CN111815773B - Three-dimensional complex geologic model label manufacturing method suitable for machine learning algorithm - Google Patents
Three-dimensional complex geologic model label manufacturing method suitable for machine learning algorithm Download PDFInfo
- Publication number
- CN111815773B CN111815773B CN202010847625.7A CN202010847625A CN111815773B CN 111815773 B CN111815773 B CN 111815773B CN 202010847625 A CN202010847625 A CN 202010847625A CN 111815773 B CN111815773 B CN 111815773B
- Authority
- CN
- China
- Prior art keywords
- model
- dimensional
- fault
- machine learning
- displacement
- 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
- 238000010801 machine learning Methods 0.000 title claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 24
- 238000004519 manufacturing process Methods 0.000 title claims description 6
- 238000000034 method Methods 0.000 claims abstract description 21
- 238000006073 displacement reaction Methods 0.000 claims description 29
- 230000006870 function Effects 0.000 claims description 22
- 238000005452 bending Methods 0.000 claims description 13
- 230000005764 inhibitory process Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 230000001629 suppression Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims 3
- 230000008021 deposition Effects 0.000 abstract description 10
- 238000012545 processing Methods 0.000 abstract description 6
- 230000004044 response Effects 0.000 description 10
- 230000015572 biosynthetic process Effects 0.000 description 9
- 238000000151 deposition Methods 0.000 description 9
- 238000011161 development Methods 0.000 description 9
- 238000005755 formation reaction Methods 0.000 description 9
- 238000010276 construction Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000005553 drilling Methods 0.000 description 4
- 239000004215 Carbon black (E152) Substances 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 229930195733 hydrocarbon Natural products 0.000 description 3
- 150000002430 hydrocarbons Chemical class 0.000 description 3
- 230000001788 irregular Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 239000000945 filler Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V20/00—Geomodelling in general
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/004—Annotating, labelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Remote Sensing (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Graphics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a three-dimensional complex geologic model label making method suitable for a machine learning algorithm, which is applied to the field of seismic data processing and aims at solving the problems that a three-dimensional geologic model constructed by the prior art does not completely accord with stratum deposition, geomechanics rules.
Description
Technical Field
The invention belongs to the field of seismic data processing, and particularly relates to a three-dimensional geological model construction technology.
Background
The depth of hydrocarbon reservoir exploration and development is increasing, and the complexity of geological structures is increasing. The conventional oil and gas exploration and development method needs to go through a plurality of links, so that the research work efficiency is low, the exploration period is long, the oil and gas reservoir description precision is low, the cost is high, and the method is getting more and more attention. Aiming at the difficult problems faced by conventional oil and gas exploration and development, the machine learning algorithm provides a new thought and means for the oil and gas reservoir exploration and development of complex geological structures. In recent years, the application of machine learning to oil and gas exploration and development is increasingly favored by geophysicists and geologists. The success of the application of the machine learning algorithm is closely related to the richness of the tag data set, and the open source data sets such as MNIST (LeCun et al, 2018) and ImageNet (Deng et al, 2009) promote the rapid progress of machine learning, so that the machine learning is exponentially increased in the past 10 years. In an ideal case, experts in the field of geography also need such a geoscience reference dataset.
In the petroleum industry, the main means for exploration and development of hydrocarbon reservoirs of complex geological structures is seismic exploration. The basis for extracting hydrocarbon reservoir related parameters from complex geologic structures from seismic data acquired from seismic exploration using machine learning is the need to define the seismic response characteristics of different geologic structures. Drilling and formation parameter measurements are one of the most effective ways to establish relationships between different types of geologic formations and seismic response characteristics. The data sets obtained by drilling and formation parameter measurement are far from meeting the requirements for solving the machine learning algorithm for the exploration and development of complex geologic structures. For processing images by applying a machine learning algorithm, a former person often performs data sample expansion on the tagged image by means of scaling, rotation, noise disturbance and the like. For seismic data, the method has the characteristics of large uncertainty factor, extremely complex internal structure, huge data volume, little known information and the like, and shows unique characteristics on the seismic data aiming at specific geological structures, stratum depositions and lithofacies, so that the method for expanding the data samples in the field of image processing is not applicable. Constructing seismic response datasets of different types of geologic structures suitable for machine learning algorithms has been a challenge for geophysicists. In particular to a broken solution oil reservoir researched in a Tarim basin in recent years, the reservoir space has uneven distribution, diversified forms, large longitudinal depth, extremely strong internal heterogeneity and very complex spatial distribution of the reservoir, and the acquisition of a seismic response sample data set corresponding to the broken solution reservoir is more difficult and expensive. The seismic response dataset construction of different types of geologic structures comprises geologic model construction conforming to actual conditions and seismic response numerical simulation. Seismic response simulation algorithms are well established. For this reason, the core of the seismic response dataset that builds the complex geologic structure model is how to build the complex geologic structure model that conforms to the formation depositional rules, geomechanics.
Aiming at the problem that the machine learning algorithm identifies the space spread of faults based on the seismic data, wu Xinming and the like (2019) train by establishing a large amount of three-dimensional fault synthesized data as data samples, and consider the identification of faults as two classification problems, and apply the faults to the actual seismic data, so that the accuracy and the efficiency of fault identification are greatly improved. Wu Xinming et al (2020) propose to construct a true geologic structure model for training convolutional neural network training, so as to further realize seismic structure interpretation, and the actual processing shows that good effects are obtained. From a mathematical and statistical theoretical perspective, the method can build countless geologic structure models. However, from the perspective of depositional and geomechanical analysis, the three-dimensional geologic model constructed by the method does not completely conform to the rules of stratum deposition and geomechanical.
Disclosure of Invention
In order to solve the technical problems, the invention provides a three-dimensional complex geologic model label manufacturing method suitable for a machine learning algorithm, which is based on the theories of stratigraphic depositions, structural geology, geologic structure mechanics and the like, considers factors such as stratigraphic depositions rules, stratum stress conditions generated by faults, and relations among karst cave, holes, cracks and faults.
The invention adopts the technical scheme that: the method for manufacturing the three-dimensional complex geologic model label suitable for the machine learning algorithm comprises the following steps:
s1, generating a three-dimensional horizontal lamellar model according to simulated stratum parameters;
s2, calculating displacement, and adding the displacement into the horizontal lamellar model to generate a three-dimensional fault model;
s3, carrying out stratum bending and stratum tilting on the basis of the three-dimensional fault model obtained in the step S2 to obtain a three-dimensional bending layered model;
and S4, adding the three-dimensional random hole-seam-hole model into the three-dimensional curved layered model generated in the step S3, and generating a final three-dimensional geological model.
Further, in step S2, the three-dimensional tomographic model expression is micro:
V f (X,Y,Z)＝v(X,Y+D y ,Z+D z )
wherein ,Dy Is displacement in the y direction; d (D) z Is z-direction displacement.
Further, D y And D z The general calculation formula of (2) is:
D＝α·(d max d)+η
wherein alpha is the inverse inhibition coefficient and d is the sum of (X 0 ,Y 0 ,Z 0 ) A three-dimensional Gaussian function as a center, namely an inverse suppression variable, eta is a constant, and d max Is the maximum displacement.
Further, the step S3 specifically includes:
s31, adding a vertical displacement variable S on the basis of the three-dimensional fault model 1 (X, Y, Z) to obtain a curved lamellar model V f (X,Y,Z+S1)；
S32, adding stratum inclination variable S 2 (X, Y, Z) to produce a tilted curved lamellar model V f (X,Y,Z+S 1 +S 2 )。
Further, S 1 The (X, Y, Z) expression is:
wherein ,as a linear scale function, Z max Is the maximum value of Z coordinate; c k ,d k ,σ k Parameters of a two-dimensional Gaussian function; b k Is a coefficient of a mixed gaussian function.
Further, S 2 The (X, Y, Z) expression is:
S 2 (X,Y,Z)＝aX+bY+c
wherein a and b are random numbers; c= -aX 0 -bY 0 ；
Further, step S3 further includes performing cubic spline interpolation processing on the oblique bending layered model.
Further, the three-dimensional random hole-seam hole model in the step S4 is generated by adopting a random medium theory.
The invention has the beneficial effects that: according to the invention, from the aspects of depositional science, structure geology and geomechanics, geological factors are comprehensively considered, three-dimensional seismic data labels conforming to stratum depositions, geological stress conditions and geological laws are simulated, a large number of three-dimensional geological models conforming to actual geological conditions are constructed, existing seismic label data can be enriched, and then a seismic wave field response simulation technology is combined, so that conditions are provided for extracting relevant features of oil and gas reservoirs in complex geological structures from seismic data by utilizing a machine learning algorithm, exploration cost and exploration period are reduced, and the success rate of drilling in the oil and gas exploration and development process is improved.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2 is a schematic view of a fault plane model;
FIG. 3 is a three-dimensional model building step;
FIG. 4 is a prior art fault model;
FIG. 5 is a fault model of the present invention;
FIG. 6 is a geologic model of the invention;
fig. 7 is a forward model of the present invention.
Detailed Description
To facilitate understanding of the technical content of the present invention by those skilled in the art, the following description is given of the geologic model building technology in the prior art:
the existing geological model building method comprises the following steps:
the technology is mainly based on solid geometry and curve fitting theory, a fault plane is fitted into a curve with obvious characteristics, and the curve is stretched and extruded to realize three-dimensional geological model construction. The implementation of the technology comprises the following steps:
step 1: establishing a three-dimensional horizontal lamellar geological model r (X, Y, Z), wherein the value of the model r is a random number between [ -1,1 ]; r (X, Y, Z) is the seismic longitudinal wave reflection coefficient of the stratum model.
Step 2: adding a geologic body vertical displacement variable S 1 (X, Y, Z) can be expressed as:
wherein ,the vertical displacement is weakened from top to bottom as a linear scale function; e is a natural base number; c k ,d k ,σ k Is a linear combination of two-dimensional Gaussian functions>Parameters of (a); b k Coefficients that are linear combinations of gaussian functions; k is the subscript of the kth two-dimensional Gaussian function parameter; n is the number of two-dimensional Gaussian functions.
Step 3: adding formation dip variable S 2 (X, Y, Z) can be expressed as:
S 2 (X,Y,Z)＝aX+bY+c (2)
wherein a and b are [ -0.25,0.25]Random numbers in between; c= -aX 0 -bY 0 So that the central track (X 0 ,Y 0 Z) does not shift; finally, a curved lamellar model r (X, Y, Z+S) is generated 1 +S 2 )；
Step 4: to make the model more realistic with the actual formation, faults are added to the model, wherein the generation of faults has the following flow:
(1) determining the inclination angle theta (theta epsilon (0 DEG, 90 DEG)) of a fault plane and trendConverting global coordinates to local coordinates (X, y, z) =r (X-X 0 '；Y-Y 0 '；Z-Z 0 ') in local coordinate system 0 ',Y 0 ',Z 0 ') is the origin of coordinates, the trend is the x-direction, the tilt angle is the y-direction, the trend is the z-direction, whichIn (a):
(2) the fault plane is further complicated, so that the smooth fault plane is bent, points around the fault plane are randomly selected under a local coordinate system, and green spline interpolation is carried out to obtain a new fault plane z=f (x, y);
(3) defining a displacement function D in the y-direction y (x,y,z)：
(f(x,y)≤z≤γ+f(x,y)),D y (x,y,z)＝λ·d(x,y；z＝0)·α(x,y,z) (4)
(f(x,y)-γ≤z≤f(x,y)),D y (x,y,z)＝(λ-1)·d(x,y；z＝0)·α(x,y,z) (5)
Wherein, gamma is reverse inhibition radius, lambda epsilon (0, 1); the d (x, y; z=0) function spreads the displacement from the fault center point around, defined herein as:
wherein ,alpha (x, y, z) is such that the displacement does not spread too far in the model, which is defined as +.>Gamma is the reverse inhibition radius;
(4) defining a displacement function D in the z direction z (x,y,z)：
D z (x,y,z)＝f(x,y+D y (x,y,z))-f(x,y) (7)
Finally, generating a fault model (x, y+D y ,z+D z ) Then converting from the local coordinate system (X, Y, Z) into the global coordinate system (X, Y, Z)
The following describes the invention in detail with reference to fig. 1 to 7:
the success of the application of the machine learning algorithm depends on the integrity of the tag dataset. A large number of tag data sets with complete characteristics can ensure the effectiveness of a machine learning algorithm, prevent overfitting and improve the robustness of the algorithm. The seismic response data sets corresponding to different geologic structure models calibrated by drilling and stratum parameter measurement are far from meeting the requirements of machine learning algorithm training for the exploration and development of complex geologic structures. Although the prior art of geologic model construction can build countless geologic structure models, the three-dimensional geologic model constructed by the prior art does not completely conform to the rules of stratum deposition and geomechanics from the perspective of theory of depositional science, geomechanics and the like.
Therefore, the three-dimensional complex geologic model label manufacturing method is suitable for a machine learning algorithm based on the theories of stratigraphic depositions, structural geology, geologic structure mechanics and the like, and considers factors such as stratigraphic depositions rules, stratum stress conditions generated by faults, relations among solution cavities, holes, cracks and faults and the like, so as to realize a large number of three-dimensional geologic model constructions which accord with actual geologic conditions (such as stratum bending, faults, cracks, irregular geometric shapes of corrosion holes, filler properties and the like).
The establishment of a large number of three-dimensional geologic models can enrich the characteristics of seismic data samples, and provides more complete information for machine learning to extract the related characteristics of oil and gas reservoirs of complex geologic structures from seismic data. The geological model is very complex to build, and the method can approximate actual seismic data to a certain extent. As shown in fig. 1, the process of establishing the three-dimensional geologic model is:
a1, establishing a three-dimensional horizontal lamellar velocity model v (X, Y, Z), wherein the value of the model v is a random value of a corresponding simulated lithology;
a2, as shown in FIG. 2, defines an angle θ (θ ε (0 °,90 °)) and an angleCenter point (X) 0 ,Y 0 ,Z 0 ) A fault plane f (X, Y, Z) is determined. Wherein->The fault trend is that the dip angle of the fault is
Next, a fault is generated in space:
D y (X,Y,Z)＝Dsin(θ) (8)
D z (X,Y,Z)＝Dcos(θ) (9)
wherein ,Dy (X, Y, Z) is a Y-direction displacement; d (D) z (X, Y, Z) is Z-direction displacement; the displacement variable D has the following calculation formula: d=α· (D max d)+η：
Wherein, alpha is a reverse inhibition coefficient; d is represented by (X) 0 ,Y 0 ,Z 0 ) A three-dimensional Gaussian function which is the center is a reverse suppression variable; lambda epsilon (0, 1); n (n)>1) The greater n is the power of the equation, the smaller the degree and extent of bending of the generated fault, and in most cases, the weaker the degree and extent of bending of the fault; gamma is the reverse inhibition radius; eta is a constant eta>d max ，d max Is the maximum displacement.
Thus fault model V f (X, Y, Z) results in the following:
V f (X,Y,Z)＝v(X,Y+D y ,Z+D z ) (12)
in the formula ,Dy Is displacement in the y direction; d (D) z Is z-direction displacement;
a3, in order to make the model approach the real data more, the fault model V is formed f Performing formation bending, formation tilting and adding a hole and slot model into the model on the basis of (X, Y and Z):
first, add droopsStraight-shift variable S 1 (X, Y, Z) in which a curved lamellar model V is obtained f (X, Y, Z+S1), which can change the flat fault plane generated before into a curved surface, reduces the green spline interpolation step in the prior art, and simplifies the model building step, which can be expressed as:
wherein ,the vertical displacement is weakened from top to bottom as a linear scale function, and the larger the beta is, the higher the stratum bending degree is; c k ,d k ,σ k Is a linear combination of two-dimensional Gaussian functions>Parameters of (a); b k Coefficients that are linear combinations of gaussian functions;
next, the formation inclination variable S is added 2 (X, Y, Z) can be expressed as:
S 2 (X,Y,Z)＝aX+bY+c (14)
wherein a and b are [ -0.25,0.25]Random numbers in between; c= -aX 0 -bY 0 So that the central track (X 0 ,Y 0 Z) does not deviate, (X) 0 ,Y 0 ,Z 0 ) Is the center channel (X) 0 ,Y 0 Points on Z);
then generating a tilted curved lamellar model V f (X,Y,Z+S 1 +S 2 ) In order to make the model lines smoother, performing cubic spline interpolation on the model;
a4, combining the relation between the hole and the fault in the actual situation according to the position generated by the fault, and using a three-dimensional random hole and fault model V generated by a random medium theory rand (X, Y, Z) is added to the model V f (X,Y,Z+S 1 +S 2 ) And (3) generating a final three-dimensional geological model.
As shown in fig. 3, an inverse fault is simulated, the fault is generated to generate weak dragging under the condition of stress, and the dragging accords with the actual geological law. Then, three-dimensional random karst cave, holes and cracks generated according to the random medium theory can be seen, the cracks penetrate through the karst cave, random distribution is shown in the whole three-dimensional space, and in the three-dimensional geological model, the karst cave, holes and cracks are distributed around the faults according to geological relations between the three-dimensional geological model and the faults, so that the geological model is more complex to construct, and the characteristics of the geological model are consistent with the rules of stratum deposition and geomechanics.
The technical effects of the present invention are described below with reference to specific examples:
the method and the device respectively use the prior method and the device to generate a fault model with three layers of stratum, then use the random medium theory to fill the karst cave, the hole and the crack of the model, and finally obtain a three-dimensional geological model and a three-dimensional convolution model.
As the prior art of FIG. 4 simulates a normal fault, when the upper disc of the normal fault is lowered in consideration of the stress factor, the stratum drag around the fault should be bent upwards, so that the tiny drag of the fault can be seen from FIG. 4 to violate the actual geological law. The inclined bending lamellar model is shown in fig. 5, the problem in fig. 4 is corrected in fig. 5, the corrected fault tiny dragging can be seen to be more in line with geological laws, and factors such as karst cave, holes, cracks and faults are considered. As shown in fig. 6, the karst cave, the holes and the cracks are added on the basis of fig. 5, the cracks can be seen to pass through the karst cave in fig. 6, and the generated random karst cave, holes and cracks are distributed around the fault to simulate the more real stratum condition,
FIG. 7 is a three-dimensional convolution model obtained by convolving a three-dimensional geologic model with a 15HZ Rake wavelet and adding random noise, wherein the fracture surface is an irregular curved surface after the stratum is bent, so that the model is more in accordance with the actual geologic law.
The invention has the following advantages:
(1) Constructing a displacement function for generating faults, so that the generation of faults accords with a geological stress rule;
(2) Constructing a reverse inhibition function with controllable fault pulling range, so that the pulling generated by the fault under the action of stress is controlled within a certain range;
(3) After the fault is generated, the stratum is bent to obtain an irregular fault plane, so that the fault plane is more in line with the actual geological condition, the green spline interpolation step in the prior art is omitted, and the model construction is simpler and more effective;
(4) The vertical bending variable of the stratum is changed, so that the bending degree of the stratum is more convenient to control;
(5) The random medium theory is introduced to be combined with the fault model, so that the information of the constructed geologic model characterization is more abundant;
those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the present invention and should be understood that the scope of the invention is not limited to such specific statements and embodiments. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (6)
1. The method for manufacturing the three-dimensional complex geologic model label suitable for the machine learning algorithm is characterized by comprising the following steps of:
s1, generating a three-dimensional horizontal lamellar model according to simulated stratum parameters; establishing a three-dimensional horizontal lamellar speed model v (X, Y, Z), wherein the value of the model v is the random value of the corresponding simulated lithology;
s2, calculating displacement, and adding the displacement into the horizontal lamellar model to generate a three-dimensional fault model;
define an angle θ, θ∈ (0 °,90 °) and an angleCenter point (X) 0 ,Y 0 ,Z 0 ) Determining a fault plane f (X, Y, Z); wherein->For the trend of the fault, the inclination angle of the fault is +.>
Next, a fault is generated in space:
D y (X,Y,Z)＝Dsin(θ)
D z (X,Y,Z)＝Dcos(θ)
wherein ,Dy (X, Y, Z) is a Y-direction displacement; d (D) z (X, Y, Z) is Z-direction displacement; the displacement variable D has the following calculation formula: d=α· (D max d)+η：
Wherein, alpha is a reverse inhibition coefficient; d is represented by (X) 0 ,Y 0 ,Z 0 ) A three-dimensional Gaussian function which is the center is a reverse suppression variable; lambda epsilon (0, 1); n (n > 1) is the power of the equation; gamma is the reverse inhibition radius; eta is a constant, |eta| > d max ，d max Is the maximum displacement;
thus fault model V f (X, Y, Z) results in the following:
V f (X,Y,Z)＝v(X,Y+D y ,Z+D z )
in the formula ,Dy Is displacement in the y direction; d (D) z Is z-direction displacement; "
S3, carrying out stratum bending and stratum tilting on the basis of the three-dimensional fault model obtained in the step S2 to obtain a three-dimensional bending layered model;
and S4, adding the three-dimensional random hole-seam-hole model into the three-dimensional curved layered model generated in the step S3, and generating a final three-dimensional geological model.
2. The method for making a three-dimensional complex geologic model label suitable for machine learning algorithms according to claim 1, wherein step S3 is specifically:
s31, adding a vertical displacement variable S on the basis of the three-dimensional fault model 1 (X, Y, Z) to obtain a curved lamellar model V f (X,Y,Z+S1)；
S32, adding stratum inclination variable S 2 (X, Y, Z) to produce a tilted curved lamellar model V f (X,Y,Z+S 1 +S 2 )。
3. The method for labeling three-dimensional complex geologic model suitable for machine learning algorithms as set forth in claim 2, wherein S 1 The (X, Y, Z) expression is:
wherein ,as a linear scale function, Z max Is the maximum value of Z coordinate; c k ,d k ,σ k Parameters of a two-dimensional Gaussian function; b k Is a coefficient of a mixed gaussian function.
4. The method for labeling three-dimensional complex geologic model suitable for machine learning algorithms as set forth in claim 2, wherein S 2 The (X, Y, Z) expression is:
S 2 (X,Y,Z)＝aX+bY+c
wherein a and b are random numbers; c= -aX 0 -bY 0 。
5. The method for labeling a three-dimensional complex geologic model for a machine learning algorithm of claim 1, wherein step S3 further comprises performing a cubic spline interpolation process on the oblique curved laminar model.
6. The method for making a three-dimensional complex geologic model label suitable for machine learning algorithms according to claim 1, wherein the three-dimensional random hole-seam hole model in step S4 is generated by adopting random medium theory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010847625.7A CN111815773B (en) | 2020-08-21 | 2020-08-21 | Three-dimensional complex geologic model label manufacturing method suitable for machine learning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010847625.7A CN111815773B (en) | 2020-08-21 | 2020-08-21 | Three-dimensional complex geologic model label manufacturing method suitable for machine learning algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111815773A CN111815773A (en) | 2020-10-23 |
| CN111815773B true CN111815773B (en) | 2023-08-25 |
Family
ID=72859531
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010847625.7A Active CN111815773B (en) | 2020-08-21 | 2020-08-21 | Three-dimensional complex geologic model label manufacturing method suitable for machine learning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111815773B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220098972A1 (en) * | 2020-09-25 | 2022-03-31 | Halliburton Energy Services, Inc. | Correcting borehole images using machine-learning models |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112766321B (en) * | 2020-12-31 | 2024-05-17 | 中国地质调查局成都地质调查中心 | Geological feature detection and recognition method and system based on deep learning |
| CN113253342A (en) * | 2021-05-11 | 2021-08-13 | 电子科技大学 | Method for constructing complex three-dimensional seismic model label by combining logging data |
| CN114355449B (en) * | 2022-01-05 | 2023-04-25 | 电子科技大学 | Structure-oriented three-dimensional seismic image enhancement method based on vector median constraint |
| CN114966840B (en) * | 2022-05-24 | 2023-04-07 | 哈尔滨工业大学 | Model building method of three-dimensional seismic wave field based on general solution neural network |
| CN115469361B (en) * | 2022-09-28 | 2023-05-09 | 北京中恒利华石油技术研究所 | Clastic rock stratum three-dimensional geological modeling method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886641A (en) * | 2014-04-18 | 2014-06-25 | 重庆市勘测院 | Three-dimensional geological model constructing and integrating method for mountain city area |
| CN109388817A (en) * | 2017-08-04 | 2019-02-26 | 中国石油化工股份有限公司 | A kind of Reservoir Fracture three-dimensional modeling method |
| CN110060342A (en) * | 2019-02-22 | 2019-07-26 | 中铁工程设计咨询集团有限公司 | A kind of three-dimension curved surface approximating method |
| CN110705168A (en) * | 2019-10-11 | 2020-01-17 | 内蒙古科技大学 | Simulation method of structural stress field |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11644589B2 (en) * | 2017-05-23 | 2023-05-09 | Schlumberger Technology Corporation | Analogue facilitated seismic data interpretation system |
-
2020
- 2020-08-21 CN CN202010847625.7A patent/CN111815773B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886641A (en) * | 2014-04-18 | 2014-06-25 | 重庆市勘测院 | Three-dimensional geological model constructing and integrating method for mountain city area |
| CN109388817A (en) * | 2017-08-04 | 2019-02-26 | 中国石油化工股份有限公司 | A kind of Reservoir Fracture three-dimensional modeling method |
| CN110060342A (en) * | 2019-02-22 | 2019-07-26 | 中铁工程设计咨询集团有限公司 | A kind of three-dimension curved surface approximating method |
| CN110705168A (en) * | 2019-10-11 | 2020-01-17 | 内蒙古科技大学 | Simulation method of structural stress field |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220098972A1 (en) * | 2020-09-25 | 2022-03-31 | Halliburton Energy Services, Inc. | Correcting borehole images using machine-learning models |
| US11898435B2 (en) * | 2020-09-25 | 2024-02-13 | Halliburton Energy Services, Inc. | Correcting borehole images using machine-learning models |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111815773A (en) | 2020-10-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111815773B (en) | Three-dimensional complex geologic model label manufacturing method suitable for machine learning algorithm | |
| US8838425B2 (en) | Generating facies probablity cubes | |
| US10838108B2 (en) | Method and system for interpolating discontinuous functions in a subsurface model | |
| Sech et al. | Three-dimensional modeling of a shoreface-shelf parasequence reservoir analog: Part 1. Surface-based modeling to capture high-resolution facies architecture | |
| US20170051598A1 (en) | System For Hydraulic Fracturing Design And Optimization In Naturally Fractured Reservoirs | |
| CN105651676A (en) | Reservoir heterogeneity characterization method under regular development well pattern of horizontal wells | |
| EP3631528A1 (en) | Constructing structural models of the subsurface | |
| CN109884710B (en) | Micro-logging tomography method aiming at excitation well depth design | |
| AU2005262438A1 (en) | Hydrodynamics-based gridding geologic modeling (Hydro-Gridding) | |
| US10387583B2 (en) | Rotations from gradient directions | |
| CN111612906B (en) | Method and system for generating three-dimensional geological model and computer storage medium | |
| CN115220101B (en) | Modeling method for deep clastic rock sand body deposition structure | |
| CN107966732A (en) | The seismic properties change rate acquiring method being oriented to based on space structure | |
| CN109143337A (en) | A kind of lithologic sand body reservoir modeling methodologies | |
| US20210405250A1 (en) | Conditioning of Surface-Based Geologic Models | |
| Chen et al. | Upscaled modeling of well singularity for simulating flow in heterogeneous formations | |
| CN114755721A (en) | Unet-based river channel sand body top surface prediction method | |
| CN106908838A (en) | The method for building target area stratum inelastic attenuation quality factor three-dimensional model | |
| CN113253342A (en) | Method for constructing complex three-dimensional seismic model label by combining logging data | |
| MacDonald et al. | Stochastic modeling of incised valley geometries | |
| CN111751886A (en) | Shale gas reservoir fracture modeling method based on microseism monitoring data | |
| CN115292771B (en) | Pseudo 2.5D simulation method for resistivity logging while drilling response | |
| CN115963569A (en) | Method and device for establishing phase-control initial model in phase-control inversion | |
| CN117420599A (en) | River simulation and deep learning method for building river sand stratum grillwork | |
| Das et al. | An Integrated Reservoir Modelling Approach in Constructing a Representative Thin, Heterogeneous Clastic Reservoir-Case History |
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 |