is checked using the test set data.
Compared with the prior art, the invention has the following beneficial effects:
1. the injection and production connectivity analysis method based on machine learning of the double parallel neural networks is simple and practical, compared with the traditional connectivity analysis method, the injection and production connectivity analysis method has the advantages of high calculation speed, low economic cost, high representation precision, no need of considering geological static parameters and the like, and meanwhile, the network can be used for production prediction and has important guiding significance for adjusting injection and production relations in an oil field, adopting water shutoff and profile control measures, well pattern optimization and the like;
2. through historical injection and production data, the evolution process of injection and production dynamic connectivity and the total communication condition in production history can be accurately calculated and analyzed, two kinds of connectivity analysis are integrated into a whole, and the actual application requirements can be met;
3. the trained neural network model can calculate and judge the communication condition of the injection and production wells, can be applied to yield prediction, ensures the calculation precision, is simple and efficient in solving method, and overcomes the defects of the prior art. The method provides important guidance for making and adjusting development schemes, adopting technological measures such as injection-production optimization, water shutoff and profile control and the like, greatly saves labor and time cost, and has important guidance significance for quantitative description of residual oil distribution and deployment of oilfield development schemes.
Drawings
FIG. 1 is a flow chart of a method for determining injection-production connectivity based on machine learning of a dual parallel neural network;
FIG. 2 is a schematic view of an injection-production model of an oilfield block to be studied;
FIG. 3 is a block diagram of a generic artificial network model;
FIG. 4 is a block diagram of a built artificial network model;
FIG. 5 is a connectivity evaluation heatmap;
fig. 6 is an injection-production communication diagram drawn according to the injection-production communication coefficient obtained by calculation.
Detailed Description
The average effective thickness of a certain block of stratum to be researched is 4m, the length and the width of each grid are 30m, and the total number of the grids is 31 multiplied by 1 to 961. A five-injection four-extraction model was established as shown in fig. 2. The oil reservoir only simulates oil-water two-phase flow, and the oil wells produce oil at constant pressure. The five water injection wells are respectively named as I1, I2, I3, I4 and I5; the four production wells are named P1, P2, P3 and P4 respectively.
The injection-production connectivity determination method based on machine learning of the double parallel neural networks has a brief flow as shown in fig. 1, and specifically comprises the following steps:
the method comprises the following steps: acquiring input and output data of a double parallel neural network model, and performing data preprocessing, wherein the specific steps are as follows:
step 1.1, collecting production data of each water injection well and oil production well of an oil field to be analyzed, comprising the following steps of: daily injection amount (unit: m) of five water injection wells of I1, I2, I3, I4 and I53D) as input to the dual parallel neural network, denoted X ═ X1,x2,x3,x4,x5}; daily production (unit: m) of four production wells, P1, P2, P3 and P43D), denoted as Y ═ Y1,y2,y3,y4,y5As an ideal output of the network;
step 1.2, normalizing initial injection-production data samples formed by the collected daily injection quantity X and the daily production data Y of the production well to map values between [0 and 1], wherein the conversion function is as follows:
X*＝(X-Xmin)/(X-Xmax) (1)
Y*＝(Y-Ymin)/(Y-Ymax) (2)
in the formula, XminAnd XmaxRespectively representing the minimum value and the maximum value in the input set X; y isminAnd YmaxRespectively representing the minimum value and the maximum value in the output set Y; x*And Y*And normalizing the values of X and Y to serve as a standard sample set for machine learning.
Step two: constructing a double parallel neural network and carrying out model training, wherein the specific steps are as follows:
step 2.1, building a neural network, in this example, a three-layer neural network model composed of 5 input nodes, 10 hidden nodes and 4 output nodes is represented by vectors respectively as follows: [ x ] of1，x2，...xi，...x5]，[h1，h2，...hk，...h10]，[y1，y2，...yj，...y4]；
Step 2.2, according to the input sample set X obtained in the step one*And output sample set Y*And the ratio of 8: 2 dividing the sample set into training sets X* tr,Y* trAnd test set X* te,Y* te；
Step 2.3, actual daily output Y based on production welljAnd neural network prediction value
The method comprises the following steps of establishing an objective function as a loss function of training and learning of the double parallel neural network by using the mean square sum of errors, specifically:
wherein F represents an error of the neural network, and Y representsjThe yield after the normalization is shown as,
representing the predicted value of the neural network, wherein J represents the jth sample point, and J represents the number of the sample points;
step 2.4, updating the gradient of the parameters in the double parallel neural network through the target function in the step 2.3, and calculating the error until the error meets the requirement;
step three: based on the trained double parallel neural network, connectivity analysis is carried out, and the specific steps are as follows:
step 3.1, based on the weight matrix directly connected between the input layer and the output layer of the double parallel neural network, connectivity analysis can be performed, the matrix is a 4 × 5 non-negative matrix W, transposition operation is performed on the matrix to obtain a 5 × 4 non-negative matrix, rows 1, 2, …, i, … and 5 in the matrix represent water injection wells, rows 1, 2, …, j … and 4 represent production wells, and the element W in the matrix represents a production wellijThe corresponding value is the connectivity evaluation value between the ith water injection well and the jth production well, WijThe larger the value of (a), the better the connectivity between the corresponding water injection well and the production well;
and 3.2, drawing a connectivity evaluation heat map and an injection-production connectivity map based on the weight matrix obtained in the step 3.1, wherein the deeper the color block in the graph 5 is, the larger the corresponding connectivity numerical value is, and the stronger the connectivity relation of the corresponding injection-production well is. The thicker the arrow in fig. 5, the larger the corresponding connectivity value, indicating a stronger connectivity between the two wells.
Step four: the generalization ability of the model is checked by using the test set data, and the specific steps are as follows:
based on the network model trained in the step two, the test set data X is utilized* te,Y* teInstead of X* tr,Y* trThe network error is recalculated according to equation (3). If the error of the test set is slightly larger than the error of the training set, the network has better generalization capability; if the error of the test set is far larger than the error of the training set, it is indicated that the network may be over-fitted, and the network training needs to be performed again by returning to the step two.
According to the trained double parallel neural network, the transpose matrix of the weight matrix directly connected with the input layer and the output layer is used as the evaluation index of the injection-production connectivity, and the result is shown in the table I:
TABLE 1 quantitative evaluation of injection-production connectivity
The connectivity analysis result obtained by the double parallel neural networks is shown in fig. 5, wherein the darker the color block is, the larger the corresponding connectivity numerical value is, and the stronger the connectivity relation of the corresponding injection well is. In this example, as can be seen from the permeability field distribution of fig. 2, I1-P1 is located on the primary hyperosmotic channel (1000md), I3-P4 is located on the secondary hyperosmotic channel (500md), and the permeability between the remaining injection and production wells (5md) is much lower than both channels. Therefore, I1-P1 and I3-P4 in FIG. 5 accurately reflect the strong connectivity between two pairs of injection and production wells, and the connectivity analysis results of the rest of the weak connectivity well groups are all small and accord with the permeability field distribution. The situation of the voidage replacement connectivity in this example can be more accurately and intuitively illustrated by fig. 6, wherein the thicker the arrow, the stronger the connectivity. Through the two connectivity analysis result graphs, the dual parallel neural network can accurately invert the connectivity between injection wells and production wells, and has high credibility.
Claims (5)
1. A method for determining injection-production connectivity based on machine learning of a double parallel neural network is characterized by comprising the following steps:
the method comprises the following steps: acquiring input and output data of the double parallel neural network model, and performing data preprocessing
Step two: constructing double parallel neural networks and carrying out model training
Step three: performing connectivity analysis based on the trained dual parallel neural network
Step four: the generalization ability of the model is checked using the test set data.
2. The method of claim 1, wherein: the method comprises the following specific steps:
step 1.1, collecting production data of each water injection well and oil production well of an oil field to be analyzed, comprising the following steps of: daily injection amount (unit: m) of five water injection wells I1, I2 and … IN3D) as input to the dual parallel neural network, denoted X ═ X1,x2,…,xN}; daily production (unit: m) of four production wells of P1, P2, … PM3D), denoted as Y ═ Y1,y2,…,yMAs an ideal output of the network;
step 1.2, normalizing initial injection-production data samples formed by the collected daily injection quantity X and the daily production data Y of the production well to map values between [0 and 1], wherein the conversion function is as follows:
X*＝(X-Xmin)/(X-Xmax) (1)
Y*＝(Y-Ymin)/(Y-Ymax) (2)
in the formula, XminAnd XmaxRespectively representing the minimum value and the maximum value in the input set X; y isminAnd YmaxRespectively representing the minimum value and the maximum value in the output set Y; x*And Y*And normalizing the values of X and Y to serve as a standard sample set for machine learning.
3. The method according to claims 1-2, wherein: the second step comprises the following concrete steps:
step 2.1, build up a neural network, in this example, of NThe three-layer neural network model formed by the input nodes, the Q hidden layer nodes and the M output nodes is respectively expressed by vectors as follows: [ x ] of1,x2,…,xN]，[h1,h2,…,hQ],[y1,y2,…,yM]；
Step 2.2, according to the input sample set X obtained in the step one*And output sample set Y*And the ratio of 8: 2 dividing the sample set into training sets X* tr,Y* trAnd test set X* te,Y* te；
Step 2.3, actual daily output Y based on production welljAnd neural network prediction value
The method comprises the following steps of establishing an objective function as a loss function of training and learning of the double parallel neural network by using the mean square sum of errors, specifically:
wherein F represents an error of the neural network, and Y representsjThe yield after the normalization is shown as,
representing the predicted value of the neural network, wherein J represents the jth sample point, and J represents the number of the sample points;
and 2.4, performing gradient updating on the parameters in the double parallel neural networks through the target function in the step 2.3, and calculating errors until the errors meet the requirements.
4. The method according to claims 1-3, wherein: the second step comprises the following concrete steps:
step 3.1, connectivity analysis can be performed based on a weight matrix directly connected between the input layer and the output layer of the double parallel neural network, the matrix being an MxN non-negative matrix W, a pairTransposing to obtain an NxM non-negative matrix, wherein each row 1, 2, …, i, …, N in the matrix represents a water injection well, each row 1, 2, …, j …, each column M represents a production well, and an element W in the matrixijThe corresponding value is the connectivity evaluation value between the ith water injection well and the jth production well, WijThe larger the value of (a), the better the connectivity between the corresponding water injection well and the production well;
step 3.2, drawing a connectivity evaluation heat map and an injection-production connectivity map based on the weight matrix obtained in the step 3.1, wherein the deeper the color block color is, the larger the corresponding connectivity numerical value is, and the stronger the connectivity relation of the corresponding injection-production well is; the thicker the arrow in the injection-production connectivity graph is, the larger the corresponding connectivity value is, and the stronger the connectivity between the two wells is.
5. The method according to claims 1-4, wherein: the third step is as follows:
based on the network model trained in the step two, the test set data X is utilized* te,Y* teInstead of X* tr,Y* trRecalculating the network error according to equation (3); if the error of the test set is slightly larger than the error of the training set, the network has better generalization capability; if the error of the test set is far larger than the error of the training set, it is indicated that the network may be over-fitted, and the network training needs to be performed again by returning to the step two.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011339272.6A CN112541571A (en) | 2020-11-25 | 2020-11-25 | Injection-production connectivity determination method based on machine learning of double parallel neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011339272.6A CN112541571A (en) | 2020-11-25 | 2020-11-25 | Injection-production connectivity determination method based on machine learning of double parallel neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112541571A true CN112541571A (en) | 2021-03-23 |
Family
ID=75017050
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011339272.6A Pending CN112541571A (en) | 2020-11-25 | 2020-11-25 | Injection-production connectivity determination method based on machine learning of double parallel neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112541571A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065705A (en) * | 2021-04-07 | 2021-07-02 | 中海石油(中国)有限公司 | Offshore thin interbed sandstone reservoir multilayer commingled production encrypted well recoverable reserve prediction method |
| CN113177319A (en) * | 2021-04-30 | 2021-07-27 | 中国石油大学(华东) | Method and system for judging connectivity among wells based on neural network sensitivity analysis |
| CN114562236A (en) * | 2022-02-15 | 2022-05-31 | 中海油能源发展股份有限公司 | Geological engineering integrated lost circulation real-time early warning method based on integrated learning model |
| CN114737928A (en) * | 2022-06-13 | 2022-07-12 | 中煤科工集团西安研究院有限公司 | Nuclear learning-based coalbed methane intelligent drainage and mining method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017188858A1 (en) * | 2016-04-28 | 2017-11-02 | Schlumberger Canada Limited | Reservoir performance system |
| CN110096755A (en) * | 2019-04-08 | 2019-08-06 | 沈阳工业大学 | The online temperature flexible measurement method of solid heat storage furnace high-temperature heating element and system |
| CN111027249A (en) * | 2019-12-10 | 2020-04-17 | 北京科技大学 | Machine learning-based inter-well connectivity evaluation method |
-
2020
- 2020-11-25 CN CN202011339272.6A patent/CN112541571A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017188858A1 (en) * | 2016-04-28 | 2017-11-02 | Schlumberger Canada Limited | Reservoir performance system |
| CN110096755A (en) * | 2019-04-08 | 2019-08-06 | 沈阳工业大学 | The online temperature flexible measurement method of solid heat storage furnace high-temperature heating element and system |
| CN111027249A (en) * | 2019-12-10 | 2020-04-17 | 北京科技大学 | Machine learning-based inter-well connectivity evaluation method |
Non-Patent Citations (3)
| Title |
|---|
| EMRE ARTUN: "Characterizing interwell connectivity in waterflooded reservoirs using data-driven and reduced-physics models: a comparative study", 《NEURAL COMPUT & APPLIC》 * |
| 刘巍 等: "利用卡尔曼滤波和人工神经网络相结合的油藏井间连通性研究", 《油气地质与采收率》 * |
| 朱庆杰 等: "人工神经网络模型在石油资源预测中的应用", 《资源调查与环境》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065705A (en) * | 2021-04-07 | 2021-07-02 | 中海石油(中国)有限公司 | Offshore thin interbed sandstone reservoir multilayer commingled production encrypted well recoverable reserve prediction method |
| CN113065705B (en) * | 2021-04-07 | 2023-12-08 | 中海石油(中国)有限公司 | Multi-layer production encryption well recoverable reserve prediction method for offshore thin interbed sandstone reservoir |
| CN113177319A (en) * | 2021-04-30 | 2021-07-27 | 中国石油大学(华东) | Method and system for judging connectivity among wells based on neural network sensitivity analysis |
| CN114562236A (en) * | 2022-02-15 | 2022-05-31 | 中海油能源发展股份有限公司 | Geological engineering integrated lost circulation real-time early warning method based on integrated learning model |
| CN114737928A (en) * | 2022-06-13 | 2022-07-12 | 中煤科工集团西安研究院有限公司 | Nuclear learning-based coalbed methane intelligent drainage and mining method and system |
| CN114737928B (en) * | 2022-06-13 | 2022-09-06 | 中煤科工集团西安研究院有限公司 | Nuclear learning-based coalbed methane intelligent drainage and mining method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109447532B (en) | Oil reservoir inter-well connectivity determination method based on data driving | |
| CN112541571A (en) | Injection-production connectivity determination method based on machine learning of double parallel neural networks | |
| CN108240208B (en) | Oil field water-flooding classification well group development effect benchmarking method | |
| CN109543828B (en) | Water absorption profile prediction method based on small sample condition | |
| CN111625922B (en) | Large-scale oil reservoir injection-production optimization method based on machine learning agent model | |
| CN104533400B (en) | Method for reconstructing logging curve | |
| CN109214026A (en) | A kind of shale gas horizontal well initial productivity prediction technique | |
| CN108843312B (en) | Oilfield reservoir in-layer heterogeneity integrated evaluating method | |
| CN101533000A (en) | Method for constructing water eutrophication risk analysis model | |
| CN102880905A (en) | Atmospheric overhead oil dry point on-line soft measurement method | |
| CN104899411A (en) | Method and system for establishing reservoir capacity prediction model | |
| CN108595803A (en) | Shale gas well liquid loading pressure prediction method based on recurrent neural network | |
| CN111173507A (en) | High-water-cut oil field residual oil prediction method | |
| CN113297740A (en) | Horizontal well gel profile control parameter optimization method | |
| CN114004100B (en) | Oil reservoir assisted history fitting and optimization simulation method | |
| CN110988997A (en) | Hydrocarbon source rock three-dimensional space distribution quantitative prediction technology based on machine learning | |
| CN114580261A (en) | Recoverable storage prediction model establishing method, recoverable storage prediction method and recoverable storage prediction device | |
| CN111622724A (en) | Method for calculating and optimizing injection allocation amount of layered water injection layer section | |
| CN108843296A (en) | A kind of individual well refracturing effect prediction method based under multifactor impact | |
| CN107355200A (en) | One kind receives micron particles dispersion improving ecology well choosing method | |
| CN112049629A (en) | Fracture-cavity type oil reservoir recovery ratio prediction method based on A-type water drive characteristic curve | |
| Zhang et al. | An automatic history matching method of reservoir numerical simulation based on improved genetic algorithm | |
| CN115270623A (en) | Method for predicting embankment settlement | |
| Liu et al. | Application of Water Injection Profile Recognition Based on Machine Learning Method in F Oilfield | |
| CN115422789B (en) | Prediction method and system for optimizing water drive recovery ratio of fault block oil reservoir in consideration of whole process |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210323 |