CN104992178A - Tight sandstone fluid type identification method based on support vector machine simulation cross plot - Google Patents
Tight sandstone fluid type identification method based on support vector machine simulation cross plot Download PDFInfo
- Publication number
- CN104992178A CN104992178A CN201510333998.1A CN201510333998A CN104992178A CN 104992178 A CN104992178 A CN 104992178A CN 201510333998 A CN201510333998 A CN 201510333998A CN 104992178 A CN104992178 A CN 104992178A
- Authority
- CN
- China
- Prior art keywords
- support vector
- vector machine
- fluid type
- plot
- machine simulation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a tight sandstone fluid type identification method based on a support vector machine simulation cross plot. Parameters best reflecting oil gas and water layer characteristics in a reservoir layer are firstly selected to be taken as fluid sample data; a support vector machine algorithm is applied then, and a proper penalty parameter and a proper kernel function are selected to construct a classification prediction model; vector data points obtained by the classification prediction model are projected to a cross plot plane so that a support vector machine simulation cross plot plate is formed; and to-be-recognized sample data are input into the support vector machine simulation cross plot plate to form projection distribution points, distances between the various projection distribution points and various fluid type center points in the support vector machine simulation cross plot are calculated, and the fluid type represented by a shortest-distance center point is taken as a to-be-identified fluid type. The method can integrate multiple characteristic parameters of research objects to perform accurate division on types of the objects, and limitation that the cross plot only can integrate two characteristic parameters to perform type division is made up.
Description
Technical field
The present invention relates to a kind of tight sand fluid type recognition methods based on support vector machine simulation X plot, belong to technical field.
Background technology
Tight sand hydrocarbon-bearing pool hides type as a kind of important continuity unconventionaloil pool, is distributed widely in each basin of China, has good fair exploration.But because reservoir permeability is low, porosity type is various, nonuniformity is strong, electrical property feature is caused to affect larger by physical property and rock skeleton, logging response character is complicated, conventional X plot method identification oil, gas, water layer difficulty are large, easy omission is wrong identification tight sand reservoir fluid type even, affects the computational accuracy of tight sand hydrocarbon-bearing pool reserves.
X plot recognition technology is widely used in oil-gas exploration, it in inspection logging data quality, select interpretation parameters, determine lithology, inspection explanation results and evaluating in formation fluid type plays an important role.In petrophysics, rock physics template can be made by X plot, utilize template to carry out lithology prediction; In seismic AVO (Amplitude Versus Offset, the change of amplitude offset distance) technology application aspect, by crossplot technique, by AVO attribute (λ ρ-μ ρ, I
p-I
sdeng) project on X plot, utilize the lithology of different reservoir and the abnormal feature of occupying zones of different in X plot plane of fluid type, carry out dividing anomaly; In reservoir and fluid explanation, crossplot technique not only can be used for attribute optimization, can be applied to that Reservoir type divides, the qualitative recognition of the evaluation of reservoir logging in water flooded layer and reservoir fluid simultaneously.Tradition X plot is generally select the two kind parameters relevant to research object or attribute, and XY coordinate plane builds interpretation chart, and this method just seems unable to do what one wishes when the attribute of object is more, lacks operability and accuracy.Meanwhile, in traditional crossplot analysis and its, rough description is generally adopted to the division of X plot inner region or manually sketches, there is very large uncertainty in the of method own, particularly when data point is more, the sample point of different attribute overlaps, and is difficult to judge the classification belonging to sample point quickly and accurately and identify.
Summary of the invention
The object of this invention is to provide a kind of tight sand fluid type recognition methods based on support vector machine simulation X plot, be difficult to the accurately quick problem that classification belonging to sample point is identified when data point is more to solve traditional X plot.
The present invention for solve the problems of the technologies described above and provide a kind of based on support vector machine simulation X plot the recognition methods of tight sand fluid type, this recognition methods comprises the following steps:
1) select to reflect that the parameter of oil-water-layer feature and fluid type corresponding to oil test data are as fluid sample the parameter calculated from well logging or logging trace;
2) build supporting vector machine model, according to selected fluid sample, supporting vector machine model is trained, to obtain the support vector cassification forecast model of error in allowed band;
3) to form support vector machine simulation crossplot on the vector data spot projection obtained by support vector cassification forecast model to X plot plane, and the central point of phasor as each fluid type in support vector machine simulation X plot is exported using the target of forecast model of classifying;
4) sample data to be identified is input in formed support vector machine simulation crossplot to form projective distribution point, calculate each projective distribution point and support vector machine and simulate distance in X plot between each fluid type central point, the shortest fluid type representated by central point of distance is the fluid type of sample to be identified.
Described step 2) in adopt cross-validation method to choose punishment parameter and the kernel function of supporting vector machine model when building supporting vector machine model.
Described cross-validation method refers to and is divided into groups by original input data, a part is as training set, another part is as checking collection, with training set, sorter is trained, the model of training and obtaining tested by recycling checking collection, performance index using the classification accuracy obtained as classification of assessment device, choose optimum punishment parameter and kernel function with this.
Described support vector cassification forecast model is dual output layer.
Described step 2) in when training supporting vector machine model, need fluid sample to be normalized.
During described normalized fluid sample property value be normal distribution carry out conventional normalization, fluid sample property value be skewed distribution carry out logarithm normalization.
Conveniently support vector cassification forecast model exports and shows in the plane, need carry out digitizing to support vector cassification forecast model Output rusults.
The invention has the beneficial effects as follows: first the present invention selects to reflect that the parameter of oil-gas-water layer feature in reservoir is as fluid sample data; Then use algorithm of support vector machine, select suitable punishment parameter and kernel function to build classification forecast model; The vector data spot projection obtained by classification forecast model simulates crossplot to form support vector machine on X plot plane; Sample data to be identified is input in formed support vector machine simulation crossplot to form projective distribution point, calculate each projective distribution point and support vector machine and simulate distance in X plot between each fluid type central point, and using the fluid type representated by the shortest central point of distance as fluid type to be identified.Algorithm of support vector machine and crossplot chart recognition technology have been merged in the present invention, transform traditional X plot, achieve the identification to tight sand multiple fluid type.The supporting vector machine model that the present invention utilizes sample training to obtain, can the classification of various features parameter to object of synthetic study object accurately divide, compensate for X plot can only comprehensive two kinds of characteristic parameters to carry out the limitation of category division.
And the present invention adopt support vector machine carry out tight sand fluid type classification prediction time highly versatile, can in very wide various collections of functions constructed fuction; Robustness is good, does not need fine setting; Calculate simple, only need to utilize simple optimisation technique; Perfect in theory, based on the framework of VC generalization theory.
Accompanying drawing explanation
Fig. 1 is the architectural schematic of support vector machine;
Fig. 2 support vector machine simulation X plot algorithm schematic diagram;
Fig. 3 cross validation point-score parameter optimization two dimensional surface result schematic diagram;
Fig. 4 cross-validation method parameter optimization 3 D stereo result schematic diagram;
Fig. 5 support vector machine simulation X plot fluid identification plate schematic diagram;
Fig. 6 support vector machine simulation crossplot classification fluid type of reservoir through result schematic diagram;
Fig. 7 Sandstone Gas Reservoir fluid type recognition result schematic diagram.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is further described.
Support vector machine (SVM, Support Vector Machine) is first proposed by Vapnik, and its main thought sets up an Optimal Separating Hyperplane as decision-making curved surface, and the isolation edge between positive example and counter-example is maximized.The theoretical foundation of support vector machine is Statistical Learning Theory, and or rather, support vector machine is the approximate realization of structural risk minimization.This principle is based on such fact: the error rate of Learning machine in test data (i.e. extensive error rate) depend on training error rate and one item of VC dimension (Vapnik-Chervonenkisdimension) and for boundary, can in merotype situation, support vector machine is zero for the value of last item, and Section 2 is minimized.Therefore, although the field internal problem of its not Utilizing question, the Generalization Capability that support vector function provides on pattern classification problem, this attribute is that support vector machine is distinctive.The architecture of support vector machine as shown in Figure 1, X in figure
1, X
2..., X
nthe input value of model, Y
1, Y
2be the predicted value of model, K is kernel function.As can be seen from the figure, when supporting vector machine model can regard a nonlinear function m as, algorithm just have expressed the Function Mapping relation from n independent variable to m dependent variable, therefore, can solve the type identification problem of multiparameter (attribute) object with it.
From statistics viewpoint, well logging fluid identification is actually a classification forecasting problem, algorithm of support vector machine can be adopted to set up disaggregated model come, therefore the invention provides a kind of tight sand fluid type recognition methods based on support vector machine simulation X plot, as shown in Figure 2, the method merges algorithm of support vector machine and crossplot chart recognition technology, transforms, achieve the identification to tight sand multiple fluid type to traditional X plot.First select to reflect that the parameter of oil-water-layer feature and fluid type corresponding to oil test data are as fluid sample the parameter calculated from well logging or logging trace; Then build supporting vector machine model, according to selected fluid sample, supporting vector machine model is trained, to obtain the support vector cassification forecast model of error in allowed band; To form support vector machine simulation crossplot on the vector data spot projection obtained by support vector cassification forecast model to X plot plane, and export the central point of phasor as each fluid type in support vector machine simulation X plot using the target of forecast model of classifying; Finally sample data to be identified is input in formed support vector machine simulation crossplot to form projective distribution point, calculate each projective distribution point and support vector machine and simulate distance in X plot between each fluid type central point, the shortest fluid type representated by central point of distance is the fluid type of sample to be identified.
The present invention props up provided vector machine of holding and simulates the ultimate principle that X plot recognition methods combines algorithm of support vector machine and X plot identification, and the key of method is the degree of accuracy of model construction and the distance exam of X plot data point.Forecast model input end is multiparameter (many logging trace), output terminal is arranged to the (x of two dimension, y), the effect of model is exactly set up mapping relations between the two, four kinds of corresponding four kinds of fluid types of fluid type, make input end export bivector after model, reprojection is to plane.
The specific implementation process of the method is as follows:
1. the choosing of sample data
In well logging interpretation, the authenticity of learning sample, representativeness and generalization are the keys determining classifying quality, sample data in the present invention is formed primarily of formation testing result and one group of corresponding with it log response value, select to reflect that the parameter of oil-water-layer feature and fluid type corresponding to oil test data are as fluid sample, using this sample as the input vector of supporting vector machine model and target phasor the parameter that the present embodiment calculates from well logging or logging trace.
2. the process of sample data
Negative effect is caused to predicting the outcome for avoiding each parameter dimension difference, the property value of the learning sample as input vector is normalized, as natural gamma curve, resistivity curve, factor of porosity, oil saturation etc., property value be normal distribution carry out conventional normalization, here conventional normalization can adopt minimax normalization mode, as f:x → y=(x-xmin)/(xmax-xmin), normalized effect is that raw data is arrived in [0,1] scope by regular; Property value be skewed distribution carry out logarithm normalization.
The output layer of support vector cassification forecast model is two-layer, output for the ease of forecast model is projected on two dimensional surface coordinate system, digitizing need be carried out to support vector cassification forecast model Output rusults, oil reservoir code is set as (0.25 in this enforcement, 0.25), oil-water common-layer code is set to (0.75, 0.25), water layer is set to (0.75, 0.75), dried layer is set to (0.25, 0.75), using these four fluid centers as object vector, namely the object of model maps on these aspects, but it is final due to global error, oil reservoir logging trace sample through the possibility of result that model calculation obtains be (0.2, 0.25), just not necessarily (0.25, 0.25).
3. structure supporting vector machine model
According to the structure of system inputoutput data the characteristic study algorithm of support vector machine, utilize punishment parameter and the kernel function of cross-validation method Support Vector Machines Optimized, as shown in Figure 3 and Figure 4, original input data divides into groups by this process, a part is as training set, another part is as checking collection, with training set, sorter is trained, the model of training and obtaining tested by recycling checking collection, performance index using the classification accuracy obtained as classification of assessment device, screen optimum SVM with this and punish parameter and kernel function.In training set, sample is used to make model, this sample point is many, and test set is last model carry out after, verification model accuracy, sample data selected by utilization is trained support vector cassification forecast model, in the training process according to the precision of model predictive error adjustment model, as shown in Figure 5.
4. testing classification forecast model
The disaggregated model set up is utilized to predict the fluid sample obtained by formation testing result and well-log information and error is returned and sentenced, the classifying quality of test model.
5. make support vector machine simulation X plot plate, utilize the support vector cassification forecast model trained, to classify the fluid flow parameter of unknown fluid type layer position, and be projected in planimetric coordinates and fasten, use Euclidean distance measuring and calculating subpoint apart from the distance at each fluid center, obtain the fluid type of this layer of position, as shown in Figure 6 and Figure 7.
In order to verify that the present invention can the classification of various features parameter to object of synthetic study object accurately divide, be described for the recessed band tight sand of Biyang Sag ring below.
Biyang Sag ring recessed band tight sand oil and gas reserves scale is estimated at about 2,000 ten thousand tons, and wherein the longitudinal oil-bearing strata position of peace canopy Series of Deep tight sand is many, and continuity distributes, and plane distribution area is large, shape distribution in flakes.But because Sandstone Gas Reservoir permeability is low, porosity type is various, nonuniformity is strong, electrical property feature is caused to affect larger by physical property and rock skeleton, logging response character is complicated, and conventional logging crossplot method identification oil, gas, water layer difficulty are comparatively greatly, low with field conduct result rate of coincideing.
The log data that the support vector machine simulation crossplot utilizing the present invention to set up is corresponding to peace canopy Series of Deep tight sand 20 well layer depth segment carries out back sentencing, as shown in table 1, by analysis, the number of the sample that this result conforms to the formation testing means of production is 19, what be not inconsistent is 2, recognition result is corresponding with formation testing conclusion very well, accuracy reaches 90%, show that the method can be good at differentiating the fluid type in Sandstone Gas Reservoir, compared with traditional X plot method, there is obvious advantage.
The differentiation result of table 120 check sample
| Sequence number | Well-name | SH | DT | POR | RT | DEN | Formation testing conclusion | Well logging interpretation | SVM explains |
| 1 | AN3003 | 10.8 | 181.1 | 4.0 | 132.9 | 2.6 | Dried layer | Dried layer | Dried layer |
| 2 | AN2050 | 11.3 | 182.2 | 4.2 | 82.1 | 2.6 | Dried layer | Dried layer | Dried layer |
| 3 | AN2160 | 11.5 | 200.7 | 3.2 | 65.2 | 2.6 | Dried layer | Water layer | Dried layer |
| 4 | B212 | 5.7 | 182.1 | 4.0 | 286.1 | 2.6 | Dried layer | Dried layer | Dried layer |
| 5 | B213 | 6.9 | 182.1 | 3.8 | 203.3 | 2.6 | Dried layer | Oil reservoir | Dried layer |
| 6 | AN2071 | 6.6 | 216.7 | 8.8 | 10.8 | 2.6 | Water layer | Water layer | Water layer |
| 7 | B353 | 10.5 | 207.7 | 6.5 | 30.1 | 2.7 | Water layer | Dried layer | Water layer |
| 8 | B116 | 5.6 | 203.5 | 6.1 | 29.7 | 2.7 | Water layer | Water layer | Water layer |
| 9 | AN3002 | 6.7 | 199.8 | 6.8 | 38.2 | 2.4 | Water layer | Water layer | Water layer |
| 10 | AN3006 | 5.3 | 186.9 | 3.6 | 66.8 | 2.5 | Water layer | Dried layer | Water layer |
| 11 | B255 | 7.3 | 188.8 | 4.1 | 208.4 | 2.4 | Oil reservoir | Dried layer | Oil reservoir |
| 12 | AN84 | 8.5 | 192.3 | 5.5 | 165.3 | 2.4 | Oil reservoir | Oil reservoir | Oil reservoir |
| 13 | AN2019 | 8.8 | 199.8 | 4.0 | 300.2 | 2.3 | Oil reservoir | Dried layer | Oil reservoir |
| 14 | AN2031 | 4.4 | 194.0 | 6.1 | 289.2 | 2.6 | Oil reservoir | Oil reservoir | Oil reservoir |
| 15 | AN89 | 27.9 | 206.9 | 5.9 | 186.7 | 2.6 | Oil reservoir | Oil reservoir | Oil reservoir |
| 16 | AN2028 | 16.2 | 203.8 | 4.5 | 148.2 | 2.4 | Oil-water common-layer | Oil reservoir | Oil-water common-layer |
| 17 | AN2051 | 25.9 | 193.4 | 4.2 | 210.6 | 2.6 | Oil-water common-layer | Oil reservoir | Oil reservoir |
| 18 | AN29 | 28.7 | 209.5 | 4.6 | 115.0 | 2.2 | Oil-water common-layer | Oil-water common-layer | Oil-water common-layer |
| 19 | B246 | 13.3 | 198.6 | 7.2 | 166.4 | 2.5 | Oil-water common-layer | Oil reservoir | Oil reservoir |
| 20 | B254 | 7.9 | 190.9 | 4.4 | 124.6 | 2.7 | Oil-water common-layer | Oil-water common-layer | Oil-water common-layer |
Claims (7)
1., based on a tight sand fluid type recognition methods for support vector machine simulation X plot, it is characterized in that, this recognition methods comprises the following steps:
1) select to reflect that the parameter of oil-water-layer feature and fluid type corresponding to oil test data are as fluid sample the parameter calculated from well logging or logging trace;
2) build supporting vector machine model, according to selected fluid sample, supporting vector machine model is trained, to obtain the support vector cassification forecast model of error in allowed band;
3) to form support vector machine simulation crossplot on the vector data spot projection obtained by support vector cassification forecast model to X plot plane, and the central point of phasor as each fluid type in support vector machine simulation X plot is exported using the target of forecast model of classifying;
4) sample data to be identified is input in formed support vector machine simulation crossplot to form projective distribution point, calculate each projective distribution point and support vector machine and simulate distance in X plot between each fluid type central point, the shortest fluid type representated by central point of distance is the fluid type of sample to be identified.
2. the tight sand fluid type recognition methods based on support vector machine simulation X plot according to claim 1, it is characterized in that, described step 2) in adopt cross-validation method to choose punishment parameter and the kernel function of supporting vector machine model when building supporting vector machine model.
3. the tight sand fluid type recognition methods based on support vector machine simulation X plot according to claim 2, it is characterized in that, described cross-validation method refers to and is divided into groups by original input data, a part is as training set, another part, as checking collection, is trained sorter with training set, and the model of training and obtaining tested by recycling checking collection, performance index using the classification accuracy obtained as classification of assessment device, choose optimum punishment parameter and kernel function with this.
4. the tight sand fluid type recognition methods based on support vector machine simulation X plot according to claim 3, it is characterized in that, described support vector cassification forecast model is dual output layer.
5. the recognition methods of tight sand fluid type based on support vector machine simulation X plot according to claim 4, is characterized in that, described step 2) in when training supporting vector machine model, need fluid sample to be normalized.
6. the tight sand fluid type recognition methods based on support vector machine simulation X plot according to claim 5, it is characterized in that, during described normalized fluid sample property value be normal distribution carry out conventional normalization, fluid sample property value be skewed distribution carry out logarithm normalization.
7. the tight sand fluid type recognition methods based on support vector machine simulation X plot according to claim 5, it is characterized in that, conveniently support vector cassification forecast model exports and shows in the plane, need carry out digitizing to support vector cassification forecast model Output rusults.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510333998.1A CN104992178A (en) | 2015-06-16 | 2015-06-16 | Tight sandstone fluid type identification method based on support vector machine simulation cross plot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510333998.1A CN104992178A (en) | 2015-06-16 | 2015-06-16 | Tight sandstone fluid type identification method based on support vector machine simulation cross plot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104992178A true CN104992178A (en) | 2015-10-21 |
Family
ID=54303991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510333998.1A Pending CN104992178A (en) | 2015-06-16 | 2015-06-16 | Tight sandstone fluid type identification method based on support vector machine simulation cross plot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104992178A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105715253A (en) * | 2016-01-30 | 2016-06-29 | 上海大学 | Prediction method for flowing bottomhole pressure of gas well |
| CN106545337A (en) * | 2016-11-25 | 2017-03-29 | 西南石油大学 | A kind of sedimentary micro Logging Identification Method based on support vector machine |
| CN106570524A (en) * | 2016-10-25 | 2017-04-19 | 中国石油天然气股份有限公司 | Reservoir fluid type identifying method and device |
| CN107367755A (en) * | 2016-05-11 | 2017-11-21 | 中国石油化工股份有限公司 | A kind of improved multi-parameter crossplot method for drafting |
| CN109993219A (en) * | 2019-03-21 | 2019-07-09 | 长江大学 | Dividing elements method is seeped in braided stream tight sand storage based on support vector machines |
| CN110552693A (en) * | 2019-09-19 | 2019-12-10 | 中国科学院声学研究所 | layer interface identification method of induction logging curve based on deep neural network |
| CN110717528A (en) * | 2019-09-25 | 2020-01-21 | 中国石油大学(华东) | Support vector machine-based sedimentary microfacies identification method using conventional logging information |
| CN110717301A (en) * | 2019-09-19 | 2020-01-21 | 中国石油大学(华东) | Flow unit information classification and identification method based on support vector machine algorithm |
| CN110717301B (en) * | 2019-09-19 | 2024-06-28 | 中国石油大学(华东) | Flow unit information classification and identification method based on support vector machine algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050049985A1 (en) * | 2003-08-28 | 2005-03-03 | Mangasarian Olvi L. | Input feature and kernel selection for support vector machine classification |
| CN102518428A (en) * | 2011-11-11 | 2012-06-27 | 中国石油天然气股份有限公司 | Oily water layer recognition method and device |
| CN102830441A (en) * | 2012-05-16 | 2012-12-19 | 陶冶 | Comprehensive interpretation and evaluation method of hydrocarbon reservoir including data mining |
| CN104200217A (en) * | 2014-08-07 | 2014-12-10 | 哈尔滨工程大学 | Hyperspectrum classification method based on composite kernel function |
-
2015
- 2015-06-16 CN CN201510333998.1A patent/CN104992178A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050049985A1 (en) * | 2003-08-28 | 2005-03-03 | Mangasarian Olvi L. | Input feature and kernel selection for support vector machine classification |
| CN102518428A (en) * | 2011-11-11 | 2012-06-27 | 中国石油天然气股份有限公司 | Oily water layer recognition method and device |
| CN102830441A (en) * | 2012-05-16 | 2012-12-19 | 陶冶 | Comprehensive interpretation and evaluation method of hydrocarbon reservoir including data mining |
| CN104200217A (en) * | 2014-08-07 | 2014-12-10 | 哈尔滨工程大学 | Hyperspectrum classification method based on composite kernel function |
Non-Patent Citations (2)
| Title |
|---|
| 梁丽梅 等: ""神经网络模拟交会图在低阻油层流体识别中的应用"", 《石油工业计算机应用》 * |
| 边会媛 等: ""利用弹性参数识别致密砂岩储层流体性质"", 《地球科学与环境学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105715253A (en) * | 2016-01-30 | 2016-06-29 | 上海大学 | Prediction method for flowing bottomhole pressure of gas well |
| CN107367755A (en) * | 2016-05-11 | 2017-11-21 | 中国石油化工股份有限公司 | A kind of improved multi-parameter crossplot method for drafting |
| CN106570524A (en) * | 2016-10-25 | 2017-04-19 | 中国石油天然气股份有限公司 | Reservoir fluid type identifying method and device |
| CN106545337A (en) * | 2016-11-25 | 2017-03-29 | 西南石油大学 | A kind of sedimentary micro Logging Identification Method based on support vector machine |
| CN109993219A (en) * | 2019-03-21 | 2019-07-09 | 长江大学 | Dividing elements method is seeped in braided stream tight sand storage based on support vector machines |
| CN110552693A (en) * | 2019-09-19 | 2019-12-10 | 中国科学院声学研究所 | layer interface identification method of induction logging curve based on deep neural network |
| CN110717301A (en) * | 2019-09-19 | 2020-01-21 | 中国石油大学(华东) | Flow unit information classification and identification method based on support vector machine algorithm |
| CN110717301B (en) * | 2019-09-19 | 2024-06-28 | 中国石油大学(华东) | Flow unit information classification and identification method based on support vector machine algorithm |
| CN110717528A (en) * | 2019-09-25 | 2020-01-21 | 中国石油大学(华东) | Support vector machine-based sedimentary microfacies identification method using conventional logging information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104992178A (en) | Tight sandstone fluid type identification method based on support vector machine simulation cross plot | |
| CN110674841B (en) | Logging curve identification method based on clustering algorithm | |
| CN113901681B (en) | Three-dimensional compressibility evaluation method for dual desserts of shale gas reservoir in whole life cycle | |
| CN105760673B (en) | A kind of fluvial depositional reservoir seismic-sensitive parameterized template analysis method | |
| CN107728231B (en) | A kind of prediction nuclear magnetic resonance log lateral relaxation time T2The method of distribution | |
| CN102176052B (en) | Hierarchical sequence analysis method oriented to generation of three-dimensional hierarchical grids | |
| CN109214026A (en) | A kind of shale gas horizontal well initial productivity prediction technique | |
| CN109345007B (en) | Advantageous reservoir development area prediction method based on XGboost feature selection | |
| CN108897066A (en) | Carbonate fracture density quantitative forecasting technique and device | |
| CN103336305B (en) | A kind of method dividing Sandstone Gas Reservoir high water cut based on gray theory | |
| CN103376468A (en) | Reservoir parameter quantitative characterization method based on neural network function approximation algorithm | |
| CN105447762A (en) | Calculation method for low permeability reservoir flooding information of fluid replacement | |
| CN110097069A (en) | A kind of support vector machines Lithofacies Identification method and device based on depth Multiple Kernel Learning | |
| CN104850867A (en) | Object identification method based on intuitive fuzzy c-means clustering | |
| CN105467449A (en) | Deep thin interbedded reservoir quantitative characterization method based on seismic grading sensitive attribute fusion | |
| CN105093331A (en) | Method for obtaining rock matrix bulk modulus | |
| CN103527172B (en) | Variable rock electric coupling index water saturation computational methods | |
| CN106569272A (en) | Earthquake attribute fusion method based on data property space ascending dimension | |
| CN109425889B (en) | Method for depicting ancient karst underground river | |
| Lu et al. | Identifying flow units by FA-assisted SSOM—An example from the Eocene basin-floor-fan turbidite reservoirs in the Daluhu Oilfield, Dongying Depression, Bohai Bay Basin, China | |
| CN112576238A (en) | System, method and application for determining position and content of residual oil in low-permeability reservoir | |
| CN113484907B (en) | Method for predicting distribution on different types of reservoir planes | |
| CN104880737A (en) | Multivariate Logistic method using logging information to identify type of underground fluid | |
| Chen et al. | Seismic facies visualization analysis method of SOM corrected by uniform manifold approximation and projection | |
| CN114202036A (en) | Static sounding automatic soil layer dividing technology based on pattern recognition technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151021 |
|
| RJ01 | Rejection of invention patent application after publication |