CN112906242B - Geophysical modeling method based on combination of naive Bayes method and proximity classification method - Google Patents

Geophysical modeling method based on combination of naive Bayes method and proximity classification method Download PDF

Info

Publication number
CN112906242B
CN112906242B CN202110292022.XA CN202110292022A CN112906242B CN 112906242 B CN112906242 B CN 112906242B CN 202110292022 A CN202110292022 A CN 202110292022A CN 112906242 B CN112906242 B CN 112906242B
Authority
CN
China
Prior art keywords
point
attribute
points
dimensional space
naive bayes
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
Application number
CN202110292022.XA
Other languages
Chinese (zh)
Other versions
CN112906242A (en
Inventor
冯晅
刘乾
侯贺晟
赵鹏飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN202110292022.XA priority Critical patent/CN112906242B/en
Publication of CN112906242A publication Critical patent/CN112906242A/en
Application granted granted Critical
Publication of CN112906242B publication Critical patent/CN112906242B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

The invention provides a geophysical modeling method based on combination of a naive Bayes method and an adjacent classification method, which utilizes a modeling method based on combination of an adjacent method and a naive Bayes method, firstly utilizes a known profile to carry out naive Bayes classification, then finds out N known points closest to unknown points according to the adjacent method, selects a part of points with larger variance to train, represents that the points are positioned at the boundary surfaces of strata, and has more difficult attribute judgment, while the points with smaller variance are positioned between the boundary surfaces of the strata, have the same attribute, do not need to be classified, and then utilizes the rest points to train, thereby learning the existing geological model and modeling data, combining with the existing geological knowledge, leading the modeling to be more objective on the premise of reducing human intervention, further constructing the geological model in large areas and in greater depths, and leading the constructed model to accord with the existing geological knowledge and accord with the actual geological conditions.

Description

Geophysical modeling method based on combination of naive Bayes method and proximity classification method
Technical Field
The invention relates to the field of geophysical exploration, in particular to a geophysical modeling method based on combination of a naive Bayes method and an adjacent classification method.
Background
The traditional geological modeling method can only be applied to a small area, has extremely shallow depth and more seismic sections, and consumes more manpower and material resources, for example, in the area with the large scale of the northern area of the Songliao basin and only a few seismic sections, the traditional modeling method has great limitations, cannot really build a geological model of the area, and cannot combine geological conditions of different areas.
Disclosure of Invention
In order to solve the problems, the invention provides a geophysical modeling method based on the combination of a naive Bayes method and an adjacent classification method, and provides support conditions for three-dimensional geological research and deep resource potential evaluation.
The technical scheme provided by the invention is as follows: a geophysical modeling method based on combination of a naive Bayes method and a proximity classification method comprises the following steps:
1) Extracting geological information in a region range, performing geological interpretation on the geological information, performing data processing on an obtained geological interpretation data result, projecting the data geological interpretation result into a three-dimensional space, and classifying unknown stratums by using a naive Bayes proximity method;
2) Retrieving K known points around one point in the three-dimensional space in the step 1), and extracting the attributes of the K known points, wherein the total types of the attributes are N, and the attribute of one point in the three-dimensional space is A n The attribute of the ith point around it is expressed as
Figure GDA0003844081500000011
Wherein i represents the ith point, N belongs to (1, 2 \8230; N);
3) Learning the known points according to a naive Bayes method, using the formula as follows:
Figure GDA0003844081500000012
wherein N represents any number of 1 \ 8230, and each N does not influence each other, wherein
Figure GDA0003844081500000021
Representing the property of the ith point around this known point, P (A) n ) The attribute representing this point is A n The probability of (a) of (b) being,
Figure GDA0003844081500000022
represents the attribute of the ith point as
Figure GDA0003844081500000023
The probability of (a) of (b) being,
Figure GDA0003844081500000024
representing the attribute of K points around it, respectively
Figure GDA0003844081500000025
In the case of (2), this point attribute is A n The probability of (d); by learning, it can be knownIn point A n Probability of (A), i.e. P (A) n ) The ith point attribute is
Figure GDA0003844081500000026
Probability of (2)
Figure GDA0003844081500000027
And when this point attribute is A n When the measured data is obtained, the attributes of K points around the measured data are respectively
Figure GDA0003844081500000028
The above three probabilities:
Figure GDA0003844081500000029
as a prior probability;
4) Applying the result of step 3) to the known points remaining in step 2), the posterior probability being as follows:
Figure GDA00038440815000000210
wherein, the first and the second end of the pipe are connected with each other,
Figure GDA00038440815000000211
representing the properties of each point around the known point
Figure GDA00038440815000000212
In the case of (2), this point attribute is A n When a probability of
Figure GDA00038440815000000213
When the maximum value is taken, the attribute of the point is determined to be A n ；
5) Different K values are brought into the step 2) for learning, the results obtained in the step 4) are compared with the attributes of the known points to judge the learning accuracy, the learning accuracy at different K values is compared to select the K value K with the highest accuracy, and the prior probability obtained at the moment is the prior probability with the best learning effect
Figure GDA00038440815000000214
6) Selecting the nearest k points of each unknown point and then utilizing a prior probability formula:
Figure GDA00038440815000000215
simultaneously, a naive Bayes method formula is utilized:
Figure GDA00038440815000000216
classifying prior probability formula and naive Bayes method formula, and taking
Figure GDA0003844081500000031
Is the attribute A of the unknown point n Thus, the stratum attribute of each point is obtained, and finally, a three-dimensional geological model of the region is constructed.
Preferably, the geological information in step 1) includes: geophysical information, well drilling data, physical property data, gravity, magnetism, electric shock data and seismic profile interpretation results.
Further preferably, in the step 1), the geological interpretation is subjected to data conversion by using a matlab identification function and filtering smoothing processing.
Further preferably, the geological interpretation result obtained by the step 1) comprises the formation characteristic, the formation physical characteristic and the formation lithology characteristic of the stratum.
The invention provides a geophysical modeling method based on the combination of a naive Bayes method and an adjacent classification method, which adopts the combination of the naive Bayes method and the adjacent classification method, compared with the traditional geological modeling method, the method introduces a machine learning method for modeling, saves a large amount of manpower and material resources, and leads the geological modeling to be more convenient and rapid;
in addition, the method overcomes the limitations of small area, less general geological profile of the area, more data and shallow stratum depth in the traditional geological modeling method, and realizes the modeling of large area and deep stratum under the condition of less data;
in the modeling process, subjective judgment is reduced on the basis of the existing stratum knowledge, and the method enables the result of geological modeling to be more objective and accurate.
Drawings
FIG. 1 is a graph showing the variation of the accuracy of the naive Bayes method and the proximity method with N in the method of the present invention;
FIG. 2 is a plot of a Songliao basin modeling area and a Pinke two-well longitudinal wave log in an embodiment of the method of the present invention;
FIG. 3 is a diagram illustrating the results of seismic profiling and time-depth conversion in an embodiment of the method of the present invention;
FIG. 4 is a diagram of a three-dimensional geological geophysical modeling result of the Songliao basin in the embodiment of the method.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a geophysical modeling method based on combination of a naive Bayes method and a proximity classification method, which comprises the following steps of:
1) The geological interpretation result is digitized, the geological stratification and stratum thickness of the region are known through the image recognition function of matlab, filtering smoothing and other processing, the geological interpretation result is digitized, the physical properties are extracted, the digitized result is projected into a three-dimensional space, a proximity method is used for learning the data by combining a naive Bayes method, and the stratum of an unknown region is classified;
2) K known points around one point in the three-dimensional space in the step 1), wherein K takes 7-50 points, the attributes of the K known points are extracted, the total types of the attributes are N, and the attribute of one point in the three-dimensional space is A n The attribute of the ith point around it is expressed as
Figure GDA0003844081500000041
Wherein i represents the ith point, N belongs to (1, 2 \8230; N);
3) Learning the known points according to a naive Bayes method, using the formula as follows:
Figure GDA0003844081500000042
wherein N represents any number of 1 \ 8230, and each N does not influence each other, wherein
Figure GDA0003844081500000043
Representing the property of the ith point around this known point, P (A) n ) The attribute representing this point is A n The probability of (a) of (b) being,
Figure GDA0003844081500000044
represents the attribute of the ith point as
Figure GDA0003844081500000045
The probability of (a) of (b) being,
Figure GDA0003844081500000046
representing the attribute of K points around it, respectively
Figure GDA0003844081500000047
In the case of (2), this point attribute is A n The probability of (d); through learning, A in the known point can be obtained n Probability of (A), i.e. P (A) n ) The ith point attribute is
Figure GDA0003844081500000048
Probability of (2)
Figure GDA0003844081500000049
And when this point attribute is A n When the measured data is obtained, the attributes of K points around the measured data are respectively
Figure GDA00038440815000000410
The above three probabilities:
Figure GDA0003844081500000051
as a prior probability;
4) Applying the result of step 3) to the known points remaining in step 2), the posterior probability being as follows:
Figure GDA0003844081500000052
wherein, the first and the second end of the pipe are connected with each other,
Figure GDA0003844081500000053
representing the properties of each point around the known point
Figure GDA0003844081500000054
In the case of (2), this point attribute is A n When a probability of
Figure GDA0003844081500000055
When the maximum value is taken, the attribute of the point is determined to be A n ；
5) Different K values are brought into the step 2) for learning, the results obtained in the step 4) are compared with the attributes of the known points to judge the learning accuracy, the learning accuracy at different K values is compared to select the K value K with the highest accuracy, and the prior probability obtained at the moment is the prior probability with the best learning effect
Figure GDA0003844081500000056
6) Selecting the nearest k points of each unknown point and then utilizing a prior probability formula:
Figure GDA0003844081500000057
simultaneously, a naive Bayes method formula is utilized:
Figure GDA0003844081500000058
classifying prior probability formula and naive Bayes method formula, and taking
Figure GDA0003844081500000059
Is the attribute A of the unknown point n Thus, the stratum attribute of each point is obtained, and finally, a three-dimensional geological model of the region is constructed.
The method firstly uses the known profile to carry out naive Bayes classification, then finds out N known points which are nearest to the unknown points according to the proximity method, selects a part of points with larger variance to train, then uses the rest points to train, and when half of the known points are trained and the rest are tested, the accuracy is as follows: 70% -72% (N = 50) 88% -90% (N = 10), accuracy when all known points around the nearest N points with a variance greater than 0.5 (S > 0.5) are tested (method above): 25% -35% (N = 50) 45% -55% (N = 10).
Table 1: selecting the correct value of two methods between 7 and 50 from the value range of N
Figure GDA0003844081500000061
As shown in FIG. 1, are the accuracy rate curves of N from 7-50 for both methods with variance greater than 0.5; as can be seen from table 1, the combination of the proximity method and the naive bayes method has higher accuracy than the proximity method in comparison of the two methods.
Examples
Taking the northern part of the Songliaopelvic area as an example, as shown in FIG. 2, geological information, geophysical information, drilling data, physical property data and gravity magnetic electric seismic data of the area are collected, seven seismic sections with pine depth and seven lines are interpreted to obtain accurate and reliable geological interpretation results of the underground part of the Songliaopelvic area, the seven seismic interpretation sections are digitalized, the interpretation results of the seven deep reflection seismic sections are digitalized through the image recognition function of matlab, the filtering smoothing and other processing, the interpretation results of the seven deep reflection seismic sections are digitalized, the digitalized sections have pine Wushuangliao fracture and some small structures, granite invasion and other special structures know the geological stratification, stratum thickness and physical property of the area through well drilling data, the seismic sections in the shallow part are subjected to time-depth conversion through a longitudinal wave logging curve, and the deep seismic sections are subjected to time-depth conversion through a speed structure diagram of the geological section in Suisui Fenghe river in Manchurian, so as to unify modeling units.
Firstly, converting the position of each survey line and a model construction area into geodetic coordinates, wherein the used survey lines are seven loose-depth lines, converting an interpretation result of a seismic profile into data to obtain depth data of each stratum and fault in each profile, projecting a datamation result into a three-dimensional space, learning the data by using a nearest neighbor method and a naive Bayes combined method, and classifying the stratums of an unknown area, wherein the classification process comprises the following steps:
(1) The nearest K known points (K is 7-50) around a part of the known points are searched, as shown in FIG. 3, the attributes of the stratum only change at the stratum interface, and in other places, the attributes are the same, and learning will affect the accuracy of the learning result, so we select the points located at the stratum interface for learning, and these points are characterized by large variance, so we select the points with large variance for learning and training:
(2) The known point is learned by a naive Bayes method, in the learning process, if only a few KNN methods subject to majority are adopted, the accuracy of the learning result is influenced, so that a naive Bayes method is adopted, the attribute of the point is calculated by means of prior information, namely, the point is covered with curtains with various attributes, the result is more reasonable, for example, the attribute of the point is a fourth system, only 3 points of 10 surrounding points are the fourth system, six points are the third system, and one point is granite, if the point is judged to belong to the third system according to the KNN method, the point is inconsistent with the real attribute of the point, the modeling accuracy is influenced, if the naive Bayes method is adopted, the obtained result is that the probability of the point belonging to the fourth system is the greatest under the condition, so that the point belongs to the fourth system, the method is based on the KNN as a frame, and naive Bayes is introduced in the classifying process for training, and therefore, a new classification and accuracy method is achieved;
(3) Applying the learning result to the known points which are not learned;
(4) Comparing the learning accuracy rates of different K values, and selecting the K value K with the highest accuracy rate, wherein K =11, namely the training effect obtained when K is 11 is the best;
(5) Selecting k nearest points of each known point and learning by using a naive Bayes method;
(6) And selecting the nearest k points of each unknown point, and classifying by using the learning result so as to obtain the stratum attribute of each point, and constructing a three-dimensional geological model of the northern region of the Songliao basin as shown in FIG. 4.
The earth-geology physical model of the northern region of the Songliao basin is constructed through the process, and a good foundation is laid for further researching the geological structure of the Songliao basin on the regional geological background.
By the method, three-dimensional geological geophysical modeling is carried out on the Songliao basin, and the following conclusion is obtained:
(1) The machine learning method is used for large-area three-dimensional geological-geophysical model modeling, the data volume is small, the depth is deep, the modeling is reliable, and good effect can be achieved;
(2) Through comparison, a method combining the proximity method and naive Bayes classification is found to obtain a better effect;
(3) A geological-geophysical model from the earth surface to the underground Mohuo surface of the Songliao basin is established, and a reliable basis is provided for the detection of underground geological resources of the Songliao basin and the forward inversion simulation experiment.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.

Claims (4)

1. A geophysical modeling method based on the combination of a naive Bayes method and a proximity classification method is characterized by comprising the following steps:
1) Extracting geological information in a region range, performing geological interpretation on the geological information, performing data processing on an obtained geological interpretation data result, projecting the data geological interpretation result into a three-dimensional space, and classifying unknown stratums by using a naive Bayes proximity method;
2) Retrieving K known points around one point in the three-dimensional space in the step 1), and extracting the attributes of the K known points, wherein the total types of the attributes are N, and the attribute of the point in the three-dimensional space is A n The attribute of the ith point around it is expressed as
Figure FDA0004012477680000011
Wherein i represents the ith point, N belongs to (1, 2 \8230; N);
3) Learning the points in the three-dimensional space in step 2) according to a naive Bayes method, using the formula:
Figure FDA0004012477680000012
wherein N represents any number of 1 \ 8230, and each N is not influenced with each other,
Figure FDA0004012477680000013
representing an attribute of an ith point around a point in said three-dimensional space, P (A) n ) An attribute representing a point in said three-dimensional space is A n The probability of (a) of (b) being,
Figure FDA0004012477680000014
represents the attribute of the ith point as
Figure FDA0004012477680000015
The probability of (a) of (b) being,
Figure FDA0004012477680000016
representing K known point attributes around a point in said three-dimensional space, respectively
Figure FDA0004012477680000017
The point attribute in the three-dimensional space is A n The probability of (d); through learning, A in the point in the three-dimensional space can be obtained n Probability of (A), i.e. P (A) n ) The ith known point attribute is
Figure FDA0004012477680000018
Probability of (2)
Figure FDA0004012477680000019
And when the point attribute in the three-dimensional space is A n When the data is processed, the attributes of K known points around the data are respectively
Figure FDA00040124776800000110
The above three probabilities: p (A) n )，
Figure FDA00040124776800000111
As a prior probability;
4) Applying the results of step 3) to the points in the three-dimensional space remaining in step 2), the posterior probabilities being as follows:
Figure FDA00040124776800000112
wherein the content of the first and second substances,
Figure FDA00040124776800000113
representing each known point attribute around a point in said three-dimensional space is known
Figure FDA0004012477680000021
In the case of (2), the point attribute in the three-dimensional space is A n Probability of when
Figure FDA0004012477680000022
When the maximum value is taken, determining the attribute of the point in the three-dimensional space as A n ；
5) Bringing different K values into the step 2) for learning, comparing the result obtained in the step 4) with the attributes of the points in the three-dimensional space in the step 2) to judge the learning accuracy, comparing the learning accuracy at different K values, and selecting the K value K with the highest accuracy, wherein the prior probability obtained at the moment is the prior probability P (A) with the best learning effect n )，
Figure FDA0004012477680000023
6) Selecting k points closest to the unknown points of the unknown stratum attributes in the area range and then utilizing a prior probability formula:
P(A n )，
Figure FDA0004012477680000024
simultaneously, a naive Bayes method formula is utilized:
Figure FDA0004012477680000025
classifying the prior probability formula and the naive Bayes method formula, and taking
Figure FDA0004012477680000026
Is the attribute A of the unknown point of the unknown stratum attribute in the region range n Therefore, the stratum attribute of each unknown point is obtained, and finally, a three-dimensional geological model of the region is constructed.
2. The geophysical modeling method based on a naive bayes method combined with a neighbor classification as claimed in claim 1, wherein said geological information in step 1) comprises: geophysical information, well drilling data, physical property data, gravity, magnetism, electric shock data and seismic profile interpretation results.
3. The geophysical modeling method based on the combination of the naive Bayes method and the neighbor classification as claimed in claim 1, wherein the geological interpretation in the step 1) is characterized by datamation of geological interpretation results through a matlab recognition function and a filtering smoothing process.
4. The geophysical modeling method based on the combination of the naive Bayes method and the adjacent classification method as claimed in claim 1, wherein the geological interpretation result of the step 1) data comprises the formation characteristics, the formation physical characteristics and the formation lithology characteristics of the stratum.
CN202110292022.XA 2021-03-18 2021-03-18 Geophysical modeling method based on combination of naive Bayes method and proximity classification method Active CN112906242B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110292022.XA CN112906242B (en) 2021-03-18 2021-03-18 Geophysical modeling method based on combination of naive Bayes method and proximity classification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110292022.XA CN112906242B (en) 2021-03-18 2021-03-18 Geophysical modeling method based on combination of naive Bayes method and proximity classification method

Publications (2)

Publication Number Publication Date
CN112906242A CN112906242A (en) 2021-06-04
CN112906242B true CN112906242B (en) 2023-02-28

Family

ID=76105434

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110292022.XA Active CN112906242B (en) 2021-03-18 2021-03-18 Geophysical modeling method based on combination of naive Bayes method and proximity classification method

Country Status (1)

Country Link
CN (1) CN112906242B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20150080111A (en) * 2013-12-30 2015-07-09 연세대학교 산학협력단 System and method for initial profit prediction model for overseas construction project
CN105425777A (en) * 2015-12-11 2016-03-23 渤海大学 Chemical process fault monitoring method based on active learning
CN109214025A (en) * 2017-07-06 2019-01-15 中国石油化工股份有限公司 Reservoir parameter predication method and system based on Bayes's classification
CN109577972A (en) * 2018-12-21 2019-04-05 西南石油大学 Sandy gravel materials rock mechanics parameters Logging Evaluation Method based on lithology breakdown
CN110442709A (en) * 2019-06-24 2019-11-12 厦门美域中央信息科技有限公司 A kind of file classification method based on model-naive Bayesian
CN111340057A (en) * 2018-12-19 2020-06-26 杭州海康威视数字技术股份有限公司 Classification model training method and device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110852441B (en) * 2019-09-26 2023-06-09 温州大学 Fire disaster early warning method based on improved naive Bayes algorithm

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20150080111A (en) * 2013-12-30 2015-07-09 연세대학교 산학협력단 System and method for initial profit prediction model for overseas construction project
CN105425777A (en) * 2015-12-11 2016-03-23 渤海大学 Chemical process fault monitoring method based on active learning
CN109214025A (en) * 2017-07-06 2019-01-15 中国石油化工股份有限公司 Reservoir parameter predication method and system based on Bayes's classification
CN111340057A (en) * 2018-12-19 2020-06-26 杭州海康威视数字技术股份有限公司 Classification model training method and device
CN109577972A (en) * 2018-12-21 2019-04-05 西南石油大学 Sandy gravel materials rock mechanics parameters Logging Evaluation Method based on lithology breakdown
CN110442709A (en) * 2019-06-24 2019-11-12 厦门美域中央信息科技有限公司 A kind of file classification method based on model-naive Bayesian

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于机器学习的松辽盆地三维地质-地球物理建模研究;刘乾等;《中国地球科学联合学术年会 2018》;20181231;第750-751页 *

Also Published As

Publication number Publication date
CN112906242A (en) 2021-06-04

Similar Documents

Publication Publication Date Title
CN106526693B (en) crack identification method and device
KR101618713B1 (en) 3-Dimensional Space Modeling Method based on the Geotechnical Information
CN108802812A (en) Well-seismic fusion stratum lithology inversion method
CN109085663A (en) A kind of tight sandstone reservoir stratification seam recognition methods
CN103437760B (en) Method for rapidly evaluating oil-water layer by using array induction data
CN111596978A (en) Web page display method, module and system for lithofacies classification by artificial intelligence
CN105510993A (en) Foreland basin deep buried and compressed type complex gypsum-salt rock identification and distribution prediction method
CN109541685B (en) River channel sand body identification method
CN108804728B (en) Horizontal well stratum reservoir classification analysis method and computer readable storage medium
CN103345001A (en) Ancient lake depth measuring method
CN109164485A (en) A kind of quantitative analysis method influencing low order fault accuracy of identification
CN104280771A (en) Three-dimensional seismic data waveform semi-supervised clustering method based on EM algorithm
CN108952699A (en) A kind of complicated geological drilling process formation lithology intelligent identification Method
CN107356965A (en) Reflectance factor inverting method for predicting reservoir based on weighted superposition Noise Elimination strategy
CN111767674A (en) Well logging lithology identification method based on active domain adaptation
CN111622201B (en) Soil layer equivalent shear wave velocity direct calculation method based on surface wave frequency dispersion characteristics
CN114386464B (en) Deep learning extraction method for transient electromagnetic excitation information
CN107831540A (en) The direct new method for extracting of reservoir physical parameter
CN108459358A (en) Novel method for restoring top surface morphology of sedimentary basin substrate and predicting deep ore body positioning
WO2020161518A1 (en) Method of detection of hydrocarbon horizontal slippage passages
Liu* et al. Seismic lithofacies computation method based on deep learning
CN117075221A (en) Reservoir fluid identification method based on combination of seismic elastic properties, resistivity and polarization rate
CN109459791A (en) A kind of method and system determining river location using log
CN112906242B (en) Geophysical modeling method based on combination of naive Bayes method and proximity classification method
Sarkheil et al. The fracture network modeling in naturally fractured reservoirs using artificial neural network based on image loges and core measurements

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