CN110118994B - Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning - Google Patents
Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning Download PDFInfo
- Publication number
- CN110118994B CN110118994B CN201910440723.6A CN201910440723A CN110118994B CN 110118994 B CN110118994 B CN 110118994B CN 201910440723 A CN201910440723 A CN 201910440723A CN 110118994 B CN110118994 B CN 110118994B
- Authority
- CN
- China
- Prior art keywords
- machine learning
- organic matter
- matter content
- lithology
- learning network
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/40—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging
- G01V1/44—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging using generators and receivers in the same well
- G01V1/48—Processing data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/40—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging
- G01V1/44—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging using generators and receivers in the same well
- G01V1/48—Processing data
- G01V1/50—Analysing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention relates to a land-phase hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning, which is used for predicting the spatial distribution and organic matter content of land-phase hydrocarbon source rocks in a certain area. Compared with the prior art, the method has high prediction accuracy.
Description
Technical Field
The invention relates to a hydrocarbon source rock prediction method, in particular to a continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning.
Background
The existing earthquake evaluation technology of the hydrocarbon source rock is mainly based on qualitative analysis of earthquake facies or direct conversion relation between the impedance of post-stack waves and the content of organic matters, the technologies are difficult to be used for evaluating the hydrocarbon source rock with strong land heterogeneity, on one hand, the hydrocarbon source rock facies under the land deposition environment are narrow, the thickness change of the hydrocarbon source rock is large, the earthquake facies analysis is difficult to accurately describe the hydrocarbon source rock, on the other hand, because the mapping relation of earthquake elasticity-lithology-organic matter content under the land deposition environment is very fuzzy, a very complex nonlinear mapping relation is formed between the elasticity parameters or the earthquake attributes and the evaluation parameters (organic matter content) of the hydrocarbon source rock, and the physical characteristics of the earthquake rock are difficult to be comprehensively and accurately described by utilizing simple linear model driving.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a method for quantitatively predicting a continental-phase source rock based on seismic inversion and machine learning.
The purpose of the invention can be realized by the following technical scheme:
a continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning is used for predicting the spatial distribution and organic matter content of continental facies hydrocarbon source rocks in a certain area and comprises a training stage and a prediction stage, wherein,
the training phase comprises the following steps:
a1, preferably elastic properties sensitive to lithologic differentiation and organic matter content of the land sedimentary formation sand mudstone;
a2, training a first machine learning network for predicting lithology, wherein the input of the first machine learning network is the elastic property of a continental facies sedimentary stratum, and the output of the first machine learning network is lithology, and the lithology comprises mudstone and sandstone;
a3, training a second machine learning network for predicting the organic matter content, wherein the input of the second machine learning network is the elastic attribute corresponding to the mudstone layer, and the output is the organic matter content;
the prediction phase comprises:
b1, performing prestack elastic parameter inversion on the prestack seismic data of the area to be predicted to obtain the elastic property corresponding to the step A1;
b2, predicting lithology by adopting a first machine learning network, and obtaining spatial distribution of a mudstone layer;
and B3, predicting the organic matter content of the mudstone layer by adopting a second machine learning network, wherein the spatial distribution and the organic matter content of the mudstone layer are the spatial distribution and the organic matter content of the continental facies hydrocarbon source rock.
The training phase step A1 is preceded by obtaining training data, wherein the training data comprises logging data, drilling or logging lithology calibration data, and organic matter content data of geochemical test.
The elastic properties comprise any one or more of longitudinal wave velocity, transverse wave velocity, density, longitudinal wave impedance and transverse wave impedance.
Step A1 quantitatively evaluates the correlation between elastic property and lithology and between elastic property and organic matter content, and when the correlation coefficient is greater than 0.5, selects the corresponding elastic property as the sensitive elastic property.
The first machine learning network is a machine learning network based on a random forest algorithm.
The second machine learning network is a machine learning network based on a random forest algorithm.
And in the training stage, corresponding first machine learning networks are respectively trained for different depth sections of sedimentary strata, and then in the prediction stage, the lithology is predicted by adopting the corresponding first machine learning networks for different depth sections of sedimentary strata.
And in the training stage, corresponding second machine learning networks are respectively trained for the mudstone layers of different depth sections, and then the organic matter content is predicted by adopting the corresponding second machine learning networks for the mudstone layers of different depth sections in the prediction stage.
Compared with the prior art, the invention has the following advantages:
(1) according to the method, the sensitive elastic property for predicting the lithology and the organic matter content is preferably selected through a seismic rock physical response mechanism of determining the lithology-organic matter content-elastic property of the continental facies sedimentary stratum, and the reliability of quantitative prediction of the continental facies hydrocarbon source rock is enhanced;
(2) according to the method, a two-step strategy of firstly predicting lithology and then predicting organic matter content is adopted, and a mudstone stratum is firstly screened out (the mudstone stratum is likely to be a hydrocarbon source rock), so that the uncertainty of quantitative prediction of the hydrocarbon source rock of the continental facies sedimentary basin is reduced;
(3) aiming at the characteristic of fuzzy mapping relation of lithology-organic matter content-elastic property of a continental facies sedimentary stratum, the nonlinear mapping relation of the elastic property, the organic matter content and the lithology is effectively represented by using a random forest algorithm, so that the accuracy of quantitative prediction of the hydrocarbon source rock is improved;
(4) the invention adopts the idea of subsection training and subsection prediction for sedimentary strata at different depth sections, can effectively remove the influence of compaction effect on lithology-elastic property and organic matter content-elastic property, and further improves the prediction accuracy.
Drawings
FIG. 1 is a flow chart of a continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning according to the invention;
FIG. 2 is the well logging data of the A well of the east China sea continental facies sedimentary basin in the embodiment;
FIG. 3 is a graph showing the effect of compaction on the lithological elastic characteristics of sand mudstone at different depth intervals in well A;
FIG. 4 is a first machine learning network for establishing longitudinal wave velocity and density and sandstone lithology based on a random forest algorithm, and adopting a segmented training and segmented prediction method to predict the lithology of the sandstone of the well A, wherein the left graph is the real lithology calibrated by a drilling core, and the right graph is the lithology predicted by machine learning;
FIG. 5 is a second machine learning network based on random forest algorithm for establishing longitudinal wave impedance and organic matter content, and adopting a segmented training and segmented prediction method to predict the organic matter content of the well A, wherein the upper segment (a) represents a stratum with lower organic matter content of the well A, the lower segment (b) represents a stratum with higher organic matter content of the well A, the black line is the organic matter content calibrated by geochemical data, and the gray line is the organic matter content predicted by machine learning;
FIG. 6 is a pre-stack elastic parameter inversion result based on pre-stack seismic data for well A, (a) is a longitudinal velocity inversion result, and (b) is a density inversion result;
FIG. 7 is a sand shale lithology (white sandstone) prediction based on a random forest algorithm;
fig. 8 is the organic matter content prediction based on a random forest algorithm.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. Note that the following description of the embodiments is merely a substantial example, and the present invention is not intended to be limited to the application or the use thereof, and is not limited to the following embodiments.
Examples
As shown in fig. 1, a method for quantitative prediction of continental facies source rocks based on seismic inversion and machine learning is used for predicting the spatial distribution and organic matter content of the continental facies source rocks in a certain area, and the method comprises a training phase and a prediction phase, wherein,
the training phase comprises the following steps:
a1, preferably elastic properties sensitive to lithologic differentiation and organic matter content of the land sedimentary formation sand mudstone;
a2, training a first machine learning network for predicting lithology, wherein the input of the first machine learning network is the elastic property of a continental facies sedimentary stratum, and the output of the first machine learning network is lithology, and the lithology comprises mudstone and sandstone;
a3, training a second machine learning network for predicting the organic matter content, wherein the input of the second machine learning network is the elastic attribute corresponding to the mudstone layer, and the output is the organic matter content;
the prediction phase comprises:
b1, performing prestack elastic parameter inversion on the prestack seismic data of the area to be predicted to obtain the elastic property corresponding to the step A1;
b2, predicting lithology by adopting a first machine learning network, and obtaining spatial distribution of a mudstone layer;
and B3, predicting the organic matter content of the mudstone layer by adopting a second machine learning network, wherein the spatial distribution and the organic matter content of the mudstone layer are the spatial distribution and the organic matter content of the continental facies hydrocarbon source rock.
The training phase step A1 is preceded by obtaining training data, wherein the training data comprises logging data, drilling or logging lithology calibration data, and organic matter content data of geochemical test.
Step A1 is to clarify the seismic and rock physical response mechanism of sandstone and mudstone differentiation and hydrocarbon source rock quality evaluation (organic matter content) in the exploration area from the lithological calibration, geochemical test data and logging data of the drilling or logging, preferably the elastic property sensitive to sandstone and mudstone lithological differentiation and organic matter content of the land-facies sedimentary stratum, wherein the elastic property comprises any one or more of longitudinal wave velocity, transverse wave velocity, density, longitudinal wave impedance and transverse wave impedance. In this embodiment, the elastic property and lithology and the correlation between the elastic property and organic matter content are quantitatively evaluated, and when the correlation coefficient is greater than 0.5, the corresponding elastic property is selected as the sensitive elastic property, where the quantitative evaluation may be analyzed and evaluated by a large amount of existing data. In the embodiment, the longitudinal wave velocity and the density are preferably sensitive elastic properties distinguished by lithological characters of the continental facies sedimentary basin, and the longitudinal wave impedance is preferably sensitive elastic properties predicted by organic matter content.
The first machine learning network is a machine learning network based on a random forest algorithm, and specifically, the machine learning network based on the random forest algorithm has the following specific training flow:
(1) based on lithology-elasticity characteristic data of well logging, resampling in a place-to-place manner by using a Bootstrap method, and randomly generating T training sets S1,S2,…,ST。
Let the set S contain n different samples x1,x2,…,xnGet back every timeExtracting a sample from the set S for a total of n times to form a new set S, wherein the set S does not contain a sample xiThe probability of (i ═ 1,2, …, n) is:
when n → ∞ there are:
therefore, although the total number of samples in the new set S is equal to the total number of samples in the original set S, the new set may include duplicate samples, and the new set S includes only about 1-0.368 × 100% or 63.2% of the samples in the original set S, excluding the duplicate samples.
(2) Using each training set, a corresponding decision tree C is generated1,C2,…,CT(ii) a Before selecting attributes on each non-leaf node, randomly draw M (0) from a total of M elastic attributes<m<M) attributes are used as a splitting attribute set of the current node, and the node is split in the optimal splitting mode in the M attributes. The splitting criterion is the impurity degree, and how to split is determined by comparing the impurity degree values before and after splitting, and the more the impurity degree is reduced after splitting, the better the classifying effect is. The information gain, information gain ratio or kini coefficient is usually selected to quantify the variation in the purities, and different selection methods form different decision tree methods (including ID3, C4,5, CART).
(3) Each tree was allowed to grow completely without pruning.
(4) For the test set sample X, each decision tree is used for testing to obtain a corresponding category C1(X),C2(X),…,CT(X)。
(5) And adopting a voting method to take the category with the most output in the T decision trees as the lithology of the test set sample X.
Similarly, for the logging data of the organic matter content of the mudstone part calibrated by the geochemistry test, a second machine learning network for representing the evaluation parameters (organic matter content) of the source rock and the corresponding elastic characteristics is obtained by using a related machine learning algorithm (random forest algorithm), namely the second machine learning network is also a machine learning network based on the random forest algorithm. The concrete training process of the second machine learning network is substantially consistent with the process of the first machine learning network, and only the lithology prediction obtained by the voting method in the step (5) is converted into the organic matter content prediction obtained by an averaging method.
And removing the influence of the compaction effect on the lithology-elasticity characteristics, respectively training corresponding first machine learning networks on sedimentary strata of different depth sections in the training stage, and then predicting the lithology of the sedimentary strata of different depth sections in the prediction stage by adopting the corresponding first machine learning networks. And simultaneously, the training stage respectively trains corresponding second machine learning networks for the mudstone layers of different depth sections, and then the prediction stage predicts the organic matter content by adopting the corresponding second machine learning networks for the mudstone layers of different depth sections.
Fig. 2 is well logging data of a well a in a land sedimentary basin in the east sea of this embodiment, where the first column is sandstone and sandstone according to the lithology of a drill core, the second column from left to right is mudstone, the second column from left to right is organic matter content predicted according to the logR method, and a dot is a result of organic matter content calibration measured according to the geochemistry method (Rock-Eval). The calibrated data provides sample data for machine learning of the sandstone lithology-elasticity characteristic and the organic matter content-elasticity characteristic.
FIG. 3 is the elastic characteristics of sand shale of sedimentary formations of different depth sections of the well A, and sensitive elastic properties of longitudinal wave velocity and density which are differentiated for the lithology of the land phase sedimentary basin can be preferably selected. Meanwhile, by adopting the idea of subsection training and subsection prediction for sedimentary strata of different depth sections, the influence of compaction effect on lithology-elasticity characteristics can be effectively removed.
FIG. 4 is a machine learning network 1 for establishing two elastic parameters of longitudinal wave velocity and density and sandstone lithology based on a random forest algorithm, and a lithology prediction result of the sandstone of the well A is obtained by adopting a segmented training and segmented prediction method. Wherein 50% of data are randomly selected for training each section of stratum, 50% of data are detected, and the lithology prediction accuracy rate reaches 93.5%. The training network also lays a foundation for subsequent sand shale space distribution prediction by utilizing the prestack elastic parameter inversion result.
FIG. 5 is a machine learning network 2 for establishing longitudinal wave impedance and organic matter content based on a random forest algorithm, and adopting a segmented training and segmented prediction method to predict the organic matter content of the well A, wherein the upper segment (a) represents a stratum with lower organic matter content in the well A, and the lower segment (b) represents a stratum with higher organic matter content in the well A. The black line is the organic content calibrated by the geochemical data, and the gray line is the organic content predicted by machine learning. And the upper-section stratum and the lower-end stratum are trained by selecting 50% of data and monitored by 50% of data. The organic matter content predicted by the machine learning network has a good matching relation with the organic matter content calculated and calibrated by logging data, and the machine learning network can effectively depict the spatial change of the organic matter content in the longitudinal direction.
FIG. 6 shows the inversion results of the prestack elastic parameters based on the prestack seismic data from well A, where the upper graph is the inversion result of the longitudinal velocity and the lower graph is the inversion result of the density.
FIG. 7 shows the result of the sand shale lithology spatial distribution prediction obtained by combining the machine learning network 1 and the pre-stack seismic inversion result. The method can be seen in the good corresponding relation between the sand shale seismic lithology prediction result of the well section A and the actual sand shale lithology distribution calibrated by the drilling data, namely, the stratum of the Yangtze river group takes sandstone distribution as the main part, and the mudstone distribution of the MFR-1 section and the MRF-2 section is most concentrated. Meanwhile, the lithological distribution of the sand shale predicted by the earthquake is more consistent with the sedimentary facies result of the land facies sedimentary basin, namely, the sand shale mutual layer shallow lake facies/half-deep lake facies of the MRF-6 section, the MRF-5 section and the MRF-4 section are gradually transited into the half-deep lake/deep lake facies which mainly comprise the mudstones of the MRF-3 section, the MRF-2 section and the MRF-1 section.
Fig. 8 is the result of organic matter content spatial distribution prediction obtained by combining the machine learning network 2 and the pre-stack seismic inversion result. It can be seen that the organic matter content prediction result of the well section A has a better corresponding relation with the organic matter content calculated by the geochemical calibration through the logging data: the organic matter content is gradually reduced from the MRF-6 section and the MRF-5 section to the MRF-4 section, and is increased in the MRF-3 section, the MRF-2 section and the MRF-1 section, and the organic matter content can be well depicted on earthquakes at local high points on a logging curve.
The above embodiments are merely examples and do not limit the scope of the present invention. These embodiments may be implemented in other various manners, and various omissions, substitutions, and changes may be made without departing from the technical spirit of the present invention.
Claims (6)
1. A continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning is used for predicting the spatial distribution and organic matter content of continental facies hydrocarbon source rocks in a certain area and is characterized by comprising a training stage and a prediction stage, wherein,
the training phase comprises the following steps:
a1, preferably elastic properties sensitive to lithologic differentiation and organic matter content of the land sedimentary formation sand mudstone;
a2, training a first machine learning network for predicting lithology, wherein the input of the first machine learning network is the elastic property of a continental facies sedimentary stratum, and the output of the first machine learning network is lithology, and the lithology comprises mudstone and sandstone;
a3, training a second machine learning network for predicting the organic matter content, wherein the input of the second machine learning network is the elastic attribute corresponding to the mudstone layer, and the output is the organic matter content;
the prediction phase comprises:
b1, performing prestack elastic parameter inversion on the prestack seismic data of the area to be predicted to obtain the elastic property corresponding to the step A1;
b2, predicting lithology by adopting a first machine learning network, and obtaining spatial distribution of a mudstone layer;
b3, predicting the organic matter content of the mudstone layer by adopting a second machine learning network, wherein the spatial distribution and the organic matter content of the mudstone layer are the spatial distribution and the organic matter content of the continental facies hydrocarbon source rock;
respectively training corresponding first machine learning networks for different depth sections of sedimentary strata in the training stage, and then predicting lithology of the sedimentary strata in the predicting stage by adopting the corresponding first machine learning networks;
and in the training stage, corresponding second machine learning networks are respectively trained for the mudstone layers of different depth sections, and then the organic matter content is predicted by adopting the corresponding second machine learning networks for the mudstone layers of different depth sections in the prediction stage.
2. The method for quantitative prediction of continental facies source rocks based on seismic inversion and machine learning of claim 1, wherein the training phase step a1 is preceded by the acquisition of training data comprising logging data, drilling or logging lithology calibration data, and geochemical test organic matter content data.
3. The method of claim 1, wherein the elastic properties comprise any one or more of compressional velocity, shear velocity, density, compressional impedance and shear impedance.
4. The method for quantitative prediction of the continental facies source rock based on seismic inversion and machine learning as claimed in claim 1 or 3, wherein step A1 is specifically: quantitatively evaluating the correlation between the elastic property and lithology and between the elastic property and organic matter content, and selecting the corresponding elastic property as the sensitive elastic property when the correlation coefficient is greater than 0.5.
5. The method for quantitative prediction of the continental facies source rock based on seismic inversion and machine learning as claimed in claim 1, wherein the first machine learning network is a machine learning network based on a random forest algorithm.
6. The method for quantitative prediction of continental facies hydrocarbon source rocks based on seismic inversion and machine learning as claimed in claim 1, wherein the second machine learning network is a machine learning network based on a random forest algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910440723.6A CN110118994B (en) | 2019-05-24 | 2019-05-24 | Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910440723.6A CN110118994B (en) | 2019-05-24 | 2019-05-24 | Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110118994A CN110118994A (en) | 2019-08-13 |
| CN110118994B true CN110118994B (en) | 2020-10-02 |
Family
ID=67523176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910440723.6A Active CN110118994B (en) | 2019-05-24 | 2019-05-24 | Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110118994B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110988997A (en) * | 2019-12-27 | 2020-04-10 | 中国海洋石油集团有限公司 | Hydrocarbon source rock three-dimensional space distribution quantitative prediction technology based on machine learning |
| US11719851B2 (en) | 2020-09-02 | 2023-08-08 | Saudi Arabian Oil Company | Method and system for predicting formation top depths |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105807320A (en) * | 2016-03-17 | 2016-07-27 | 成都创源油气技术开发有限公司 | Shale gas reservoir seismic prediction method |
| CN107437229A (en) * | 2016-05-26 | 2017-12-05 | 中国石油化工股份有限公司 | The construction method and Forecasting Methodology of content of organic carbon of hydrocarbon source rock forecast model |
| WO2018137747A1 (en) * | 2017-01-30 | 2018-08-02 | Entreprise Nationale De Geophysique | Method for estimating total organic carbon (cot) using acoustic velocities and total porosity (φt) in a bedrock |
| CN109342697A (en) * | 2018-12-07 | 2019-02-15 | 福建农林大学 | Based on random forest-normal stabilizing pile soil organic carbon prediction technique |
| CN109507733A (en) * | 2018-12-11 | 2019-03-22 | 中国石油天然气股份有限公司大港油田分公司 | A kind of method and device for predicting organic matter abundance in hydrocarbon source rock |
| CN109709608A (en) * | 2018-12-07 | 2019-05-03 | 中国石油天然气股份有限公司 | The mixed product hydrocarbon source rock TOC content of one kind and lithology compound mensuration method and application |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7349806B2 (en) * | 2004-09-15 | 2008-03-25 | United States Of America As Represented By The Secretary Of The Navy | System and method for extracting optical properties from environmental parameters in water |
| CN1239921C (en) * | 2004-10-22 | 2006-02-01 | 石油大学(北京) | Method for predicting collapse pressure and bursting pressure for borehole to be drilled section using earthquake record |
| US10353093B2 (en) * | 2017-07-27 | 2019-07-16 | International Business Machines Corporation | Multi-scale manifold learning for full waveform inversion |
| CN108897042B (en) * | 2018-08-28 | 2020-03-10 | 中国石油天然气股份有限公司 | Organic matter content earthquake prediction method and device |
| CN109765609A (en) * | 2019-01-30 | 2019-05-17 | 中国石油大学(北京) | A kind of Sand-body Prediction method and system based on target zone Yu adjacent layer seismic properties |
-
2019
- 2019-05-24 CN CN201910440723.6A patent/CN110118994B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105807320A (en) * | 2016-03-17 | 2016-07-27 | 成都创源油气技术开发有限公司 | Shale gas reservoir seismic prediction method |
| CN107437229A (en) * | 2016-05-26 | 2017-12-05 | 中国石油化工股份有限公司 | The construction method and Forecasting Methodology of content of organic carbon of hydrocarbon source rock forecast model |
| WO2018137747A1 (en) * | 2017-01-30 | 2018-08-02 | Entreprise Nationale De Geophysique | Method for estimating total organic carbon (cot) using acoustic velocities and total porosity (φt) in a bedrock |
| CN109342697A (en) * | 2018-12-07 | 2019-02-15 | 福建农林大学 | Based on random forest-normal stabilizing pile soil organic carbon prediction technique |
| CN109709608A (en) * | 2018-12-07 | 2019-05-03 | 中国石油天然气股份有限公司 | The mixed product hydrocarbon source rock TOC content of one kind and lithology compound mensuration method and application |
| CN109507733A (en) * | 2018-12-11 | 2019-03-22 | 中国石油天然气股份有限公司大港油田分公司 | A kind of method and device for predicting organic matter abundance in hydrocarbon source rock |
Non-Patent Citations (2)
| Title |
|---|
| 利用地震资料预测和评价烃源岩;顾礼敬 等;《天然气地球科学》;20110630;第22卷(第3期);第554-560页 * |
| 利用随机森林回归算法预测总有机碳含量;冯明刚 等;《矿物岩石地球化学通报》;20180531;第37卷(第3期);第475-481页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110118994A (en) | 2019-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109061765B (en) | Trap evaluation method for heterogeneous thin sandstone interbed reservoir | |
| CN105182424B (en) | A kind of method and apparatus based on patchy saturation quantitative forecast reservoir porosity | |
| CN105277982B (en) | A kind of mud shale total content of organic carbon earthquake prediction method | |
| CN108897975A (en) | Coalbed gas logging air content prediction technique based on deepness belief network | |
| CN113553780B (en) | Stratum pore pressure prediction method based on machine learning | |
| CN110954944A (en) | Fault trap oil-containing height earthquake prediction method | |
| CN109543915B (en) | Method for identifying total organic carbon content of hydrocarbon source rock in whole well section based on logging information | |
| CN112034521B (en) | Method for predicting overpressure of under-compacted and hydrocarbon-production mixed formation | |
| CN105807320A (en) | Shale gas reservoir seismic prediction method | |
| CN110118994B (en) | Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning | |
| CN111767674A (en) | Well logging lithology identification method based on active domain adaptation | |
| CN114529110A (en) | Lithofacies inversion method and system based on deep neural network model | |
| CN116305850A (en) | Stratum thermal conductivity prediction method based on random forest model | |
| CN107870368A (en) | A kind of total content of organic carbon spatial distribution Forecasting Methodology based on seismic properties | |
| Abubakar et al. | Deep learning applications for wind farms site characterization and monitoring | |
| CN110988997A (en) | Hydrocarbon source rock three-dimensional space distribution quantitative prediction technology based on machine learning | |
| CN116168224A (en) | Machine learning lithology automatic identification method based on imaging gravel content | |
| CN111948718B (en) | Method and device for predicting total organic carbon content of shale gas reservoir | |
| CN114114414A (en) | Artificial intelligence prediction method for 'dessert' information of shale reservoir | |
| CN106568918B (en) | Shale organic carbon content TOC prediction method | |
| CN113608258B (en) | Self-consistent deep learning method for constructing high-resolution wave impedance inversion label | |
| CN110568149B (en) | Fine and rapid quantitative simulation method for hydrocarbon generation and discharge history of sedimentary basin hydrocarbon source rock | |
| CN112394392B (en) | Method and device for evaluating distribution condition of hydrocarbon source rock | |
| Nath et al. | Prediction and analysis of geomechanical properties using deep learning: A Permian Basin case study | |
| CN111594156A (en) | Method and system for calculating saturation of natural gas hydrate |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |