CN114575834A - Electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning - Google Patents
Electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning Download PDFInfo
- Publication number
- CN114575834A CN114575834A CN202011392517.1A CN202011392517A CN114575834A CN 114575834 A CN114575834 A CN 114575834A CN 202011392517 A CN202011392517 A CN 202011392517A CN 114575834 A CN114575834 A CN 114575834A
- Authority
- CN
- China
- Prior art keywords
- machine learning
- prediction
- saturation field
- saturation
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000010801 machine learning Methods 0.000 title claims abstract description 30
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 239000012530 fluid Substances 0.000 claims abstract description 12
- 230000035699 permeability Effects 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 9
- 230000015572 biosynthetic process Effects 0.000 claims description 7
- 230000005284 excitation Effects 0.000 claims description 7
- 230000003068 static effect Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 4
- 230000008030 elimination Effects 0.000 claims description 3
- 238000003379 elimination reaction Methods 0.000 claims description 3
- 230000010287 polarization Effects 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 abstract description 6
- 239000003345 natural gas Substances 0.000 abstract description 3
- 239000003209 petroleum derivative Substances 0.000 abstract description 3
- 238000012360 testing method Methods 0.000 abstract description 3
- 238000000605 extraction Methods 0.000 abstract description 2
- 239000002356 single layer Substances 0.000 abstract description 2
- 239000010410 layer Substances 0.000 description 8
- 238000011161 development Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 239000007789 gas Substances 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000012706 support-vector machine Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 208000035126 Facies Diseases 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000003921 oil Substances 0.000 description 1
- 239000004533 oil dispersion Substances 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Classifications
-
- 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
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
-
- 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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mining & Mineral Resources (AREA)
- Geology (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Environmental & Geological Engineering (AREA)
- Mathematical Physics (AREA)
- Fluid Mechanics (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to the technical field of oil extraction engineering in petroleum and natural gas, in particular to an electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning. The method combines a machine learning algorithm with a well-earth electromagnetic method to test the saturation field, introduces reservoir flow correlation parameters based on single-layer saturation field distribution obtained by the well-earth electromagnetic method, predicts longitudinal full-area saturation distribution of the reservoir by using different regression algorithms in machine learning, and finally forms an accurate high water-cut oil reservoir fluid prediction result.
Description
Technical Field
The invention relates to the technical field of oil extraction engineering in petroleum and natural gas, in particular to an electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning.
Background
Petroleum and natural gas are important strategic resources of the country and are important life lines for national economic development. With the expansion of the field of oil and gas exploration, the high-efficiency development of oil reservoirs with high water cut period becomes one of the main battlefields for improving the recovery ratio of each large oil field in China.
The borehole electromagnetic method is used as a non-seismic geophysical technology and is applied to monitoring reservoir fluid parameters in recent years, research shows that the resistivity is sensitive to parameters such as saturation, permeability and the like of an oil-gas reservoir, and the electromagnetic technology has the advantages of economy, effectiveness, large detection range and the like when being used for detecting the boundary and saturation distribution of oil, gas and water. However, for reservoirs with a large number of longitudinal layers, especially under the influence of factors such as construction cost and the like under the current low oil price, the borehole electromagnetic technology usually only monitors the main force target layer position, which also brings difficulty for fine identification of residual oil of the oil reservoir in the high water-cut period and restricts the well position deployment work in the later period of oil reservoir development. Machine learning models are increasingly applied to various industries as a leading-edge technology of a project, and are considered as an effective means for solving the problem of complex index prediction, and the linbertau system analyzes professional applications of technologies from upstream exploration, development, production throughout to downstream operation, sales and artificial intelligence investment and specific requirements for practitioners.
Focusing on the field of petroleum exploration and development, WANG establishes a sample library by using parameters such as a well inclination angle, an azimuth angle, well depth, fluid properties, fracture attributes, a production period and the like, and optimizes a horizontal well fracturing index by applying a neural network algorithm. The Wangzaghui chapter selects 12 rock physical parameters sensitive to reaction of volcanic lithology and pore structure as classification characteristic quantities, and six algorithms of decision tree, support vector machine, logistic regression, AdaBoost-decision tree, AdaBoost-support vector machine and AdaBoost-logistic regression are adopted to identify and classify the volcanic lithology. Jung analyzes the similarity of the data sets through a clustering analysis method, and predicts the production capacity of the shale reservoir by adopting a Random Forest (RF), a gradient boosting tree (GBM) and a Support Vector Machine (SVM) supervised learning model. The kingdom utilizes Principal Component Analysis (PCA) to carry out multidimensional data analysis, utilizes red, green and blue (RGB) fusion technology to establish a sedimentary facies diagram, utilizes fuzzy self-organizing map (FSOM) to carry out unsupervised classification, and generates a seismic facies classification result of a reservoir stratum.
Chinese patent application CN108241785A discloses a method for fine characterization of heterogeneous reservoir saturation field, comprising the following steps: 1) establishing a saturation formula; 2) and (3) substituting the saturation formula obtained in the step 1) by combining the height distribution condition of the oil column on the basis of the porosity model and the permeability model of the oil field, and realizing the construction of a saturation field model by using geological model modeling software. The method is based on the capillary pressure data of the actual core sample of the heterogeneous reservoir, comprehensively considers the characteristics of geological oil reservoirs, deeply analyzes the relationship between the oil saturation and the capillary pressure and the reservoir quality coefficient, firstly establishes a saturation empirical formula, and secondly constructs a saturation field geological model, thereby realizing reasonable, quantitative and fine characterization of the saturation field.
Chinese patent application CN111832227A is a shale gas saturation determination method based on deep learning, which comprises the following steps: acquiring the fracture permeability of a target stratum and a target moment; converting the fracture permeability into equivalent matrix permeability to obtain equivalent permeability data of the target stratum; determining a target equivalent permeability field map of the target formation according to the equivalent permeability data; according to the target equivalent permeability field map and the target time, determining a saturation field map of the target stratum at the target time by using a target deep convolution decoding network; the target deep convolution de-coding network is used for reducing the resolution of a target equivalent permeability field graph by using an encoder to obtain a plurality of characteristic graphs and calculating according to the plurality of characteristic graphs by using a decoder to obtain a saturation field graph at a target moment.
The Chinese invention patent CN107044277B discloses a method for evaluating the repeated fracturing yield-increasing potential of a horizontal well of a low-permeability heterogeneous reservoir, which sequentially comprises the following steps: 1) establishing a fracture expansion model by using Meyer software, and inverting the parameters of the primary fracturing hydraulic fracture; 2) establishing an oil reservoir heterogeneous geological model by using an oil reservoir numerical simulation software Eclipse, implanting the initial fracturing hydraulic fracture parameters into the heterogeneous geological model, and performing production dynamic history fitting to obtain a residual oil saturation field and stratum pressure field distribution; 3) quantitatively evaluating the repeated fracturing yield-increasing potential of the horizontal well according to the distribution of the residual oil saturation field and the formation pressure field, classifying the primary fracturing hydraulic fractures, and providing a targeted repeated fracturing mode.
However, there are few reports on prediction of residual oil saturation field by machine learning, especially on saturation field prediction method capable of reflecting recognition characteristics of electromagnetic method in well.
Disclosure of Invention
The invention mainly aims to provide a machine learning-based electromagnetic method high water-cut oil reservoir saturation field prediction method.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides an electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning, which comprises the following steps: measuring a saturation field by a well electromagnetic method; establishing a machine learning prediction model; screening machine learning prediction indexes; and predicting a saturation field.
Further, the determination method of the saturation field of the well by an electromagnetic method comprises the following steps: two excitation points are deployed at the top and the bottom for excitation, after the collected signals are subjected to denoising, casing influence elimination and other processing, an oil-gas-water interface is determined through the spatial distribution of local polarization imbalance, oil layer resistivity field distribution is obtained through logging and field constraint inversion of seismic data, and finally the distribution of the residual oil saturation field of the reservoir is determined by using an Archie formula. The principle of predicting the reservoir fluid distribution by the borehole electromagnetic method is mainly that a series of currents with different frequencies are emitted by an excitation electrode positioned in a borehole, a receiving electrode is deployed on the ground along a measuring line, the influence caused by reservoir heterogeneity is eliminated through differential processing, the resistivity distribution of oil gas with electrical abnormality is obtained, and the monitoring of the residual oil and water saturation distribution in a work area is realized.
Further, saturation field prediction of the strong regressor is realized through an AdaBoost algorithm.
Further, the specific implementation process of the algorithm is as follows:
(1) firstly, a basis regression algorithm and a training set are given: { (x)1,y1),…,(xN,yN)}
Wherein x1、xN∈X,y1、yNE, Y, and X and Y respectively represent a characteristic data set;
(2) initializing a weight vector of training data:
whereini=1,2…,N;The weight vector of the first base regression is obtained, and N is the Nth characteristic parameter set;
(3) for n basis regressors set by AdaBoost, when m is 1,2, …, n, a weight vector w is used(m)Learning the training data set to obtain the regression error rate of the training data set and the judgment of the base regressor at the sampleThe other results are as follows:
in the formula, emError rate for the previous step based regressor; h ism(xi) The judgment result of the basis regression is taken as the judgment result of the basis regression; a ismIs the weight coefficient of the basis regressor;
updating the weight vector distribution of the sample set:
(4) constructing a final strong regressor:
in the formula, CkThe basis regressor obtained after the k-th training, and H (x) is the final regressor.
Furthermore, in the machine learning prediction index screening process, five parameters of the thickness, the permeability, the porosity, the oil-containing boundary and the dip angle of the stratum are screened as geological factors for controlling the distribution of the residual oil. The main factors influencing the distribution of the saturation field can be divided into geological factors and reservoir factors, for the geological factors, reservoir lithofacies are complex, physical heterogeneity is strong, thickness change is large, the main reasons for residual oil dispersion are all the factors, and meanwhile, a micro-structural zone of a reservoir is also an important factor for controlling the enrichment of residual oil, so five parameters of stratum thickness, permeability, porosity, oil-containing boundary and stratum inclination angle are screened as the geological factors for controlling the distribution of the residual oil.
Further, in the screening process of the machine learning prediction index, the accumulated water passing amount, the instantaneous water passing amount and the fluid flow days are used as dynamic indexes for measuring the saturation degree of the residual oil. From the development dynamic perspective, the water passing amount of different areas determines the degree of the reservoir stratum washed by water injection; meanwhile, the accumulated time of the fluids participating in flowing in different areas in the whole development process is also an important index for reflecting the change of the saturation, the fluids in part of dead oil areas do not participate in flowing, the oil saturation in the areas is still the original oil saturation, and the accumulated water passing amount, the instantaneous water passing amount and the days of fluid flowing are taken as dynamic indexes for measuring the residual oil saturation after comprehensive consideration.
Further, in the process of establishing the feature data set, 8 indexes including 5 static parameters and 3 dynamic parameters are screened to establish a saturation field prediction feature data set.
Furthermore, in the saturation field prediction process, an index prediction data set is constructed by using dynamic and static index data points, a prediction model is established by using an AdaBoost machine learning algorithm, the fitting degree of the machine learning model is evaluated, and the model hyper-parameters are optimized by using grid search, so that an optimized model and a prediction result are finally obtained.
Compared with the prior art, the invention has the following advantages:
the method combines a machine learning algorithm with a well-earth electromagnetic method to test the saturation field, introduces reservoir flow correlation parameters based on single-layer saturation field distribution obtained by the well-earth electromagnetic method, predicts longitudinal full-area saturation distribution of the reservoir by using different regression algorithms in machine learning, and has more accurate prediction result on the finally formed high water-cut oil reservoir fluid. The method can accurately, quickly and effectively quantitatively predict the distribution of the saturation field.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a plot of residual oil saturation from a borehole electromagnetic test in accordance with an embodiment of the present invention;
fig. 2 is a static index data point distribution diagram according to an embodiment of the invention: a is the permeability data point distribution, b is the porosity data point distribution, c is the oil-containing boundary data point distribution, d is the formation dip angle data point distribution, and e is the formation thickness data point distribution.
Fig. 3 is a distribution diagram of dynamic index data points according to an embodiment of the invention: a is the cumulative water content data point distribution, b is the instantaneous water content data point distribution, and c is the fluid flow days data point distribution.
Fig. 4 is a diagram of a small layer saturation prediction model of Ng2-3-1 according to an embodiment of the present invention.
Detailed Description
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of the stated features, steps, operations, and/or combinations thereof, unless the context clearly indicates otherwise.
In order to make the technical solutions of the present invention more clearly understood by those skilled in the art, the technical solutions of the present invention will be described in detail below with reference to specific embodiments.
Example 1
The method for predicting the saturation field of the oil reservoir in the high water cut stage by using an electromagnetic method based on machine learning comprises the following steps:
Step 2, establishing a machine learning prediction model:
and realizing saturation field prediction of the strong regressor by using an AdaBoost algorithm. The specific implementation process of the algorithm is as follows:
(1) firstly, a basis regression algorithm and a training set are given: { (x)1,y1),…,(xN,yN)}
Wherein x1、xN∈X,y1、yNAnd E is Y, and X and Y respectively represent the characteristic data set.
(2) Initializing a weight vector of training data:
whereini=1,2…,N;The weight vector of the first basis regressor is N, and N is the Nth characteristic parameter set.
(3) For n basis regressors set by AdaBoost, when m is 1,2, …, n, a weight vector w is used(m)Learning the training data set to obtain the regression error rate of the training data set and the discrimination result of the base regressor at the sample:
in the formula, emError rate for the previous step based regressor; h ism(xi) The judgment result of the basis regression is taken as the judgment result of the basis regression; a ismIs the weight coefficient of the basis regressor;
updating the weight vector distribution of the sample set:
(4) constructing a final strong regressor:
in the formula, CkThe basis regressor obtained after the k-th training, and H (x) is the final regressor.
And 3, screening machine learning prediction indexes: screening five parameters of the thickness, the permeability, the porosity, the oil-containing boundary and the formation inclination angle of the stratum as geological factors for controlling the distribution of the residual oil; the accumulated water excess, the instantaneous water excess and the fluid flow days are used as dynamic indexes for measuring the saturation of the residual oil. In the process of establishing the characteristic data set, 8 indexes of the 5 static parameters and the 3 dynamic parameters are screened to establish a saturation field prediction characteristic data set.
And 4, predicting a saturation field: the method comprises the steps of constructing an index prediction data set by using dynamic and static index data points, establishing a prediction model by using the mentioned AdaBoost machine learning algorithm, evaluating the fitting degree of the machine learning model, and optimizing the super-parameters of the model by using grid search to finally obtain an optimized model and a prediction result.
Example 2
The method described in example 1 was used to predict the saturation field of a real block reservoir. The panel Nm3-4-2 small layer saturation field detection plot is obtained by the borehole electromagnetic method, as shown in FIG. 1. Through data screening, 8-parameter prediction index data sets are established, as shown in fig. 2 and 3. The 8 parameters are predicted through an AdaBoost machine learning model, a data relation between the 8 parameters and a Nm3-4-2 small-layer electromagnetic method detection saturation degree model is established, and a prediction result of the Ng2-3-1 small-layer saturation degree is formed, and is shown in figure 4.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (8)
1. The method for predicting the saturation field of the oil reservoir in the high water cut period by the electromagnetic method based on machine learning is characterized by comprising the following steps of: measuring a saturation field by a well electromagnetic method; establishing a machine learning prediction model; screening machine learning prediction indexes; and predicting a saturation field.
2. The method of claim 1, wherein the determination of the saturation field is determined by the electromagnetic method of the well: two excitation points are deployed at the top and the bottom for excitation, after the collected signals are subjected to denoising, casing influence elimination and other processing, an oil-gas-water interface is determined through the spatial distribution of local polarization imbalance, oil layer resistivity field distribution is obtained through logging and field constraint inversion of seismic data, and finally the distribution of the residual oil saturation field of the reservoir is determined by using an Archie formula.
3. The method of claim 1, wherein the prediction of the saturation field of the strong regressor is performed by an AdaBoost algorithm.
4. The method according to claim 3, wherein the algorithm is implemented by:
(3) firstly, a basis regression algorithm and a training set are given: { (x)1,y1),…,(xN,yN) Where x1、xN∈X,y1、yNE, Y, and X and Y respectively represent a characteristic data set;
(4) initializing a weight vector of training data:
wherein The weight vector of the first base regression is obtained, and N is the Nth characteristic parameter set;
(3) for n basis regressors set by AdaBoost, when m is 1,2, …, n, a weight vector w is used(m)Learning the training data set to obtain the regression error rate of the training data set and the discrimination result of the base regressor at the sample:
in the formula, emError rate for the previous step based regressor; h is a total ofm(xi) The judgment result of the basis regression is taken as the judgment result of the basis regression; a ismIs the weight coefficient of the basis regressor;
updating the weight vector distribution of the sample set:
(4) constructing a final strong regressor:
in the formula, CkFor the basis regressor obtained after the kth training,h (x) is the final regressor.
5. The method of claim 1, wherein five parameters of formation thickness, permeability, porosity, oil-bearing boundary and formation dip are screened as geological factors for controlling the distribution of remaining oil during the machine-learned predictor screening process.
6. The method of claim 1, wherein the cumulative water excess, instantaneous water excess, and fluid flow days are used as dynamic metrics to measure remaining oil saturation during the machine learning prediction index screening process.
7. The method according to claim 5 or 6, wherein in the process of establishing the feature data set, 8 indexes including 5 static parameters and 3 dynamic parameters are screened to establish a saturation field prediction feature data set.
8. The method as claimed in claim 1, wherein in the saturation field prediction process, dynamic and static index data points are used to construct an index prediction data set, an AdaBoost machine learning algorithm is used to establish a prediction model, the fitness of the machine learning model is evaluated, and grid search is used to optimize the hyper-parameters of the model, so as to finally obtain an optimized model and a prediction result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011392517.1A CN114575834A (en) | 2020-12-01 | 2020-12-01 | Electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011392517.1A CN114575834A (en) | 2020-12-01 | 2020-12-01 | Electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114575834A true CN114575834A (en) | 2022-06-03 |
Family
ID=81767069
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011392517.1A Pending CN114575834A (en) | 2020-12-01 | 2020-12-01 | Electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114575834A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117005846A (en) * | 2023-07-25 | 2023-11-07 | 成都理工大学 | Resistivity logging invasion correction method and system under tight sandstone oil and gas reservoir condition |
-
2020
- 2020-12-01 CN CN202011392517.1A patent/CN114575834A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117005846A (en) * | 2023-07-25 | 2023-11-07 | 成都理工大学 | Resistivity logging invasion correction method and system under tight sandstone oil and gas reservoir condition |
CN117005846B (en) * | 2023-07-25 | 2024-02-23 | 成都理工大学 | Resistivity logging invasion correction method and system under tight sandstone oil and gas reservoir condition |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Aminian et al. | Application of artificial neural networks for reservoir characterization with limited data | |
Lin et al. | Automatic lithology identification by applying LSTM to logging data: A case study in X tight rock reservoirs | |
Wang et al. | Marcellus shale lithofacies prediction by multiclass neural network classification in the Appalachian Basin | |
Zhou* et al. | Evaluating gas production performances in marcellus using data mining technologies | |
CN111783825A (en) | Well logging lithology identification method based on convolutional neural network learning | |
WO2018152147A1 (en) | Well performance classification using artificial intelligence and pattern recognition | |
CN111596978A (en) | Web page display method, module and system for lithofacies classification by artificial intelligence | |
CN112684497B (en) | Seismic waveform clustering method and device | |
EP2271954A2 (en) | System and method for interpretation of well data | |
CN105572747A (en) | Method for identifying lithology of high-salt potassium-rich region water-flooding oil reservoir clastic rocks | |
CN104570109B (en) | A kind of method of reservoir hydrocarbons prediction | |
CN110412662A (en) | Method for prediction of reservoirs of thin interbeded based on seismic multi-attribute deep learning | |
CN116127675A (en) | Prediction method for maximum recoverable reserve of shale oil horizontal well volume fracturing | |
CN114529110A (en) | Lithofacies inversion method and system based on deep neural network model | |
Lu et al. | Lithology identification using graph neural network in continental shale oil reservoirs: A case study in Mahu Sag, Junggar Basin, Western China | |
Hatampour et al. | Hydraulic flow units' estimation from seismic data using artificial intelligence systems, an example from a gas reservoir in the Persian Gulf | |
CN113033648A (en) | Method for realizing logging interpretation by using machine learning algorithm | |
CN110927814A (en) | Crack prediction method based on lithofacies configuration | |
CN116168224A (en) | Machine learning lithology automatic identification method based on imaging gravel content | |
CN114575834A (en) | Electromagnetic method high water cut oil reservoir saturation field prediction method based on machine learning | |
Lu et al. | Identifying flow units by FA-assisted SSOM—An example from the Eocene basin-floor-fan turbidite reservoirs in the Daluhu Oilfield, Dongying Depression, Bohai Bay Basin, China | |
Djebbas et al. | Integrating hydraulic flow unit concept and adaptive neuro-fuzzy inference system to accurately estimate permeability in heterogeneous reservoirs: Case study Sif Fatima oilfield, southern Algeria | |
Sarkheil et al. | The fracture network modeling in naturally fractured reservoirs using artificial neural network based on image loges and core measurements | |
Liu et al. | Identification of tight sandstone reservoir lithofacies based on CNN image recognition technology: A case study of Fuyu reservoir of Sanzhao Sag in Songliao Basin | |
CN103048688A (en) | Seismic attribute optimal selection method based on three-step method |
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 |