CN108952699A - A kind of complicated geological drilling process formation lithology intelligent identification Method - Google Patents
A kind of complicated geological drilling process formation lithology intelligent identification Method Download PDFInfo
- Publication number
- CN108952699A CN108952699A CN201810853102.6A CN201810853102A CN108952699A CN 108952699 A CN108952699 A CN 108952699A CN 201810853102 A CN201810853102 A CN 201810853102A CN 108952699 A CN108952699 A CN 108952699A
- Authority
- CN
- China
- Prior art keywords
- lithology
- ecoc
- drilling process
- value
- formula
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 230000015572 biosynthetic process Effects 0.000 title claims abstract description 25
- 238000005553 drilling Methods 0.000 title claims abstract description 21
- 230000008569 process Effects 0.000 title claims abstract description 21
- 239000011159 matrix material Substances 0.000 claims abstract description 26
- 238000005457 optimization Methods 0.000 claims abstract description 23
- 238000010606 normalization Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 2
- 238000013138 pruning Methods 0.000 abstract description 8
- 239000011435 rock Substances 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 229910052500 inorganic mineral Inorganic materials 0.000 description 4
- 239000011707 mineral Substances 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000002790 cross-validation Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 238000013524 data verification Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- TVEXGJYMHHTVKP-UHFFFAOYSA-N 6-oxabicyclo[3.2.1]oct-3-en-7-one Chemical compound C1C2C(=O)OC1C=CC2 TVEXGJYMHHTVKP-UHFFFAOYSA-N 0.000 description 1
- 241000239290 Araneae Species 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 241001074085 Scophthalmus aquosus Species 0.000 description 1
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 239000006071 cream Substances 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000013525 hypothesis research Methods 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 239000011780 sodium chloride Substances 0.000 description 1
- 238000013106 supervised machine learning method Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- 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)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mining & Mineral Resources (AREA)
- Geology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Databases & Information Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a kind of complicated geological drilling process formation lithology intelligent identification Methods, including acquisition log data, analyze the log data of acquisition, select the data being closely related with lithology, the sample data set after being denoised with wavelet de-noising techniques;It is normalized on the basis of obtaining sample data set, establish ECOC encoder matrix, and ECOC encoder matrix is deleted using concurrent Pruning algorithm, classified as base classifier to lithology using core Fisher linear discriminant, and utilizes the nuclear parameter and threshold parameter of bat optimization algorithm optimization base classifier;Final lithology classification output is completed by Hamming distance decoding.More classification problems are converted multiple two classification problems by the present invention, lithology similar in logging character can preferably be distinguished, it can overcome the problems, such as that the category information to differ greatly in the more classification methods of core Fisher linear discriminant covers the lesser category information of difference again, lay a good foundation for complicated drilling process intelligent control.
Description
Technical field
The present invention relates to geology drilling process field of intelligent control more particularly to a kind of complicated geological drilling process formation rocks
Property intelligent identification Method.
Background technique
Currently, demand of the China to large mineral products still constantly increases, existing resource imbalance between supply and demand becomes increasingly conspicuous, externally according to
Degree of depositing continues to increase.But in fact, Mineral Resources in China type is more, contain abundant.Continuous with new mineral resources is verified,
Certainty is explored and be developed into geology of deep part.
Although deep resource energy development potentiality are big, due to the bad environments of stratum depths, drilling process is frequently encountered
The various origin causes of formation such as the stratum such as sensibility shale formation, rheological formation, various salts are different, the different bad ground of feature.These
Stratum due to itself Rock-forming Mechanism is easier that undergauge occurs relative to other stratum, collapses or ruptures.As well slough mostly occurs in
In mud shale stratum, 90% or more is accounted for.Undergauge mostly occurs in saline bed, saliferous cream ooze layer, in sandstone, requires especially to adjust
Whole mud density guarantees that the borehole wall is not damaged.So identifying stratum in drilling process, real-time detection current formation when boring in fact is
Realize the basis of safe and efficient drilling.
Lithology refers to some attributes of reflection rock characteristic, such as color, ingredient, structure, construction.Lithology Discrimination is basis
Different characteristic and attribute existing for rock, classification of reservoir rock at the process of different units.Carrying out formation lithology identification is stratum
The first step of three pressure modeling, the distribution characteristics for accurately identifying subterranean strata are the passes realizing deep mineral resources and breaking through
Key.
Include formation lithology information abundant in well-log information, is the basic data of lithologic analysis.Utilize log data
Lithology Discrimination modeling method can be mainly divided into two kinds: first is that Man Graphics method;Second is that intelligent modeling method.Man Graphics side
Method identifies there are many kinds of the methods of lithology, including cross-plot, spider diagram method and staircase chart method etc..Intelligent modeling method is most
Using supervised machine learning method, the intelligent modeling method for formation lithology identification include neural network, support vector machine,
The methods of ambiguous theory, core fisher linear discriminant.Lithology Discrimination, which is carried out, using intelligent modeling method has become logging technique and brill
The important directions of well technology development.
When carrying out Lithology Discrimination using intelligent modeling method, most Research Hypothesis lithology distributions uniformly and are obtained
Log data is balanced, in fact, the unbalanced feature of data is widely present between stratum and stratum.In response to this, have
Necessity carries out unbalanced data Research on classifying method for log data.Merge the intelligent recognition of the processing method of unbalanced data
The more efficient ground-to-ground layer lithology of method energy is identified.
Summary of the invention
In view of this, the embodiment provides one kind, to establish complicated geological drilling process formation lithology adaptive
Uneven framework of classifying, can effectively improve the precision of Lithology Discrimination model, beat for complicated geological drilling process intelligent control more
Under good basis complicated geological drilling process formation lithology intelligent identification Method.
The embodiment of the present invention provides a kind of complicated geological drilling process formation lithology intelligent identification Method, including following step
It is rapid:
S1. log data is acquired, the log data of acquisition is analyzed, selects the data being closely related with lithology, is transported
Sample data set after being denoised with wavelet de-noising techniques;
S2. it is normalized on the basis of obtaining sample data set, establishes ECOC encoder matrix, and using concurrently
Pruning algorithm deletes ECOC encoder matrix;
S3. the ECOC encoder matrix obtained based on step S2, using core Fisher linear discriminant as base classifier to rock
Property classify, and using bat optimization algorithm optimization base classifier nuclear parameter and threshold parameter;
S4. final lithology classification is completed by Hamming distance decoding to export.
It further, include density log, acoustic logging, naturally electricity with the data that lithology is closely related in the step S1
Position well logging, depth, shallow laterolog 3, micronormal and microinverse.
Further, in the step S2, the formula of normalized are as follows:
In formula: xmaxWith xminThe respectively maxima and minima of characteristic, characteristic are closely related with lithology
Data, xnormFor the value after normalization, xnorm∈[0,1]。
Further, in the step S3, specifically includes the following steps:
S3.1. core fisher base classifier selects Gaussian kernel, using bat optimization algorithm, using Gmean value as fitness letter
Count the nuclear parameter K according to following formula optimization each core fisher base classifier;
Gmean=(TP/ (TP+FN) × TN/ (TN+FP))1/2
K=exp | | x-xc||^2/(2*δ^2}
In formula: TP, FP, FN, TN respectively represent positive class and correctly classify quantity, and prediction is positive class but is really negative class, in advance
Survey is negative class but is really positive class, and negative class is correctly classified quantity, and x is certain point in space, xcFor kernel function center, δ is letter
Several width parameters;
S3.2. according to following formula, using bat optimization algorithm, optimize threshold value ginseng by fitness function of Gmean value
Number λ;
μ=λ × m1+(1-λ)×m2
In formula: m1And m2It is to be divided into the average value after the projection of two classes after ECOC- core fisher is encoded, if threshold parameter λ
It changes, threshold value μ can move between mean1 and mean2.
Further, in the step S4, specifically includes the following steps:
S4.1. ECOC is calculated according to the following formula simplifies the Hamming distance encoded between predicted value;
In formula: x, y are the coding and predicted value of ECOC matrix respectively, and i is the i-th bit of ECOC coding,Indicate or;
S4.2. line number corresponding to the minimum value of H, that is, the lithology label predicted are searched for.
Compared with prior art, the invention has the following advantages:
(1) ECOC algorithm and core Fisher algorithm are combined, multiple two classification problems is converted by more classification problems, both played
The advantages of core Fisher linear discriminant, lithology similar in logging character can be preferably distinguished, and core Fisher can be overcome
The category information that differs greatly in the more classification methods of linear discriminant covers the problem of difference lesser category information;
(2) structure for optimizing ECOC coding, can obtain higher modeling accuracy, model is enable to obtain more rapidly more
Accurate effect;
(3) on the basis of establishing base classifier using ECOC coding, optimize nuclear parameter and threshold with bat optimization algorithm
Value parameter, it can be considered that the influence of unbalanced data, adaptively adjusts threshold value, and know using the practical logging data verification lithology
Other method is conducive to apply in actual production.
Detailed description of the invention
The well log of log data is acquired in Fig. 1 one embodiment of the invention.
The flow chart of concurrent Pruning algorithm in Fig. 2 one embodiment of the invention.
Original ECOC encoder matrix schematic diagram in Fig. 3 one embodiment of the invention.
ECOC encoder matrix precedence diagram is deleted in Fig. 4 one embodiment of the invention.
ECOC-BAKFDA experimental configuration schematic diagram in Fig. 5 one embodiment of the invention.
ECOC encoder matrix length and classification accuracy relational graph in Fig. 6 one embodiment of the invention.
KFDA algorithm One Dimensional Projection distribution map in Fig. 7 one embodiment of the invention.
An a kind of flow chart of complicated geological drilling process formation lithology intelligent identification Method of Fig. 8 present invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is further described.
Referring to FIG. 8, the embodiment provides a kind of complicated geological drilling process formation lithology intelligent recognition sides
Method, comprising the following steps:
S1. log data is acquired as shown in Figure 1, and the log data of acquisition is analyzed, selects and is closely related with lithology
Data, the sample data set after being denoised with wavelet de-noising techniques;It include that density is surveyed with the data that lithology is closely related
Well, acoustic logging, nutural potential logging, depth, shallow laterolog 3, micronormal and microinverse.
S2. it is normalized on the basis of obtaining sample data set, establishes ECOC encoder matrix (such as Fig. 3 institute
Show), and ECOC encoder matrix (as shown in Figure 4) is deleted using concurrent Pruning algorithm;
The formula of normalized are as follows:
In formula: xmaxWith xminThe respectively maxima and minima of characteristic, characteristic are closely related with lithology
Data, xnormFor the value after normalization, xnorm∈[0,1]。
As shown in Fig. 2, concurrent Pruning algorithm:
List all combinations of ECOC coding;
With training data training base classifier;
To each base classifier Ci;
To each input sample;
If assembled classifier classification error and base classifier classification error;
Metrici=Metrici -2;
If assembled classifier classification error and base classifier are classified correct;
Metrici=Metrici+2;
If assembled classifier classification is correct and the classification of base classifier is correct;
Metrici=Metrici+1;
It is ranked up according to Metrici value, selects the highest base classifier of Metrici value as selected first base point
Class device;
To each remaining base classifier;
Select the highest preceding a base classifier of b=1/2 (adjustable) of Metrici value, calculate itself and selected base classifier it
Between separating degree k;
H represents the different numbers of each base classifier coding;
Select the base classifier for possessing highest k value in remaining base classifier;
Circulation is until selected base classifier reaches requirement number;
S3. the ECOC encoder matrix obtained based on step S2, using core Fisher linear discriminant as base classifier to rock
Property classify, and using bat optimization algorithm optimization base classifier nuclear parameter and threshold parameter;
Specifically includes the following steps:
S3.1. core fisher base classifier selects Gaussian kernel, using bat optimization algorithm, using Gmean value as fitness letter
Count the nuclear parameter K according to following formula optimization each core fisher base classifier;
Gmean=(TP/ (TP+FN) × TN/ (TN+FP))1/2
K=exp | | x-xc||^2/(2*δ^2}
In formula: TP, FP, FN, TN respectively represent positive class and correctly classify quantity, and prediction is positive class but is really negative class, in advance
Survey is negative class but is really positive class, and negative class is correctly classified quantity, and x is certain point in space, xcFor kernel function center, δ is letter
Several width parameters;;
S3.2. according to following formula, using bat optimization algorithm, optimize threshold value ginseng by fitness function of Gmean value
Number λ;
μ=λ × m1+(1-λ)×m2
In formula: m1And m2It is to be divided into the average value after the projection of two classes after ECOC- core fisher is encoded, if threshold parameter λ
It changes, threshold value μ can move between mean1 and mean2.
S4. final lithology classification is completed by Hamming distance decoding to export.
Specifically includes the following steps:
S4.1. ECOC is calculated according to the following formula simplifies the Hamming distance encoded between predicted value;
In formula: x, y are the coding and predicted value of ECOC matrix respectively, and i is the i-th bit of ECOC coding,Indicate or;
S4.2. line number corresponding to the minimum value of H, that is, the lithology label predicted are searched for.
Present invention combination ECOC algorithm and core Fisher algorithm, convert multiple two classification problems for more classification problems, both
The advantages of having played core Fisher linear discriminant can preferably distinguish lithology similar in logging character, and can overcome core
The category information that differs greatly in the more classification methods of Fisher linear discriminant covers the problem of difference lesser category information;It optimizes
The structure of ECOC coding, can obtain higher modeling accuracy, model is enable to obtain more rapidly more accurate effect;It is utilizing
ECOC coding is established on the basis of base classifier, optimizes nuclear parameter and threshold parameter with bat optimization algorithm, it can be considered that not
The influence of equalization data adaptively adjusts threshold value, and utilizes the practical logging data verification Lithology Identification Methods, is conducive in reality
It is applied in the production of border.
Embodiment 1
A kind of complicated geological drilling process formation lithology intelligent identification Method, firstly, to the log data obtained by Fig. 1
It is analyzed, selects the data being closely related with lithology, the sample data set after being denoised with wavelet de-noising techniques will obtain
Input of the 7 kinds of parameters obtained as formation lithology identification model establishes encoder matrix with improved ECOC algorithm, and utilizes core
Fisher linear discriminant method establishes base classifier, while utilizing bat algorithm optimization nuclear parameter and threshold parameter.The method makes
Simulating, verifying is carried out with practical logging data.Specific step is as follows:
(1) raw sample data is by density log, acoustic logging, nutural potential logging, depth, shallow laterolog 3, micro- electricity
Totally 153 groups of log datas form with seven kinds of micro gradient logging for position, share 4 kinds of different lithologies, are indicated respectively with label 1,2,3,4,
Ratio is respectively 35:52:50:16, and lithology data is unbalanced.Table 1 is part initial data, is obtained with one-dimensional wavelet de-noising techniques
Sample data set after must denoising;
(2) step (1) obtained 7 kinds of parameters totally 132 groups of sample data sets are based on, takes in four kinds of lithology and arbitrarily compares respectively
The data for the number that example is 15:25:25:5 are training set, the training set for taking the data of remaining any 10:10:10:4 and being chosen
Intersection as verifying collect measure classifier accuracy, using remaining data as test set.Establish the ECOC coding of 4 kinds of lithology
It is as shown in Figure 2 to delete ECOC matrix Pruning algorithm flow chart using concurrent Pruning algorithm for matrix.Wherein, it is deleted according to Metric value
The length and accuracy rate relationship of the ECOC matrix subtracted are as shown in Figure 6.Fig. 3 is original ECOC encoder matrix combination.Fig. 4 is to delete
Later ECOC encoder matrix.For each column of encoder matrix, it is converted to one two classification.ECOC after for deleting
The classification that code 1 represents can be regarded as positive class by coding, and the classification that code -1 represents regards negative class as, and the classification sample that code 0 represents is ignored.
The original log parameter table in 1 part of table
(3) the ECOC encoder matrix obtained based on step (2), using core Fisher linear discriminant as base classifier over the ground
Layer lithology is classified, and using Gmean value as objective function, using bat optimization algorithm optimize base classifier nuclear parameter and
Threshold parameter;
(4) it is decoded to obtain final result using Hamming distance.And its accuracy rate is calculated with ten folding cross validations
Average value, as shown in Figure 5.Pruning algorithm improvement is not carried out the IECOC-BAKFDA method of proposition and the more classification methods of KFDA and
ECOC-BAKFDA algorithm and the ECOC-KFDA algorithm that is not optimised compare, be as a result separately recorded in table 2-4.It is proposed
Method is superior to other methods on nicety of grading, generalization ability and predetermined speed;
Table 2 does not utilize the BA-KFDA confusion matrix of Error Correction of Coding
Final discrimination: 76%.
Table 3 utilizes the formation lithology recognition accuracy (ten folding cross validations) of bat algorithm optimization
Table 4 does not utilize the formation lithology recognition result (ten folding cross validations) of bat algorithm optimization
In the absence of conflict, the feature in embodiment and embodiment herein-above set forth can be combined with each other.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (5)
1. a kind of complicated geological drilling process formation lithology intelligent identification Method, which comprises the following steps:
S1. log data is acquired, the log data of acquisition is analyzed, selects the data being closely related with lithology, utilization is small
Wave noise-removed technology denoised after sample data set;
S2. it is normalized on the basis of obtaining sample data set, establishes ECOC encoder matrix, and utilize and concurrently delete
Algorithm deletes ECOC encoder matrix;
S3. the ECOC encoder matrix obtained based on step S2, using core Fisher linear discriminant as base classifier to lithology into
Row classification, and utilize the nuclear parameter and threshold parameter of bat optimization algorithm optimization base classifier;
S4. final lithology classification is completed by Hamming distance decoding to export.
2. complicated geological drilling process formation lithology intelligent identification Method according to claim 1, which is characterized in that described
It include density log, acoustic logging, nutural potential logging, depth, shallow three side to survey with the data that lithology is closely related in step S1
Well, micronormal and microinverse.
3. complicated geological drilling process formation lithology intelligent identification Method according to claim 1, which is characterized in that described
In step S2, the formula of normalized are as follows:
In formula: xmaxWith xminThe respectively maxima and minima of characteristic, characteristic are the numbers being closely related with lithology
According to xnormFor the value after normalization, xnorm∈ [0,1].
4. complicated geological drilling process formation lithology intelligent identification Method according to claim 1, which is characterized in that described
In step S3, specifically includes the following steps:
S3.1. core fisher base classifier selects Gaussian kernel, using bat optimization algorithm, using Gmean value as fitness function root
According to the nuclear parameter K of following formula optimization each core fisher base classifier;
Gmean=(TP/ (TP+FN) × TN/ (TN+FP))1/2
K=exp | | x-xc||^2/(2*δ^2}
In formula: TP, FP, FN, TN respectively represent positive class and correctly classify quantity, and prediction is positive class but is really negative class, are predicted as
Negative class is still really positive class, and negative class is correctly classified quantity, and x is certain point in space, xcFor kernel function center, δ is function
Width parameter;
S3.2. according to following formula, using bat optimization algorithm, optimize threshold parameter λ by fitness function of Gmean value;
μ=λ × m1+(1-λ)×m2
In formula: m1And m2It is to be divided into the average value after the projection of two classes after ECOC- core fisher is encoded, if threshold parameter λ occurs
Variation, threshold value μ can move between mean1 and mean2.
5. complicated geological drilling process formation lithology intelligent identification Method according to claim 1, which is characterized in that described
In step S4, specifically includes the following steps:
S4.1. ECOC is calculated according to the following formula simplifies the Hamming distance encoded between predicted value;
In formula: x, y are the coding and predicted value of ECOC matrix respectively, and i is the i-th bit of ECOC coding,Indicate or;
S4.2. line number corresponding to the minimum value of H, that is, the lithology label predicted are searched for.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810853102.6A CN108952699B (en) | 2018-07-30 | 2018-07-30 | Intelligent stratum lithology identification method in complex geological drilling process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810853102.6A CN108952699B (en) | 2018-07-30 | 2018-07-30 | Intelligent stratum lithology identification method in complex geological drilling process |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108952699A true CN108952699A (en) | 2018-12-07 |
CN108952699B CN108952699B (en) | 2021-07-13 |
Family
ID=64466460
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810853102.6A Active CN108952699B (en) | 2018-07-30 | 2018-07-30 | Intelligent stratum lithology identification method in complex geological drilling process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108952699B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109740636A (en) * | 2018-12-12 | 2019-05-10 | 华北电力大学 | Formation oil water layer recognition methods based on AdaBoost combination GRU |
CN112085109A (en) * | 2020-09-14 | 2020-12-15 | 电子科技大学 | Phase-controlled porosity prediction method based on active learning |
CN112612935A (en) * | 2020-12-16 | 2021-04-06 | 浙大城市学院 | Complete logging data acquisition method based on self-reasoning model |
CN112686259A (en) * | 2020-12-16 | 2021-04-20 | 中国石油大学(北京) | Rock image intelligent identification method and device based on deep learning and storage medium |
CN112766309A (en) * | 2020-12-30 | 2021-05-07 | 中国地质调查局成都地质调查中心 | Lithology identification method and system |
CN113047859A (en) * | 2021-04-12 | 2021-06-29 | 上海应用技术大学 | Shield tunneling parameter control method based on local Fisher soil layer identification |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060074825A1 (en) * | 2003-01-24 | 2006-04-06 | Piotr Mirowski | System and method for inferring geological classes |
US8374974B2 (en) * | 2003-01-06 | 2013-02-12 | Halliburton Energy Services, Inc. | Neural network training data selection using memory reduced cluster analysis for field model development |
CN105243428A (en) * | 2015-09-07 | 2016-01-13 | 天津市市政工程设计研究院 | Bus arrival time prediction method through optimizing support vector machine based on bat algorithm |
CN105550426A (en) * | 2015-12-08 | 2016-05-04 | 东北大学 | Multi-scale binary tree blast furnace fault diagnosis method based on sample segmentation |
CN107272705A (en) * | 2017-07-31 | 2017-10-20 | 中南大学 | A kind of multiple neural network controlling planning method of robot path under intelligent environment |
CN107766880A (en) * | 2017-10-09 | 2018-03-06 | 江苏大学 | The flexible measurement method of Fermentation by Photosynthetic Bacteria process key parameter based on BA LSSVM |
-
2018
- 2018-07-30 CN CN201810853102.6A patent/CN108952699B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8374974B2 (en) * | 2003-01-06 | 2013-02-12 | Halliburton Energy Services, Inc. | Neural network training data selection using memory reduced cluster analysis for field model development |
US20060074825A1 (en) * | 2003-01-24 | 2006-04-06 | Piotr Mirowski | System and method for inferring geological classes |
CN105243428A (en) * | 2015-09-07 | 2016-01-13 | 天津市市政工程设计研究院 | Bus arrival time prediction method through optimizing support vector machine based on bat algorithm |
CN105550426A (en) * | 2015-12-08 | 2016-05-04 | 东北大学 | Multi-scale binary tree blast furnace fault diagnosis method based on sample segmentation |
CN107272705A (en) * | 2017-07-31 | 2017-10-20 | 中南大学 | A kind of multiple neural network controlling planning method of robot path under intelligent environment |
CN107766880A (en) * | 2017-10-09 | 2018-03-06 | 江苏大学 | The flexible measurement method of Fermentation by Photosynthetic Bacteria process key parameter based on BA LSSVM |
Non-Patent Citations (3)
Title |
---|
王兴振: "热红外高光谱(TASI)数据岩性识别方法研究——以甘肃柳园花黑滩研究区为例", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
魏峻: "基于蝙蝠算法的支持向量机参数优化", 《暴击文理学院学报(自然科学版)》 * |
魏聪等: "最小二乘支持向量机在储层流体识别中的应用", 《石油天然气学报(江汉石油学院学报)》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109740636A (en) * | 2018-12-12 | 2019-05-10 | 华北电力大学 | Formation oil water layer recognition methods based on AdaBoost combination GRU |
CN112085109A (en) * | 2020-09-14 | 2020-12-15 | 电子科技大学 | Phase-controlled porosity prediction method based on active learning |
CN112612935A (en) * | 2020-12-16 | 2021-04-06 | 浙大城市学院 | Complete logging data acquisition method based on self-reasoning model |
CN112686259A (en) * | 2020-12-16 | 2021-04-20 | 中国石油大学(北京) | Rock image intelligent identification method and device based on deep learning and storage medium |
CN112612935B (en) * | 2020-12-16 | 2022-03-29 | 浙大城市学院 | Complete logging data acquisition method based on self-reasoning model |
CN112686259B (en) * | 2020-12-16 | 2023-09-26 | 中国石油大学(北京) | Rock image intelligent recognition method and device based on deep learning and storage medium |
CN112766309A (en) * | 2020-12-30 | 2021-05-07 | 中国地质调查局成都地质调查中心 | Lithology identification method and system |
CN113047859A (en) * | 2021-04-12 | 2021-06-29 | 上海应用技术大学 | Shield tunneling parameter control method based on local Fisher soil layer identification |
Also Published As
Publication number | Publication date |
---|---|
CN108952699B (en) | 2021-07-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108952699A (en) | A kind of complicated geological drilling process formation lithology intelligent identification Method | |
CN109736790A (en) | Recognition of Weil Logging Lithology method based on deepness belief network | |
Yan et al. | A real-time intelligent lithology identification method based on a dynamic felling strategy weighted random forest algorithm | |
CN113642698B (en) | Geophysical well logging intelligent interpretation method, system and storage medium | |
CN111596978A (en) | Web page display method, module and system for lithofacies classification by artificial intelligence | |
CN102830441A (en) | Comprehensive interpretation and evaluation method of hydrocarbon reservoir including data mining | |
CN104047598A (en) | Method for predicating productivity of nonhomogeneity ancient karst carbonate reservoir | |
US11727583B2 (en) | Core-level high resolution petrophysical characterization method | |
CN112712025A (en) | Complex lithology identification method based on long-term and short-term memory neural network | |
Xueqing et al. | Bi-LSTM deep neural network reservoir classification model based on the innovative input of logging curve response sequences | |
CN111767674A (en) | Well logging lithology identification method based on active domain adaptation | |
CN105986811A (en) | Five-dimension lithologic character judgment and recognition method based on conventional logging data | |
CN109613623A (en) | A kind of lithology prediction method based on residual error network | |
Lu et al. | Lithology identification using graph neural network in continental shale oil reservoirs: A case study in Mahu Sag, Junggar Basin, Western China | |
Brown et al. | Machine learning on Crays to optimize petrophysical workflows in oil and gas exploration | |
Chang et al. | SegLog: Geophysical logging segmentation network for lithofacies identification | |
Guoqiang et al. | Construction of well logging knowledge graph and intelligent identification method of hydrocarbon-bearing formation | |
Cheng et al. | Petroleum lithology discrimination based on PSO-LSSVM classification model | |
CN116796231A (en) | Method and device for automatically dividing lithology and comparing lithology based on logging curve | |
Emelyanova et al. | Detecting specific facies in well-log data sets using knowledge-driven hierarchical clustering | |
Fan et al. | Semi-supervised learning–based petrophysical facies division and “Sweet Spot” identification of low-permeability sandstone reservoir | |
Palmer | Using AI and Machine Learning to Indicate Shale Anisotropy and Assist in Completions Design | |
US11719851B2 (en) | Method and system for predicting formation top depths | |
Kuchin et al. | Machine learning methods for identifying composition of uranium deposits in Kazakhstan | |
Kakouei et al. | Lithological facies identification in Iranian largest gas field: A comparative study of neural network methods |
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 |