CN120972281A - A Method for Target Selection and Resource Prediction in Karst Areas Based on 3D Geological Modeling - Google Patents

Abstract

This invention relates to the field of petroleum exploration, specifically disclosing a method for target optimization and resource prediction in karst areas based on three-dimensional geological modeling. The method first collects seismic data, well logging data, and core analysis results from karst areas, performs data normalization and time-depth conversion, and constructs a three-dimensional geological model integrating structural and physical constraints. Then, it selects multiple attribute indicators such as coherence volume, curvature, amplitude attenuation, deep lateral resistivity, and sonic transit time, and calculates the karst development index through principal component analysis. Combining parameters such as reservoir thickness, porosity, permeability, trap integrity, and hydrocarbon shows, it uses a multi-index weighted scoring method to screen favorable exploration blocks. Finally, it calculates reserves based on the volumetric method, obtaining resource prediction results for different target areas. This method achieves three-dimensional fusion modeling of multi-source data, quantitative optimization of multiple indicators, and chain-like calculation, providing quantifiable and traceable technical support for petroleum exploration deployment in karst areas.

Description

Karst region petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling

Technical Field

The invention relates to the technical field of petroleum exploration, in particular to a karst region petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling.

Background

Karst reservoirs are one of the important types of carbonate rock hydrocarbon reservoirs, particularly at the edges of sea-phase carbonate terrains or ancient landforms, where reservoir space is often controlled by erosion and additive modification of structural fractures. Compared with sandstone reservoirs, karst reservoirs have the characteristics of various pore types, strong heterogeneity, complex distribution rule and the like, and reservoir parameters change severely in space, so that the difficulty of optimizing exploration targets and predicting resource quantity is high. Particularly in the region with strong karst development, the variation range of physical parameters such as reservoir thickness, porosity, permeability and the like in the plane and vertical direction is obvious, so that single-well data are difficult to accurately reflect the integral characteristics of the reservoir, and the three-dimensional heterogeneous distribution of the reservoir is difficult to be revealed by traditional two-dimensional seismic interpretation.

In the prior art, the preferred method for the karst region exploration target mainly relies on two-dimensional seismic section interpretation, well drilling, well logging single well analysis and other means, and the spatial position of a karst reservoir is estimated by identifying seismic responses such as abnormal amplitude, waveform distortion and the like. However, the method has three limitations that firstly, the two-dimensional interpretation lacks continuous space constraint, the morphological boundary of the karst reservoir is difficult to accurately describe, secondly, the quantitative relation between attribute inversion and physical parameters is not accurate enough, a unified multi-source data fusion modeling framework is lacking, thirdly, in the resource quantity prediction, a single factor volumetric method or an analogy method is mostly adopted, and multiple indexes such as the spatial distribution characteristics of the reservoir, the physical parameters, the karst development degree and the like cannot be comprehensively coupled, so that the uncertainty of a prediction result is larger.

In recent years, the development of three-dimensional geologic modeling and multi-attribute seismic inversion techniques provides a new idea for the research of complex carbonate reservoirs. By fusing the seismic data, the logging data and the core analysis result, a high-precision three-dimensional geological model can be constructed, and spatial visualization of parameters such as a reservoir top-bottom interface, thickness variation, porosity, permeability and the like can be realized. Meanwhile, the scientificity and the prediction accuracy of target optimization can be obviously improved by combining the multi-index comprehensive evaluation and the quantitative resource amount calculation formula. However, in the existing research, the combination of three-dimensional geologic modeling and karst reservoir evaluation is still not tight enough, and in particular, the method has the defects of algorithm formula chaining, parameter interpretation consistency and traceability of a calculation process.

Based on the problems, there is an urgent need to design a karst region petroleum exploration target optimization and resource amount prediction method based on three-dimensional geologic modeling to solve the problems.

Disclosure of Invention

The invention aims to solve the technical problems in the background art and provide a karst region petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling, and the aims of the invention are realized in the following way:

the karst region petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling comprises the following steps:

S1, multi-source data collection and standardization processing, namely acquiring a three-dimensional seismic attribute body, a logging curve, drilling core description data, geological profile data and production dynamic data which cover a target karst area, removing noise and abnormal values, carrying out standardization processing on different source data according to a maximum value D _max and a minimum value D _min to form a standardized data set D ', wherein D ' comprises density, impedance, acoustic time difference and natural gamma attribute parameters, and the D ' is used as basic data input of the subsequent time deep conversion and space alignment steps;

S2, deep conversion and space alignment are carried out, namely, based on a regional speed model v (t, r), seismic data are converted from a time domain tau (r) to a depth domain z (r), well logging curves and core data are resampled to a three-dimensional grid consistent with a seismic body, integrated registration of the seismic well data is realized, and a modeling point set P (D ', z) containing attribute data D' and depth information z (r) is obtained and is used as input of three-dimensional geological modeling;

S3, three-dimensional geological modeling, namely under the fault constraint function g (x, y) and horizon control conditions, taking a modeling point set P (D', Z) as input, completing structural modeling, sedimentary facies modeling and physical modeling, and obtaining three-dimensional distribution of reservoir interface elevation distribution Z (x, y) and corresponding porosity phi (x, y, Z) and permeability K (x, y, Z), wherein the results are used as input data for identifying karst development units;

S4, identifying a karst development unit, namely extracting karst related attributes based on standardized attribute data D' and a three-dimensional physical model, calculating karst development indexes I _karst, marking karst cave and crack development units in a three-dimensional grid, wherein the obtained I _karst and physical parameters are used together for optimizing and evaluating an exploration target;

S5, optimizing and evaluating an exploration target, namely taking reservoir thickness H, porosity phi, permeability K, karst development index I _karst, construction trap integrity C and oil-gas content display O as inputs, establishing a multi-index comprehensive evaluation system, calculating comprehensive scores S, and sequencing and optimizing the exploration target area according to the scoring result;

s6, calculating the effective area and the thickness, namely calculating the effective area A and the effective reservoir thickness H _eff according to the comprehensive score S and the karst development index I _karst of the optimal target area, wherein the result is used for calculating the water saturation and the resource amount;

S7, calculating the water saturation, namely estimating the water saturation S _w of the target area by using the porosity distribution phi, and combining the effective area A and the effective thickness H _eff as input of resource quantity calculation;

S8, calculating the resource quantity, namely substituting an effective area A, an effective thickness H _eff, a porosity phi, a water saturation S _w, a crude oil density rho _o and a stratum volume coefficient B _o into a volumetric method calculation formula to obtain an original geological resource quantity Q;

And S9, visual display and closed-loop optimization, namely, jointly displaying the comprehensive score S and the resource quantity Q on a three-dimensional visual platform to generate an exploration deployment scheme, backfilling new drilling production data to the step S1 to update and optimize a model, and realizing iterative optimization closed loop of exploration target optimization and resource quantity prediction.

As a preferable technical scheme of the invention, the normalization processing and time depth conversion formulas of the S1 and the S2 are as follows:

wherein, D _ij is the attribute value of the j-th sampling point of the i-th log, D _max and D _min are the maximum value and the minimum value of the dataset, v (t, r) is the velocity function at the position r, τ (r) is the modeling point set P (D ', z) formed by the output D' and z (r) during the double journey trip.

As a preferable technical scheme of the invention, the structural modeling elevation calculation formula of the S3 is as follows:

Wherein z _i (D ', z) is the depth of the i-th point in the modeling point set P (D', z), λ _i is the interpolation weight, μ is the fault constraint coefficient, g (x, y) is the fault constraint function, and g ₀ is the reference plane function.

As a preferred technical solution of the present invention, the interpolation weight λ _i is calculated by the kriging method:

C(P)λ_i=c(P)

Wherein C (P) is the covariance matrix of points inside the modeling point set P (D', z), and C (P) is the covariance vector between the predicted point and the known point.

As a preferable technical scheme of the invention, the karst development index calculation formula is as follows:

wherein A _j is the measured value of the j-th karst related attribute, Sigma _A is the standard deviation and m is the number of karst related attributes for its mean.

As a preferable technical scheme of the invention, the comprehensive scoring formula is as follows:

Wherein X _k is the weight of reservoir thickness H, porosity phi, permeability K, karst development index I _karst, construction trap integrity C and oil and gas content display O in sequence, alpha _k is the K index, and

As a preferable technical scheme of the invention, the calculation formulas of the effective area A and the effective thickness H _eff are as follows:

Wherein Ω _S is the target zone range for which the composite score is greater than the set threshold, and H (x, y, z) is the reservoir thickness profile.

As a preferable technical scheme of the invention, the calculation formula of the water saturation S _w is as follows:

wherein phi is porosity, a, m and n are empirical coefficients, and R _w is formation water resistivity.

As a preferable technical scheme of the invention, the calculation formula of the original geological resource Q is as follows:

Q=7758·A·H_eff·φ·(1-S_w)·ρ_o·B_o

Wherein A is the effective area, H _eff is the effective thickness, phi is the porosity, S _w is the water saturation, ρ _o is the crude oil density, and B _o is the formation volume factor.

As a preferred technical solution of the present invention, the closed loop optimization process implements parameter update by the following formula:

wherein θ is a parameter set of the three-dimensional geological model and the evaluation index system, η is a learning rate, Is a loss function based on the newly added drilling data D _new.

Firstly, the technical scheme realizes the depth fusion of multi-source geology and geophysical data, seismic attributes, logging curves and core analysis results are introduced simultaneously in the three-dimensional geological modeling process, and the accuracy and the continuity of spatial characterization of a karst reservoir are obviously improved through methods such as construction constraint kriging interpolation, multi-attribute inversion and the like. Compared with the traditional two-dimensional section interpretation, the method can completely present the three-dimensional distribution of the reservoir morphological boundary and physical parameters, and reduces interpretation deviation caused by insufficient local information.

Secondly, the technical scheme introduces a multi-index comprehensive evaluation system in the exploration target optimization link, and performs standardized treatment and weighted calculation on multi-dimensional parameters such as karst development index, reservoir thickness, porosity, permeability, trap integrity, oil gas display and the like to form a quantitative and adjustable optimization scoring mechanism. The mechanism has flexibility in setting the weight coefficient and the threshold value, can be dynamically adjusted according to geological backgrounds and exploration phases of different blocks, realizes closed-loop optimization of the geological recognition evaluation result, and greatly improves the scientificity and adaptability of target optimization.

Finally, the technical scheme is spatially coupled with the three-dimensional geological model, so that the resource quantity calculation not only considers the spatial variation of physical parameters of the reservoir, but also considers the inhomogeneous characteristics of karst development, thereby ensuring the precision and simultaneously ensuring the maintainability and the reliability, and providing a solid quantitative basis for the design of the follow-up development scheme and investment decision.

Drawings

FIG. 1 is a flow chart of a method for predicting the oil exploration target optimization and the resource amount of a karst region based on three-dimensional geological modeling.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

With reference to fig. 1, the implementation of the present invention is described in detail below in connection with a specific embodiment.

The invention provides a karst region petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling, which forms a closed-loop optimization flow from data to exploration deployment scheme through multi-source data fusion, geological modeling, karst development analysis, comprehensive evaluation and resource quantity prediction.

The method is particularly suitable for complex structural feature identification and optimal target area evaluation of karst reservoirs, and can remarkably improve exploration precision and reliability of resource quantity prediction.

In a specific implementation process, firstly, multi-source data covering a target karst region is obtained in step S1, including three-dimensional seismic attribute body, logging curve, drilling core description data, geological section data and production dynamic data, noise rejection and outlier removal processing are performed, and data from different sources are normalized by adopting a maximum value and minimum value normalization method to form a normalized data set D', wherein the calculation formula is as follows:

wherein D _ij is the attribute value of the jth sampling point of the ith log, and D _max and D _min are the maximum value and the minimum value of the dataset, respectively.

In step S2, the seismic data is converted from the time domain τ (r) to the depth domain z (r) based on the regional velocity model v (t, r), as follows:

where τ (r) is the speed function when traveling in double-pass, v (t, r). The converted depth data and the standardized attribute data D 'form a modeling point set P (D', z), and spatial registration of the seismic well data under the three-dimensional grid is realized.

In step S3, using modeling point set P (D', Z0) as input, and under fault constraint function g (x, y) and horizon control conditions, obtaining reservoir interface elevation distribution Z (x, y; P) by using a construction modeling method:

Where z _i (D', z) is the depth of the i-th point in the modeling point set, λ _i is the interpolation weight, μ is the fault constraint coefficient, g (x, y) is the fault constraint function, and g ₀ is the reference plane function. The interpolation weight λ is determined by the kriging method:

C(P)λ=c(P)

C (P) is the covariance matrix of the known points, and C (P) is the covariance vector of the predicted points and the known points.

In step S4, karst-related attributes are extracted based on the three-dimensional physical model, and a karst development index I is calculated _karst

Wherein A _j is the j-th karst related attribute value,Sigma _A is standard deviation, and m is the karst correlation attribute number. The index is used for describing the development degree of the karst reservoir and provides an important reference for target preference.

In step S5, multiple indexes including reservoir thickness H, porosity Φ, permeability K, karst development index I _karst, construction trap integrity C, oil and gas content indicator O and the like are synthesized, and a weighted scoring method is used to calculate a comprehensive score S:

Wherein X _k corresponds to each index, and alpha _k is a weight coefficient and satisfies The high scoring area is determined to be the preferred survey target area.

In step S6, the target area range Ω _S is extracted according to the comprehensive scoring result, and the effective area a and the effective reservoir thickness H _eff are calculated:

wherein H (x, y, z) is a three-dimensional reservoir thickness distribution function.

In step S7, the water saturation is estimated using the porosity phi and empirical coefficients S _w, where a, m, n are empirical coefficients and R _w is the formation water resistivity. In step S8, the metric unit volume method is used to calculate the original geological resource Q:

Q=A·H_eff·φ·(1-S_w)·ρ_o/B_o

Wherein, A is square meter, H _eff is meter, phi and S _w are dimensionless fractions, rho _o is crude oil density, B _o is stratum volume coefficient (dimensionless), and the calculated result Q is ton.

In step S9, the comprehensive score S and the resource quantity Q are subjected to space superposition display in a three-dimensional visual platform, and an exploration deployment scheme is generated by combining geological information such as faults, trap and karst units. The new drilling and production data D _new will be backfilled to step S1, achieving closed-loop optimization of the model by the following parameter update formula:

Q=7758·A·H_eff·φ·(1-S_w)·ρ_o·B_o

Wherein A is the effective area, H _eff is the effective thickness, phi is the porosity, S _w is the water saturation, ρ _o is the crude oil density, and B _o is the formation volume factor.

Through the implementation mode, the whole-process chain derivation from multi-source geological data processing, three-dimensional modeling, karst development evaluation to exploration target optimization and resource quantity prediction is realized, the input-output consistency and dimension uniformity among formulas are ensured, and the method has good practicability and popularization.

Taking a karst zone (ZK block) at the edge of a certain sea-phase carbonate bench as an example, the three-dimensional seismic coverage area is about 18 square miles, and 7 wells are provided with logging data (density, acoustic wave, deep lateral resistivity, natural gamma) and 2 rock core analysis data.

Firstly, performing data cleaning and normalization processing, and adopting a maximum and minimum normalization formula D '_ij＝(D_ij-D_min)/(D_max-D_min), for example, the seismic amplitude of a certain well at a target horizon is d=1.90, and when the horizon statistics reaches D _min＝0.40,D_max =2.80, the normalization result is D' = (1.9-0.40)/(2.80-0.40) =0.625.

And then performing time-depth conversion, adopting a joint calibration speed model, carrying out one-pass travel time tau=0.80 s at a certain point, obtaining depth z (r) =v×tau=7600 ft at the speed v (t, r) =9500 ft/s, and resampling the logging and core attribute to the seismic three-dimensional grid to form a modeling point set P (D', z).

In three-dimensional geological modeling, a structural constraint kriging interpolation method is adopted on the top surface of a reservoir under the condition of fault constraint function g (x, y) and main horizon control, three adjacent well point depths 7520, 7610 and 7680ft are taken as examples, weight lambda= [0.22,0.53,0.25] is obtained based on a variation function, and the depth of a predicted point is Z=0.22×7520+0.53×7610+0.25×7680 approximately equal to 7,614ft.

The porosity phi (x, y, z) and permeability K (x, y, z) were then inverted and constrained in a 50X 1ft grid, and the statistics showed that the target zone phi was 0.12-0.19 and K was 40-300mD. Karst development identification selects five attributes of coherent body (COH), curvature (CUR), amplitude Attenuation (ATT), deep lateral resistivity (LLD) and acoustic time Difference (DT), and a group of samples [0.80,0.30, -0.20,0.50,0.10] are obtained after calculation of standardized values.

Principal component analysis yielded a eigenvalue λ _j = [3.0,1.2,0.5,0.2,0.1], normalized weights w _j = 0.625,0.25,0.104,0.042,0.021]. Substituting karst development index formula

The calculation result is I _karst approximately equal to 0.577, and the high value area is mainly distributed on the southwest side terrace edge band.

The targets preferably employ Gaussian membershipThe weight coefficient alpha= [0.35,0.25,0.20,0.15,0.03,0.02] and the index of the candidate area A is statistically that the layer thickness H=50 ft, the average porosity phi=0.16, the permeability K=180 mD, the karst index I _karst =0.58, the trap integrity C=0.80 and the oil gas display O=0.60. The composite score S.apprxeq.0.664 is calculated item by item. And screening grid cells with S being more than or equal to 0.65 and I _karst being more than or equal to 0.55 in a whole area, wherein the cell area is 2.50acre, and 360 cells are obtained, so that the effective area A=900 acre and the average effective thickness H _eff =48 ft are obtained.

The water saturation is calculated using the alchi formula,

Taking a=0.62, m=2.0, n=2.0, r _w =0.1Ω·m, Φ=0.16, obtaining S _w ≡0.29, and taking the formation volume coefficient B _o =1.20, substituting into the resource amount formula of the claims:

Q=7758·A_acre·H_ft·φ·(1-S_w)/B_o

the calculation process is as follows:

1.A·H=900×48=43,200acre-ft;

2.7758×43,200=335,145,600;

3. multiplied by the porosity:. X0.16 = 53,623,296;

4. Oil phase saturation: ×0.71= 38,072,540.16;

5. divide by B _o = 1.20:q≡ 31,727,117STB;

The final predicted resource amount is about 3,170 ten thousand barrels, accords with the expected magnitude of the carbonate karst reservoir under the conditions of medium effective area, thickness and porosity, and has engineering feasibility and conservation. And superposing the prediction result and the three-dimensional model, preferentially arranging 2 development wells and 1 evaluation well in a high-scoring area, and carrying out iterative updating on model weight and threshold value by combining subsequent production data to realize closed-loop optimization of data-model-one deployment-feedback.

The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.

Claims (10)

1. The karst region petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling is characterized by comprising the following steps of:

S1, multi-source data collection and standardization processing, namely acquiring a three-dimensional seismic attribute body, a logging curve, drilling core description data, geological profile data and production dynamic data which cover a target karst area, removing noise and abnormal values, carrying out standardization processing on different source data according to a maximum value D _max and a minimum value D _min to form a standardized data set D ', wherein D ' comprises density, impedance, acoustic time difference and natural gamma attribute parameters, and the D ' is used as basic data input of the subsequent time deep conversion and space alignment steps;

S2, deep conversion and space alignment are carried out, namely, based on a regional speed model v (t, r), seismic data are converted from a time domain tau (r) to a depth domain z (r), well logging curves and core data are resampled to a three-dimensional grid consistent with a seismic body, integrated registration of the seismic well data is realized, and a modeling point set P (D ', z) containing attribute data D' and depth information z (r) is obtained and is used as input of three-dimensional geological modeling;

S3, three-dimensional geological modeling, namely under the fault constraint function g (x, y) and horizon control conditions, taking a modeling point set P (D', Z) as input, completing structural modeling, sedimentary facies modeling and physical modeling, and obtaining three-dimensional distribution of reservoir interface elevation distribution Z (x, y) and corresponding porosity phi (x, y, Z) and permeability K (x, y, Z), wherein the results are used as input data for identifying karst development units;

S4, identifying a karst development unit, namely extracting karst related attributes based on standardized attribute data D' and a three-dimensional physical model, calculating karst development indexes I _karst, marking karst cave and crack development units in a three-dimensional grid, wherein the obtained I _karst and physical parameters are used together for optimizing and evaluating an exploration target;

S5, optimizing and evaluating an exploration target, namely taking reservoir thickness H, porosity phi, permeability K, karst development index I _karst, construction trap integrity C and oil-gas content display O as inputs, establishing a multi-index comprehensive evaluation system, calculating comprehensive scores S, and sequencing and optimizing the exploration target area according to the scoring result;

s6, calculating the effective area and the thickness, namely calculating the effective area A and the effective reservoir thickness H _eff according to the comprehensive score S and the karst development index I _karst of the optimal target area, wherein the result is used for calculating the water saturation and the resource amount;

S7, calculating the water saturation, namely estimating the water saturation S _w of the target area by using the porosity distribution phi, and combining the effective area A and the effective thickness H _eff as input of resource quantity calculation;

S8, calculating the resource quantity, namely substituting an effective area A, an effective thickness H _eff, a porosity phi, a water saturation S _w, a crude oil density rho _o and a stratum volume coefficient B _o into a volumetric method calculation formula to obtain an original geological resource quantity Q;

And S9, visual display and closed-loop optimization, namely, jointly displaying the comprehensive score S and the resource quantity Q on a three-dimensional visual platform to generate an exploration deployment scheme, and backfilling new drilling production data to the step S1 for model updating and optimization.

2. The method for predicting the oil exploration target preference and resource amount of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the normalization processing and time depth conversion formulas of the S1 and the S2 are as follows:

wherein, D _ij is the attribute value of the j-th sampling point of the i-th log, D _max and D _min are the maximum value and the minimum value of the dataset, v (t, r) is the velocity function at the position r, τ (r) is the modeling point set P (D ', z) formed by the output D' and z (r) during the double journey trip.

3. The method for predicting the oil exploration target preference and resource amount of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the structural modeling elevation calculation formula of S3 is:

Wherein z _i (D ', z) is the depth of the i-th point in the modeling point set P (D', z), λ _i is the interpolation weight, μ is the fault constraint coefficient, g (x, y) is the fault constraint function, and g ₀ is the reference plane function.

4. The method for predicting the oil exploration target preference and resource amount of a karst region based on three-dimensional geologic modeling according to claim 3, wherein the interpolation weight lambda _i is calculated by a kriging method:

C(P)λ_i=c(P)

Wherein C (P) is the covariance matrix of points inside the modeling point set P (D', z), and C (P) is the covariance vector between the predicted point and the known point.

5. The method for predicting the oil exploration target preference and resource amount of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the karst development index calculation formula is as follows:

wherein A _j is the measured value of the j-th karst related attribute, Sigma _A is the standard deviation and m is the number of karst related attributes for its mean.

6. The method for predicting the oil exploration target preference and resource amount of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the comprehensive scoring formula is:

Wherein X _k is the weight of reservoir thickness H, porosity phi, permeability K, karst development index I _karst, construction trap integrity C and oil and gas content display O in sequence, alpha _k is the K index, and

7. The method for predicting the oil exploration target preference and resource amount of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the calculation formula of the effective area A and the effective thickness H _eff is as follows:

Wherein Ω _S is the target zone range for which the composite score is greater than the set threshold, and H (x, y, z) is the reservoir thickness profile.

8. The method for predicting the oil exploration target preference and resource amount of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the water saturation S _w is calculated according to the following formula:

wherein phi is porosity, a, m and n are empirical coefficients, and R _w is formation water resistivity.

9. The method for predicting the oil exploration target preference and resource quantity of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the calculation formula of the original geologic resource quantity Q is as follows:

Q=7758·A·H_eff·φ·(1-S_w)·ρ_o·B_o

Wherein A is the effective area, H _eff is the effective thickness, phi is the porosity, S _w is the water saturation, ρ _o is the crude oil density, and B _o is the formation volume factor.

10. The method for predicting the oil exploration target preference and resource quantity of a karst region based on three-dimensional geologic modeling according to claim 1, wherein the closed-loop optimization process realizes parameter updating by the following formula:

wherein θ is a parameter set of the three-dimensional geological model and the evaluation index system, η is a learning rate, Is a loss function based on the newly added drilling data D _new.