Method based on digital core for predicting reservoir permeability

Abstract

Disclosed is a method based on a digital core for predicting reservoir‘permeability.The method based.on adigital core for predicting reservoir permeability includes: acquiring a CT scanning image of a<:ore; building a three—dimensional digital core model, and.extracting a pore space model; extracting pore structure parameters, and dividing pore space types; calculating hole structure parameters and crack structure parameters;multiplyingtheholestructureparametersbyeahole rate to obtain weighted. hole structure parameters, and multiplying the crack structure parameters by a crack rate to obtain. weighted. crack structure parameters; building an optimizeclmultiple regressionlnodel with fitting relationship among the weighted hole structure parameters, the weighted crack structure parameters and permeability, and obtaining a permeability prediction model; and inputting sample data to be detected into the permeability prediction model, and obtaining predicted permeability.

Description

TECHNICAL FIELD [Ol] The present invention relates to the theoretical and technical field of influence of porous medium characteristics on_formation_fluitipercolation,ijiparticulartoeamethodioased on a digital core for predicting reservoir permeability. BACKGROUND ART

[02] The physical properties and fluid percolation characteristics of reservoirs with complex structures are controlled by the microscopic pore structure of rock and the crackhole configuration relationship. Reservoir permeability is a key parameter for reservoir evaluation, petroleum and gas production prediction, etc.

[03] Digital core represents the real rock skeleton.and.pores i11thecorevjjidigitalvoxelthat(xn1berecognizedknlcomputer, and further constructseapore network1nodel.that can accurately reflect the pore space distribution characteristics of core and also reflect the percolation characteristics of fluid in the core. Threedimensional digital core model can. more intuitively show the influence of pore structure on reservoir percolation,clarifythechanginglawtetweenpermeabilityand pore structure parameters, and deeply explore the influence of different reservoir types on reservoir permeability.

[04] At present, the dualmedium porosity and permeability model, which is widely used in complex structure reservoirs and unconventional reservoirs, faces the problems of low accuracy and poor generalization in the actual reservoir permeability evaluation process.

[05] Therefore, it is an urgent problem for those skilled in the art to propose a method based on a digital core for predicting reservoir permeability to solve the difficulties existing in the prior art. SUMMARY

[06] In view of this, the present invention provides a method basedcn1a.digital core xrpredicting'reservoirpermeability, which may explores influence factors of a pore structure on reservoir permeability fronlalnicroscopic level, such that the reservoirpermeabilityijsmore accuratelyrpredictedq andeanew idea is provided for effective evaluation of a petroleum reservoir.

[07] In order to achieve the above objective, the present invention uses the following technical solution:

[08] The method. based. on. a digital core for predicting reservoir permeability includes:

[09] S1, acquiring a CT scanning image of a core;

[10] S2, building a threedimensional digital core model according to the CT scanning image of the core, and.extracting a pore space model;

[11] S3, extracting pore structure parameters with the pore space model, and dividing pore space types;

[12] S4, calculating hole structure parameters and crack structure parameters separately according to the pore space types;

[13] S5, multiplying the hole structure parameters by a hole rate to obtain weighted. hole structure parameters, and multiplying the crack structure parameters by a crack rate to obtain weighted crack structure parameters;

[14] S6, building an.optimizec1multiple regressionlnodel with fitting relationship among the weighted hole structure parameters, the weighted. crack structure parameters and permeability, and obtaining a permeability prediction model; and

[15] S7, inputting' sample data. to be detected. into the permeability prediction model, and obtaining predicted permeability.

[16] In the above method, alternatively, in S1, by performing a.CT scanning experimentcnia.core sample, the CT:scanning image of the core is acquired.

[17] In.the abovelnethod, alternativelyy in S2, avizo software is used as a platform, a segmentation threshold of the image is determined based on porosity measured by the core, and the image is subjected to binarization treatment with threshold segmentation.

[18] A. formula of the image subjected to threshold segmentation is:

[19] F(n m)={1'H(n rn) (1), 0,H(n m)2t

[20] in the formula, t is the segmentation threshold of the image, (n1n) is orientations of pixels, H(n,1n) is gray values of the pixels, and F(n,1n) is gray values of pixels of the image after segmentation.

[21] The threedimensional digital core model composed of a pore space and a skeleton space is constructed, and the pore space model is obtained by extraction.

[22] In.the abovelnethod, alternatively, in S3, the pore space is divided into pores and throats by a maximum ball method, and.when the pore throat relationship is characterized, basic pore throat geometric parameters are obtained, including pore volume and a pore surface area.

[23] Sphericity parameters are calculated, and a calculation formula of sphericity is:

[24] S==ÎËËËÎ (2),

[25] i11the formula, S representsijuasphericity, ineadecimal fraction; V represents the pore volume, in unit of um?; and Arepresents the pore surface area, in unit of um2.

[26] According to industry standards, the pores are divided into two types of holes and cracks.

[27] 1n_the above method, alternatively, pore space geometric parametersareobtainedbasecnitheporespacemodel,including all hole pore sizes, local crack apertures, crack areas, apparent volumes of the core model, connected pore volumes, connected hole volumes and connected crack volumes.

[28] .Average pore size is calculated, and expression thereof is:

[29] E:? (3),

[30] in the formula, Ë represents the average pore size, in unit of um; N represents the total number of holes, which is dimensionless; and Ri.represents pore size of an ith hole, in unit of um.

[31] Average crack aperture is calculated, and expression thereof is: -sxw-

[32] V:213%!) (4),

[33] in.the formula, VV represents the average crack:aperture, in unit of um; Airepresents the crack area of an ith part, in unit of um?; and MG represents the local crack aperture of the ith part, in unit of um.

[34] Connectedporosityijscalculated,andexpressionthereof is:

[35] «FZ° (5), a

[36] in.the formula, @ represents the connected.porosity, in percentage; %,represents the connected pore volume, in unit of um3;enui1§ represents the apparent volume of the corelnodel, in unit of um3.

[37] The hole rate is calculated, and expression thereof is:

[38] g:}; (6),

[39] iJltheformula, @ representsiuaholerate, inpercentage; [@ represents the connected pore volume, in unit of um?; and IQ represents the connected hole volume, in unit of um3.

[40] The crack rate is calculated, and expression thereof is:

[41] 1ÿ==ëâ (7),

[42] in the formula, qv represents the crack rate, in percentage; %,represents the connected pore volume, in unit of um?; and W represents the connected crack volume, in unit of um3.

[43] In the above method, alternatively, in S5, the weighted hole structure parameters, the weighted. crack structure parameters are calculated.

[44] Weighted.average pore size is calculated, and expression thereof is:

[45] Ë=Ë><rb (8),

[46] in the formula, Ê represents the weighted average pore size, in unit of um; Ë represents average pore size, in unit of um; and @ represents the hole rate, in percentage.

[47] Weighted.average aperture is calculated, and expression thereof is:

[48] [7 / 21 / 7er (9),

[49] 111 the formula, ÜÜ represents the weighted. average aperture, in unit of um; W7 represents the average crack aperture, in unit of um; and quepresents the crack rate, in percentage.

[50] In the above method, alternatively, in S6, the weighted average pore sizepermeability model and the weighted average aperturepermeability model is used, and the optimized binary regression model is built.

[51] The optimal fitting relationship between the weighted average pore size and logarithmic permeability is a secondorder polynomial, the optimal fitting relationship between the weighted average aperture and logarithmic permeability is logarithmic, and relational expression is obtained, that is, an optimized binary regression model:

[52] 1nK,=a*§2+b*1nM= / +c (10),

[53] in the formula, Kiis measured permeability, in unit of mD; Ê represents the weighted average pore size, in unit of um; ÜÜ represents theVNeighted.average aperture, in unit ofimu and a, b and c are constants.

[54] According to the technical solution, compared with the prior art, the present invention provides the , which has the following' beneficial effects: the present invention performs binary regression analysis on the weighted average pore size, theVNeighted.average aperture and.the permeability, and the influences of the proportion of reservoir space types and pore structure characteristics on the permeability are combinedndcroscopically.Aimingat:theproblanthataacommonly used dualmedium porosity and permeability model may not be well applied. to reservoirs with complex structures and unconventional reservoirs, the weighted average pore size and the weighted average aperture are introduced to build a permeability model more suitable for the reservoirs with complex structures and the unconventional reservoirs, such that the accuracy of reservoir permeability prediction is improved.Thepermeabilitymodelbuiltbythepresentinvention has higher accuracy than the dualmedium. porosity and permeabilitylnodel, and1nay]oe11sed.asaifeasible and.effective tool for petroleum reservoir exploitation. BRIEF DESCRIPTION OF THE DRAWINGS

[55] In order to describe the examples of the present disclosurecnrthe technical solutions intjuapriorart clearer, and the accompanying drawings required by the examples or description of the prior art are briefly described below. Obviously, the accompanying drawings in the following description show merely examples of the present disclosure, and a person of ordinary skill in the art would also be able to derive other accompanying drawings from. the provided accompanying drawings without creative efforts.

[56] FIG. 1 is a flow chart of a provided by the present invention;

[57] FIG. 2 is a CT scanning image of a core provided by the present invention;

[58] FIG. 3 is a diagram of a pore space model extracted by the core provided by the present invention;

[59] FIG. 4.is51diagranlshowing'relationship>betweeereighted average pore size and logarithmic permeabilityïprovideclby the present invention;

[60] FIG. 5.is51diagranlshowing'relationship>betweeereighted average aperture and.the logarithmic permeability provided by the present invention;

[61] FIG. 6 is a diagram showing relationship between hole porosity and the logarithmic permeability provided by the present invention;

[62] FIG.7:Msadiagramshowingtherelationshipbetweencrack porosity and the logarithmic permeability provided by the present invention;

[63] FIG. 8 is a diagram. showing relationship between permeability and core permeability fitted by a weighted pore structure parameter model provided by the present invention; and

[64] FIG. 9 is a diagram. showing relationship between permeability and core permeability fitted by a dualmedium porosity and permeability model provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[65] The technical solution in the examples of the present invention is clearly and completely described below with reference to the accompanying drawings in the examples of the present invention, Apparently, the<iescribed.examples are some rather than.all of the examples of the present invention, Based ontheexamplesofthepmesentinvention,alltheotherexamples obtainajbythoseofcmdinaryskilliJltheartwithoutinventive effort are within the scope of protection of the present invention.

[66] In the present application, relational terms such as first and second are merely used to distinguish one entity or one operation from another entity or another operation, and do not necessarily require or imply any such actual relationship or order among these entities or operations. The terms "comprising", "including" or any other variant thereof are intended to cover nonexclusive inclusion, such that a process, a method, an article or a device including a series of elements not only include those elements, but include other elements not listed clearly, or further include elements inherent to such process, method, article or apparatus. In the caseofIMDmorelimitations,theelementlimitajbythesentence "including a..." does not exclude that there exists another same element in the process, method, article or device including the element.

[67] .As shown in FIG. 1, a method.based on a digital core for predicting reservoir permeability is disclosed by the present invention and includes:

[68] S1, a CT scanning image of a core is acquired;

[69] S2, a threedimensional digital core model is built according to the CT scanning image of the core, andEìpore space model is extracted;

[70] S3, pore structure parameters are extracteclwith.the pore space model, and pore space types are divided;

[71] S4, hole structure parameters and crack structure parameters are calculated separately according to the pore space types;

[72] S5,theholestructureparametersarenmltipliedbyaahole rate to obtain weighted hole structure parameters, and the crack structure parameters are multiplied by a crack rate to obtain weighted crack structure parameters;

[73] S6, an optimized.multiple regression.model is built with fitting relationship among the weighted hole structure parameters, the weighted. crack structure parameters and permeability,andzapermeabilitypredictiondeel isobtained; and

[74] S7, sample data to be detected are input into the permeabilityïprediction model, and.predicteclpermeability are obtained.

[75] Further, in S1, by performing a CT scanning experiment oneicore sample, the CT scanning image of the core is acquired.

[76] Further, in S2, avizo software is used as a platform, a segmentation threshold of the image is determined based on porosity measured by the core, and the image is subjected to binarization treatment with threshold segmentation.

[77] A formula of the image subjected to threshold segmentation is:

[78] F(n, m)={1'H(n m)<t (1), 0,H(n, m)2t

[79] in the formula, t is the segmentation threshold of the image, (n1n) is orientations of pixels, H(n,1n) is gray values of the pixels, and F(n,1n) is gray values of pixels of the image after segmentation.

[80] The threedimensional digital core model composed of a pore space and a skeleton space is constructed, and the pore space model is obtained by extraction.

[81] Further, in S3, the pore space is divided into pores and throats by a maximum ball method, and when the pore throat relationship is characterized, basic pore throat geometric parameters are obtained, including pore volume and a pore surface area.

[82] Sphericity parameters are calculated, and.a calculation formula of sphericity is:

[83] S==ÈËËËÎ (2),

[84] in.the formula, S represents the sphericity, inEìdecimal fraction; V represents the pore volume, in unit of um?; and A represents the pore surface area, in unit of um2.

[85] According to industry standards, the pores are divided into two types of holes and cracks.

[86] Further, pore space geometric parameters are obtained based on the pore space model, including all hole pore sizes, local crack apertures, crack areas, apparent volumes of the core model, connectecìpore volumes, connected hole volumes and connected crack volumes.

[87] Specifically, average pore size is calculated, and expression thereof is:

[88] E:? (3),

[89] in the formula, R represents the average pore size, in unit of um; N represents the total number of holes, which is dimensionless; and Ri.represents pore size of an ith hole, in unit of um.

[90] Average crack aperture is calculated, and expression thereof is: -sxw-

[91] V:213.81) (4),

[92] in.the formula, VV represents the average crack:aperture, in unit of um; Airepresents the crack area of an ith part, in unit of um?; and MG represents the local crack aperture of the i-th part, in unit of um.

[93] Connectedporosityijscalculated,andexpressionthereof is:

[94] (p=? (5), a

[95] in.the formula, @ represents the connected.porosity, in percentage; %,represents the connected pore volume, in unit of um3;enui1§ represents the apparent volume of the core]nodel, in unit of um3.

[96] The hole rate is calculated, and expression thereof is:

[97] rb=:: (6),

[98] iJltheformula, @ representsijuaholerate, inpercentage; [@ represents the connected pore volume, in unit of um?; and 1% represents the connected hole volume, in unit of um}; and

[99] The crack rate is calculated, and.expression.thereof is:

[100] 1ÿ==%£ (7),

[101] in. the formula, qv represents the crack rate, in percentage; %,represents the connected pore volume, in unit of um?; and W represents the connected crack volume, in unit of um3.

[102] Furthermore, in S5, the weighted hole structure parameters, the weighted. crack structure parameters are calculated.

[103] ]Neighted.average pore size is calculated, and.expression thereof is:

[104] §=R><rb (8),

[105] in the formula, Ë represents the weighted average pore size, in unit of um; Ë represents average pore size, in unit of um; and @ represents the hole rate, in percentage.

[106] Weighted.average aperture is calculated, and expression thereof is:

[107] I / T / =W> <rf (9),

[108] i11 the formula, ÜÜ represents the weighted. average aperture, in unit of um; W7 represents the average crack aperture, in unit of um; and qrepresents the crack rate, in percentage.

[109] Further, in S6, the weighted average pore sizepermeability model and the weighted average aperturepermeability model is used, and the optimized binary regression model is built.

[110] The optimal fitting relationship between the weighted average pore size and logarithmic permeability is polynomial (second order), the optimal fitting relationship between the weighted average aperture and logarithmic permeability is logarithmic, and relational expression is obtained, that is, an optimized binary regression model:

[111] 1nK,-=a*§2+b*1nM= / +c (10),

[112] in the formula, Kiis measured permeability, in unit of mD; Ê represents the weighted average pore size, in unit of um; ÜÜ represents the]Neighted.average aperture, in.unit ofimu and a, b and c are constants.

[113] In.a specific example, core CT scanning image data of 30 core samples in the same horizon in a certain area of Sichuan Basin are taken as an example:

[114] Z&3 shown ij] FIG. 2, by performing the (IF scanning experiment on the core sample, a series of core CT scanning image data are obtained. The avizo software is used as the platform the segmentation.thresholdcïfthe imageiisdetermined based on the porosity measured by the core, and the image is subjected to binarization treatment with threshold segmentation.

[115] As shown in FIG. 3, the threedimensional digital core model composed of the pore space and the skeleton space is constructed, and the pore space model is obtained by extraction.

[116] The pore space is divided into the pores and.the throats by the maximum.ball method, when the pore throat relationship is characterized, the basic pore throat geometric parameters are(obtained, including the pore volume, theepore surface area, etc., and the sphericity parameters are calculated.

[117] According to the industry standards, the pores are divided into two types of the holes and the cracks.

[118] Table 1 Pore space type division standard table Geometrical characteristics types Hole and sphericity greater than or equal to 0.43 _ Crack sphericity less than 0.43 _

[119] Pore space geometric parameters are obtained based on the pore space model, including all the hole pore sizes, the local crack apertures, the crack areas, the apparent volumes of the core model, the connected pore volumes, the connected hole volumes, the connected crack volumes, etc. the weighted hole structure parameters, the weighted crack structure parameters are calculated. In addition, in order to compare with a dualmedium porosity and permeability model, hole porosity is calculated, and expression thereof is:

[120] (pb =% (ll)

[121] in the formula, (pb represents the hole porosity, in percentage; Vb represents the connected hole volume, in unit of um3; and V;, represents the apparent volume of the core model, in unit of um3.

[122] Crack porosity is calculated, and expression thereof is: V

[123] «pf = Vf (12)

[124] in the formula, (pf represents the crack porosity, in percentage; V} represents the connected pore volume, in unit of um3; and V;, represents the apparent volume of the core model, in unit of um3.

[125] As shown in Table 2, the weighted average pore size, the weighted average aperture, the hole porosity, the crack porosity and corresponding permeability are calculated for 20 groups of digital core models.

[126] Table 2 Digital core pore structure parameter table Weighted . Hole Crack average Weighted Core Permeability _ _ average Number K(mD) porOSity porSSity pore aperture VT / (um) 2221:) (Pf(o) (PHG) ___-__ ___-__ ___-__ ___-__ ___-__

[127] With.reference to FIG. 4, FIG. 5, FIG. 6 and.FIG. 7, based on.the above 20 groups of data, the weighted.average pore size, the weighted.average aperture, the hole porosity and.the crack porosity corresponding to the core are intersected with the permeability separately, and.the fitting relationship between the parameters and the permeability is obtained.

[128] It can be seen from FIG. 4 and FIG. 5 that the optimal fitting relationship between the weighted average pore size and the logarithmic permeability is the secondorder polynomial, and the optimal fitting relationship between the weighted average aperture and the logarithmic permeability is logarithmic. The optimizeclbinary:regressionanodel.is studied, and relational expression of a weighted. pore structure parameter model is obtained by means of regression analysis:

[129] ln K,- = 70.826 * ? +0.709 * ln [Î / 0.362

[130] in the formula, Kiis the measured.permeability, in unit ofnU and Ërepresentsijuaweighted.average pore size,iJ]unit of um; ÜÜ represents the weighted average aperture, in unit of um.

[131] It can be seen from FIG. 6 and FIG. 7 that the optimal fitting relationship between the hole porosity and the logarithmic permeability is linear, and the optimal fitting relationship between the crack porosity and the logarithmic permeability is logarithmic. The optimized.binary regression model is studied, and relational expression of a dualmedium porosity and permeability model is obtained by means of the regression analysis:

[132] an]=-10500¢b03061n¢743008

[133] in.the formula, Kiis the measured.permeability, in unit of mD; and wb represents the hole porosity, in percentage; wf represents the crack porosity, in unit of um.

[134] The weighted average pore size, the weighted average aperture, the hole porosity and.the crack:porosityfof the other 10 digital cores are calculated in the above way, which are used as a prediction set, and the permeability is calculated separately according to the built weighted pore structure parameter model and dualmedium porosity and permeability model, as shown in Table 3.

[135] Table 3 Permeability prediction results Weighted Weighted Predicted Predicted poros1ty poros1ty ber K(mD) O 0 size aperture new models old models (pb (pf Emm) VT / (um) KN(mD) KL(mD) ___-__ ______ ______

[136] Data used for modeling are used as a back determination set, and a total of 30 groups of models are obtained to calculate the permeability.

[137] As shown in FIG. 8, a diagram showing the relationship between permeability fitted by the weighted pore structure parameter model and core permeability is obtained by intersecting the permeability calculated by the weighted pore structure parameter model with the core permeability.

[138] As shown in FIG. 9, a diagram showing the relationship between permeability fitted by the dualmedium porosity and permeability model and core permeability is obtained by intersecting the permeability calculated by the dualmedium porosity and permeability model with the core permeability.

[139] FIG.E3andFIG.SBcancompareadvantagesanddisadvantages of the two permeability models, and consistency between the fitting permeability in FIG. 8 and the core permeability is higher than that in FIG. 9.

[140] In addition, the advantages and.the disadvantages of the two permeability models can be compared from the perspective of relative errors, and. the relative errors of the two permeability models are shown in Table 4.

[141] Table 4 Comparison of relative errors of permeability prediction Model error Relative error of Relative error of weighted pore dualmedium structure porosity and Sample type parameter model permeability model set

[142] It can be seen from error analysis in Table 4 that the relative error of the weighted.pore structure parameter model is reduced by 225% compared with a total sample set of the dualmedium.porosity and permeability model, which can prove that fitting permeability accuracy of the weighted pore structure parameter model is higher than.that of a traditional dualmedium porosity and permeability model.

[143] The examples in the specification are described in a progressive manner, the examples may refer to one another for the same and similar portions, and each example focuses on differenceswithanotherexample.Especially,asforthesystem or the system.example, since this example is basically similar to a method example, description is relatively simple, and it is possible to refer to part of description of the method embodiment for relevant contents. The system and the system example described. above are merely schematic. The unit illustratedaseaseparatecomponentnykxaphysicallyseparated or not. The component shown as a unit may be a physical unit ornot.Thatis,theymaybelocatediíloneplace,ordistributed over a plurality of network units. Part or all of the modules may be selected according to actual needs, to implement the solution to the embodiment. A.person of ordinary skill in the artcanunderstandandimplementthemmthoutcreativeefforts.

[144] The above description of the disclosed examples enables those skilled.in the art to make or use the present invention. Many modifications to these examples would have been obvious to those skilled.in the art, and.the general principles defined herein can.be implemented.in other examples without departing from.the spirit or scope of the present invention. Therefore, the present invention is not to be limited to the examples described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.< / rf>

Claims

1. A method based on a digital core for the predicting reservoir permeability, including: 81, acquiring a CT scan image of a nucleus; 52, Building a Three-Dimensional Digital Core model based on the CT scan image of the nucleus, and the extracting a pore space model; 53, Extracting pore structure parameters with the pore space model, and the division of pore space types; 54, the separate calculation of hole structure parameters ters and crack structure parameters based on the pore room types; S5, multiplying the hole structure parameters with a hole degree to calculate weighted hole structure parameters get, and multiplying the crack structure para meters with a crack degree to determine weighted crack structure parameters available to obtain; 56, building an optimized multi-purpose regression model with appropriate relationship between the weighted hole structure structure parameters, the weighted crack structure parameters and per meability, and obtaining a permeability for spelling model; and S7, the input into the permeability prediction model of sample data to be detected, and obtaining predicted permeability.

2. Method based on a digital core for the predicting reservoir permeability according to claim 1, in which in 81, by performing a CT scan experiment on a core sample, the CT scan image of the core is to recruit.

3. Method based on a digital core for the predicting reservoir permeability according to claim 1, where in 52, avizo software is used as a flat form, a segmentation threshold of the image is determined based on porosity measured through the core, and the image is subjected to binarization treatment with drem peel segmentation; where a formula of the image subjected to threshold segmentation is as follows: Lfíru rn t F(n, m)={0,HÊn "à; (1), where, in the formula, t is the segmentation threshold of the image is, UL WO orientations of pixels are, H(n,1n) gray values ​​of the pixels are, and F(n, "Ü gray values ​​of pixels of the image after segmentation; and where the three-dimensional digital core model is composed of a pore space and a skeletal space constructed, and the pore space model is obtained by extraction.

4. Method based on a digital core for the predicting reservoir permeability according to claim 1, where in 53, the pore space is divided into pores and necks by a maximum ball method, and where, when the pore neck relationship is characterized, basis pore neck geometric parameters are obtained, including pore volume and a pore surface area; where sphericity parameters are calculated, and a calculation formula for sphericity is: S=Ê (2), where, lll is the formula, 5 represents the kbl-shape tares; D represents the pore volume, in um?; and 14 the pore surface area represents, in um?; and where, based on industry standards, the pores are divided into two types of holes and cracks.

5. Method based on a digital core for the predicting reservoir permeability according to claim 1, where in 54, pore space geometric parameters are obtained based on the pore space model, including all pore dimensions, local crack apertures, crack areas, apparent volumes of the core model, connected pore volumes, connected hole volumes and connected crack volumes; where average pore size is calculated, and ex pressure thereof is: Me? (3), where, in the formula, R is the average pore size represents, in um; N represents the total number of holes sentert, which is dimensionless; and Ripore dimension of an i the hole represents, in um; where the average crack aperture is calculated, and ex pressure thereof is: -sxw- M / =:Ä%ZrQ (4), where, in the formula, M / is the average crack aperture represents, in um; Ai the crack area of ​​an ith part represents, in um?; and M4 the local crack aperture of the ide part represents, in um; where connected porosity is calculated, and expressly one of which is: <p=:: (5), where, in the formula, @ is the associated porosity re represents, in percentage; Dg the connected pore volume re presents, in umê; and thet apparent volume of the core model represents, in um3; where hole degree is calculated, and expression thereof is: ms;í (6), where, in the formula, 1% represents the hole degree, in percentage; 1% represents the 'connected pore volume', in um?; and Vb represents the connected hole volume, in um?; and where the crack degree is calculated, and expression of which is: 3 zëí (7), @ where, in the formula, que represents the degree of cracking, in percentage; %,represents the connected pore volume, in um?; en.D} represents the connected crack volume, in um3.

6. Method based on a digital core for the predicting reservoir permeability according to claim 5, where in 55, the weighted hole structure parameters, the weighted crack structure parameters are calculated, where weighted average pore size is calculated, and an expression thereof is: Ê=Ë> <rb (8),where, in the formula, Ê is the weighted average pore size represents, in (your Ë average pore size represents, in um; and rb represents the hole degree, in percentage; and where weighted average aperture is calculated, and expression of which is: VI: / :Wer (9), where, in the formula, W is the weighted average aperture represents, in um; WV the average crack aperture repre senteert, in um; and & represents the degree of tear, in per percentage.

7. Method based on a digital core for the predicting reservoir permeability according to claim 6, where in 56, the weighted average pore size permea model and the weighted average aperture permeabi ity model are used, and where the optimized binary regression model is being built; and where the optimal fitting relationship between the weighted average pore size and logarithmic permeability a second order polynomial.is, where the optimal fitting relation between the weighted average aperture and logarithmic per meability is logarithmic, and where relational expression is obtained, being an optimized binary regression model: 1nK,=a*Ë2+b*anT / +c (10), where, in the formula, Kigemet permeability is, in mD; Ê represents the weighted average pore size, in um; W represents the weighted average aperture, in um; and a, b and c are constants. 000 S1 Acquire a CT scanning image of a core S2 Build a three-dimensional digital core model according to the CT scanning image of the core, and extract a pore space model S3 Extract pore structure parameters with the pore space model, and divide pore space types S4 Calculate hole structure parameters and crack structure parameters separately according to the pore space types Multiply the hole structure parameters by a hole rate to obtain S5 weighted hole structure parameters, and multiply the crack structure parameters by a crack rate to obtain weighted crack structure parameters Build an optimized multiple regression model with fitting relationship S6 among the weighted hole structure parameters, the weighted crack structure parameters and permeability, and obtain a permeability prediction model S7 Input sample data to be detected into the permeability prediction model, and obtain predicted permeability FIG. 1 FIG.2 FIG.3< / rb>