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

Abstract

The invention relates to the field of oil exploration, and particularly discloses a karst area oil exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling. The method comprises the steps that firstly, seismic data, logging data and a rock core analysis result of a karst area are collected, data normalization and time-depth conversion are carried out, and a three-dimensional geologic model fusing structural constraints and physical property constraints is constructed; then multi-attribute indexes such as a coherent body, curvature, amplitude attenuation, deep lateral resistivity and interval transit time are selected, and karst development indexes are calculated through principal component analysis; in combination with parameters such as reservoir thickness, porosity, permeability, trap integrity and oil and gas display, utilizing a multi-index weighted scoring method to screen exploration favorable blocks; and finally, calculating reserves according to a volumetric method to obtain resource quantity prediction results of different target areas. According to the method, three-dimensional fusion modeling, multi-index quantitative optimization and chained calculation of multi-source data are realized, and quantifiable and traceable technical support is provided for oil exploration deployment in a karst area.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil exploration, in particular to a karst area oil exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling. BACKGROUND

[0002] Karst reservoir is one of the important types of carbonate rock oil and gas reservoirs, especially on the platform edge or ancient landform high of marine carbonate rock, the reservoir space is often controlled by the superimposed modification of dissolution and tectonic fracture. Compared with sandstone reservoir, karst reservoir has the characteristics of diverse pore types, strong heterogeneity, complex distribution law, etc., and the reservoir parameters change dramatically in space, which makes it difficult to optimize the exploration target and predict the resource quantity. Especially in the area with strong karst development, the variation range of physical parameters such as reservoir thickness, porosity and permeability in the plane and vertical direction is significant, so that the single well data is difficult to accurately reflect the overall characteristics of the reservoir, and the traditional two-dimensional seismic interpretation is difficult to reveal the three-dimensional heterogeneous distribution of the reservoir.

[0003] In the prior art, the karst area exploration target optimization method mainly relies on two-dimensional seismic profile interpretation, drilling and logging single well analysis, etc., and the spatial position of the karst reservoir is inferred by identifying abnormal amplitude, waveform distortion and other seismic responses. However, this method has three limitations: first, two-dimensional interpretation lacks continuous spatial constraints, and it is difficult to accurately depict the shape boundary of the karst reservoir; second, the quantitative relationship between attribute inversion and physical parameters is not accurate, and there is a lack of unified multi-source data fusion modeling framework; third, in resource quantity prediction, single factor volume method or analogy method is mostly used, and the multi-index comprehensive coupling of reservoir spatial distribution characteristics, physical parameters and karst development degree is not achieved, resulting in large uncertainty of the prediction result.

[0004] In recent years, the development of three-dimensional geological modeling and multi-attribute seismic inversion technology provides a new idea for the research of complex carbonate reservoir. By integrating seismic data, logging data and core analysis results, a high-precision three-dimensional geological model can be constructed to realize the spatial visualization of reservoir top and bottom interfaces, thickness changes, porosity and permeability parameters. At the same time, combined with multi-index comprehensive evaluation and quantitative resource quantity calculation formula, the scientificity of target optimization and the accuracy of prediction can be significantly improved. However, in the existing research, the combination of three-dimensional geological modeling and karst reservoir evaluation is still not close enough, especially in the aspects of algorithm formula chain, parameter interpretation consistency and traceability of calculation process.

[0005] Based on the above problems, it is urgent to design a karst area oil exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling to solve the above problems. SUMMARY

[0006] The present application aims to solve the technical problems raised in the background art, and provides a karst area oil exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling.

[0007] The karst area oil exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling comprises the following steps:

[0008] S1 multi-source data collection and standardization processing: obtain three-dimensional seismic attribute bodies, well logging curves, drilling core description data, geological profile data and production dynamic data covering the target karst area, remove noise and outliers, and normalize different source data according to the maximum value D max and the minimum value D min to form a standardized data set D', wherein D' contains density, impedance, acoustic time difference and natural gamma attribute parameters; the D' is used as the basic data input for the subsequent time-depth conversion and spatial alignment steps;

[0009] S2 time-depth conversion and spatial alignment: convert the seismic data from the time domain τ(r) to the depth domain z(r) based on the regional velocity model v(t,r), and resample the well logging curves and core data to the three-dimensional grid consistent with the seismic body to realize the integrated registration of seismic and well data, and obtain the modeling point set P(D',z) containing attribute data D' and depth information z(r) as the input for three-dimensional geological modeling;

[0010] S3 three-dimensional geological modeling: under the fault constraint function g(x,y) and the horizon control condition, complete the structural modeling, sedimentary facies modeling and physical property modeling by taking the modeling point set P(D',z) as the input, and obtain the three-dimensional distribution of the reservoir interface elevation distribution Z(x,y) and the corresponding porosity φ(x,y,z) and permeability K(x,y,z), which are used as the input data for karst development unit identification;

[0011] S4 karst development unit identification: based on the standardized attribute data D' and the three-dimensional physical property model, extract karst-related attributes and calculate the karst development index I karst , and mark the cave and fracture development units in the three-dimensional grid; the obtained I karst is used together with the physical property parameters for exploration target optimization evaluation;

[0012] S5 exploration target optimization evaluation: take the reservoir thickness H, porosity φ, permeability K, karst development index I karst , structural trap integrity C and oil and gas show O as the input, establish a multi-index comprehensive evaluation system and calculate the comprehensive score S, and sort and optimize the exploration target area according to the score results; the obtained high-score target area will enter the effective area and thickness calculation step;

[0013] S6 Effective area and thickness calculation: according to the comprehensive score S of the preferred target area and the karst development index I karst , the effective area A and the effective reservoir thickness H eff are calculated eff , and the results will be used for water saturation and resource quantity calculation;

[0014] S7 Water saturation calculation: the water saturation S of the target area is estimated using the porosity distribution φ w , and the effective area A and the effective thickness H eff are combined as inputs for resource quantity calculation;

[0015] S8 Resource quantity calculation: the effective area A, the effective thickness H eff , the porosity φ, the water saturation S w , the crude oil density ρ o and the formation volume factor B o are substituted into the volumetric method calculation formula to obtain the original geological resource quantity Q;

[0016] S9 Visual display and closed-loop optimization: the comprehensive score S and the resource quantity Q are jointly displayed on a three-dimensional visualization platform to generate an exploration deployment plan, and new drilling production data is backfilled to step S1 for model updating and optimization, realizing iterative optimization and closed loop of exploration target optimization and resource quantity prediction.

[0017] As a preferred technical solution of the present application, the standardization processing and time-depth conversion formula of S1 and S2 are:

[0018]

[0019] Wherein, D ij is the attribute value of the jth sampling point of the ith well logging curve, D max and D min are the maximum and minimum values of the data set respectively, v(t,r) is the velocity function at position r, τ(r) is the two-way travel time, and the output D' and z(r) constitute the modeling point set P(D',z).

[0020] As a preferred technical solution of the present application, the construction modeling elevation calculation formula of S3 is:

[0021]

[0022] Wherein, z i (D',z) is the depth of the ith 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 g0 is the reference surface function.

[0023] As a preferred technical solution of the present application, the interpolation weight λi The covariance matrix of the points in the set P(D',z) is calculated by the Kriging method:

[0024] C(P)λ i =c(P)

[0025] Wherein, C(P) is the covariance matrix of the points in the set P(D',z), and c(P) is the covariance vector between the prediction point and the known points.

[0026] As a preferred technical solution of the present application, the karst development index calculation formula is:

[0027]

[0028] Wherein, A j is the measured value of the jth karst-related attribute, is the mean value, and σ A is the standard deviation, and m is the number of karst-related attributes.

[0029] As a preferred technical solution of the present application, the comprehensive score formula is:

[0030]

[0031] Wherein, X k is the reservoir thickness H, porosity φ, permeability K, karst development index I karst , structural trap integrity C, and oil and gas show O, respectively, α k is the weight of the kth index, and

[0032] As a preferred technical solution of the present application, the calculation formula of the effective area A and the effective thickness H eff is:

[0033]

[0034] Wherein, Ω S is the range of the target area with a comprehensive score greater than a set threshold, and H(x,y,z) is the reservoir thickness distribution.

[0035] As a preferred technical solution of the present application, the water saturation S w calculation formula is:

[0036]

[0037] Wherein, φ is the porosity, a, m, and n are empirical coefficients, and R w is the formation water resistivity.

[0038] As a preferred technical solution of the present application, the original geological resource quantity Q calculation formula is:

[0039] Q = 7758 AH eff φ (1 - S w ) ρ o B o

[0040] Where A is the effective area, H eff is the effective thickness, φ is the porosity, S w is the water saturation, ρ o is the crude oil density, and B o is the formation volume factor.

[0041] As a preferred technical solution of the present application, the closed-loop optimization process updates parameters through the following formula:

[0042]

[0043] Where θ is the parameter set of the three-dimensional geological model and the evaluation index system, η is the learning rate, is the loss function based on the newly added drilling data D new .

[0044] Beneficial effects: First, this technical solution realizes the deep integration of multi-source geological and geophysical data. In the process of three-dimensional geological modeling, seismic attributes, well logging curves and core analysis results are introduced at the same time. Through methods such as structural constraint kriging interpolation and multi-attribute inversion, the accuracy and continuity of karst reservoir space description are significantly improved. Compared with traditional two-dimensional profile interpretation, this method can fully present the three-dimensional distribution of reservoir shape boundaries and physical parameters, reducing the interpretation deviation caused by insufficient local information.

[0045] Second, this technical solution introduces a multi-index comprehensive evaluation system in the exploration target optimization link. The karst development index, reservoir thickness, porosity, permeability, trap integrity and oil and gas show are standardized and weighted to form a quantitative and adjustable optimization scoring mechanism. This mechanism has flexibility in setting weight coefficients and threshold values, which can be dynamically adjusted according to the geological background and exploration stage of different blocks, realizing closed-loop optimization from geological understanding to evaluation results, and greatly improving the scientificity and adaptability of target optimization.

[0046] Finally, through spatial coupling with the three-dimensional geological model, the resource quantity calculation not only considers the spatial variation of reservoir physical parameters, but also takes into account the heterogeneous characteristics of karst development, ensuring accuracy while being conservative and reliable, providing a solid quantitative basis for subsequent development plan design and investment decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1A method flowchart of a karst area petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling is provided for the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present application more clear and explicit, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0049] Reference Figure 1 The implementation of the present application is described in detail below in combination with specific examples.

[0050] The present application provides a karst area petroleum exploration target optimization and resource quantity prediction method based on three-dimensional geological modeling. Through multi-source data fusion, geological modeling, karst development analysis, comprehensive evaluation and resource quantity prediction, a closed-loop optimization process from data to exploration deployment scheme is formed.

[0051] The present method is particularly suitable for complex structural feature identification and target area evaluation of karst reservoirs, and can significantly improve the exploration accuracy and reliability of resource quantity prediction.

[0052] In the specific implementation process, first, in step S1, multi-source data covering the target karst area is obtained, including three-dimensional seismic attribute volume, well logging curve, drilling core description data, geological profile data and production dynamic data, and noise elimination and abnormal value removal processing is performed. Different sources of data are standardized by using the maximum and minimum normalization method to form a standardized data set D', and the calculation formula is:

[0053]

[0054] Wherein, D ij is the attribute value of the jth sampling point of the ith well logging curve, D max and D min are the maximum and minimum values of the data set, respectively.

[0055] 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), and the formula is:

[0056]

[0057] Wherein, τ(r) is the two-way travel time, and v(t,r) is the velocity function. The converted depth data and the standardized attribute data D' constitute the modeling point set P(D',z), and the spatial registration of the seismic and well data in the three-dimensional grid is realized.

[0058] In step S3, the reservoir interface elevation distribution Z(x, y; P) is obtained by using the structure modeling method under the fault constraint function g(x, y) and the horizon control condition with the set P(D', z0) as input:

[0059]

[0060] wherein z i (D', z) is the depth of the i-th point in the set of modeling points, λ i is the interpolation weight, μ is the fault constraint coefficient, g(x, y) is the fault constraint function, and g0is the reference surface function. The interpolation weight λ is determined by the Kriging method:

[0061] C(P) λ = c(P)

[0062] C(P) is the known point covariance matrix, and c(P) is the covariance vector of the prediction point and the known point.

[0063] In step S4, the karst-related attributes are extracted based on the three-dimensional physical property model, and the karst development index I karst

[0064]

[0065] wherein A j is the j-th karst-related attribute value, is the mean value, σ A is the standard deviation, and m is the number of karst-related attributes. The index is used to describe the development degree of the karst reservoir, and provides an important reference for target optimization.

[0066] In step S5, the comprehensive score S is calculated by using the weighted scoring method based on the multiple indexes such as the reservoir thickness H, the porosity φ, the permeability K, the karst development index I karst , the structural trap integrity C, and the oil and gas show O:

[0067]

[0068] wherein X k corresponds to each index, α k is the weight coefficient and satisfies The high-score area is determined as the preferred exploration target area.

[0069] In step S6, the target area range Ω S is extracted according to the comprehensive score result, and the effective area A and the effective reservoir thickness H eff are calculated:

[0070]

[0071] Where H(x,y,z) is the three-dimensional reservoir thickness distribution function.

[0072] In step S7, the water saturation S is estimated using porosity φ and empirical coefficients. w Where a, m, and n are empirical coefficients, and R0 w The resistivity of the formation water is given. In step S8, the original geological resource quantity Q is calculated using the metric unit volume method.

[0073] Q = A·H eff ·φ·(1-S w )·ρ o / B o

[0074] Where A is in square meters, H eff The unit is meters, φ and S w ρ is a dimensionless fraction. o B represents the density of crude oil. o The volume factor (dimensionless) is Q, and the calculated result Q is in tons.

[0075] In step S9, the comprehensive score S and resource quantity Q are spatially overlaid and displayed on a 3D visualization platform. Combined with geological information such as faults, traps, and karst units, an exploration deployment plan is generated. New drilling and production data D... new The data will be backfilled to step S1, and the closed-loop optimization of the model will be achieved by updating the formula using the following parameters:

[0076] Q = 7758·A·H eff ·φ·(1-S w )·ρ o ·B o

[0077] Where A is the effective area, H eff Where φ is the effective thickness, S is the porosity, and S is the effective thickness. w ρ represents the water saturation level. o B represents the density of crude oil. o This is the formation volume coefficient.

[0078] Through the above implementation methods, the present invention realizes the chain derivation of the entire process from multi-source geological data processing, three-dimensional modeling, karst development evaluation to exploration target selection and resource prediction, ensuring the consistency of input and output and the uniformity of dimensions between formulas, and has good practicality and extensibility.

[0079] Taking the karst zone (ZK block) at the edge of a marine carbonate platform as an example, the 3D seismic coverage area is about 18 square miles, and there are already 7 wells with logging data (density, sonic, deep lateral resistivity, natural gamma) and 2 core analysis data.

[0080] Firstly, data cleaning and normalization are carried out, and the maximum and minimum value normalization formula D′ ij =(D ij -D min ) / (D max -D min ) is adopted, for example, the seismic amplitude of a well in the target layer is D=1.90, and the statistical value in the layer is D min =0.40, D max =2.80, and the normalization result is D′= (1.9-0.40) / (2.80-0.40)=0.625.

[0081] Then, time-depth conversion is carried out, a joint calibration velocity model is adopted, the one-way travel time of a point is τ=0.80 s, the velocity v(t,r)=9500 ft / s, the depth z(r)=v×τ=7600 ft is obtained, the logging and core properties are resampled to the seismic three-dimensional grid, and the modeling point set P(D′,z) is formed.

[0082] In three-dimensional geological modeling, the structural constraint Kriging interpolation method is adopted for the reservoir top surface under the condition of fault constraint function g(x,y) and main layer control, taking the depths of three adjacent well points 7520, 7610 and 7680 ft as an example, the weight λ=[0.22,0.53,0.25] is obtained based on the variation function, and the depth of the prediction point is Z=0.22×7520+0.53×7610+0.25×7680≈7,614 ft.

[0083] Then, porosity φ(x,y,z) and permeability K(x,y,z) are inverted and constrained in 50×50×1 ft grid, and the statistical results show that the target area φ is 0.12-0.19, and K is 40-300 mD. Five attributes of coherence (COH), curvature (CUR), amplitude attenuation (ATT), deep lateral resistivity (LLD) and acoustic travel time (DT) are selected for karst development identification, and a group of samples [0.80,0.30,-0.20,0.50,0.10] is obtained after calculating the normalized values.

[0084] The eigenvalues λ j =[3.0,1.2,0.5,0.2,0.1] are obtained by principal component analysis, and the normalized weights w j =0.625,0.25,0.104,0.042,0.021]. The karst development index formula

[0085]

[0086] The calculation result is I karst ≈0.577, and the high value area is mainly distributed in the southwest side of the platform edge.

[0087] The target preferably adopts a Gaussian membership The weight coefficient a = [0.35, 0.25, 0.20, 0.15, 0.03, 0.02], the candidate area A index statistics are: layer thickness H = 50 ft, average porosity φ = 0.16, permeability K = 180 mD, karst index I karst = 0.58, trap integrity C = 0.80, oil and gas show O = 0.60. The comprehensive score S ≈ 0.664 can be obtained by calculating item by item. In the whole area, the grid units with S ≥ 0.65 and I karst ≥ 0.55 are screened, the unit area is 2.50 acres, there are 360 units, the effective area A = 900 acres, the average effective thickness H eff = 48 ft.

[0088] The water saturation is calculated by using the Archie formula,

[0089] Taking a = 0.62, m = 2.0, n = 2.0, R w = 0.11 Ω·m, φ = 0.16, S w ≈ 0.29 is obtained, the formation volume factor B o = 1.20 is taken, and the resource amount formula in the claim is substituted:

[0090] Q = 7758·A acre ·H ft ·φ·(1-S w ) / B o

[0091] The calculation process is:

[0092] 1. A·H = 900 × 48 = 43,200 acre-ft;

[0093] 2. 7758 × 43,200 = 335,145,600;

[0094] 3. Multiply the porosity: × 0.16 = 53,623,296;

[0095] 4. Multiply the oil phase saturation: × 0.71 = 38,072,540.16;

[0096] 5. Divide by B o = 1.20: Q ≈ 31,727,117 STB;

[0097] The final predicted resource volume is about 31.7 million barrels, which is consistent with the expected magnitude of carbonate karst reservoirs under the conditions of medium effective area, thickness and porosity, and the results are feasible and conservative. The predicted results are superimposed on the three-dimensional model, two development wells and one evaluation well are preferentially arranged in the high-score area, and the model weight and threshold are iteratively updated combined with subsequent production data, realizing the closed-loop optimization of data-model-deployment-feedback.

[0098] The above merely describes preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, and improvement within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for selecting optimal oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling, characterized in that, Includes the following steps: S1 Multi-Source Data Acquisition and Standardization: Acquire 3D seismic attribute volumes, well logging curves, drilling core description data, geological profile data, and production dynamic data covering the target karst area; remove noise and outliers; and categorize data from different sources according to their maximum value D. max With minimum value D min The data is normalized to form a standardized dataset D′, which contains density, impedance, acoustic transit time, and natural gamma attribute parameters. This D′ serves as the basic data input for subsequent time-depth conversion and spatial alignment steps. S2 Time-Depth Conversion and Spatial Alignment: Based on the regional velocity model v(t,r), seismic data is converted from the time domain τ(r) to the depth domain z(r), and well logging curves and core data are resampled to a three-dimensional grid consistent with the seismic body to achieve integrated registration of seismic and well data, resulting in a modeling point set P(D′,z) containing attribute data D′ and depth information z(r), which serves as the input for three-dimensional geological modeling; S3 3D geological modeling: Under the conditions of fault constraint function g(x,y) and stratigraphic control, and with the modeling point set P(D′,z) as input, structural modeling, sedimentary facies modeling and physical property modeling are completed, and the reservoir interface elevation distribution Z(x,y) and its corresponding three-dimensional distribution of porosity φ(x,y,z) and permeability K(x,y,z) are obtained. These results will be used as input data for the identification of karst development units. S4 Karst Development Unit Identification: Based on standardized attribute data D′ and a three-dimensional physical property model, karst-related attributes were extracted and the karst development index I was calculated. karst In the 3D mesh, the development units of caves and fissures are labeled; the resulting I karst It will be used in conjunction with physical property parameters for the optimal evaluation of exploration targets; S5 Exploration Target Selection and Evaluation: Based on reservoir thickness H, porosity φ, permeability K, and karst development index I karst Using structural trap integrity (C) and hydrocarbon show (O) as inputs, a multi-index comprehensive evaluation system is established and a comprehensive score (S) is calculated. Based on the score results, exploration target areas are ranked and optimized. The high-scoring target areas will then proceed to the effective area and thickness calculation steps. S6 Effective Area and Thickness Calculation: Based on the comprehensive score S and karst development index I of the preferred target area. karst Calculate its effective area A and effective reservoir thickness H. eff This result will be used to calculate water saturation and resource quantity; S7 Water Saturation Calculation: Estimating the water saturation S of the target area using the porosity distribution φ. w And combining the effective area A and the effective thickness H eff As input for resource quantity calculation; S8 resource quantity calculation: Effective area A, effective thickness H... eff Porosity φ, water saturation S w Crude oil density ρ o and formation volume factor B o Substituting into the volumetric calculation formula, we obtain the original geological resource quantity Q; S9 Visualization and Closed-Loop Optimization: The comprehensive score S and resource quantity Q are jointly displayed on the 3D visualization platform to generate an exploration deployment plan, and the new drilling production data is backfilled into step S1 for model update and optimization.

2. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling as described in claim 1, characterized in that, The standardization and time-depth conversion formulas for S1 and S2 are as follows: Among them, D ij Let D be the attribute value of the j-th sampling point of the i-th logging curve. max With D min These are the maximum and minimum values ​​of the dataset, respectively. v(t,r) is the velocity function at position r, and τ(r) is the output of the modeling point set P(D′,z) when traveling two miles.

3. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling according to claim 1, characterized in that, The formula for calculating the elevation of the S3 construction model is as follows: Among them, z i (D′,z) represents the depth of the i-th point in the modeling point set P(D′,z), and λ i denoted as interpolation weight, μ as fault constraint coefficient, g(x,y) as fault constraint function, and g0 as reference surface function.

4. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling as described in claim 3, characterized in that, The interpolation weight λ i Calculated using the Kriging method: C(P)λ i =c(P) Where 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 selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling according to claim 1, characterized in that, The formula for calculating the karst development index is as follows: Among them, A j Let j be the measured value of the j-th karst-related attribute. Its mean, σ A denoted as standard deviation, and m as the number of karst-related attributes.

6. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling according to claim 1, characterized in that, The comprehensive scoring formula is as follows: Among them, X k The parameters are, in order: reservoir thickness H, porosity φ, permeability K, and karst development index I. karst Structural trap integrity (C) and hydrocarbon indications (O, α) k Let be the weight of the k-th indicator, and 7. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling according to claim 1, characterized in that, The effective area A and the effective thickness H eff The calculation formula is: Among them, Ω S H(x,y,z) represents the target area range where the comprehensive score is greater than the set threshold, and H(x,y,z) represents the reservoir thickness distribution.

8. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling according to claim 1, characterized in that, The water saturation S w The calculation formula is: Where φ is porosity, a, m, and n are empirical coefficients, and R w This represents the resistivity of the formation water.

9. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling according to claim 1, characterized in that, The formula for calculating the original geological resource quantity Q is: Q=7758·A·H eff ·φ·(1-S w )·r o ·B o Where A is the effective area, H eff Where φ is the effective thickness, S is the porosity, and S is the effective thickness. w ρ represents the water saturation level. o B represents the density of crude oil. o This is the formation volume coefficient.

10. The method for selecting oil exploration targets and predicting resource quantity in karst areas based on three-dimensional geological modeling according to claim 1, characterized in that, The closed-loop optimization process updates parameters using the following formula: Where θ is the parameter set of the three-dimensional geological model and evaluation index system, and η is the learning rate. Based on newly added drilling data D new The loss function.

Images