CN106570524A - Reservoir fluid type identifying method and device - Google Patents
Reservoir fluid type identifying method and device Download PDFInfo
- Publication number
- CN106570524A CN106570524A CN201610940183.4A CN201610940183A CN106570524A CN 106570524 A CN106570524 A CN 106570524A CN 201610940183 A CN201610940183 A CN 201610940183A CN 106570524 A CN106570524 A CN 106570524A
- Authority
- CN
- China
- Prior art keywords
- reservoir
- type
- fluid
- training
- reservoir fluid
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
The invention provides a reservoir fluid type identifying method and device. The method comprises that a reservoir fluid training set which includes reservoir fluids of determined types and training characteristic values associated with the reservoir fluids respectively is obtained; according to the reservoir fluid training set, a reservoir fluid type prediction model is obtained by training, and for each reservoir fluid in the reservoir fluid training set, the reservoir fluid type prediction model satisfies that the reservoir fluid type predicted according to the training characteristic value of the reservoir fluid is consistent with the determined reservoir fluid type; a target characteristic value of a reservoir fluid to be identified is obtained; and the reservoir fluid type prediction model is used to predict the target characteristic value, and obtains the type of the reservoir fluid to be identified. The reservoir fluid type identifying method and device can improve the identification precision of the reservoir fluid type.
Description
Technical field
The application is related to reservoir exploration and development technical field, the recognition methodss of more particularly to a kind of fluid type of reservoir through and
Device.
Background technology
At present, fluid type of reservoir through recognition methodss are more, are broadly divided into three classes：It is identified using ordinary curve, is utilized
Mathematical method is identified and is identified using new Logging Technology.Yet with originals such as shale content height or complex pore structures
Cause, can cause accuracy of identification not high.
Support vector cassification Forecasting Methodology is a kind of machine learning method that the nineties grow up, its operation letter
Just, many distinctive advantages are shown in small sample, the identification of non-linear and high dimensional pattern is solved.At present to reservoir fluid class
When type is identified, the features of logging curve value that can read the various fluid type samples of reservoir is supported vector machine research,
But existing method does not account for the impact that geologic characteristics differentiate to support vector machine fluid type, this impact cannot be with often
Gamma ray curve, spontaneous potential curve, interval transit time curve, density curve of rule etc. are characterized, therefore method and reality at present
There is gap in geological condition.
Additionally, existing method is most from this parameter of natural potential when support vector machine differentiate.But by measurement process
Control, spontaneous potential curve affected by F and formation water salinity, and the geological Significance of its absolute value is little, so as to cannot
Accurately recognize fluid type of reservoir through.
It should be noted that the introduction of technical background is intended merely to above the convenient technical scheme to the application carry out it is clear,
Complete explanation, and facilitate the understanding of those skilled in the art and illustrate.Can not be merely because these schemes be the application's
Background section is set forth and thinks that above-mentioned technical proposal is known to those skilled in the art.
The content of the invention
The purpose of the application embodiment is recognition methodss and the device for providing a kind of fluid type of reservoir through, it is possible to increase
The accuracy of identification of fluid type of reservoir through.
For achieving the above object, on the one hand the application provides a kind of recognition methodss of fluid type of reservoir through, methods described bag
Include：Obtain reservoir fluid training set, the reservoir fluid training set include having determined that multiple reservoir fluids of type and with
The training characteristics value that each described reservoir fluid is respectively associated；Reservoir fluid class is obtained according to reservoir fluid training set training
Type forecast model, wherein, for each reservoir fluid in the reservoir fluid training set, the fluid type of reservoir through predicts mould
Type is satisfied by：The fluid type of reservoir through obtained according to the prediction of the training characteristics value of the reservoir fluid and the fixed reservoir
The type of fluid is consistent；Obtain the object feature value of reservoir fluid to be identified；Using the fluid type of reservoir through forecast model pair
The object feature value is predicted, and obtains the type of the reservoir fluid to be identified.
Further, the training characteristics value at least includes Reservoir type mark, reservoir natural gamma value, natural potential width
Degree ratio, layer depth resistivity, interval transit time, neutron porosity and density of earth formations.
Further, the Reservoir type mark is obtained in such a way：From default Reservoir Classification set determine with
The Reservoir type that the type of reservoir fluid is adapted, and obtain the Reservoir type mark being associated with the Reservoir type for determining.
Further, after the step of obtaining reservoir fluid training set, methods described also includes：It is right according to the following equation
Each training characteristics value in the reservoir fluid training set is normalized：
Wherein, B represents the training characteristics value after normalized, and A represents the training characteristics value before normalization, AminTable
Show the minima in such training characteristics value, AmaxRepresent the maximum in such training characteristics value.
Further, obtain fluid type of reservoir through forecast model according to reservoir fluid training set training to specifically include：
According to initial reservoir fluid type forecast model, prediction obtains corresponding with the training characteristics value in the reservoir fluid training set
Reservoir fluid training type；Error pair between type and fixed fluid type of reservoir through is trained according to the reservoir fluid
Preset function in the initial reservoir fluid type forecast model is corrected, so that according to the reservoir fluid class after correction
Type forecast model prediction obtain reservoir fluid training type and fixed fluid type of reservoir through between error be less than or
Equal to predetermined threshold value, the preset function includes penalty and/or kernel function；Fluid type of reservoir through after the correction is pre-
Survey model and be defined as training the fluid type of reservoir through forecast model for obtaining.
Further, after the type for obtaining the reservoir fluid to be identified, methods described also includes：When the institute for obtaining
When the type for stating reservoir fluid to be identified meets pre-conditioned, checking treatment is carried out to the reservoir fluid to be identified, and according to
The result of checking treatment determines again the type of the reservoir fluid to be identified.
Further, carry out checking treatment to the reservoir fluid to be identified to specifically include：By the reservoir stream to be identified
The height above sea level of body is compared respectively with the height above sea level of default oil-water interfaces and the height above sea level of default gasoil horizon, and root
Determine the actual type of the reservoir fluid to be identified according to comparative result.
For achieving the above object, on the other hand the application provides a kind of identifying device of fluid type of reservoir through, described device
Including：Training set acquiring unit, for obtaining reservoir fluid training set, the reservoir fluid training set includes having determined that type
Multiple reservoir fluids and the training characteristics value that is respectively associated with reservoir fluid each described；Forecast model acquiring unit, uses
Fluid type of reservoir through forecast model is obtained in training according to the reservoir fluid training set, wherein, for reservoir fluid instruction
Practice each reservoir fluid concentrated, the fluid type of reservoir through forecast model is satisfied by：It is special according to the training of the reservoir fluid
The fluid type of reservoir through that value indicative prediction is obtained is consistent with the type of the fixed reservoir fluid；Object feature value obtains single
Unit, for obtaining the object feature value of reservoir fluid to be identified；Type prediction unit, for pre- using the fluid type of reservoir through
Survey model to be predicted the object feature value, obtain the type of the reservoir fluid to be identified.
Further, described device also includes normalized unit, and the normalized unit is used for according to following
Formula is normalized to each training characteristics value in the reservoir fluid training set：
Wherein, B represents the training characteristics value after normalized, and A represents the training characteristics value before normalization, AminTable
Show the minima in such training characteristics value, AmaxRepresent the maximum in such training characteristics value.
Further, the forecast model acquiring unit is specifically included：Initial predicted module, for according to initial reservoir stream
Body type prediction model, prediction obtains the reservoir fluid training corresponding with the training characteristics value in the reservoir fluid training set
Type；Correction module, for training the error pair between type and fixed fluid type of reservoir through according to the reservoir fluid
Preset function in the initial reservoir fluid type forecast model is corrected, so that according to the reservoir fluid class after correction
Type forecast model prediction obtain reservoir fluid training type and fixed fluid type of reservoir through between error be less than or
Equal to predetermined threshold value, the preset function includes penalty and/or kernel function；Model determining module, for by the correction
Fluid type of reservoir through forecast model afterwards is defined as training the fluid type of reservoir through forecast model for obtaining.
The technical scheme provided from above the application embodiment, the application is by by defining natural potential amplitude
Than judging nicety rate can be caused to increase.Additionally, taking into full account the shadow that reservoir characteristic is recognized to fluid type of reservoir through in identification
Ring, and using Reservoir type as one of training characteristics value, make result more meet research area's practical situation.Furthermore, it is pre- by training
Model is surveyed, the subjectivity in fluid type of reservoir through identification process can be reduced, so as to improve the standard of fluid type of reservoir through identification
True rate.
With reference to explanation hereinafter and accompanying drawing, the particular implementation of the application is disclose in detail, specify the original of the application
Reason can be in adopted mode.It should be understood that presently filed embodiment is not so limited in scope.In appended power
In the range of the spirit and terms that profit is required, presently filed embodiment includes many changes, modifications and equivalent.
The feature for describing for a kind of embodiment and/or illustrating can be in same or similar mode one or more
It is combined with the feature in other embodiment used in individual other embodiment, or substitute the feature in other embodiment.
It should be emphasized that term "comprises/comprising" refers to the presence of feature, one integral piece, step or component when using herein, but and
It is not excluded for the presence of one or more further features, one integral piece, step or component or additional.
Description of the drawings
Included accompanying drawing is used for providing being further understood from the application embodiment, which constitutes the one of description
Part, for illustrating presently filed embodiment, and comes together to explain the principle of the application with word description.It should be evident that
Drawings in the following description are only some embodiments of the application, for those of ordinary skill in the art, are not being paid
On the premise of going out creative labor, can be with according to these other accompanying drawings of accompanying drawings acquisition.In the accompanying drawings：
A kind of recognition methodss flow chart of fluid type of reservoir through that Fig. 1 is provided for the application embodiment；
A kind of functional block diagram of the identifying device of fluid type of reservoir through that Fig. 2 is provided for the application embodiment.
Specific embodiment
In order that those skilled in the art more fully understand the technical scheme in the application, below in conjunction with the application reality
The accompanying drawing in mode is applied, the technical scheme in the application embodiment is clearly and completely described, it is clear that described
Embodiment is only a part of embodiment of the application, rather than the embodiment of whole.Based on the embodiment party in the application
Formula, all other embodiment that those of ordinary skill in the art are obtained under the premise of creative work is not made all should
When the scope for belonging to the application protection.
A kind of recognition methodss flow chart of fluid type of reservoir through that Fig. 1 is provided for the application embodiment.Although hereafter retouching
State multiple operations that flow process includes occurring with particular order, but it should be clearly understood that these processes can include it is more or more
Few operation, these operations can be performed sequentially or executed in parallel (such as using parallel processor or multi-thread environment).Such as Fig. 1
Shown, methods described may comprise steps of.
Step S1：Reservoir fluid training set is obtained, the reservoir fluid training set includes having determined that multiple storages of type
Layer fluid and the training characteristics value being respectively associated with reservoir fluid each described.
In the present embodiment, the plurality of storage can be determined by being analyzed to static data and dynamic data
The type of layer fluid.Wherein, the static data can include oil test data, rock core information, landwaste data etc..The dynamic money
Material can include individual well day oil-producing, daily output liquid, moisture content etc..In the present embodiment, the type of reservoir fluid can be divided into five
Class：Gas-bearing formation, oil reservoir, low-resistivity reservoir, oil-water common-layer and water layer.Wherein, low-resistivity reservoir can be resistance in same petroleum system
Rate with close on and Reservoir type identical water layer ratio less than 2 oil reservoir.
In the present embodiment, oil test data can directly give the oiliness conclusion of reservoir, such that it is able to directly give storage
The type of layer fluid.For rock core information, fluid type of reservoir through, rock core oil bearing grade can be analyzed by observing oil bearing grade
Be generally divided into be full of oil, rich in oil, oil immersion, oil mark, oil stain, fluorescence this several class, in the present embodiment, under reservoir oil-containing
Oil stain is defined to, therefore oil bearing grade is oil reservoir in oil stain and can be assumed that for the above in core observation；Oil bearing grade is in oil
It can be assumed that being water layer below mark.For dynamic data, can be as follows to the standard that fluid type of reservoir through judges：Produce after penetrating out
Tolerance is defined as gas-bearing formation more than default aerogenesis threshold value, reservoir of the producing water ratio less than 5%；Not aqueous or moisture content is less than after penetrating out
5% reservoir is judged to oil reservoir；Reservoir of the rear moisture content between 5～90% will be penetrated open and be defined as oil-water common-layer；Will be aqueous big
Reservoir in 90% is defined as water layer.
In the present embodiment, the training characteristics value can at least include Reservoir type mark, reservoir natural gamma value,
Natural potential Amplitude Ratio, layer depth resistivity, interval transit time, neutron porosity and density of earth formations.Wherein, the Reservoir type mark
Knowledge can be obtained in such a way：The reservoir class being adapted with the type of reservoir fluid is determined from default Reservoir Classification set
Type, and obtain the Reservoir type mark being associated with the Reservoir type for determining.
In the present embodiment, the Reservoir type can be obtained according to the analysis of lithology or physical characterization data.Specifically,
Rock core information, landwaste data, core analysis chemical examination data, log etc. can be analyzed, it is considered to lithology, grain graininess, hole
The geologic characteristics such as type, reservoir tightness, reservoir properties, classify to reservoir.General is 4 classes by reservoir division, and this 4
The mark of class Reservoir type can be respectively 1 to 4, and the division of the Reservoir type can be as shown in table 1.
The Reservoir Classification standard of table 1
Rock core oiliness in table 1 can represent the fluid type of reservoir through.So, determine fluid type of reservoir through it
Afterwards, the Reservoir type being adapted with the type of reservoir fluid just can be determined from above-mentioned Reservoir Classification standard, such that it is able to
Determine that Reservoir type is identified.
In a detailed embodiment, at 2523～2525m of X-2 wells, formation testing result is 705 barrels of oil-producing of individual well day, no
Aqueous, reservoir average resistivity is 37.27ohmm, more than 2 times neighbouring of Reservoir type identical water layer resistivity, so as to can
To determine that it is conventional oil reservoir.Well log interpretation porosity is 15.9%, and well log interpretation permeability is 77.9mD, and density curve shows
Average density is 2.41g/cm3, and interval transit time meansigma methodss are 72.85us/ft, such that it is able to according to the synthetic determination of table 1, it is II
Class reservoir, Reservoir type is designated 2.
In another embodiment, at 2465～2468m of X-4 wells, oil-containing sign is not found in core observation, it is purple
Fluorescence is shown after outside line irradiation, due to being defined to oil stain under reservoir oil-containing, therefore its oil bearing grade is set to fluorescence, is judged to water
Layer.Well log interpretation porosity is 13.9%, and well log interpretation permeability is 31.2mD, and density curve shows that average density is 2.45g/
Cm3, interval transit time meansigma methodss are 65.8us/ft, and it is III class reservoir to synthetic determination, and Reservoir type is designated 3.
In another embodiment, at 2619～2624m of X-30 wells, reservoir penetrates out rear 550 barrels of individual well day oil-producing,
Not aqueous, reservoir average resistivity is 26.8ohmm, less than 2 times neighbouring of Reservoir type identical water layer resistivity, is judged
It is low-resistivity reservoir.Well log interpretation porosity is 18.1%, and well log interpretation permeability is 61.6mD, and density curve shows average close
Spend for 2.38g/cm3, interval transit time meansigma methodss are 76.6us/ft, and it is I class reservoir to synthetic determination, and Reservoir type is designated 1.
In the present embodiment, Reservoir type mark, reservoir natural gamma, layer depth resistivity, interval transit time, middle sub-aperture
Porosity and density of earth formations can directly read according to Reservoir type achievement in research and log.Natural potential Amplitude Ratio can be
The ratio of this layer of natural potential amplitude difference and reservoir natural potential amplitude difference maximum under same well same deposition background.Specifically, X-1
Reservoir type is designated 1 at 2566～2573m of well, reservoir natural gamma 59API, layer depth resistivity be 30Ohmm, sound wave
The time difference is 72 us/ft, neutron porosity is 0.082 and density of earth formations is 2.38 g/cm3.This layer of natural potential amplitude difference be
37mV, the well same deposition background reservoir natural potential amplitude difference maximum is 80 mV, therefore its natural potential Amplitude Ratio is
0.46。
In the present embodiment, after the training characteristics value for getting each reservoir fluid association, can be according to following public affairs
Formula is normalized to each training characteristics value in the reservoir fluid training set：
Wherein, B represents the training characteristics value after normalized, and A represents the training characteristics value before normalization, AminTable
Show the minima in such training characteristics value, AmaxRepresent the maximum in such training characteristics value.
Step S2：Fluid type of reservoir through forecast model is obtained according to reservoir fluid training set training, wherein, for institute
Each reservoir fluid in reservoir fluid training set is stated, the fluid type of reservoir through forecast model is satisfied by：According to the reservoir
The fluid type of reservoir through that the training characteristics value prediction of fluid is obtained is consistent with the type of the fixed reservoir fluid.
In the present embodiment, can be based on support vector machine come to the training characteristics value in the reservoir fluid training set
It is trained, so as to obtain the fluid type of reservoir through forecast model for recognizing fluid type of reservoir through.Specifically, can adopt
Libsvm graders being trained to training characteristics value, so as to the corresponding reservoir fluid class of the training characteristics value for judging be input into
Type.Because normal grader is often classified to two class data, and the fluid type of reservoir through in present embodiment is 5
Class, so as to need to be extended to normal grader.Specifically, voting mechanism can be taken to support to carry out various data
Classification.Existing 5 class data, can choose the i-th wherein different classes and jth class data, constitute a grader, so just can be with
2 classification problems for solving the i-th class and jth class data will be converted into the problem that 5 class data are classified.When the number for judging input
During according to belonging to the i-th class, the i-th class poll adds 1；When the data for judging input belong to jth class, jth class poll adds 1.Finally can be with
The most class of poll will be possessed and be defined as final judged result.
In the present embodiment, can predict and obtain and the reservoir stream according to initial reservoir fluid type forecast model
Training characteristics in body training set are worth corresponding reservoir fluid training type.In the initial reservoir fluid type forecast model
Parameter may be inaccurate, so as to cause to predict that the reservoir fluid for obtaining training type and actual fluid type of reservoir through are not inconsistent.
As such, it is possible to train the error between type and fixed fluid type of reservoir through to the initial storage according to the reservoir fluid
Preset function in layer fluid type prediction model is corrected, so that according to the fluid type of reservoir through forecast model after correction
The reservoir fluid that prediction is obtained trains the error between type and fixed fluid type of reservoir through less than or equal to default threshold
Value.In the present embodiment, the preset function includes penalty and/or kernel function.Specifically, penalty C and core letter
The best of breed of number r can be [C, r]=[1342,0.157], and so, fluid type of reservoir through forecast model after correction is tested
Card precision is up to 92.5%.So just the fluid type of reservoir through forecast model after the correction can be defined as training what is obtained
Fluid type of reservoir through forecast model.
The fluid type of reservoir through forecast model for training can obtain right according to the training characteristics value prediction of input exactly
That what is answered has determined the reservoir fluid of type.
Step S3：Obtain the object feature value of reservoir fluid to be identified.
In the present embodiment, after training obtains the higher fluid type of reservoir through forecast model of precision, just can obtain
Take the object feature value of reservoir fluid to be identified.The object feature value can be with the type of the training characteristics value in training process
It is identical.In the present embodiment, if be normalized to training characteristics value in the training process, then can be to described
Object feature value is normalized in the same way.
Step S4：The object feature value is predicted using the fluid type of reservoir through forecast model, obtains described
The type of reservoir fluid to be identified.
In the present embodiment, the object feature value can be input in the fluid type of reservoir through forecast model, from
And can predict and obtain the fluid type of reservoir through corresponding with the object feature value.
In the present embodiment, in order to further ensure that the accuracy of the fluid type of reservoir through that identification is obtained, can work as
When the type of the reservoir fluid described to be identified for obtaining meets pre-conditioned, the reservoir fluid to be identified is carried out at verification
Reason, and the type of the reservoir fluid to be identified is determined again according to the result of checking treatment.
Specifically, if the fluid type of reservoir through that identification is obtained is gas-bearing formation, oil reservoir or low-resistivity reservoir, oil can be utilized respectively
Matching type, Production development method are studied reservoir fluid between water termination method, well, so as to judge the reasonability of recognition result.If
The contradiction such as matching type, Production development method between recognition result and oil-water interfaces method, well, then it is assumed that recognition result is unreliable, it is impossible to enter
Row differentiates；If there is not contradiction, then it is assumed that recognition result reliability.
Oil-water interfaces method is by by the height above sea level of the height above sea level of the reservoir fluid to be identified and default oil-water interfaces
It is compared respectively with the height above sea level of default gasoil horizon, and the reality of the reservoir fluid to be identified is determined according to comparative result
Border type.Specifically, when reservoir fluid to be identified is located on oil-water interfaces, it is believed that it is oil reservoir；When reservoir stream to be identified
When body is located under oil-water interfaces, it is believed that it is water layer；When reservoir fluid to be identified is located at oil-water interfaces predeterminable range model nearby
When enclosing interior, it is believed that it is oil-water common-layer.Similarly, when reservoir fluid to be identified is located on gasoil horizon, it is believed that it is gas
Layer；When reservoir fluid to be identified is located under gasoil horizon, it is believed that it is oil reservoir.
Matching type is judged reservoir fluid to be identified by offset well identical layer position reservoir oiliness between well.For example, it is single
Judge that X-29 well E14 2613～2624m of substratum sand bodies middle part improves with reservoir property from features of logging curve, ground layer depth resistivity
Reduce on the contrary, with reference to additive method low-resistivity reservoir can not be judged to, need comprehensive items reservoir geology data to be analyzed.Should
Fault block oil-water interfaces are 2115m, find that the suspicious layer bottom surface height above sea level depth is 2101m after hole-deviation correction, and whole oil reservoir is equal
On oil-water interfaces, from E14 substratum top surfaces it can also be seen that it is within oil area, therefore judge that it is oil reservoir.It is another
Aspect, pair with the connection of X-29 wells sand body and the Production development of adjacent X-30 is analyzed, it is found that the well penetrates out the initial stage after the layer
It is day oil-producing 110.6d/t, not aqueous, 1.5 ten thousand tons of oil-producing is tired out at present.X-29 and its resistivity curve feature similarity, and construction location
It is slightly higher, judge that it is oil reservoir.To sum up, confirm that the suspicious layer is low-resistivity reservoir.
Rear Production development is penetrated out according to a certain layer position of individual well to judge other layer of position of same well.Such as X-S-2 well thickness
The block sand body of 32.5m point two sections, epimere reservoir properties are good, shale content is low, and hypomere reservoir properties are poor, shale content
Height, apparently higher than hypomere, lower two intersegmental nothings stablize interlayer to epimere ground layer depth resistivity.After to penetrating out upper strata, occur 10 days it is anhydrous
The oil recovery phase, day at initial stage oil-producing 150.5t, can assert that it is oil reservoir, and bottom is without adjacent water body, therefore bottom is to be all oil reservoir.
The application also provides a kind of identifying device of fluid type of reservoir through.Fig. 2 is referred to, described device includes：
Training set acquiring unit 100, for obtaining reservoir fluid training set, the reservoir fluid training set includes true
The multiple reservoir fluids for determining type and the training characteristics value being respectively associated with reservoir fluid each described；
Forecast model acquiring unit 200, it is pre- for obtaining fluid type of reservoir through according to reservoir fluid training set training
Model is surveyed, wherein, for each reservoir fluid in the reservoir fluid training set, the fluid type of reservoir through forecast model is equal
Meet：The fluid type of reservoir through obtained according to the prediction of the training characteristics value of the reservoir fluid and the fixed reservoir fluid
Type it is consistent；
Object feature value acquiring unit 300, for obtaining the object feature value of reservoir fluid to be identified；
Type prediction unit 400, for being carried out to the object feature value using the fluid type of reservoir through forecast model
Prediction, obtains the type of the reservoir fluid to be identified.
In the present embodiment, described device also includes normalized unit, and the normalized unit is used to press
Each training characteristics value in the reservoir fluid training set is normalized according to following formula：
Wherein, B represents the training characteristics value after normalized, and A represents the training characteristics value before normalization, AminTable
Show the minima in such training characteristics value, AmaxRepresent the maximum in such training characteristics value.
In the present embodiment, the forecast model acquiring unit 200 is specifically included：
Initial predicted module, for according to initial reservoir fluid type forecast model, prediction to be obtained and the reservoir fluid
Training characteristics in training set are worth corresponding reservoir fluid training type；
Correction module, for training the error between type and fixed fluid type of reservoir through according to the reservoir fluid
Preset function in the initial reservoir fluid type forecast model is corrected, so that according to the reservoir fluid after correction
Error between reservoir fluid that type prediction model prediction is obtained training type and fixed fluid type of reservoir through be less than or
Person is equal to predetermined threshold value, and the preset function includes penalty and/or kernel function；
Model determining module, for being defined as training the storage for obtaining by the fluid type of reservoir through forecast model after the correction
Layer fluid type prediction model.
It should be noted that the specific implementation of above-mentioned each functional module is consistent with the description in step S1 to S4,
Here become and repeat no more.
Therefore, the application is by by defining natural potential Amplitude Ratio, causing judging nicety rate to increase.This
Outward, the impact that reservoir characteristic is recognized to fluid type of reservoir through is taken into full account in identification, and using Reservoir type as training characteristics
One of value, makes result more meet research area's practical situation.Furthermore, by training forecast model, fluid type of reservoir through can be reduced
Subjectivity in identification process, so as to improve the accuracy rate of fluid type of reservoir through identification.
Description to the various embodiments of the application above is supplied to those skilled in the art with the purpose for describing.It is not
Be intended to exhaustion or be not intended to limit the invention to single disclosed embodiment.As described above, the application's is various
Substitute and change will be apparent for above-mentioned technology one of ordinary skill in the art.Therefore, although specifically beg for
The embodiment of some alternatives has been discussed, but other embodiment will be apparent, or those skilled in the art are relative
Easily draw.The application is intended to be included in this all replacement of the invention for having discussed, modification and change, and falls
Other embodiment in the spirit and scope of above-mentioned application.
Each embodiment in this specification is described by the way of progressive, identical similar between each embodiment
Part mutually referring to what each embodiment was stressed is the difference with other embodiment.Especially, it is right
For device embodiments, because it is substantially similar to method embodiment, so description is fairly simple, related part ginseng
The part explanation of square method embodiment.
Although depicting the application by embodiment, it will be appreciated by the skilled addressee that the application has many deformations
With change without deviating from spirit herein, it is desirable to which appended claim includes these deformations and changes without deviating from the application
Spirit.
Claims (10)
1. a kind of recognition methodss of fluid type of reservoir through, it is characterised in that methods described includes：
Obtain reservoir fluid training set, the reservoir fluid training set include having determined that multiple reservoir fluids of type and with
The training characteristics value that each described reservoir fluid is respectively associated；
Fluid type of reservoir through forecast model is obtained according to reservoir fluid training set training, wherein, for the reservoir fluid
Each reservoir fluid in training set, the fluid type of reservoir through forecast model is satisfied by：According to the training of the reservoir fluid
The fluid type of reservoir through that eigenvalue prediction is obtained is consistent with the type of the fixed reservoir fluid；
Obtain the object feature value of reservoir fluid to be identified；
The object feature value is predicted using the fluid type of reservoir through forecast model, obtains the reservoir stream to be identified
The type of body.
2. method according to claim 1, it is characterised in that the training characteristics value at least include Reservoir type mark,
Reservoir natural gamma value, natural potential Amplitude Ratio, layer depth resistivity, interval transit time, neutron porosity and density of earth formations.
3. method according to claim 2, it is characterised in that the Reservoir type mark is obtained in such a way：
The Reservoir type being adapted with the type of reservoir fluid is determined from default Reservoir Classification set, and is obtained and the storage for determining
The associated Reservoir type mark of channel type.
4. method according to claim 1, it is characterised in that described after the step of obtaining reservoir fluid training set
Method also includes：Each training characteristics value in the reservoir fluid training set is normalized according to the following equation：
Wherein, B represents the training characteristics value after normalized, and A represents the training characteristics value before normalization, AminRepresenting should
Minima in class training characteristics value, AmaxRepresent the maximum in such training characteristics value.
5. method according to claim 1, it is characterised in that reservoir stream is obtained according to reservoir fluid training set training
Body type prediction model is specifically included：
According to initial reservoir fluid type forecast model, prediction is obtained and the training characteristics value phase in the reservoir fluid training set
Corresponding reservoir fluid trains type；
Error between type and fixed fluid type of reservoir through is trained to the initial reservoir stream according to the reservoir fluid
Preset function in body type prediction model is corrected, so that being predicted according to the fluid type of reservoir through forecast model after correction
Error between the reservoir fluid training type for obtaining and fixed fluid type of reservoir through is less than or equal to predetermined threshold value, institute
Preset function is stated including penalty and/or kernel function；
Fluid type of reservoir through forecast model after the correction is defined as training the fluid type of reservoir through forecast model for obtaining.
6. method according to claim 1, it is characterised in that after the type for obtaining the reservoir fluid to be identified,
Methods described also includes：
When the type of the reservoir fluid described to be identified for obtaining meets pre-conditioned, school is carried out to the reservoir fluid to be identified
Process is tested, and the type of the reservoir fluid to be identified is determined again according to the result of checking treatment.
7. method according to claim 6, it is characterised in that checking treatment is carried out to the reservoir fluid to be identified concrete
Including：
By the height above sea level of the reservoir fluid to be identified and the height above sea level of default oil-water interfaces and the sea of default gasoil horizon
Degree of lifting is compared respectively, and determines the actual type of the reservoir fluid to be identified according to comparative result.
8. a kind of identifying device of fluid type of reservoir through, it is characterised in that described device includes：
Training set acquiring unit, for obtaining reservoir fluid training set, the reservoir fluid training set includes having determined that type
Multiple reservoir fluids and the training characteristics value that is respectively associated with reservoir fluid each described；
Forecast model acquiring unit, for obtaining fluid type of reservoir through forecast model according to reservoir fluid training set training,
Wherein, for each reservoir fluid in the reservoir fluid training set, the fluid type of reservoir through forecast model is satisfied by：Root
The type of the fluid type of reservoir through and fixed reservoir fluid for obtaining is predicted according to the training characteristics value of the reservoir fluid
Unanimously；
Object feature value acquiring unit, for obtaining the object feature value of reservoir fluid to be identified；
Type prediction unit, for being predicted to the object feature value using the fluid type of reservoir through forecast model, is obtained
To the type of the reservoir fluid to be identified.
9. device according to claim 8, it is characterised in that described device also includes normalized unit, described to return
One change processing unit is used to according to the following equation be normalized each training characteristics value in the reservoir fluid training set
Process：
Wherein, B represents the training characteristics value after normalized, and A represents the training characteristics value before normalization, AminRepresenting should
Minima in class training characteristics value, AmaxRepresent the maximum in such training characteristics value.
10. device according to claim 8, it is characterised in that the forecast model acquiring unit is specifically included：
Initial predicted module, for according to initial reservoir fluid type forecast model, prediction to obtain being trained with the reservoir fluid
The training characteristics of concentration are worth corresponding reservoir fluid training type；
Correction module, for the error between type and fixed fluid type of reservoir through to be trained according to the reservoir fluid to institute
The preset function stated in initial reservoir fluid type forecast model is corrected, so that according to the fluid type of reservoir through after correction
The reservoir fluid that forecast model prediction is obtained trains the error between type and fixed fluid type of reservoir through to be less than or wait
In predetermined threshold value, the preset function includes penalty and/or kernel function；
Model determining module, for being defined as training the reservoir stream for obtaining by the fluid type of reservoir through forecast model after the correction
Body type prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610940183.4A CN106570524A (en) | 2016-10-25 | 2016-10-25 | Reservoir fluid type identifying method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610940183.4A CN106570524A (en) | 2016-10-25 | 2016-10-25 | Reservoir fluid type identifying method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106570524A true CN106570524A (en) | 2017-04-19 |
Family
ID=58536277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610940183.4A Pending CN106570524A (en) | 2016-10-25 | 2016-10-25 | Reservoir fluid type identifying method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106570524A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110056348A (en) * | 2019-04-25 | 2019-07-26 | 中国海洋石油集团有限公司 | A kind of method and system of measurement formation fluid composition and property |
| CN112862139A (en) * | 2019-11-27 | 2021-05-28 | 北京国双科技有限公司 | Fluid type prediction model construction method, fluid type prediction method and device |
| CN115961952A (en) * | 2023-02-21 | 2023-04-14 | 成都理工大学 | Reservoir fluid comprehensive discrimination method based on combination parameters in oil and gas reservoir |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7454321B1 (en) * | 2002-01-07 | 2008-11-18 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Robust, optimal subsonic airfoil shapes |
| 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 |
| CN104992178A (en) * | 2015-06-16 | 2015-10-21 | 中国石油化工股份有限公司 | Tight sandstone fluid type identification method based on support vector machine simulation cross plot |
-
2016
- 2016-10-25 CN CN201610940183.4A patent/CN106570524A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7454321B1 (en) * | 2002-01-07 | 2008-11-18 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Robust, optimal subsonic airfoil shapes |
| 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 |
| CN104992178A (en) * | 2015-06-16 | 2015-10-21 | 中国石油化工股份有限公司 | Tight sandstone fluid type identification method based on support vector machine simulation cross plot |
Non-Patent Citations (2)
| Title |
|---|
| 刘英才 等: "阿达油田低阻、低对比油层的成因与识别", 《江汉石油学院学报》 * |
| 张银德 等: "支持向量机方法在低阻油层流体识别中的应用", 《石油物探》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110056348A (en) * | 2019-04-25 | 2019-07-26 | 中国海洋石油集团有限公司 | A kind of method and system of measurement formation fluid composition and property |
| CN110056348B (en) * | 2019-04-25 | 2021-05-11 | 中国海洋石油集团有限公司 | Method and system for determining formation fluid composition and properties |
| CN112862139A (en) * | 2019-11-27 | 2021-05-28 | 北京国双科技有限公司 | Fluid type prediction model construction method, fluid type prediction method and device |
| CN115961952A (en) * | 2023-02-21 | 2023-04-14 | 成都理工大学 | Reservoir fluid comprehensive discrimination method based on combination parameters in oil and gas reservoir |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101930082B (en) | Method for distinguishing reservoir fluid type by adopting resistivity data | |
| Grana et al. | Quantitative log interpretation and uncertainty propagation of petrophysical properties and facies classification from rock-physics modeling and formation evaluation analysis | |
| Hammes et al. | Regional assessment of the Eagle Ford Group of South Texas, USA: Insights from lithology, pore volume, water saturation, organic richness, and productivity correlations | |
| Alnahwi et al. | Mineralogical composition and total organic carbon quantification using x-ray fluorescence data from the Upper Cretaceous Eagle Ford Group in southern Texas | |
| CN103993871B (en) | Method and device for processing well logging information of thin interbed stratums in standardization mode | |
| Scheidt et al. | Probabilistic falsification of prior geologic uncertainty with seismic amplitude data: Application to a turbidite reservoir case | |
| Corina et al. | Automatic lithology prediction from well logging using kernel density estimation | |
| CN109653725A (en) | A layer water flooding degree log interpretation method is stored up based on sedimentary micro and the mixed of rock phase | |
| Amosu et al. | Effective machine learning identification of TOC-rich zones in the Eagle Ford Shale | |
| Wang et al. | Improved pore structure prediction based on MICP with a data mining and machine learning system approach in Mesozoic strata of Gaoqing field, Jiyang depression | |
| Grana et al. | Petroelastic and geomechanical classification of lithologic facies in the Marcellus Shale | |
| Liu et al. | Petrophysical characteristics and log identification of lacustrine shale lithofacies: A case study of the first member of Qingshankou Formation in the Songliao Basin, Northeast China | |
| Masoudi et al. | Net pay determination by artificial neural network: Case study on Iranian offshore oil fields | |
| CN113033648A (en) | Method for realizing logging interpretation by using machine learning algorithm | |
| CN106570524A (en) | Reservoir fluid type identifying method and device | |
| Allen et al. | Geologically constrained electrofacies classification of fluvial deposits: An example from the Cretaceous Mesaverde Group, Uinta and Piceance Basins | |
| Barach et al. | Development and identification of petrophysical rock types for effective reservoir characterization: Case study of the Kristine Field, Offshore Sabah | |
| Caf et al. | Dolomitization geometry and reservoir quality from supervised Bayesian classification and probabilistic neural networks: Midland Basin Leonardian Wichita and Clear Fork Formations | |
| CN109826623A (en) | Knowledge method is sentenced in a kind of geophysical log of tight sandstone reservoir stratification seam | |
| Davies | Permeability Modelling of a Sandstone Reservoir in Parts of the Niger Delta | |
| US11592593B2 (en) | Modeling hydrocarbon reservoirs using rock fabric classification at reservoir conditions | |
| Emelyanova et al. | Detecting specific facies in well-log data sets using knowledge-driven hierarchical clustering | |
| Ghanbarnejad Moghanloo et al. | Integrating watershed segmentation algorithm and supervised Bayesian classification for the assessment of petrophysical parameters, pore properties, and lithofacies: a case study from Abadan Plain, SW Iran | |
| Bello et al. | Multiuse of seismic and attribute mapping for field appraisal and development in the Eagle Ford Shale: Mapping TOC, porosity and seal integrity | |
| Lu et al. | Prediction of the pore structure by machine learning techniques in the carbonate reservoirs in Iraq H oilfield |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170419 |