CN103485770B - A kind of method and system asking for oil saturation based on artificial neural network - Google Patents
A kind of method and system asking for oil saturation based on artificial neural network Download PDFInfo
- Publication number
- CN103485770B CN103485770B CN201310244257.7A CN201310244257A CN103485770B CN 103485770 B CN103485770 B CN 103485770B CN 201310244257 A CN201310244257 A CN 201310244257A CN 103485770 B CN103485770 B CN 103485770B
- Authority
- CN
- China
- Prior art keywords
- neural network
- network model
- curve
- log
- rock
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to oil saturation technical field, disclose a kind of method and system asking for oil saturation based on artificial neural network.Wherein, method includes: rock core is carried out rock-electric test, obtains log data；A and b is entered as 1, and inverse obtains m and n；Using the log of sign lithology, physical property and oiliness as input layer, select m and n as output layer, set up BP neural network model；M ', n ' is obtained by substituting into model with the log data corresponding to m, the n obtained by rock-electric test；M, n of being obtained by rock-electric test are compared with m ', the n ' obtained by model, obtains the model that error is minimum；The model that data on the log of corresponding input layer substitute into error minimum is obtained cementation factor curve and saturation exponent curve；Curve is substituted into Archie formula and carries out the calculating of oil saturation.The present invention improves the computational accuracy being obtained oil saturation by Archie formula.
Description
Technical field
The present invention relates to oil saturation technical field, be primarily adapted for use in the method and system asking for oil saturation based on artificial neural network.
Background technology
When utilizing well-log information identification oil, gas and water layer, Archie formula always asks for the main method of reservoir oil-containing (gas) saturation, its application effect is mainly affected by litho-electric parameters such as lithology factor a, lithology constant b, cementation factor m, saturation exponents, the particularly value of cementation factor m and saturation exponent directly determines Archie formula and calculates the accuracy of oil saturation (for lithology factor a and lithology constant b the two parameter, often suppose that they changes are little, and approximation is taken as 1).In current Archie formula, the numerical value of litho-electric parameters m, n is mainly obtained by experiment average value measured method and region empirical value method.
Both approaches is all based on the litho-electric parameters result that laboratory is finely measured, and takes the meansigma methods of some rock-electric test results or directly invoke this region litho-electric parameters empirical value and carry out oil saturation calculating in a certain region or interval.The method has carried out equalization to the value of litho-electric parameters m, n, fails to take into full account the concrete influence factor of reservoir litho-electric parameters, and calculated oil saturation error is bigger.
Summary of the invention
The technical problem to be solved is to provide a kind of method and system asking for oil saturation based on artificial neural network, it can improve the computational accuracy of cementation factor and saturation exponent, thus improves the computational accuracy being obtained oil saturation by Archie formula.
For solving above-mentioned technical problem, the invention provides a kind of method asking for oil saturation based on artificial neural network, including:
In stratum, choose rock core, described rock core is carried out rock-electric test, obtain the log data in Archie formula；
Lithology factor a in Archie formula and lithology constant b is entered as 1, and inverse obtains cementation factor m and saturation exponent；
Using the log of sign lithology, physical property and oiliness as the input layer of described BP neural network model, select described cementation factor m and described saturation exponent as the output layer of BP neural network model；Set up the relation between described output layer and described input layer by simulation, set up BP neural network model；
Be calculated m by substituting into described BP neural network model with the log data of the depth of stratum corresponding to m, the n obtained by described rock-electric test, n,；
By m, n of being obtained by described rock-electric test and the m obtained by described BP neural network model, n, compare, obtain the BP neural network model that error is minimum；
The BP neural network model that data on the described log of corresponding input layer substitute into described error minimum is carried out cementation factor curve and the saturation exponent curve that node-by-node algorithm is dynamically changed；
Data on the cementation factor curve of the described dynamic change obtained and saturation exponent curve are substituted into Archie formula and carries out the calculating of oil saturation.
Further, the node number in the intermediate layer of described BP neural network model is 2N+1, and wherein N is the nodes of described input layer, and N is the integer more than 1.
Further, described by m, n of being obtained by rock-electric test and the m obtained by BP neural network model,, n, compare, the concrete grammar obtaining the minimum BP neural network model of error includes: regulate the span of the input layer log of described BP neural network model, obtain different m, n, value；By m, n of being obtained by described rock-electric test and the m obtained by BP neural network model, n, carry out size and compare；According to the relative coefficient of the BP neural network model corresponding with comparative result, obtain the BP neural network model that described error is minimum.
Further, the relative coefficient of the BP neural network model that described basis is corresponding with comparative result, obtain the BP neural network model that error is minimum, including: if the relative coefficient of described BP neural network model is equal to 0.85, then corresponding with relative coefficient 0.85 BP neural network model is the BP neural network model that described error is minimum.
Further, the log of described sign lithology includes gamma ray curve, spontaneous potential curve and CAL；The log of described sign physical property includes interval transit time curve, porosity curve, neutron curve, density curve and permeability curve；The log of described sign oiliness includes resistivity curve and stratum shallow resistivity curve in true formation resistivity curve, stratum.
Present invention also offers a kind of system asking for oil saturation based on artificial neural network, including:
Rock-electric test module, for choosing rock core in stratum, carries out rock-electric test to described rock core, obtains the log data in Archie formula；
First computing module, for the lithology factor a in Archie formula and lithology constant b is entered as 1, inverse obtains cementation factor m and saturation exponent；
Oneth BP neural network model generation module, for will characterize lithology, physical property and oiliness log as the input layer of described BP neural network model, select described cementation factor m and described saturation exponent as the output layer of BP neural network model；Set up the relation between described output layer and described input layer by simulation, set up BP neural network model；
Second computing module, for being calculated m by substituting into described BP neural network model with the log data of the depth of stratum corresponding to m, the n obtained by described rock-electric test, n,；
2nd BP neural network model generation module, for by m, n of being obtained by described rock-electric test and the m obtained by described BP neural network model, n, compare, obtain the BP neural network model of error minimum；
3rd computing module, for carrying out, by the BP neural network model that the data on the described log of corresponding input layer substitute into described error minimum, cementation factor curve and the saturation exponent curve that node-by-node algorithm is dynamically changed；
4th computing module, carries out the calculating of oil saturation for the data on the cementation factor curve of the described dynamic change obtained and saturation exponent curve are substituted into Archie formula.
Further, in a described BP neural network model generation module, the node number in the intermediate layer of described BP neural network model is 2N+1, and wherein N is the nodes of described input layer, and N is the integer more than 1.
Further, described 2nd BP neural network model generation module, including:
Index acquiring unit, for regulating the span of input layer log of described BP neural network model, obtains different m, n, value；
Comparing unit, for by m, n of being obtained by described rock-electric test and the m obtained by BP neural network model, n, carry out size and compare；
BP neural network model signal generating unit, for the relative coefficient according to the BP neural network model corresponding with comparative result, obtains the BP neural network model that described error is minimum.
Further, described BP neural network model signal generating unit, if the relative coefficient specifically for described BP neural network model is equal to 0.85, then corresponding with relative coefficient 0.85 BP neural network model is the BP neural network model that described error is minimum.
Further, the log of described sign lithology includes gamma ray curve, spontaneous potential curve and CAL；The log of described sign physical property includes interval transit time curve, porosity curve, neutron curve, density curve and permeability curve；The log of described sign oiliness includes resistivity curve and stratum shallow resistivity curve in true formation resistivity curve, stratum.
The beneficial effects of the present invention is:
The method and system asking for oil saturation based on artificial neural network that the present invention provides, first pass through the log data that rock-electric test obtains in Archie formula, then a and b are entered as 1, and inverse goes out m and n；Again m, n of being obtained by rock-electric test are compared with m ', the n ' obtained by BP neural network model, obtain the BP neural network model that error is minimum；The cementation factor curve dynamically changed by the BP neural network model that error is minimum again and saturation exponent curve, substituted into Archie formula by the cementation factor dynamically changed and saturation exponent and carry out the calculating of oil saturation.On the basis of present invention log data in known Archie formula, by a and b is entered as 1, inverse goes out m and n, therefore overcomes the calculating defect of m and n equalization in conventional method, improves the computational accuracy of oil saturation.Compared with conventional method, the present invention is sensitiveer for the reaction of oil-gas Layer by cementation factor and the saturation exponent calculated oil saturation curve of dynamically change, and its variation tendency can react the oil-bearing information of Reservoir Section more realistically.
Accompanying drawing explanation
The flow chart of the method asking for oil saturation based on artificial neural network that Fig. 1 provides for the embodiment of the present invention；
Fig. 2 asks for, based on artificial neural network, the actual processing effect figure that the method for oil saturation obtains for provided by the embodiment of the present invention.
Detailed description of the invention
By technological means and effect that the present invention by reach predetermined goal of the invention taked is expanded on further, below in conjunction with accompanying drawing and preferred embodiment, to propose according to the present invention ask for the detailed description of the invention of method and system of oil saturation based on artificial neural network and operation principle is described in detail.
As shown in Figure 1, what the embodiment of the present invention provided ask for the method for oil saturation based on artificial neural network includes:
In stratum, choose rock core, rock core is carried out rock-electric test, obtain the log data in Archie formula；Concrete, Selecting research region in the earth formation, in survey region, choose rock core, rock core is carried out rock-electric test, obtain the log data in Archie formula；Wherein, selected survey region to ensure the data of at least rock-electric test of 20 pieces of samples.
Lithology factor a in Archie formula and lithology constant b is entered as 1, and inverse obtains cementation factor m and saturation exponent；
Concrete, due to known formation resistivity factor, wherein, Φ is porosity, and a is lithology factor, and m is cementation factor；Lithology factor a is entered as 1, cementation factor m can be obtained with inverse.In like manner, due to known resistance enhancement coefficient, wherein, SwFor oil saturation, b is lithology constant, and n is saturation exponent；Lithology constant b is entered as 1, saturation exponent can be obtained with inverse.
Using characterize lithology, physical property and oiliness log as the input layer of BP neural network model, the node number in the intermediate layer of BP neural network model is 2N+1, and wherein N is the nodes of input layer, and N is the integer more than 1.Select cementation factor m and saturation exponent as the output layer of BP neural network model；Set up the relation between output layer and input layer by simulation, set up BP neural network model；
Log data with the depth of stratum corresponding to m, the n obtained by rock-electric test is substituted into BP neural network model and is calculated m ', n '；
M, n of being obtained by rock-electric test are compared with m ', the n ' obtained by BP neural network model, obtains the BP neural network model that error is minimum；Concrete, the span of the input layer log of regulation BP neural network model, obtain different m ', n ' value；With m ', the n ' obtained by BP neural network model, m, n of being obtained by rock-electric test are carried out size compare；According to the relative coefficient of the BP neural network model corresponding with comparative result, obtain the BP neural network model that error is minimum.Further, if the relative coefficient of the BP neural network model obtained is equal to 0.85, then corresponding with relative coefficient 0.85 BP neural network model is the BP neural network model that error is minimum.
The BP neural network model that data on the log of corresponding input layer substitute into error minimum is carried out cementation factor curve and the saturation exponent curve that node-by-node algorithm is dynamically changed；
Data on the cementation factor curve dynamically changed obtained and saturation exponent curve are substituted into Archie formula and carries out the calculating of oil saturation.
Preferably, the log characterizing lithology includes gamma ray curve, spontaneous potential curve and CAL；The log characterizing physical property includes interval transit time curve, porosity curve, neutron curve, density curve and permeability curve；The log characterizing oiliness includes resistivity curve and stratum shallow resistivity curve in true formation resistivity curve, stratum.In the present embodiment, the log characterizing lithology is gamma ray curve；The log characterizing physical property includes interval transit time curve and porosity curve；The log characterizing oiliness is true formation resistivity curve.
Present invention also offers a kind of system asking for oil saturation based on artificial neural network, including:
Rock-electric test module, for choosing rock core in stratum, carries out rock-electric test to rock core, obtains the log data in Archie formula；Further, rock-electric test module, specifically for Selecting research region in the earth formation, chooses rock core in survey region, and rock core is carried out rock-electric test, obtains the log data in Archie formula；Wherein, selected survey region to ensure the data of at least rock-electric test of 20 pieces of samples.
First computing module, for the lithology factor a in Archie formula and lithology constant b is entered as 1, inverse obtains cementation factor m and saturation exponent；
Concrete, due to known formation resistivity factor, wherein, Φ is porosity, and a is lithology factor, and m is cementation factor；Lithology factor a is entered as 1, cementation factor m can be obtained with inverse.In like manner, due to known resistance enhancement coefficient, wherein, SwFor oil saturation, b is lithology constant, and n is saturation exponent；Lithology constant b is entered as 1, saturation exponent can be obtained with inverse.
Oneth BP neural network model generation module, for will characterize lithology, physical property and oiliness log as the input layer of BP neural network model, the node number in the intermediate layer of BP neural network model is 2N+1, and wherein N is the nodes of input layer, and N is the integer more than 1.Select cementation factor m and saturation exponent as the output layer of BP neural network model；Set up the relation between output layer and input layer by simulation, set up BP neural network model；
Second computing module, is calculated m ', n ' for the log data with the depth of stratum corresponding to m, the n obtained by rock-electric test is substituted into BP neural network model；
2nd BP neural network model generation module, for m, n of being obtained by rock-electric test being compared with m ', the n ' obtained by BP neural network model, obtains the BP neural network model that error is minimum；Specifically, the 2nd BP neural network model generation module, including index acquiring unit, for regulating the span of the input layer log of BP neural network model, obtain different m, n, value；Comparing unit, for by m, n of being obtained by rock-electric test and the m obtained by BP neural network model, n, carry out size and compare；BP neural network model signal generating unit, for according to by comparing the relative coefficient of the BP neural network model obtained, obtaining the BP neural network model that error is minimum.Further, BP neural network model signal generating unit, if the relative coefficient of the BP neural network model specifically for obtaining is equal to 0.85, then corresponding with relative coefficient 0.85 BP neural network model is the BP neural network model that error is minimum.
3rd computing module, for carrying out, by the BP neural network model that the data on the log of corresponding input layer substitute into error minimum, cementation factor curve and the saturation exponent curve that node-by-node algorithm is dynamically changed；
4th computing module, carries out the calculating of oil saturation for the data on the cementation factor curve of the dynamically change obtained and saturation exponent curve are substituted into Archie formula.
Preferably, the log characterizing lithology includes gamma ray curve, spontaneous potential curve and CAL；The log characterizing physical property includes interval transit time curve, porosity curve, neutron curve, density curve and permeability curve；The log characterizing oiliness includes resistivity curve and stratum shallow resistivity curve in true formation resistivity curve, stratum.
Embodiment
1, choosing Changqing oilfields Ji's plateau area is survey region, and survey region is carried out STRATIGRAPHIC DIVISION.Wherein, No. 1 survey region includes 12 mouthfuls of core holes and 29 mouthfuls of offset wells, collects the rock-electric test result of 65 pieces of core samples altogether for long 8 and long 9 two research layer positions.
2, the lithology factor a and lithology constant b of 65 pieces of core sample experiments are entered as 1, use Archie formula inverse to go out m, n value.
3, interval transit time (AC), natural gamma (GR), porosity (Por) and input layer that true formation resistivity (Rt) is BP neural network model are chosen, the nodes in intermediate layer is 9, using m, n rock-electric test result of the corresponding degree of depth as output layer, set up the relation between output layer and input layer by simulation, set up long by 8, m, n computation model on long 9 stratum respectively.
4, the log data with the depth of stratum corresponding to m, the n obtained by rock-electric test is substituted into BP neural network model and be calculated m ', n '.
5, model accuracy is carried out self-checking, constantly adjust the span of the input layer log of BP neural network model, make the dependency of assay reach maximum, thus obtain the BP neural network model that error is minimum.
6, the BP neural network model that the data on the log of corresponding input layer are substituted into long by 8, long 9 interval errors respectively minimum carries out node-by-node algorithm, the available cementation factor curve dynamically changed with the degree of depth and saturation exponent curve, the final oil saturation using dynamic cementation factor curve and the saturation exponent curve negotiating Archie formula corresponding degree of depth of calculating.Seeing Fig. 2,8 layers of 2551.4m to 2563.2m depth segment of this well length, the average oil saturation utilizing conventional saturation computation method to obtain is 43%, and conclusion is oil-containing water layer；The average oil saturation tried to achieve by the present invention is 53%, and upper and lower two sections of oiliness gaps are relatively big, conclusion be top be oil-water common-layer, bottom is dried layer.After actual perforation and fracturing reform, this interval day oil-producing 5.44t, produce water 15.8m daily3, formation testing result is oil-water-layer, reaches commercial oil, and formation testing result does not corresponds with the conclusion obtained by conventional saturation computation method, and consistent with the conclusion obtained by the present invention.It follows that the present invention oil saturation is asked for more accurate.It should be noted that oil-containing water layer is the water layer containing oil, its oil saturation is relatively low.Oil-water-layer is both to have contained oil, contains again the stratum of water, and compared with oil-containing water layer, the oil saturation of oil-water-layer is higher.Therefore, oil-containing water layer and oil-water-layer are different because of the difference of oil saturation, are 2 different concepts.
The method and system asking for oil saturation based on artificial neural network that the embodiment of the present invention provides, first pass through the log data that rock-electric test obtains in Archie formula, then a and b are entered as 1, and inverse goes out m and n；Again m, n of being obtained by rock-electric test are compared with m ', the n ' obtained by BP neural network model, obtain the BP neural network model that error is minimum；The cementation factor curve dynamically changed by the BP neural network model that error is minimum again and saturation exponent curve, substituted into Archie formula by the cementation factor dynamically changed and saturation exponent and carry out the calculating of oil saturation.On the basis of present invention log data in known Archie formula, by a and b is entered as 1, inverse goes out m and n, therefore overcomes the calculating defect of cementation factor and saturation exponent equalization in conventional method, improves the computational accuracy of oil saturation.Compared with conventional method, the present invention is sensitiveer for the reaction of oil-gas Layer by cementation factor and the saturation exponent calculated oil saturation curve of dynamically change, and its variation tendency can react the oil-bearing information of Reservoir Section more realistically.
It should be noted last that, above detailed description of the invention is only in order to illustrate technical scheme and unrestricted, although the present invention being described in detail with reference to example, it will be understood by those within the art that, technical scheme can be modified or equivalent, without deviating from the spirit and scope of technical solution of the present invention, it all should be contained in the middle of scope of the presently claimed invention.
Claims (10)
1. the method asking for oil saturation based on BP neutral net, it is characterised in that including:
In stratum, choose rock core, described rock core is carried out rock-electric test, obtain the log data in Archie formula；
Lithology factor a in Archie formula and lithology constant b is entered as 1, and inverse obtains cementation factor m and saturation exponent；
Using the log of sign lithology, physical property and oiliness as the input layer of BP neural network model, select described cementation factor m and described saturation exponent as the output layer of BP neural network model；Set up the relation between described output layer and described input layer by simulation, set up BP neural network model；
Log data with the depth of stratum corresponding to m, the n obtained by described rock-electric test is substituted into described BP neural network model and is calculated m,、n,；
By m, n of being obtained by described rock-electric test and the m obtained by described BP neural network model,、n,Compare, obtain the BP neural network model that error is minimum；
The BP neural network model that data on the described log of corresponding input layer substitute into described error minimum is carried out cementation factor curve and the saturation exponent curve that node-by-node algorithm is dynamically changed；
Data on the cementation factor curve of the described dynamic change obtained and saturation exponent curve are substituted into Archie formula and carries out the calculating of oil saturation.
2. the method asking for oil saturation based on BP neutral net as claimed in claim 1, it is characterised in that the node number in the intermediate layer of described BP neural network model is 2N+1, and wherein N is the nodes of described input layer, and N is the integer more than 1.
3. the method as claimed in claim 2 asking for oil saturation based on BP neutral net, it is characterised in that described by m, n of being obtained by rock-electric test and the m obtained by BP neural network model,、n,Comparing, the concrete grammar obtaining the minimum BP neural network model of error includes: regulates the span of the input layer log of described BP neural network model, obtains different m,、n,Value；By m, n of being obtained by described rock-electric test and the m obtained by BP neural network model,、n,Carry out size to compare；According to the relative coefficient of the BP neural network model corresponding with comparative result, obtain the BP neural network model that described error is minimum.
4. the method asking for oil saturation based on BP neutral net as claimed in claim 3, it is characterized in that, the relative coefficient of the BP neural network model that described basis is corresponding with comparative result, obtain the BP neural network model that error is minimum, including: if the relative coefficient of described BP neural network model is equal to 0.85, then corresponding with relative coefficient 0.85 BP neural network model is the BP neural network model that described error is minimum.
5. the method asking for oil saturation based on BP neutral net as described in any one in claim 1-4, it is characterised in that the log of described sign lithology includes gamma ray curve, spontaneous potential curve and CAL；The log of described sign physical property includes interval transit time curve, porosity curve, neutron curve, density curve and permeability curve；The log of described sign oiliness includes resistivity curve and stratum shallow resistivity curve in true formation resistivity curve, stratum.
6. the system asking for oil saturation based on BP neutral net, it is characterised in that including:
Rock-electric test module, for choosing rock core in stratum, carries out rock-electric test to described rock core, obtains the log data in Archie formula；
First computing module, for the lithology factor a in Archie formula and lithology constant b is entered as 1, inverse obtains cementation factor m and saturation exponent；
Oneth BP neural network model generation module, for will characterize lithology, physical property and oiliness log as the input layer of described BP neural network model, select described cementation factor m and described saturation exponent as the output layer of BP neural network model；Set up the relation between described output layer and described input layer by simulation, set up BP neural network model；
Second computing module, is calculated m for the log data with the depth of stratum corresponding to m, the n obtained by described rock-electric test is substituted into described BP neural network model,、n,；
2nd BP neural network model generation module, for by m, n of being obtained by described rock-electric test and the m obtained by described BP neural network model,、n,Compare, obtain the BP neural network model that error is minimum；
3rd computing module, for carrying out, by the BP neural network model that the data on the described log of corresponding input layer substitute into described error minimum, cementation factor curve and the saturation exponent curve that node-by-node algorithm is dynamically changed；
4th computing module, carries out the calculating of oil saturation for the data on the cementation factor curve of the described dynamic change obtained and saturation exponent curve are substituted into Archie formula.
7. the system asking for oil saturation based on BP neutral net as claimed in claim 6, it is characterized in that, in a described BP neural network model generation module, the node number in the intermediate layer of described BP neural network model is 2N+1, wherein N is the nodes of described input layer, and N is the integer more than 1.
8. the system asking for oil saturation based on BP neutral net as claimed in claim 7, it is characterised in that described 2nd BP neural network model generation module, including:
Index acquiring unit, for regulating the span of the input layer log of described BP neural network model, obtains different m,、n,Value；
Comparing unit, for by m, n of being obtained by described rock-electric test and the m obtained by BP neural network model,、n,Carry out size to compare；
BP neural network model signal generating unit, for the relative coefficient according to the BP neural network model corresponding with comparative result, obtains the BP neural network model that described error is minimum.
9. the system asking for oil saturation based on BP neutral net as claimed in claim 8, it is characterized in that, described BP neural network model signal generating unit, if the relative coefficient specifically for described BP neural network model is equal to 0.85, then corresponding with relative coefficient 0.85 BP neural network model is the BP neural network model that described error is minimum.
10. the system asking for oil saturation based on BP neutral net as described in any one in claim 6-9, it is characterised in that the log of described sign lithology includes gamma ray curve, spontaneous potential curve and CAL；The log of described sign physical property includes interval transit time curve, porosity curve, neutron curve, density curve and permeability curve；The log of described sign oiliness includes resistivity curve and stratum shallow resistivity curve in true formation resistivity curve, stratum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310244257.7A CN103485770B (en) | 2013-06-19 | 2013-06-19 | A kind of method and system asking for oil saturation based on artificial neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310244257.7A CN103485770B (en) | 2013-06-19 | 2013-06-19 | A kind of method and system asking for oil saturation based on artificial neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103485770A CN103485770A (en) | 2014-01-01 |
| CN103485770B true CN103485770B (en) | 2016-08-10 |
Family
ID=49826315
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310244257.7A Active CN103485770B (en) | 2013-06-19 | 2013-06-19 | A kind of method and system asking for oil saturation based on artificial neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103485770B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104635265B (en) * | 2014-12-31 | 2017-07-07 | 中国石油天然气集团公司 | A kind of method of controllable neural network predicting reservoir permeability |
| CN105488248B (en) * | 2015-11-18 | 2017-02-15 | 山东科技大学 | Judgment method for deep ore stratum and rock stratum |
| CN107742031B (en) * | 2017-10-24 | 2021-05-28 | 重庆科技学院 | Displacement experiment artificial rock core analysis preparation method based on experiment and mathematical algorithm |
| CN109162696B (en) * | 2018-10-08 | 2022-06-03 | 陕西延长石油(集团)有限责任公司研究院 | Method for calculating stratum oil saturation by using compensation sound wave |
| US11906695B2 (en) | 2020-03-12 | 2024-02-20 | Saudi Arabian Oil Company | Method and system for generating sponge core data from dielectric logs using machine learning |
| CN111520134B (en) * | 2020-04-03 | 2023-06-06 | 中国科学院大学 | Method and device for acquiring original oil saturation, electronic equipment and storage medium |
| CN111665560B (en) * | 2020-04-23 | 2023-06-30 | 中国石油天然气股份有限公司 | Hydrocarbon reservoir identification method, apparatus, computer device and readable storage medium |
| CN113064347B (en) * | 2021-03-15 | 2022-09-23 | 贵州大学 | PMSM chaotic system self-adaptive control method considering asymmetric input and output constraints |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5251286A (en) * | 1992-03-16 | 1993-10-05 | Texaco, Inc. | Method for estimating formation permeability from wireline logs using neural networks |
| US5444619A (en) * | 1993-09-27 | 1995-08-22 | Schlumberger Technology Corporation | System and method of predicting reservoir properties |
| CN101358943A (en) * | 2007-07-31 | 2009-02-04 | 中国石油天然气集团公司 | Method for correcting electrical experiment parameter m and n of undersaturated core |
| CN101487390A (en) * | 2009-02-23 | 2009-07-22 | 大庆油田有限责任公司 | Archie mode method for confirming initial oil saturation of oil layer |
| CN101762829A (en) * | 2010-01-18 | 2010-06-30 | 赵庆辉 | Analog measurement method and device of oil saturation in strata |
| CN102434152A (en) * | 2011-12-05 | 2012-05-02 | 中国石油天然气股份有限公司 | Method for calculating oil saturation of reservoir |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090182693A1 (en) * | 2008-01-14 | 2009-07-16 | Halliburton Energy Services, Inc. | Determining stimulation design parameters using artificial neural networks optimized with a genetic algorithm |
-
2013
- 2013-06-19 CN CN201310244257.7A patent/CN103485770B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5251286A (en) * | 1992-03-16 | 1993-10-05 | Texaco, Inc. | Method for estimating formation permeability from wireline logs using neural networks |
| US5444619A (en) * | 1993-09-27 | 1995-08-22 | Schlumberger Technology Corporation | System and method of predicting reservoir properties |
| CN101358943A (en) * | 2007-07-31 | 2009-02-04 | 中国石油天然气集团公司 | Method for correcting electrical experiment parameter m and n of undersaturated core |
| CN101487390A (en) * | 2009-02-23 | 2009-07-22 | 大庆油田有限责任公司 | Archie mode method for confirming initial oil saturation of oil layer |
| CN101762829A (en) * | 2010-01-18 | 2010-06-30 | 赵庆辉 | Analog measurement method and device of oil saturation in strata |
| CN102434152A (en) * | 2011-12-05 | 2012-05-02 | 中国石油天然气股份有限公司 | Method for calculating oil saturation of reservoir |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103485770A (en) | 2014-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103485770B (en) | A kind of method and system asking for oil saturation based on artificial neural network | |
| Tadeusiewicz | Neural networks in mining sciences–general overview and some representative examples | |
| CN102789003B (en) | Method for inverting stratum parameters by array induction logging data and device thereof | |
| US8364447B2 (en) | Method, program and computer system for conciliating hydrocarbon reservoir model data | |
| Zhao et al. | Flow-path tracking strategy in a data-driven interwell numerical simulation model for waterflooding history matching and performance prediction with infill wells | |
| CN111665560B (en) | Hydrocarbon reservoir identification method, apparatus, computer device and readable storage medium | |
| CN106294282B (en) | Black oil reservoir simulation method and device | |
| CN103760082A (en) | Method and device for confirming effective motion space of shale strata tight reservoir crude oil | |
| CN106503295B (en) | A kind of method and device for explaining oil field Water Flooding Layer using state-space model | |
| Tan et al. | Quantitative evaluation methods for water-flooded layers of conglomerate reservoir based on well logging data | |
| Zhu et al. | Inversion of the permeability of a tight gas reservoir with the combination of a deep Boltzmann kernel extreme learning machine and nuclear magnetic resonance logging transverse relaxation time spectrum data | |
| CN103867195A (en) | Quantitative evaluation method of water production rate of low permeability reservoir | |
| EP3281043A1 (en) | Oilfield reservior saturation and permeability modeling | |
| Correia et al. | Multiscale integration for karst-reservoir flow-simulation models | |
| Khazaeni et al. | Intelligent production modeling using full field pattern recognition | |
| Zhang et al. | Uncertainty Analysis and Assisted History Matching Workflow in Shale Oil Reservoirs | |
| CN112861890A (en) | Reservoir evaluation model construction method and reservoir identification method | |
| Akhmetov et al. | Grouping of objects using a limited number of parameters characterizing geological and physical properties of layers | |
| CN109655394B (en) | Nuclear magnetic resonance T2 spectrum permeability calculation method under constraint of throat structure parameters | |
| Aliouane et al. | Permeability prediction using artificial neural networks. A comparative study between back propagation and Levenberg–Marquardt learning algorithms | |
| CN113530533B (en) | Dessert region prediction method and system based on three quality parameters | |
| Batir et al. | Reservoir analysis for deep direct-use feasibility study in East Texas | |
| CN107015289A (en) | Trap evaluation stock number determines method and device | |
| Litvak et al. | Uncertainty Estimation in Production Predictions Constrained by Production History and Time-Lapse Seismic in a GOM Oil Field | |
| Li et al. | Application of Petrel software in reserve estimation in Xinli-Xinbei oilfield, Jilin Province, China |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |