CN109164491A - A kind of seismic facies recognition methods and system based on category support vector machines - Google Patents
A kind of seismic facies recognition methods and system based on category support vector machines Download PDFInfo
- Publication number
- CN109164491A CN109164491A CN201811194757.3A CN201811194757A CN109164491A CN 109164491 A CN109164491 A CN 109164491A CN 201811194757 A CN201811194757 A CN 201811194757A CN 109164491 A CN109164491 A CN 109164491A
- Authority
- CN
- China
- Prior art keywords
- seismic
- classification
- training sample
- lithofacies
- support vector
- 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
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
- G01V1/50—Analysing data
Landscapes
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- Remote Sensing (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The present invention provides a kind of seismic facies recognition methods and system based on category support vector machines, comprising: the log data of the lithofacies classification and lithofacies classification that will acquire is as training sample data;Supporting vector machine model is trained according to the training sample data, generates SVM prediction model;Inverting is carried out to the prestack seismic gather of acquisition, generates seismic properties combination;Seismic properties combination is inputted into the SVM prediction model and carries out seismic facies identification, generates the corresponding seismic facies recognition result of the log data.The application has the beneficial effect for improving using support vector machines as the pre- flow gauge of the seismic facies of classification tool and playing support vector machines technical advantage under Small Sample Size.
Description
Technical field
The present invention relates to technical field of geophysical exploration more particularly to a kind of earthquake rocks based on category support vector machines
Phase identification method and system.
Background technique
Support vector machines is as a kind of machine Learning Theory, based on Statistical Learning Theory, has been mature on the whole and more
It is used widely and develops in a field.Support vector regression algorithm by insensitive loss parameter by support vector machines from from
Classification field is dissipated to be brought into continuous recurrence field.The construction and Choice Theory of kernel function have researched and solved support vector machines not
Applicability problem when applying in same domain, greatlies simplify computational complexity.The it is proposed of multi-class classification method in 1998 is more
The expansion that support vector machines is classified from two classes to multicategory classification is furthermore achieved, meets theoretical and application demand.
One of the important method that lithofacies prediction is seismic interpretation is carried out using seismic properties, counts identification, pattern-recognition, people
Based on artificial neural networks scheduling algorithm has the practicality, but these algorithms require mass data sample.Support vector machines is made
For the pattern classifier based on structural risk minimization principle, classification and generalization ability under small sample still keep preferably,
It solves the problems, such as with the obvious advantage in non-linear and high order modes identification, becomes the research in seismic properties petroleum-gas prediction field at this stage
With apply hot spot.
Therefore, how to carry out seismic facies identification using support vector machines is current technical problem urgently to be resolved.
Summary of the invention
In order to solve defect in the prior art, the present invention provides a kind of seismic facies based on category support vector machines
Recognition methods and system, have improve using support vector machines as the pre- flow gauge of the seismic facies of classification tool and play support to
The beneficial effect of amount machine technical advantage under Small Sample Size.
To achieve the goals above, the present invention provides a kind of seismic facies identification side based on category support vector machines
Method, this method comprises:
The log data of the lithofacies classification and lithofacies classification that will acquire is as training sample data;
Supporting vector machine model is trained according to the training sample data, generates SVM prediction model;
Inverting is carried out to the prestack seismic gather of acquisition, generates seismic properties combination;
It combines the seismic properties and inputs the SVM prediction model progress seismic facies identification, described in generation
The corresponding seismic facies recognition result of log data.
The seismic facies identifying system based on category support vector machines that the present invention also provides a kind of, the system include:
Training data generation unit, the log data of lithofacies classification and lithofacies classification for will acquire is as training
Sample data;
Training unit, for being trained according to the training sample data to supporting vector machine model, generate support to
Amount machine prediction model;
Inverting unit generates seismic properties combination for carrying out inverting to the prestack seismic gather of acquisition;
Seismic facies recognition unit is carried out for seismic properties combination to be inputted the SVM prediction model
Seismic facies identification, generates the corresponding seismic facies recognition result of the log data.
A kind of seismic facies recognition methods and system based on category support vector machines provided by the invention, comprising: will obtain
The log data of the lithofacies classification and lithofacies classification that take is as training sample data;According to the training sample data to branch
It holds vector machine model to be trained, generates SVM prediction model;Inverting is carried out to the prestack seismic gather of acquisition, is generated
Seismic properties combination;Seismic properties combination is inputted into the SVM prediction model and carries out seismic facies identification, it is raw
At the corresponding seismic facies recognition result of the log data.The application, which has, to be improved using support vector machines as classification tool
The pre- flow gauge of seismic facies and the beneficial effect for playing support vector machines technical advantage under Small Sample Size.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow chart of seismic facies recognition methods based on category support vector machines of the application;
Fig. 2 is the flow chart of the seismic facies recognition methods based on category support vector machines in one embodiment of the application;
Fig. 3 is the log data in one embodiment of the application;
Fig. 4 is the optimizing result schematic diagram of the punishment parameter in one embodiment of the application;
Fig. 5 is the prestack angle domain seismic channel set figure in one embodiment of the application;
Fig. 6 a is the p-wave impedance figure in one embodiment of the application;
Fig. 6 b is the S-wave impedance figure in one embodiment of the application;
Fig. 7 is the seismic facies recognition result figure in one embodiment of the application;
Fig. 8 is a kind of structural schematic diagram of seismic facies identifying system based on category support vector machines of the application;
Fig. 9 is the structural schematic diagram of the training data generation unit in one embodiment of the application;
Figure 10 is the structural schematic diagram of the training unit in one embodiment of the application.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
About " first " used herein, " second " ... etc., not especially censure the meaning of order or cis-position,
Also non-to limit the present invention, only for distinguishing with the element of same technique term description or operation.
It is open term, i.e., about "comprising" used herein, " comprising ", " having ", " containing " etc.
Mean including but not limited to.
About it is used herein " and/or ", including any of the things or all combination.
In view of the deficiencies in the prior art, the present invention provides a kind of seismic facies based on category support vector machines
Recognition methods, flow chart as shown in Figure 1, this method comprises:
S101: the log data of the lithofacies classification and the classification of each lithofacies that will acquire is as training sample data.
S102: being trained supporting vector machine model according to training sample data, generates SVM prediction model.
S103: inverting is carried out to the prestack seismic gather of acquisition, generates seismic properties combination.
S104: it seismic properties combines input SVM prediction model and carries out seismic facies identification, generate well logging number
According to corresponding seismic facies recognition result.
Process as shown in Figure 1 is it is found that the present invention is classified using the lithofacies obtained and the log data of lithofacies classification is to branch
It holds vector machine model to be trained, generates SVM prediction model;And it combines the seismic properties generated using inverting defeated
Enter SVM prediction model and carry out seismic facies identification, generates the corresponding seismic facies recognition result of log data.This hair
The petrographic interpretation of the bright conventional and unconventional oil and gas reservoir that can be used for different exploratory areas works, and instructs geologic interpretation and oil and gas development
Work has and improves using support vector machines as the pre- flow gauge of the seismic facies of classification tool and play support vector machines in sample
The beneficial effect of technical advantage in the case of this.
In order to make those skilled in the art be better understood by the present invention, a more detailed embodiment is set forth below,
As shown in Fig. 2, the embodiment of the present invention provides a kind of seismic facies recognition methods based on category support vector machines, this method includes
Following steps:
S201: the log data of the lithofacies classification and the classification of each lithofacies that will acquire is as training sample data.Wherein train
Sample data includes: training sample vector and training sample classification.
When it is implemented, by taking certain work area oil and gas reservoir as an example, as shown in figure 3, the log data of input includes: longitudinal wave speed
Lithofacies are classified in degree, shear wave velocity, density and Gamma log value etc. and the well logging, and it includes: cementing sandstone that lithofacies, which are classified, pure
6 kinds of sandstone, argillaceous sandstone 1, argillaceous sandstone 2, Sandy Silt and mud stone lithofacies classifications, invention is not limited thereto.
The lithofacies classification that will acquire first is as training sample classification, then according to the well logging number of each lithofacies of acquisition classification
According to generation seismic properties, and using seismic properties as training sample vector.Wherein seismic properties include: p-wave impedance and shear wave resistance
Anti- etc., invention is not limited thereto.
The present invention is classified using existing support vector machines theory, is wherein related in support vector machine classifier discriminant function
And mapping phi calculating be in the form of inner product occur, therefore introduce kernel function K (xi,xj) as shown in formula (1):
K(xi,xj)=< φ (xi),φ(xj) > (1)
Wherein, <, > are inner product, and i and j are the positive integer more than or equal to 1, φ (xi) and φ (xj) it is non-linear letter
Number.
High-dimensional operation problem is solved, is obtained shown in the discriminant function such as formula (2) of support vector machines:
Wherein sign is sign function, and b is amount of bias, αiyiFor coefficient constant.Common kernel function has: linear kernel function,
Polynomial kernel function, RBF kernel function and Sigmoid kernel function etc..
Shown in linear kernel function such as formula (3):
K (x, x ')=(xx ') (3)
Shown in Polynomial kernel function such as formula (4):
K (x, x ')=(γ xx ')P, γ > 0 (4)
Shown in RBF kernel function such as formula (5):
K (x, x ')=exp (- γ | | xx ' | |2), γ > 0 (5)
Shown in Sigmoid kernel function such as formula (6):
K (x, x ')=tanh (γ xx '+c0) (6)
Wherein, γ is the linearity factor in Polynomial kernel function, RBF kernel function and Sigmoid kernel function, for controlling number
According to the scaling after high-dimensional projection, c0For the normal ginseng in Sigmoid function.
S202: being trained supporting vector machine model according to training sample data, generates SVM prediction model.
As shown in Fig. 2, when step S202 is specifically executed the following steps are included:
S301: being trained the kernel function of each pre-selection according to training sample classification and training sample vector, generates each pre-
Select the lithofacies classification accuracy of kernel function.
Training sample classification and the input of training sample vector are trained the kernel function of each pre-selection, handed over using K-fold
The statistical analysis technique of verifying, classification of assessment device performance indicator are pitched, while avoiding overfitting and deficient learning state, is ultimately generated each
The lithofacies classification accuracy for preselecting kernel function is as shown in table 1.Wherein respectively the lithofacies classification accuracy of pre-selection kernel function includes: each pre-
Select the corresponding lithofacies classification accuracy of each dimension of the corresponding lithofacies classification accuracy of kernel function and each pre-selection kernel function.
Table 1
The kernel function of pre-selection | Lithofacies classification accuracy (%) |
Linear kernel function | 87.4390 |
Polynomial kernel function | 91.5854 |
RBF kernel function | 88.0488 |
Sigmoid kernel function | 36.9512 |
S302: raw according to preset punishment parameter range and the corresponding pre-selection kernel function of maximum lithofacies classification accuracy
At support vector machines punishment parameter.
According to table 1, the maximum lithofacies classification accuracy known to is Polynomial kernel function, while according to table
Shown in 2, the corresponding maximum dimension of lithofacies classification accuracy of Polynomial kernel function is obtained are as follows: dimension=3.
Table 2
The dimension of Polynomial kernel function | Lithofacies classification accuracy (%) |
2 | 91.4634 |
3 | 91.5854 |
4 | 85.2439 |
Wherein, the codomain range of the linearity factor γ of the control scaling of Polynomial kernel function is γ ∈ [2-4, 2-3..., 24], the value range of punishment parameter C is set as C ∈ [2-2, 2-1..., 24].γ value and punishment parameter C are according to step-length
0.5 carry out optimizing, searching process also with K-fold cross validation method, calculate corresponding to different C value and γ value to
For amount machine model to the classification accuracy of lithofacies, finding makes the maximum γ value of the lithofacies classification accuracy of the support vector machines and punishment
Parameter C, searching process are as shown in Figure 4, wherein what abscissa log2c and ordinate log2g respectively indicated C value and γ value is with 2
The logarithmic coordinates at bottom, convenient for display.As shown in figure 4, being 11.3137 by optimal γ value known to optimizing result, punishment parameter C is
5.6569, therefore support vector machines punishment parameter is 5.6569.
S303: raw according to the corresponding pre-selection kernel function of maximum lithofacies classification accuracy and support vector machines punishment parameter
At the corresponding SVM prediction model of training sample data.
According to the corresponding pre-selection kernel function of the maximum lithofacies classification accuracy i.e. polynomial function generated in step S302
And the support vector machines punishment parameter 5.6569 and kernel functional parameter γ value 11.3137 generated in step S303, generate current instruction
Practice the corresponding SVM prediction model of sample data.
S203: inverting is carried out to the prestack seismic gather of acquisition, generates seismic properties combination.
When it is implemented, as shown in figure 5, carrying out inverting, generation p-wave impedance to the prestack angle domain seismic channel set of acquisition
And S-wave impedance, wherein p-wave impedance is as shown in Figure 6 a, and S-wave impedance is as shown in Figure 6 b.
S204: it seismic properties combines input SVM prediction model and carries out seismic facies identification, generate well logging number
According to corresponding seismic facies recognition result.
When it is implemented, the seismic properties that step S203 is generated combine, the supporting vector generated in input step S202
Machine prediction model carries out seismic facies identification, the corresponding seismic facies recognition result of current log data is generated, to realize base
Seismic interpretation is quantified in the reservoir of support vector machines, seismic facies recognition result is as shown in Figure 7.
Specifically, inversion formula is using the Fatti longitudinal wave reflection coefficients R (θ) derived and P wave impedance reflection coefficient, S wave resistance
Relationship between antireflection coefficient and density reflection coefficient, as shown in formula (7), the equation is often applied to inverting wave resistance in length and breadth
Anti- section, stability is high, has the application value of preferably industry:
R (θ)=(1+tan2θ)RP+(-8γ2sin2θ)RS (7)
Wherein,α is upper and lower level velocity of longitudinal wave average value, and β is upper
Lower layer's shear wave velocity average value, ρ upper and lower level density averages, Δ α are that upper and lower level velocity of longitudinal wave is poor, and Δ β is upper and lower level shear wave
Speed difference, Δ ρ are upper and lower level density difference, and θ represents incident angle.
It is established shown in objective function such as formula (8) according to generalized linear inversion thinking:
F (m)=| | S (m)-D | | (8)
Wherein, D is actual angle trace gather, and S (m)=W*R (m) is the response of desired earthquake model, and W is seismic wavelet, R (m)
For the reflection coefficient that Fatti is derived, m=[Rp,Rs]。
The first-order partial derivative for enabling objective function Equation both ends Parameters variation amount is zero, so that objective function is obtained minimum value, obtains
To formula (9):
Δ (m)=(GTG)-1GTΔ(S(m)-D) (9)
Wherein Δ (m) represents [Rp,Rs] variable quantity, G be single order local derviation matrix.Disappear to damping factor is added in formula (9)
Except the problem that refutation process is unstable, computational efficiency is low, inversion accuracy is low, formula (10) are obtained:
Δ (m)=(GTG+λI)-1GTΔ(S(m)-D) (10)
Then final inverting is according to shown in formula (11):
M=m0+Δ(m) (11)
It is iterated calculating, obtains final wave impedance seismic inversion in length and breadth.
Conceived based on application identical with the above-mentioned seismic facies recognition methods based on category support vector machines, the present invention is also
A kind of seismic facies identifying system based on category support vector machines is provided, as described in following example.Due to the system solution
Certainly the principle of problem is similar to the above method, therefore the implementation of the system may refer to the implementation of the above method, repeats place not
It repeats again.
Fig. 8 is a kind of structural representation of seismic facies identifying system based on category support vector machines of the embodiment of the present application
Figure, as shown in figure 8, the system includes: training data generation unit 101, training unit 102, inverting unit 103 and seismic facies
Recognition unit 104.
Training data generation unit 101, the log data of lithofacies classification and the classification of each lithofacies for will acquire is as instruction
Practice sample data;
Training unit 102 generates supporting vector for being trained according to training sample data to supporting vector machine model
Machine prediction model;
Inverting unit 103 generates seismic properties combination for carrying out inverting to the prestack seismic gather of acquisition;
Seismic facies recognition unit 104 carries out earthquake for seismic properties combining input SVM prediction model
Lithofacies Identification generates the corresponding seismic facies recognition result of log data.
In one embodiment, training sample data include: training sample vector and training sample classification.
In one embodiment, as shown in figure 9, training data generation unit 101 include: classification generation module 201 and to
Measure generation module 202.
Classification generation module 201, the lithofacies classification for will acquire are used as training sample classification;
Vector generation module 202, log data for being classified according to each lithofacies of acquisition generate seismic properties, and by ground
Attribute is shaken as training sample vector.
In one embodiment, as shown in Figure 10, training unit 102 includes: accuracy rate generation module 301, punishment parameter
Generation module 302 and prediction model generation module 303.
Accuracy rate generation module 301, for the kernel function according to training sample classification and training sample vector to each pre-selection
It is trained, generates the lithofacies classification accuracy of each pre-selection kernel function;
Punishment parameter generation module 302, for according to preset punishment parameter range and maximum lithofacies classification accuracy
Corresponding pre-selection kernel function generates support vector machines punishment parameter;
Prediction model generation module 303, for according to the corresponding pre-selection kernel function of maximum lithofacies classification accuracy and branch
Vector machine punishment parameter is held, the corresponding SVM prediction model of training sample data is generated.
In one embodiment, seismic properties combination includes: p-wave impedance and S-wave impedance.
A kind of seismic facies recognition methods and system based on category support vector machines provided by the invention, it is bent using well logging
Petrofacies data in line and well is trained category support vector machines, carries out ground to pre-stack seismic inversion result on this basis
Petrographic interpretation is shaken, geologic interpretation and oil are instructed in the petrographic interpretation work of the conventional and unconventional oil and gas reservoir for different exploratory areas
Gas development, at the same have improve using support vector machines as the pre- flow gauge of the seismic facies of classification tool and play support to
The beneficial effect of amount machine technical advantage under Small Sample Size.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
Specific embodiment is applied in the present invention, and principle and implementation of the present invention are described, above embodiments
Explanation be merely used to help understand method and its core concept of the invention;At the same time, for those skilled in the art,
According to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion in this specification
Appearance should not be construed as limiting the invention.
Claims (10)
1. a kind of seismic facies recognition methods based on category support vector machines characterized by comprising
The log data of the lithofacies classification and the classification of each lithofacies that will acquire is as training sample data;
Supporting vector machine model is trained according to the training sample data, generates SVM prediction model;
Inverting is carried out to the prestack seismic gather of acquisition, generates seismic properties combination;
Seismic properties combination is inputted into the SVM prediction model and carries out seismic facies identification, generates the well logging
The corresponding seismic facies recognition result of data.
2. the seismic facies recognition methods according to claim 1 based on category support vector machines, the number of training
According to including: training sample vector and training sample classification.
3. the seismic facies recognition methods according to claim 2 based on category support vector machines, the rock that will acquire
Mutually the log data of classification and the classification of each lithofacies is as training sample data, comprising:
The lithofacies classification that will acquire is as training sample classification;
According to each lithofacies of acquisition classification log data generate seismic properties, and using the seismic properties as training sample to
Amount.
4. the seismic facies recognition methods according to claim 2 or 3 based on category support vector machines, described according to
Training sample data are trained supporting vector machine model, generate SVM prediction model, comprising:
The kernel function of each pre-selection is trained according to the training sample classification and the training sample vector, is generated each described
Preselect the lithofacies classification accuracy of kernel function;
According to preset punishment parameter range and the corresponding pre-selection kernel function of maximum lithofacies classification accuracy, supporting vector is generated
Machine punishment parameter;
According to the corresponding pre-selection kernel function of maximum lithofacies classification accuracy and the support vector machines punishment parameter, described in generation
The corresponding SVM prediction model of training sample data.
5. the seismic facies recognition methods according to claim 1 based on category support vector machines, the seismic properties group
Conjunction includes: p-wave impedance and S-wave impedance.
6. a kind of seismic facies identifying system based on category support vector machines characterized by comprising
Training data generation unit, the log data of lithofacies classification and the classification of each lithofacies for will acquire is as number of training
According to;
Training unit generates support vector machines for being trained according to the training sample data to supporting vector machine model
Prediction model;
Inverting unit generates seismic properties combination for carrying out inverting to the prestack seismic gather of acquisition;
Seismic facies recognition unit carries out earthquake for seismic properties combination to be inputted the SVM prediction model
Lithofacies Identification generates the corresponding seismic facies recognition result of the log data.
7. the seismic facies identifying system according to claim 6 based on category support vector machines, the number of training
According to including: training sample vector and training sample classification.
8. the seismic facies identifying system according to claim 7 based on category support vector machines, the training data is raw
Include: at unit
Classification generation module, the lithofacies classification for will acquire are used as training sample classification;
Vector generation module, log data for being classified according to each lithofacies of acquisition generate seismic properties, and by the earthquake
Attribute is as training sample vector.
9. the seismic facies identifying system according to claim 7 or 8 based on category support vector machines, the training unit
Include:
Accuracy rate generation module, for the kernel function according to the training sample classification and the training sample vector to each pre-selection
It is trained, generates the lithofacies classification accuracy of each pre-selection kernel function;
Punishment parameter generation module, for corresponding pre- according to preset punishment parameter range and maximum lithofacies classification accuracy
Kernel function is selected, support vector machines punishment parameter is generated;
Prediction model generation module, for according to the corresponding pre-selection kernel function of maximum lithofacies classification accuracy and it is described support to
Amount machine punishment parameter generates the corresponding SVM prediction model of the training sample data.
10. the seismic facies identifying system according to claim 6 based on category support vector machines, the seismic properties group
Conjunction includes: p-wave impedance and S-wave impedance.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811194757.3A CN109164491A (en) | 2018-10-15 | 2018-10-15 | A kind of seismic facies recognition methods and system based on category support vector machines |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811194757.3A CN109164491A (en) | 2018-10-15 | 2018-10-15 | A kind of seismic facies recognition methods and system based on category support vector machines |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109164491A true CN109164491A (en) | 2019-01-08 |
Family
ID=64878282
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811194757.3A Pending CN109164491A (en) | 2018-10-15 | 2018-10-15 | A kind of seismic facies recognition methods and system based on category support vector machines |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109164491A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110097069A (en) * | 2019-03-11 | 2019-08-06 | 西安科技大学 | A kind of support vector machines Lithofacies Identification method and device based on depth Multiple Kernel Learning |
CN111079810A (en) * | 2019-12-06 | 2020-04-28 | 中国铁路设计集团有限公司 | Tunnel surrounding rock grade prediction method based on support vector machine |
CN112329804A (en) * | 2020-06-30 | 2021-02-05 | 中国石油大学(北京) | Naive Bayes lithofacies classification integrated learning method and device based on feature randomness |
WO2021130512A1 (en) * | 2019-12-23 | 2021-07-01 | Total Se | Device and method for predicting values of porosity lithofacies and permeability in a studied carbonate reservoir based on seismic data |
CN113361209A (en) * | 2021-07-23 | 2021-09-07 | 南昌航空大学 | Quantitative analysis method for magnetic anomaly of surface defects of high-temperature alloy |
CN113419271A (en) * | 2021-05-07 | 2021-09-21 | 中铁二院工程集团有限责任公司 | Earthquake magnitude prediction method, device, equipment and readable storage medium |
CN115576028A (en) * | 2022-12-01 | 2023-01-06 | 武汉盛华伟业科技股份有限公司 | Geological feature layer prediction method and system based on support vector machine |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130080066A1 (en) * | 2011-09-28 | 2013-03-28 | Saudi Arabian Oil Company | Reservoir properties prediction with least square support vector machine |
CN105005097A (en) * | 2015-07-17 | 2015-10-28 | 中国石油化工股份有限公司 | Method for comprehensive recognition of igneous rocks by employing gravity, magnetism, electromagnetism, and earthquake data |
CN106707340A (en) * | 2016-12-13 | 2017-05-24 | 中国石油天然气股份有限公司大港油田分公司 | Method for predicting volcanic rock facies |
-
2018
- 2018-10-15 CN CN201811194757.3A patent/CN109164491A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130080066A1 (en) * | 2011-09-28 | 2013-03-28 | Saudi Arabian Oil Company | Reservoir properties prediction with least square support vector machine |
CN105005097A (en) * | 2015-07-17 | 2015-10-28 | 中国石油化工股份有限公司 | Method for comprehensive recognition of igneous rocks by employing gravity, magnetism, electromagnetism, and earthquake data |
CN106707340A (en) * | 2016-12-13 | 2017-05-24 | 中国石油天然气股份有限公司大港油田分公司 | Method for predicting volcanic rock facies |
Non-Patent Citations (4)
Title |
---|
伍雪冬: "《非线性时间序列在线预测建模与仿真》", 30 November 2015, 国防工业出版社 * |
张向君: "《信号分析与数据统计学习》", 31 January 2009, 哈尔滨工程大学出版社 * |
梁礼明 等: "《优化方法导论》", 30 September 2017, 北京理工大学出版社 * |
汪益宁 等: "基于支持向量机的致密储层岩相识别——以徐家围子断陷下白垩统沙河子组为例", 《长江大学学报(自然版)》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110097069A (en) * | 2019-03-11 | 2019-08-06 | 西安科技大学 | A kind of support vector machines Lithofacies Identification method and device based on depth Multiple Kernel Learning |
CN111079810A (en) * | 2019-12-06 | 2020-04-28 | 中国铁路设计集团有限公司 | Tunnel surrounding rock grade prediction method based on support vector machine |
WO2021130512A1 (en) * | 2019-12-23 | 2021-07-01 | Total Se | Device and method for predicting values of porosity lithofacies and permeability in a studied carbonate reservoir based on seismic data |
CN112329804A (en) * | 2020-06-30 | 2021-02-05 | 中国石油大学(北京) | Naive Bayes lithofacies classification integrated learning method and device based on feature randomness |
CN113419271A (en) * | 2021-05-07 | 2021-09-21 | 中铁二院工程集团有限责任公司 | Earthquake magnitude prediction method, device, equipment and readable storage medium |
CN113419271B (en) * | 2021-05-07 | 2023-08-29 | 中铁二院工程集团有限责任公司 | Earthquake magnitude prediction method, device, equipment and readable storage medium |
CN113361209A (en) * | 2021-07-23 | 2021-09-07 | 南昌航空大学 | Quantitative analysis method for magnetic anomaly of surface defects of high-temperature alloy |
CN113361209B (en) * | 2021-07-23 | 2022-09-27 | 南昌航空大学 | Quantitative analysis method for magnetic anomaly of surface defects of high-temperature alloy |
CN115576028A (en) * | 2022-12-01 | 2023-01-06 | 武汉盛华伟业科技股份有限公司 | Geological feature layer prediction method and system based on support vector machine |
CN115576028B (en) * | 2022-12-01 | 2023-03-14 | 武汉盛华伟业科技股份有限公司 | Geological feature layer prediction method and system based on support vector machine |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109164491A (en) | A kind of seismic facies recognition methods and system based on category support vector machines | |
US11016214B2 (en) | Dolomite reservoir prediction method and system based on well and seismic combination, and storage medium | |
Graves et al. | Broadband ground-motion simulation using a hybrid approach | |
Lorentzen et al. | History matching the full Norne field model using seismic and production data | |
van der Westhuysen | Modeling of depth‐induced wave breaking under finite depth wave growth conditions | |
CN112883564B (en) | Water body temperature prediction method and prediction system based on random forest | |
CN105629303A (en) | Prestack crack quantitative forecast method and system based on rock physics | |
Lia et al. | Uncertainties in reservoir production forecasts | |
Fahimuddin et al. | Ensemble based 4D seismic history matching: Integration of different levels and types of seismic data | |
CN107817520A (en) | The pressure coefficient Forecasting Methodology and system of marine facies mud shale stratum | |
Limber et al. | Unraveling the dynamics that scale cross‐shore headland relief on rocky coastlines: 2. Model predictions and initial tests | |
CN110389382A (en) | A kind of oil-gas reservoir reservoir characterization method based on convolutional neural networks | |
Golsanami et al. | Synthesis of capillary pressure curves from post-stack seismic data with the use of intelligent estimators: a case study from the Iranian part of the South Pars gas field, Persian Gulf Basin | |
Feng et al. | A high‐resolution 3‐D circulation model in a complex archipelago on the coastal Scotian shelf | |
CN104267430B (en) | Method and device for determining seismic fluid sensitivity factor | |
CN107193044B (en) | A kind of pre-stack seismic Multi-parameters conversion method of hybrid global optimization | |
Wu et al. | Mesh-free radial-basis-function-generated finite differences and their application in reverse time migration | |
Xu et al. | Enhanced tomography resolution by a fat ray technique | |
Kostka et al. | Assessing the role of selected constraints in Bayesian dynamic source inversion: Application to the 2017 M w 6.3 Lesvos earthquake | |
Chopra et al. | Seismic characterization of a Triassic-Jurassic deep geothermal sandstone reservoir, onshore Denmark, using unsupervised machine learning techniques | |
CN110231653A (en) | Construct the method and device of two-way constraint initial model | |
Hajian et al. | Genetic algorithm with applications in geophysics | |
Sabinin et al. | Machine learning for fracture parameter estimation in fractured reservoirs from seismic data | |
CN115685314A (en) | Seismic reservoir physical property parameter prediction method and device | |
Pereira et al. | Integrating of initial guess models into geostatistical seismic inversion methodologies |
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: 20190108 |