CN110443283A - A kind of discriminant classification method and system of tight gas reservoir horizontal well - Google Patents
A kind of discriminant classification method and system of tight gas reservoir horizontal well Download PDFInfo
- Publication number
- CN110443283A CN110443283A CN201910626269.3A CN201910626269A CN110443283A CN 110443283 A CN110443283 A CN 110443283A CN 201910626269 A CN201910626269 A CN 201910626269A CN 110443283 A CN110443283 A CN 110443283A
- Authority
- CN
- China
- Prior art keywords
- classification
- horizontal well
- discriminant
- well
- horizontal
- 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
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
Abstract
The present invention relates to a kind of discriminant classification method and system of tight gas reservoir horizontal well, this method comprises: step 1, selects the characteristic parameter analyzed horizontal well, determine the classification that all horizontal wells include using clustering method;Step 2, by Bayesian method, the discriminant classification function of all kinds of horizontal wells is obtained;Step 3, the characteristic parameter of horizontal well to be evaluated is substituted into the discriminant classification function of all kinds of horizontal wells, the maximum corresponding classification of the value of the discriminant classification function is the classification of the horizontal well to be evaluated.According to gas reservoir seepage flow and well testing principle, in conjunction with gas production data self-law, select multiple characteristic parameters of horizontal well, and the classification that horizontal well includes is determined using multiple characteristic parameters of the comprehensive horizontal well of clustering method, and the discriminant classification function of all kinds of horizontal wells is obtained using Bayesian method, specific data reference foundation is provided for the classification assessment of horizontal well to be evaluated.
Description
Technical field
The present invention relates to natural stone oil gas development fields, and in particular to a kind of discriminant classification method of tight gas reservoir horizontal well
And system.
Background technique
The exploitation of tight gas reservoir horizontal well is affected by many factors, and the production existing difference of feature also has similar, differentiation water
The production feature of horizontal well, is conducive to the horizontal well of Classification Management difference development features, so that the exploitation for giving full play to horizontal well is latent
Power.
Currently, common gas well classification method has: 1, reservoir parameter method classifies to gas well using reservoir parameter method
When, classify mostly in reference to the effective reservoir thickness of gas well to gas well.Since tight gas reservoir horizontal well natural production capacity is low,
It generally requires acidified pressure break and carries out reservoir reconstruction, be affected since improved effect produces gas to gas-bearing formation, only according to gas well
Thickness, which divide, is clearly present limitation;2, gas testing open-flow capacity method, open-flow capacity reflection is producing initial stage near wellbore formation
The seepage flow characteristics of fracturing fracture band cannot reflect the energy of remote well stratum reservoir.Gas well type is evaluated only according to gas testing open-flow capacity,
It cannot really reflect gas well yield;3, synthetical classification is used using the parameters such as formation capacity kh and gas testing open-flow capacity AOF
Fuzzy mathematics to above-mentioned parameter carry out processing formed an overall merit factor CCF, classify, but formation capacity weight and
Gas testing open-flow capacity weight is obtained by expert according to on-site verification, and the size of weight is larger by subjective impact, causes classification acceptor
Sight factor influences;4, unit covers pressure drop gas production method, covers pressure drop gas production by the unit for producing period to classification gas well difference
It is returned with corresponding production time data, carries out data and be fitted to obtain the power function curve of good relationship, with what is obtained
Power function curve is further classified.This method is for because the case where the casing pressure caused by producing system variation rises, yield declines
It can not reject, influence classifying quality.Gas wells typical for simple several mouthfuls can be distinguish, but when well number reach several hundred mouths,
The classification method will fail at thousands of mouthfuls, can not be divided;5, yield weighting set platen press, yield weighting set platen press can be effective
The influence generated by closing well to be rejected, weighting yield can effectively reflect the relationship between the casing pressure of gas well and yield, and
And the gas well of different production times can be compared, but this method fails to consider stratum firsthand information, thus can not be right
Gas well carries out exhaustive classification;6, the advantages of dynamic cataloging method, in summary gas well is classified, is to form gas well waterout classification, the classification
Method comprehensively considers gas well static, dynamic parameter and carries out processing using fuzzy mathematics and obtain coefficient of colligation, but each parameters weighting
Assignment, still influenced seriously by subjective factor.Above-mentioned various classification methods, categorised demarcation line, parameters weighting etc. by it is subjective because
Element influences serious.
Summary of the invention
The present invention for the technical problems in the prior art, provides a kind of discriminant classification side of tight gas reservoir horizontal well
Method and system.
The technical scheme to solve the above technical problems is that a kind of discriminant classification side of tight gas reservoir horizontal well
Method, the discriminant classification method include:
Step 1, the characteristic parameter analyzed horizontal well is selected, all levels are determined using clustering method
The classification that well includes;
Step 2, by Bayesian method, the discriminant classification function of all kinds of horizontal wells is obtained;
Step 3, the characteristic parameter of horizontal well to be evaluated substitutes into the discriminant classification function of all kinds of horizontal wells, described point
The maximum corresponding classification of the value of class discriminant function is the classification of the horizontal well to be evaluated.
A kind of discriminant classification system of tight gas reservoir horizontal well, the discriminant classification system include: category analysis module, sentence
Other analysis module and discriminant classification module;
Category analysis module, the characteristic parameter for selecting to analyze horizontal well are determined using clustering method
The classification that all horizontal wells include;
Discriminant analysis module, for obtaining the discriminant classification of all kinds of horizontal wells by Bayesian method
Function;
Discriminant classification module, for the characteristic parameter of horizontal well to be evaluated to be substituted into the discriminant classification of all kinds of horizontal wells
Function, the maximum corresponding classification of the value of the discriminant classification function are the classification of the horizontal well to be evaluated.
A kind of non-transient computer readable storage medium, is stored thereon with computer program, and the computer program is processed
The step of device realizes the discriminant classification method of above-mentioned tight gas reservoir horizontal well when executing.
The beneficial effects of the present invention are: according to gas reservoir seepage flow and well testing principle, in conjunction with gas production data self-law,
Multiple characteristic parameters of horizontal well are selected, and determine level using multiple characteristic parameters of the comprehensive horizontal well of clustering method
The classification that well includes, and the discriminant classification function of all kinds of horizontal wells is obtained using Bayesian method, by water to be evaluated
The corresponding characteristic parameter substitution of horizontal well, which is compared, can differentiate the classification of horizontal well to be evaluated, comment for the classification of horizontal well to be evaluated
Estimate and specific data reference foundation is provided.
Based on the above technical solution, the present invention can also be improved as follows.
Further, the characteristic parameter analyzed the horizontal well selected in the step 1 includes: no choked flow
Amount, set pressure drop gas production, unit set pressure drop rate and initial stage produce tolerance daily.
The clustering method that the classification that determining all horizontal wells include in the step 1 uses is K- mean cluster
Algorithm.
All horizontal wells are divided into the process that C is assembled class using clustering method by the step 1
Step 101, the initial center of the selection C aggregation classes;
Step 102, cycle iterative operation thereof is carried out, the characteristic parameter for calculating the horizontal well of any one is described to C
The distance for assembling the initial center of class, the horizontal well is referred to apart from the aggregation where the shortest initial center
Class updates the cluster centre value of the aggregation class;
Step 103, when being iterated the central values of the described aggregation classes of C after operation and remaining unchanged, terminate iterative process, it is no
Then continue to execute the step 102.
In the step 102, the cluster centre value of the aggregation class is updated using the method for mean value.
The step 2 includes:
Step 201, horizontal well sample x is calculated from g-th of overall posterior probability with Bayesian formula are as follows:
Wherein, m is overall number, and g indicates overall serial number number, q1,q2…qmIt is general for each overall priori
Rate, f1(x),f2(x)…fmIt (x) is each overall density function;
Step 202, it calculatesWhen determine that horizontal well sample x is overall from h-th;
Step 203, it is assumed that under conditions of covariance matrix is equal, export the discriminant classification function of all kinds of horizontal wells.
Beneficial effect using above-mentioned further scheme is, choose set pressure drop gas production, set pressure drop rate, open-flow capacity,
Initial stage produces tolerance daily as classification indicators, and deterministic process is simple and clear, tests prove that result is accurate.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the embodiment of the discriminant classification method of tight gas reservoir horizontal well provided by the invention;
Fig. 2 is a kind of structural block diagram of the embodiment of the discriminant classification system of tight gas reservoir horizontal well provided by the invention;
Fig. 3 is certain gas reservoir in a kind of Application Example of the discriminant classification method of tight gas reservoir horizontal well provided by the invention
4 area's horizontal well clustering evaluation result figures;
Fig. 4 is in a kind of Application Example of the discriminant classification method of tight gas reservoir horizontal well provided by the invention using not
With the classification results figure of method.
In attached drawing, parts list represented by the reference numerals are as follows:
1, category analysis module, 2, discriminant analysis module, 3, discriminant classification module.
Specific embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the invention.
A kind of discriminant classification method of tight gas reservoir horizontal well provided in an embodiment of the present invention, as shown in Figure 1, this method packet
It includes:
Step 1, the characteristic parameter analyzed horizontal well is selected, all horizontal well packets are determined using clustering method
The classification contained.
Step 2, by Bayesian method, the discriminant classification function of all kinds of horizontal wells is obtained.
Step 3, the characteristic parameter of horizontal well to be evaluated is substituted into the discriminant classification function of all kinds of horizontal wells, which sentences
The maximum corresponding classification of the value of other function is the classification of the horizontal well to be evaluated.
The discriminant classification method of a kind of tight gas reservoir horizontal well provided in an embodiment of the present invention, according to gas reservoir seepage flow and well testing
Principle selects multiple characteristic parameters of horizontal well in conjunction with gas production data self-law, and comprehensive using clustering method
The multiple characteristic parameters for closing the horizontal well determine the classification that horizontal well includes, and it is all kinds of to use Bayesian method to obtain
The discriminant classification function of horizontal well, the corresponding characteristic parameter substitution of horizontal well to be evaluated, which is compared, can differentiate level to be evaluated
The classification of well provides specific data reference foundation for the classification assessment of horizontal well to be evaluated.
Embodiment 1
Embodiment 1 provided by the invention is a kind of reality of the discriminant classification method of tight gas reservoir horizontal well provided by the invention
Example is applied, in the embodiment, which includes:
Step 1, the characteristic parameter analyzed horizontal well is selected, all horizontal well packets are determined using clustering method
The classification contained.
Static classification is carried out according to horizontal well storage parameter and is clearly present limitation, it is contemplated that horizontal well single well-controlled radius
Greatly, the energy of the wide feature of swept area, surface layer with production may gradually use up, and the energy on remote stratum could quilt
It utilizes, in order to carry out permanently effective management to horizontal well, horizontal well preferably considers that gas well distance reservoir energy employs situation, foundation
Dynamic creation data is classified.The characteristic parameter analyzed horizontal well selected in step 1 includes: open-flow capacity, set
Pressure drop gas production, unit set pressure drop rate and initial stage produce tolerance daily.
Further, when selecting the characteristic parameter analyzed horizontal well, the quasi-stable state production phase is reached with horizontal well
When creation data as characteristic parameter cover pressure drop gas production and unit set pressure drop rate Selecting time node.Below with densification
Reach within Horizontal Wells For Gas Reservoirs 300 days or so and illustrates for the quasi-stable state production phase.
Absolute Open Flow of Gas Wells generally can more directly reflect gas well capacity situation, but gas field gas well is exhausted in process of production
Most of not carry out flow-after-flow test, the open-flow capacity of gas well is to be obtained before operation using one point method measuring and calculation, institute
With its open-flow capacity reflection be producing initial stage near wellbore formation fracturing fracture band seepage flow characteristics.Therefore, open-flow capacity can be used as
The reflection of the fluid flow characteristics of near wellbore zone.
The variation of gas well mouth casing pressure situation can preferably reflect gas capacity of well with corresponding accumulative gas production.
Reach the quasi-stable state production phase due to tight gas reservoir horizontal well 300 days or so, the casing pressure after choosing 300 days production times
Parameter of the gas production as reflection remote well area fluid flow characteristics drops.
Set pressure drop rate can directly reflect the stable yields situation of gas well, can also reflect the energy variation on stratum indirectly.Casing pressure
Reduction of speed rate is bigger, and the consumption of horizontal well stratum energy is faster, seriously affects the stable production period of horizontal well.Pressure drop is covered in conjunction with gas well unit
The feature of corresponding change occurs with the extension of production time for gas production, using set pressure drop rate as in the reflection gas well section time
The characteristic index of fluid fluid ability.
Since the open-flow capacity testing time is short, it cannot sufficiently reflect the production feature at gas well initial stage, it is therefore, raw according to gas well
Rule is produced, selects to make up the limitation of open-flow capacity index using preceding 3 months dailys output as classification indicators.
A kind of discriminant classification method of tight gas reservoir horizontal well provided in an embodiment of the present invention determines all water in step 1
The clustering method that the classification that horizontal well includes uses is adopted in step 1 for K- means clustering algorithm (k-means algorithm)
Include: by the process that all horizontal wells are divided into C aggregation class with clustering method
Step 101, the initial center of C aggregation class is selected.
Suitable initial center is selected as needed, and the selection at the center is related to successive iterations number and result.
Step 102, cycle iterative operation thereof is carried out, the characteristic parameter for calculating the horizontal well of any one assembles class to C
The horizontal well is referred to apart from the aggregation class where shortest initial center, updates the aggregation class by the distance of initial center
Cluster centre value.
Specifically, the cluster centre value of aggregation class can be updated using the methods of mean value.
For j-th of cluster set, cluster centre ZjCriterion function be that the characteristic parameter of each horizontal well arrives in cluster set
The square distance of the cluster centre and, i.e., are as follows:
Wherein, SjFor j-th of cluster set, NjJ-th of cluster set SjIncluded in number of samples, i be cluster set SjMiddle sample
This serial number number.
Have to all K mode classes:
The selection of cluster centre should make J minimum, even if JjValue it is minimum.Have:
Above formula shows to assemble class SjCluster centre should be selected as the mean value of the aggregation class sample.
Step 103, when being iterated the central value of C class after operation and remaining unchanged, terminate the iterative process, otherwise continue
Execute step 102.
By the method for iteration, the value of each cluster centre is gradually updated, until obtaining best cluster result.
Step 2, by Bayesian method, the discriminant classification function of all kinds of horizontal wells is obtained.
Step 201, horizontal well sample x is calculated from g-th of overall posterior probability with Bayesian formula are as follows:
Wherein, m is overall number, and g indicates overall serial number number, G1,G2…GmRespectively indicate m overall, q1,q2…qm
For each overall prior probability, f1(x),f2(x)…fmIt (x) is each overall density function.
Step 202, it calculatesWhen determine that horizontal well sample x is overall from h-th.
Wherein, horizontal well sample x is misjudged to h-th of overall average loss are as follows:
Wherein, L (h/g) is to indicate to be the horizontal well sample misjudgement of g-th of totality originally to be the damage of h-th of overall loss
Lose function.
Since L (h/g) is not easy to determine in practical application, but ifThen determine that sample x comes from
H-th overall.Therefore it is presumed that finding h keeps posterior probability maximum actually of equal value in the case that the loss of various misjudgement is all equal
In keeping Misjudgement Loss minimum.That is:
Step 203, it is assumed that under conditions of covariance matrix is equal, export the discriminant classification function of all kinds of horizontal wells.
Step 3, the characteristic parameter of horizontal well to be evaluated is substituted into the discriminant classification function of all kinds of horizontal wells, which sentences
The maximum corresponding classification of other functional value is the classification of the horizontal well to be evaluated.
Embodiment 2
Embodiment 2 provided by the invention is a kind of reality of the discriminant classification system of tight gas reservoir horizontal well provided by the invention
Example is applied, as shown in Figure 2, which includes: category analysis module 1, discriminant analysis module 2 and discriminant classification module 3;
Category analysis module 1, the characteristic parameter for selecting to analyze horizontal well are determined using clustering method
The classification that all horizontal wells include;
Discriminant analysis module 2, for obtaining the discriminant classification letter of all kinds of horizontal wells by Bayesian method
Number;
Discriminant classification module 3, for the characteristic parameter of horizontal well to be evaluated to be substituted into the discriminant classification letter of all kinds of horizontal wells
Number, the maximum corresponding classification of the value of discriminant classification function are the classification 3 of horizontal well to be evaluated.
Embodiment 3
Embodiment 3 provided by the invention is a kind of tool of the discriminant classification method of tight gas reservoir horizontal well provided by the invention
Body Application Example, in the concrete application embodiment, firstly, obtaining the characteristic parameter analyzed horizontal well.
Open-flow capacity: it is obtained, can be obtained from gas field well test data using one point method measuring and calculation by gas field well testing team.
Unit covers pressure drop gas production: obtaining gas well gas production G and gas well casing pressure changing value Δ by gas field Production development data
P calculates unit set pressure drop gas production (the G/ Δ p) for obtaining gas well.
Pressure drop rate: the gas well liquid loading time t and gas well casing pressure p obtained by gas well liquid loading dynamic data is covered, when unit of account
The unit of interior gas well covers pressure drop (Δ p/ Δ t).
The initial stage daily output: according to gas well liquid loading dynamic data, the average value of 3 months daily output tolerance before gas well is calculated.
121 mouthfuls of water horizontal wells that target gas field normally produces are chosen, 4 parameters of each well: open-flow capacity, unit set are obtained
Pressure drop gas production, set pressure drop rate, the initial stage daily output.
4 characteristic parameters of analysis level well are classified 121 mouthfuls of water horizontal wells using clustering method, and determination is divided into
Several classes, 121 mouthfuls of target gas field water horizontal well are divided into 3 classes after cluster.
Discriminant analysis differentiates that its type is returned according to the various characteristic values of a certain research object under conditions of classifying determining
A kind of multivariate statistical analysis method of category problem.Certain i.e. known things has several types, respectively takes from various types now
One sample goes out a set of standard by these sample designs, takes a sample so that appointing from this things, can be by this set standard
Differentiate its type.
After clustering determines the class number of horizontal well, Bayes discriminant analysis method is further utilized, all kinds of horizontal wells are obtained
Discriminant function.
By 4 parameters (open-flow capacity, set pressure drop gas production, set pressure drop rate, the initial stage daily output of the unknown horizontal well of classification
Tolerance) obtained discriminant function is substituted into, the Y value in which formula is big, just belongs to which classification.
4 parameters of horizontal well, open-flow capacity, set pressure drop gas production, set pressure drop rate, initial stage daily gas are selected first
Amount, by clustering method, by horizontal well classification;Then by Bayes discriminant analysis method, the connection of all kinds of horizontal wells is obtained
Close the discriminant classification function of distribution map and horizontal well.
The discriminant classification function of horizontal well:
I class well: Y=-29.705+0.083 × set pressure drop gas production+830.803 × set pressure drop rate+0.152 × without hindrance
Flow+0.936 × initial stage daily output;
II class well: Y=-25.058+0.064 × set pressure drop gas production+1000.123 × set pressure drop rate+0.099 × nothing
Choked flow amount+0.148 × initial stage daily output;
Group III well: Y=-15.376+0.083 × set pressure drop gas production+642.789 × set pressure drop rate+0.050 × nothing
Choked flow amount+0.151 × initial stage daily output.
Produce the open-flow capacity of horizontal gas well to be evaluated, set pressure drop gas production, set pressure drop rate, initial stage daily tolerance 4 ginsengs
Number brings horizontal well discriminant classification function into, and the Y value in which formula is big, just belongs to which classification.
It is illustrated in figure 3 in a kind of Application Example of the discriminant classification method of tight gas reservoir horizontal well provided by the invention
4 area's horizontal well clustering evaluation result figures of certain gas reservoir, from the figure 3, it may be seen that the ratio of the total well number of all kinds of well Zhan is respectively
17.36%, 34.71% and 47.93%.Classification results show three types using this classification method, being divided into according to dynamic data
Other horizontal well (I, II, III) and the effective reservoir length and pressure break number of segment of boring chance are closely related.I, II, Group III Horizontal Well Drilling
Meeting effective reservoir length is respectively 863m, 630m and 386m, and pressure break number of segment is respectively 7.80 sections, 6.75 sections and 5.26 sections, classification
As a result reflect that Horizontal Well Drilling chance effective reservoir is longer, pressure break number of segment is more, and development effectiveness is relatively more preferable, meets the reality of production
Situation.It is illustrated in figure 4 in a kind of Application Example of the discriminant classification method of tight gas reservoir horizontal well provided by the invention and adopts
Classification results figure differently in conjunction with Field Production Data, compares two kinds of classification results, set is chosen in discovery as shown in Figure 4
Pressure drop gas production, set pressure drop rate, open-flow capacity, initial stage daily output tolerance are as classification indicators, using K mean cluster analysis method
Resulting classification results are better than the open-flow capacity classification results of field application.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
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 (8)
1. a kind of discriminant classification method of tight gas reservoir horizontal well, which is characterized in that the discriminant classification method includes:
Step 1, the characteristic parameter analyzed horizontal well is selected, all horizontal well packets are determined using clustering method
The classification contained;
Step 2, by Bayesian method, the discriminant classification function of all kinds of horizontal wells is obtained;
Step 3, the characteristic parameter of horizontal well to be evaluated is substituted into the discriminant classification function of all kinds of horizontal wells, the classification is sentenced
The maximum corresponding classification of the value of other function is the classification of the horizontal well to be evaluated.
2. discriminant classification method according to claim 1, which is characterized in that selected in the step 1 to the level
The characteristic parameter that well is analyzed includes: open-flow capacity, set pressure drop gas production, unit set pressure drop rate and initial stage daily gas
Amount.
3. discriminant classification method according to claim 1, which is characterized in that determine all levels in the step 1
The clustering method that the classification that well includes uses is K- means clustering algorithm.
4. discriminant classification method according to claim 3, which is characterized in that the step 1 will using clustering method
All horizontal wells are divided into the process that C is assembled class
Step 101, the initial center of the selection C aggregation classes;
Step 102, cycle iterative operation thereof is carried out, calculates the characteristic parameter of the horizontal well of any one to a aggregations of C
The horizontal well is referred to apart from the aggregation class where the shortest initial center by the distance of the initial center of class,
Update the cluster centre value of the aggregation class;
Step 103, when being iterated the central values of the described aggregation classes of C after operation and remaining unchanged, terminate iterative process, otherwise after
It is continuous to execute the step 102.
5. discriminant classification method according to claim 4, which is characterized in that in the step 102, using the method for mean value
Update the cluster centre value of the aggregation class.
6. discriminant classification method according to claim 1, which is characterized in that the step 2 includes:
Step 201, horizontal well sample x is calculated from g-th of overall posterior probability with Bayesian formula are as follows:
Wherein, m is overall number, and g indicates overall serial number number, q1,q2…qmFor each overall prior probability, f1
(x),f2(x)…fmIt (x) is each overall density function;
Step 202, it calculatesWhen determine that horizontal well sample x is overall from h-th;
Step 203, it is assumed that under conditions of covariance matrix is equal, export the discriminant classification function of all kinds of horizontal wells.
7. a kind of discriminant classification system of tight gas reservoir horizontal well, which is characterized in that the discriminant classification system includes: classification point
Analyse module, discriminant analysis module and discriminant classification module;
Category analysis module, the characteristic parameter for selecting to analyze horizontal well are determined all using clustering method
The classification that the horizontal well includes;
Discriminant analysis module, for obtaining the discriminant classification function of all kinds of horizontal wells by Bayesian method;
Discriminant classification module, for the characteristic parameter of horizontal well to be evaluated to be substituted into the discriminant classification letter of all kinds of horizontal wells
Number, the maximum corresponding classification of the value of the discriminant classification function are the classification of the horizontal well to be evaluated.
8. a kind of non-transient computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer journey
The step of the discriminant classification method of tight gas reservoir horizontal well as described in any one of claim 1 to 6 is realized when sequence is executed by processor
Suddenly.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910626269.3A CN110443283A (en) | 2019-07-11 | 2019-07-11 | A kind of discriminant classification method and system of tight gas reservoir horizontal well |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910626269.3A CN110443283A (en) | 2019-07-11 | 2019-07-11 | A kind of discriminant classification method and system of tight gas reservoir horizontal well |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110443283A true CN110443283A (en) | 2019-11-12 |
Family
ID=68430217
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910626269.3A Pending CN110443283A (en) | 2019-07-11 | 2019-07-11 | A kind of discriminant classification method and system of tight gas reservoir horizontal well |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110443283A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140286539A1 (en) * | 2011-10-31 | 2014-09-25 | Schlumberger Technology Corporation | Inversion-based workflow for processing nuclear density images in high-angle and horizontal wells |
CN104318109A (en) * | 2014-10-29 | 2015-01-28 | 中国石油集团川庆钻探工程有限公司 | Shale gas reservoir identification method based on support vector machine |
CN104463686A (en) * | 2014-10-29 | 2015-03-25 | 中国石油集团川庆钻探工程有限公司 | Method for identifying shale gas reservoir while drilling by utilizing discriminant analysis method |
CN107590766A (en) * | 2017-09-28 | 2018-01-16 | 东南大学 | A kind of method of discrimination of the land used combination form related to road traffic accident risk |
US20190073996A1 (en) * | 2017-09-05 | 2019-03-07 | International Business Machines Corporation | Machine training for native language and fluency identification |
-
2019
- 2019-07-11 CN CN201910626269.3A patent/CN110443283A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140286539A1 (en) * | 2011-10-31 | 2014-09-25 | Schlumberger Technology Corporation | Inversion-based workflow for processing nuclear density images in high-angle and horizontal wells |
CN104318109A (en) * | 2014-10-29 | 2015-01-28 | 中国石油集团川庆钻探工程有限公司 | Shale gas reservoir identification method based on support vector machine |
CN104463686A (en) * | 2014-10-29 | 2015-03-25 | 中国石油集团川庆钻探工程有限公司 | Method for identifying shale gas reservoir while drilling by utilizing discriminant analysis method |
US20190073996A1 (en) * | 2017-09-05 | 2019-03-07 | International Business Machines Corporation | Machine training for native language and fluency identification |
CN107590766A (en) * | 2017-09-28 | 2018-01-16 | 东南大学 | A kind of method of discrimination of the land used combination form related to road traffic accident risk |
Non-Patent Citations (3)
Title |
---|
方志雄: "《延川南煤层气田勘探开发关键技术及应用》", 30 June 2017, 地质出版社 * |
闫明 等: "基于K-均值聚类与贝叶斯判别的储层定量评价——以大安油田泉四段储层为例", 《环境与资源》 * |
陈启文 等: "苏里格致密砂岩气藏水平井分类研究", 《中国石油和化工标准与质量》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108388921B (en) | Overflow loss real-time identification method based on random forest | |
CN110346831B (en) | Intelligent seismic fluid identification method based on random forest algorithm | |
CN106980877A (en) | A kind of Prediction of Blasting Vibration method based on particle cluster algorithm Support Vector Machines Optimized | |
CN107122860B (en) | Rock burst danger level prediction method based on grid search and extreme learning machine | |
CN108897975A (en) | Coalbed gas logging air content prediction technique based on deepness belief network | |
CN105760673A (en) | Fluvial facies reservoir earthquake sensitive parameter template analysis method | |
CN103967478A (en) | Method for identifying vertical well flow patterns based on conducting probe | |
Xiongyan et al. | Computational intelligent methods for predicting complex ithologies and multiphase fluids | |
CN112529341A (en) | Drilling well leakage probability prediction method based on naive Bayesian algorithm | |
CN104570109B (en) | A kind of method of reservoir hydrocarbons prediction | |
CN114358434A (en) | Drilling machine drilling speed prediction method based on LSTM recurrent neural network model | |
CN112305441A (en) | Power battery health state assessment method under integrated clustering | |
CN115640526A (en) | Drilling risk identification model, building method, identification method and computer equipment | |
CN113378998B (en) | Stratum lithology while-drilling identification method based on machine learning | |
Wang et al. | Determine level of thief zone using fuzzy ISODATA clustering method | |
CN116303626B (en) | Well cementation pump pressure prediction method based on feature optimization and online learning | |
CN113095352A (en) | Oil-gas-containing property evaluation method for fracture-cavity type oil reservoir semi-supervised learning reservoir fused with static and dynamic characteristics | |
CN116562428A (en) | Fracturing construction parameter optimization method based on machine learning | |
CN117093922A (en) | Improved SVM-based complex fluid identification method for unbalanced sample oil reservoir | |
CN110443283A (en) | A kind of discriminant classification method and system of tight gas reservoir horizontal well | |
CN109211814B (en) | It is a kind of to be set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light | |
CN113792141A (en) | Feature selection method based on covariance measurement factor | |
CN114818493A (en) | Method for quantitatively evaluating integrity degree of tunnel rock mass | |
CN114548306A (en) | Intelligent monitoring method for early drilling overflow based on misclassification cost | |
Feder | Machine-learning approach determines spatial variation in shale decline curves |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191112 |
|
RJ01 | Rejection of invention patent application after publication |