Detection method and system for inversion of carbonate reservoir by using resistivity anomaly

Abstract

The invention provides a carbonate reservoir detection method and system based on resistivity anomaly inversion, and relates to the technical field of oil-gas exploration, and the method comprises the steps: obtaining historical data; performing multi-scale analysis and classification processing according to the resistivity data to obtain a resistivity anomaly distribution diagram; carrying out modeling processing according to the geological parameters and the carbonate rock resistance test data to obtain a geological-electrical coupling relation model; a resistivity three-dimensional distribution model is constructed according to the resistivity anomaly distribution diagram and the geology-electrical coupling relation model; performing resistivity prediction processing based on adaptive mesh generation and parallel calculation to obtain a resistivity space distribution prediction result; and performing three-dimensional geological modeling processing according to the resistivity spatial distribution prediction result to obtain an oil and gas reservoir detection result of the target area. According to the method, through multi-scale resistivity anomaly analysis, resistivity data processing is carried out by adopting high-order derivative filtering and wavelet decomposition technologies, and the resistivity anomaly region can be efficiently extracted.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploration, and in particular to a method and system for detecting carbonate oil and gas reservoirs by inverting resistivity anomalies. Background Art

[0002] In the exploration of carbonate oil and gas reservoirs, traditional methods such as seismic exploration, drilling sampling and logging usually use physical detection methods to infer the existence and distribution of underground oil and gas reservoirs. However, these methods are limited in effectiveness when faced with the complex geological conditions of carbonate rocks. Carbonate rock layers usually have significant fractures and caves, and the lithology and porosity are unevenly distributed. It is difficult for traditional methods to accurately capture these characteristics, resulting in large errors in the distribution of oil and gas reservoirs and reserve assessment. In particular, seismic wave reflection and drilling data cannot effectively reflect the impact of complex structures, resulting in insufficient reliability of exploration results.

[0003] Although some studies have attempted to improve this situation through methods such as resistivity anomaly inversion, existing technologies still have many limitations when dealing with carbonate oil and gas reservoirs. In existing applications, resistivity inversion methods rely on simple resistivity data comparison and model fitting. This method cannot effectively solve the differences in electrical characteristics caused by the complex geological structure of carbonate rocks. For example, cracks and caves in carbonate rocks often lead to significant changes in resistivity, and the electrical characteristics of these structures are difficult to accurately capture with existing technologies. In addition, existing resistivity inversion methods lack refined spatial resolution, cannot accurately depict subtle changes in the reservoir, and cannot fully consider the complex dynamic relationship between resistivity and geological parameters, which affects the accuracy of reservoir boundary identification and the reliability of reserve assessment.

[0004] In view of the above problems, the present invention proposes a method and system for detecting carbonate oil and gas reservoirs by inverting resistivity anomalies. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for detecting carbonate oil and gas reservoirs by using resistivity anomaly inversion to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] In a first aspect, the present application provides a method for detecting carbonate oil and gas reservoirs by inverting resistivity anomalies, comprising:

[0007] Acquiring historical data, the historical data including geological parameters, resistivity data, carbonate rock resistance test data, and known oil and gas reservoir information of at least one region;

[0008] Performing multi-scale analysis and classification processing on the resistivity data, extracting resistivity anomaly features and performing spatial cluster analysis to obtain a resistivity anomaly distribution map;

[0009] Modeling is performed based on the geological parameters and the carbonate rock resistance test data, and a geological-electrical coupling relationship model is obtained by simulating the dynamic response characteristics of resistivity under different geological conditions;

[0010] According to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, by introducing regularized optimization and Bayesian inversion framework, and combining the known oil and gas reservoir information to constrain and calibrate the model, a three-dimensional resistivity distribution model is constructed;

[0011] Obtaining resistivity data of the target area, and inputting the resistivity data of the target area into the resistivity three-dimensional distribution model for calculation, and obtaining resistivity spatial distribution prediction results of the target area through resistivity prediction processing based on adaptive grid generation and parallel calculation;

[0012] The oil and gas reservoir boundary identification and three-dimensional geological modeling are performed according to the resistivity spatial distribution prediction results, and the oil and gas reservoir detection results in the target area are obtained by reconstructing the three-dimensional structure of the oil and gas reservoir in the target area and quantitatively evaluating the reserve distribution.

[0013] In a second aspect, the present application also provides a carbonate oil and gas reservoir detection system using resistivity anomaly inversion, comprising:

[0014] An acquisition module, used to acquire historical data, wherein the historical data includes geological parameters, resistivity data, carbonate rock resistance test data and known oil and gas reservoir information of at least one region;

[0015] A classification module is used to perform multi-scale analysis and classification processing according to the resistivity data, and obtain a resistivity anomaly distribution map by extracting resistivity anomaly features and performing spatial cluster analysis;

[0016] A modeling module, which performs modeling processing according to the geological parameters and the carbonate rock resistance test data, and obtains a geological-electrical coupling relationship model by simulating the dynamic response characteristics of resistivity under different geological conditions;

[0017] A construction module is used to construct a three-dimensional resistivity distribution model based on the resistivity anomaly distribution map and the geological-electrical coupling relationship model by introducing regularized optimization and Bayesian inversion framework and combining the known oil and gas reservoir information to constrain and calibrate the model;

[0018] A prediction module, which obtains resistivity data of a target area, and inputs the resistivity data of the target area into the resistivity three-dimensional distribution model for calculation, and obtains a resistivity spatial distribution prediction result of the target area through resistivity prediction processing based on adaptive grid generation and parallel calculation;

[0019] The output module is used to perform oil and gas reservoir boundary identification and three-dimensional geological modeling processing according to the resistivity spatial distribution prediction result, and obtain the oil and gas reservoir detection result of the target area by reconstructing the three-dimensional structure of the oil and gas reservoir in the target area and quantitatively evaluating the reserve distribution.

[0020] The beneficial effects of the present invention are:

[0021] The present invention uses multi-scale resistivity anomaly analysis to target the complex geological features of carbonate oil and gas reservoirs, adopts high-order derivative filtering and wavelet decomposition technology to process resistivity data, and can efficiently extract resistivity anomaly areas; through multi-scale analysis, microscopic features such as cracks and caves that affect resistivity changes are captured, and reservoirs and non-reservoir areas can be accurately distinguished, thereby accurately identifying the boundaries of oil and gas reservoirs, overcoming the problem that traditional resistivity analysis cannot effectively reflect complex geological features.

[0022] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 It is a schematic flow chart of a method for detecting carbonate oil and gas reservoirs by inverting resistivity anomaly according to an embodiment of the present invention;

[0025] Figure 2 It is a structural diagram of a carbonate oil and gas reservoir detection system using resistivity anomaly inversion described in an embodiment of the present invention.

[0026] Markings in the figure: 1. Acquisition module; 2. Classification module; 21. First decomposition unit; 22. First enhancement unit; 23. First classification unit; 24. First clustering unit; 3. Modeling module; 31. First modeling unit; 32. Second modeling unit; 33. Third modeling unit; 34. Fourth modeling unit; 4. Construction module; 41. First construction unit; 42. First optimization unit; 43. First inversion unit; 44. Second construction unit; 5. Prediction module; 51. First processing unit; 52. First segmentation unit; 53. First calculation unit; 54. Second optimization unit; 6. Output module; 61. First recognition unit; 62. First extraction unit; 63. Fifth modeling unit; 64. First simulation unit. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0029] Embodiment 1:

[0030] This embodiment provides a method for detecting carbonate oil and gas reservoirs by inverting resistivity anomalies.

[0031] See also Figure 1 , the figure shows that the method includes steps S100 to S600.

[0032] Step S100, acquiring historical data, where the historical data includes geological parameters, resistivity data, carbonate rock resistance test data, and known oil and gas reservoir information of at least one region;

[0033] Specifically, first, geological parameters are obtained through field exploration and laboratory testing, including lithology distribution, porosity, permeability and fracture characteristics. These parameters are usually obtained through seismic exploration, core analysis, porosity and permeability measurement, etc., to help understand the physical properties of the reservoir and its relationship with resistivity. Resistivity data is an important data for oil and gas reservoir detection, usually obtained through ground electrical exploration and resistivity logging, and used to analyze the relationship between electrical characteristics and reservoirs, especially the effects of structures such as fractures and caves on resistivity. Carbonate rock resistivity test data is obtained through laboratory testing to simulate the resistivity response under different porosities and permeabilities. Known oil and gas reservoir information comes from historical exploration data, providing data such as the location, scale and reserves of oil and gas reservoirs for correction and optimization of models.

[0034] Step S200, performing multi-scale analysis and classification processing according to the resistivity data, extracting resistivity anomaly features and performing spatial cluster analysis to obtain a resistivity anomaly distribution map;

[0035] It can be understood that the multi-scale analysis in this step can capture the details of the resistivity anomaly area at different depths or ranges by adjusting the scale level. For example, the electrical change at a large scale may represent the electrical trend of the entire reservoir, while at a small scale, it may reflect the influence of microstructures such as fractures and caves. With the help of multi-scale analysis, the electrical change characteristics of different levels in carbonate reservoirs can be effectively processed to reveal the diversity and complexity of the reservoir. Secondly, the classification processing identifies the abnormal areas that are potentially associated with oil and gas reservoirs by grouping the resistivity anomaly characteristics. Through the classification algorithm, the abnormal signals in the resistivity data can be accurately classified, thereby providing a clearer regional division for subsequent analysis. Especially in carbonate reservoirs, resistivity anomalies are often affected by complex geological structures, porosity, permeability and other factors. The classification processing divides the resistivity anomaly areas into different categories, enhances the spatial recognition of electrical characteristics, and ensures that the oil and gas reservoir-related areas can be accurately identified from the complex electrical signals, thereby improving the accuracy and reliability of the entire detection process.

[0036] Step S300, modeling is performed according to geological parameters and carbonate rock resistance test data, and a geological-electrical coupling relationship model is obtained by simulating the dynamic response characteristics of resistivity under different geological conditions;

[0037] It should be noted that this step converts the complex relationship between geological characteristics and electrical response into an operational mathematical model, providing an accurate basis for subsequent resistivity inversion and oil and gas reservoir detection. In carbonate oil and gas reservoirs, the influence of complex structures such as fractures and caves on resistivity makes the electrical characteristics highly nonlinear and spatially heterogeneous. Therefore, the static models in the prior art have been difficult to accurately reflect the dynamic changes of resistivity. By simulating the response of resistivity under different geological conditions, this step can reveal how these complex structures in the reservoir affect the electrical characteristics, thereby enhancing the coupling accuracy of electrical properties and geological data. This modeling process not only provides reliable data support for subsequent resistivity inversion, but also helps to deeply understand the laws of reservoir electrical changes, and further improve the accuracy of oil and gas reservoir boundary identification and reserve assessment.

[0038] Step S400: According to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, regularized optimization and Bayesian inversion framework are introduced, and the model is constrained and calibrated in combination with known oil and gas reservoir information to construct a three-dimensional resistivity distribution model;

[0039] First, this step can reveal the deep connection between electrical changes and underground geological structures, especially the influence of complex structures such as fractures and caves in carbonate reservoirs, by combining the resistivity anomaly distribution map with the geological-electrical coupling model. By introducing regularization optimization, data can be smoothed during the inversion process, overfitting can be reduced, and noise interference can be effectively eliminated to ensure that the generated resistivity distribution meets the actual geological conditions. Secondly, the Bayesian inversion framework further optimizes the model parameters by combining prior oil and gas reservoir information and using probabilistic methods to ensure that reliable resistivity distribution can still be generated in areas with high uncertainty. Known oil and gas reservoir information provides accurate constraints for inversion, making the model not only spatially consistent with the actual situation, but also more accurate in the boundaries and reserves distribution of oil and gas reservoirs. This step overcomes the challenges of complex structures in carbonate oil and gas reservoirs to resistivity inversion by optimizing the inversion process. The resulting three-dimensional resistivity distribution model provides scientific and accurate data support for oil and gas reservoir boundary identification, reserve assessment and further exploration decisions.

[0040] Step S500, obtaining resistivity data of the target area, and inputting the resistivity data of the target area into the resistivity three-dimensional distribution model for calculation, and obtaining the resistivity spatial distribution prediction result of the target area through resistivity prediction processing based on adaptive grid generation and parallel calculation;

[0041] It is understandable that adaptive meshing dynamically adjusts the resolution of the grid according to the resistivity gradient and spatial characteristics, ensuring that higher spatial resolution can be provided in areas where electrical changes are more drastic (such as areas where cracks and caves are located), while reducing the grid density in areas where electrical changes are relatively stable to optimize computational efficiency. This method not only improves the prediction accuracy, but also significantly reduces the amount of calculation. In addition, the introduction of parallel computing further improves the efficiency of the calculation, especially when processing large-scale three-dimensional data, it can give full play to the advantages of computing resources and shorten the calculation time. Through parallel computing technology, the computing tasks are distributed to multiple processing units, realizing the rapid processing of large-scale data sets.

[0042] Step S600: perform reservoir boundary identification and three-dimensional geological modeling processing based on the resistivity spatial distribution prediction result, reconstruct the three-dimensional structure of the reservoir in the target area and quantitatively evaluate the reserve distribution to obtain the reservoir detection result in the target area.

[0043] It should be noted that this step can accurately identify the boundaries of oil and gas reservoirs by analyzing the spatial distribution characteristics of resistivity, especially in carbonate reservoirs, where electrical characteristics are often affected by complex geological structures such as fractures and caves. Therefore, when three-dimensional modeling is performed in combination with resistivity data and geological parameters, the spatial structure and electrical response of the reservoir are accurately reproduced to reflect the actual geological conditions of the reservoir. On this basis, the accurate calculation of oil and gas reservoir reserves is ultimately achieved by quantitatively evaluating the porosity, permeability and other parameters of the reservoir. Overall, this step reconstructs the three-dimensional structure of the oil and gas reservoir by comprehensively analyzing resistivity data and geological characteristics, and provides accurate data support for reserve assessment, thereby improving the scientificity and reliability of oil and gas reservoir detection results.

[0044] Further, step S200 includes step S210 to step S240.

[0045] Step S210, performing multi-scale decomposition processing on the resistivity data, extracting resistivity high-frequency anomaly characteristics related to the carbonate reservoir by designing a wavelet basis function matching the cave-fracture scale, and obtaining a multi-scale resistivity characteristic map;

[0046] In this step, the resistivity data contains geological information of different scales, especially microstructures such as cracks and caves in carbonate reservoirs, which have different effects on resistivity at different scales. Therefore, in order to extract these key features, multi-scale analysis methods, especially wavelet transform, are needed. Wavelet transform can perform localized analysis of signals at different scales and effectively capture high-frequency anomaly features in resistivity data, which are related to structural features such as cracks and caves in the reservoir.

[0047] Specifically, in this embodiment, assuming that the resistivity data is R(x, t), where x is the spatial position and t is the time, multi-scale decomposition can be performed through wavelet transform to extract resistivity anomaly features. In order to match the scale of cracks and caves, a wavelet basis function ψ is designed. α (x, t), so that it can adapt to different scales and accurately capture the electrical response of cracks and caves. The basic formula of wavelet transform is:

[0048]

[0049] Among them, W ψ (α,τ) represents the coefficient of wavelet transform, which indicates the local characteristics of the signal at scale α and position τ; R(x,t) represents the resistivity data of the target area; x represents the spatial position; t represents the time; α represents the scale factor, which is used to control the width of the wavelet to adapt to the different scales of caves and cracks; τ represents the translation factor, which is used to represent the local position of the signal; is the wavelet basis function ψ α The complex conjugate form of (x, t) after scaling and position transformation.

[0050] In order to better adapt to the characteristics of fractures and caves in carbonate reservoirs, we design the wavelet basis function to be scale-adaptive:

[0051]

[0052] Through the above wavelet transform, the abnormal features in the resistivity data can be extracted. In order to better extract the reservoir characteristics, especially the high-frequency abnormal features brought by fractures and caves, the threshold method can be used to process the multi-scale decomposition results. The resistivity anomaly feature map can be expressed as:

[0053]

[0054] in, represents the multi-scale resistivity characteristic map; T represents the threshold; H(·) represents the Heaviside step function, which is used to set the threshold T and filter out abnormal resistivity signals.

[0055] Step S220, performing abnormal feature enhancement processing according to the multi-scale resistivity characteristic map and the porosity distribution data in the geological parameters, fusing the multi-scale abnormal features and suppressing non-reservoir interference signals through a weighted fusion algorithm based on porosity constraints, and obtaining an enhanced resistivity abnormal feature map;

[0056] After obtaining the multi-scale resistivity characteristic map, the next step is to enhance the abnormal features. The high-frequency features extracted by the multi-scale decomposition process contain the electrical differences between the reservoir and non-reservoir areas, but may also be affected by noise or non-reservoir interference. Therefore, porosity distribution data, as one of the geological parameters, plays an important role in feature enhancement. Porosity directly affects the electrical characteristics of the reservoir. The resistivity of the area with higher porosity in the reservoir is lower, so it can provide an effective constraint for the enhancement of abnormal features. Through the weighted fusion algorithm, the multi-scale resistivity characteristic map is combined with the porosity distribution data. In the fusion process, a higher weight is given to the reservoir area and a lower weight is given to the non-reservoir area, thereby suppressing the interference signal of the non-reservoir. This weighted fusion method based on porosity constraints can enhance the abnormal signal in the reservoir area while reducing the influence of noise and non-reservoir interference. Finally, the enhanced resistivity anomaly characteristic map can more clearly present the electrical characteristics of the reservoir and the resistivity anomaly area.

[0057] Step S230: According to the enhanced resistivity anomaly characteristic map, classification processing based on the heterogeneous statistical model is performed, by constructing a carbonate formation resistivity-porosity joint probability distribution function, defining anomaly thresholds and segmenting oil and gas reservoir-related abnormal areas, and obtaining classified resistivity anomaly areas;

[0058] It should be noted that in carbonate reservoirs, abnormal changes in resistivity are not only affected by structures such as caves and fractures, but are also closely related to geological parameters such as porosity and permeability. To this end, a heterogeneous statistical model is used to describe the joint probability distribution between resistivity and porosity. The heterogeneous statistical model can accurately capture the relationship between porosity and resistivity in carbonate reservoirs, and on this basis, a resistivity-porosity joint probability distribution function is constructed. This probability distribution function helps to define the threshold of resistivity anomaly, through which the resistivity anomaly area is distinguished from the non-abnormal area. In this way, the abnormal areas related to oil and gas reservoirs can be clearly segmented to form classified resistivity anomaly areas. These areas represent the location of potential oil and gas reservoirs, providing accurate regional division for subsequent oil and gas reservoir boundary identification and three-dimensional modeling. Specifically, the form of the joint probability distribution function is:

[0059]

[0060] Where P(R,φ) represents the resistivity-porosity joint probability distribution function; R represents resistivity; φ represents porosity; Z represents the normalization constant; μ R and μ φ denote the mean values ​​of resistivity and porosity, respectively; and represent the variance of resistivity and porosity respectively; f(R,φ) represents the joint distribution function of resistivity and porosity.

[0061] The model is a Gaussian joint probability distribution function, in which the changes in resistivity and porosity conform to the normal distribution, and can effectively describe the statistical relationship between resistivity and porosity in most geological environments.

[0062] Step S240: Perform fracture-cavern spatial correlation analysis based on the classified resistivity anomaly areas. By introducing the spatial autocorrelation of carbonate fracture trends, anisotropic density clustering algorithm is used to identify the clustering pattern of abnormal areas to obtain a resistivity anomaly distribution map.

[0063] Finally, after obtaining the classified resistivity anomaly areas, a spatial correlation analysis of fractures and caves is performed. Fractures and caves in carbonate reservoirs usually have obvious spatial autocorrelation. Therefore, it is effective to use anisotropic density clustering algorithm to identify spatial clustering patterns of abnormal areas. This algorithm can take into account the spatial distribution characteristics of fractures and caves, and identify potential clustering patterns in resistivity anomaly areas through clustering analysis. By introducing the spatial autocorrelation of carbonate fracture trends, the spatial distribution of abnormal areas can be further refined and the recognition accuracy of reservoir structure can be improved. Finally, a resistivity anomaly distribution map is generated through spatial clustering analysis, which accurately depicts the electrical characteristics of oil and gas reservoirs in the target area.

[0064] Further, step S300 includes step S310 to step S340.

[0065] Step S310: Perform pore structure modeling according to the porosity data and carbonate rock resistance test data in the geological parameters, construct a porosity-resistivity relationship model based on Archie's formula, and correct the model parameters in combination with the pore structure characteristics of carbonate rocks to obtain a pore structure resistivity model;

[0066] It is understandable that Archie's Law is a classic relationship model between porosity and resistivity, which is widely used in oil and gas exploration. However, in carbonate reservoirs, due to their special pore structures (such as fractures, caves, etc.), Archie's Law usually needs to be modified to better reflect the effects of these structures on resistivity. Therefore, the improved model based on Archie's Law needs to add additional parameters to consider the heterogeneity of carbonate reservoirs. The improved porosity-resistivity relationship model is expressed as:

[0067]

[0068] Where R represents resistivity; φ represents porosity; φ0 represents the base porosity, corresponding to the situation without cracks and caves; R shIt represents the resistivity value related to heterogeneous structures such as cracks and caves; a represents a constant; m is the exponent in Archie's formula, usually between 2 and 3; β is the correction factor; n represents the index of the influence of caves or cracks on resistivity, reflecting the sensitivity of resistivity to pore structure.

[0069] This formula corrects the influence of heterogeneous structures such as fractures and caves, making the model better adapted to the complex electrical characteristics in carbonate reservoirs.

[0070] Step S320: Perform fracture network modeling according to fracture density data and carbonate rock resistance test data in geological parameters, construct a fracture-resistivity relationship model based on equivalent medium theory, and optimize model parameters in combination with spatial distribution characteristics of fractures to obtain a fracture network resistivity model;

[0071] It should be noted that the equivalent medium theory is used to deal with multiphase medium systems (such as porous media, composite materials, etc.). Its basic idea is to treat complex multiphase medium systems (such as cracks, pores, etc.) as equivalent to a single medium, and the macroscopic electrical characteristics of the medium can represent the influence of all microstructures on the electrical response. When applied to carbonate reservoirs, complex geological structures such as cracks and caves can lead to significant changes in local resistivity.

[0072] In order to more accurately reflect the spatial distribution characteristics of cracks, the model needs to be further optimized to consider the connectivity and distribution pattern of cracks. The improved crack-resistivity relationship model can combine the density and distribution characteristics of cracks, such as using a network model or self-similar distribution to describe the effect of the spatial structure of cracks on resistivity. Assuming that the spatial distribution of cracks conforms to a certain statistical distribution, such as Poisson distribution or fractal distribution, we can modify the resistivity model of cracks to consider the contribution of the distribution and connectivity of cracks to resistivity.

[0073] Preferably, this embodiment uses a model based on fractal theory to describe the impact of the fracture network, assuming that the connectivity and complexity of the fractures increase with the increase of fracture density:

[0074]

[0075] Among them, R eff represents the resistivity of the fracture network; C f and C b They represent the correction coefficients of cracks and bedrock, respectively, reflecting the influence of crack distribution; γ is the fractal dimension, indicating the complexity of the crack network; φ f and φ b Represent the volume fractions of fractures and bedrock respectively; R f and R b represent the resistivity of fracture and bedrock, respectively.

[0076] Step S330: Perform rock physics comprehensive modeling processing according to the pore structure resistivity model and the fracture network resistivity model, and combine the resistivity responses of the pore structure and the fracture network by introducing a weighted fusion algorithm to obtain a rock physics comprehensive resistivity model;

[0077] It should be noted that this comprehensive modeling method can more accurately simulate the electrical characteristics of carbonate reservoirs while avoiding the complex structural effects that may be ignored by a single model. In particular, for carbonate reservoirs with complex fractures and caves, the use of a porosity resistivity model alone may underestimate or overestimate the contribution of fractures and caves to electrical changes, while a fracture network model may also ignore the effect of porosity.

[0078] Step S340: Perform dynamic response modeling based on the rock physics comprehensive resistivity model and the formation pressure and temperature data in the geological parameters. Use the finite element numerical simulation method to simulate the dynamic changes of resistivity under different geological conditions, and calibrate the model parameters in combination with the carbonate rock resistance test data to obtain a geological-electrical coupling relationship model.

[0079] It is understandable that in geological environments, structures such as porosity, permeability, and fractures will change dynamically under different formation pressure and temperature conditions. Resistivity will also change with the changes in these conditions, especially in carbonate reservoirs, where changes in temperature and pressure will directly affect the electrical response of pore structures and fractures. Therefore, it is crucial to establish a model that can simulate the dynamic changes of resistivity under different geological conditions. Finite element numerical simulation method is a numerical calculation method commonly used to deal with complex boundary conditions and nonlinear problems. In this step, the finite element method is used to simulate how the resistivity of carbonate reservoirs changes dynamically under different formation pressure and temperature conditions. The basic idea is to divide the reservoir into multiple small elements and calculate the resistivity response in each small area based on known electrical equations and boundary conditions.

[0080] Further, step S400 includes step S410 to step S440.

[0081] Step S410: According to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, a model construction process is performed, and an initial three-dimensional resistivity distribution model is generated by using a Kriging interpolation algorithm combined with geological structure constraints;

[0082] The Kriging interpolation algorithm assumes that the spatial distribution of resistivity data has a certain degree of autocorrelation, and can estimate the unknown location based on the spatial relationship of known data points. Mathematically, the goal of Kriging interpolation is to calculate the predicted value of each grid point by minimizing the error variance. The Kriging interpolation formula is usually expressed as:

[0083]

[0084] Where Z(x0) represents the resistivity value at the data point x0 to be estimated; Z(x i ) represents a known data point x i The resistivity value at the location; p represents the number of known data points; i represents the sequence number of the data point; λ i Indicates position x i Treat the size of the contribution of the estimated point x0.

[0085] By using the Kriging interpolation algorithm, combined with geological structural constraints and resistivity data, a more accurate three-dimensional resistivity distribution model can be generated. This method can effectively capture the spatial autocorrelation of electrical property changes in carbonate reservoirs, consider the influence of geological factors such as fractures and porosity on electrical characteristics, and thus improve the accuracy of the resistivity model.

[0086] Step S420, performing regularization optimization processing according to the initial three-dimensional resistivity distribution model, and obtaining a regularized resistivity model by introducing anisotropic regularization terms of fracture direction smoothness constraint and cave region gradient constraint;

[0087] Specifically, after the three-dimensional resistivity distribution model is generated, some electrical anomalies that do not conform to the actual geological conditions will appear. These anomalies may be caused by incomplete data, measurement errors, or insufficient consideration of geological characteristics during the initial modeling process. In this case, regularized optimization methods are used to smooth and correct these anomalies. Regularized optimization can limit the changes in the model in a specific area by introducing additional physical constraints to ensure that the generated resistivity distribution is consistent with both known data and geological and electrical characteristics. The regularization terms introduced include:

[0088] Smoothness constraint on fracture direction: In carbonate reservoirs, the directionality of fractures has a great influence on the resistivity distribution. Fractures usually have obvious directional characteristics, so the resistivity distribution should maintain a certain smoothness in the direction of the fracture to reflect the ductility of the fracture. To achieve this goal, we introduce a smoothness constraint on the fracture direction to constrain the change of the resistivity value in the direction of the fracture. This means that the change of resistivity along the fracture direction should be smooth and should not fluctuate too drastically.

[0089] Gradient constraint in cave area: In carbonate reservoirs, caves are usually large pore volumes with a wide range of influence, and the resistivity is usually low in the cave area. In order to ensure that the model can accurately reflect the electrical characteristics of the cave area, we need to introduce a gradient constraint in the cave area. The gradient constraint requires that the resistivity change should have a large gradient in the cave area, while the resistivity change in other areas is relatively gentle. This constraint can effectively capture the impact of caves on resistivity and ensure that it is reasonably expressed in the model.

[0090] Anisotropic regularization term: Due to the directional characteristics of fractures and caves in carbonate reservoirs, anisotropic regularization terms are introduced to ensure that the variation of resistivity in different directions conforms to the actual geological conditions. Anisotropic regularization allows different constraints to be imposed in different directions, so that the variation of resistivity distribution in space can reflect the geological structure and electrical response of the reservoir. For example, a stronger smoothing constraint can be imposed in the direction of the fracture strike, while a larger variation of resistivity can be allowed in the direction perpendicular to the fracture.

[0091] This step can effectively optimize the initial three-dimensional resistivity distribution model by introducing fracture direction smoothing constraints, cave area gradient constraints and anisotropic regularization terms, overcoming errors caused by data noise, measurement errors or inaccuracies in the initial modeling process. The optimized resistivity model can better reflect the influence of complex geological structures such as fractures and caves in the reservoir, and improve the accuracy of subsequent reservoir boundary identification, reserve assessment and three-dimensional modeling.

[0092] Step S430, performing Bayesian inversion processing according to the regularized resistivity model, defining a priori distribution by combining the Markov chain Monte Carlo sampling algorithm with known oil and gas reservoir information, solving the three-dimensional resistivity distribution parameters with the maximum posterior probability, and obtaining the Bayesian inversion result;

[0093] Bayesian inversion is a numerical method based on Bayesian theorem, which infers the posterior distribution of parameters by combining prior distribution and observed data. In oil and gas reservoir exploration, Bayesian inversion is used to update the resistivity model. Assuming that the three-dimensional distribution of resistivity is the parameter to be estimated, Bayesian inversion solves the most likely resistivity distribution by combining the regularized resistivity model with known oil and gas reservoir information. The application of Bayesian theorem in parameter estimation can be expressed as:

[0094]

[0095] Among them, P(θ|D) is the posterior distribution of parameter θ under given observation data D, which represents the probability of the parameter; P(D|θ) is the likelihood function, which represents the probability of observing data D under parameter θ; P(θ) is the prior distribution, which represents the cognition of parameter θ before there is any observation data; P(D) is the total probability of the observation data.

[0096] In the resistivity inversion problem, θ is the three-dimensional resistivity distribution that needs to be estimated, D is the observed resistivity data (including previously obtained resistivity data and known information about oil and gas reservoirs), and P(θ) is the prior knowledge of the resistivity distribution. Through Bayes' theorem, the observed data D and the prior distribution P(θ) can be combined to obtain the posterior distribution P(θ|D), thereby estimating the most likely resistivity distribution.

[0097] Since Bayesian inversion usually involves complex high-dimensional integrals, it is often not feasible to directly solve the analytical solution of the posterior distribution. To solve this problem, the Markov chain Monte Carlo sampling algorithm is used. The Markov chain Monte Carlo sampling algorithm constructs a Markov chain to generate a sample sequence close to the target posterior distribution. These samples can be used to approximate the properties of the posterior distribution. In this process, the key steps of the Markov chain Monte Carlo sampling algorithm include:

[0098] Define target distribution: The target distribution that needs to be sampled is the posterior distribution P(θ|D) in Bayesian inversion.

[0099] Constructing a Markov chain: By designing appropriate transition rules, a series of samples are generated in the parameter space that follow the target posterior distribution.

[0100] Iterative sampling: Generate a series of samples through iteration, gradually converge to the target distribution, and finally obtain the most likely three-dimensional resistivity distribution parameters through sampling.

[0101] In this step, the most likely three-dimensional resistivity distribution parameters are derived based on known reservoir information by using the Bayesian inversion method combined with the Markov chain Monte Carlo sampling algorithm. This method not only improves the accuracy of the resistivity model, but also effectively quantifies the uncertainty of the model.

[0102] Step S440: perform model constraint calibration processing according to the Bayesian inversion results, set the resistivity threshold range in combination with the location and reserve information of known oil and gas reservoirs, and use iterative least squares method to adjust the model parameters to construct a three-dimensional resistivity distribution model.

[0103] It should be noted that the Bayesian inversion results provide a preliminary estimate of the three-dimensional resistivity distribution, but these estimates may still have certain uncertainties or errors. Therefore, the purpose of model constraint calibration is to further optimize and adjust the model by combining the location and reserve information of known oil and gas reservoirs on the basis of Bayesian inversion. Specifically, firstly, by combining the location and reserve information of known oil and gas reservoirs, the threshold range of resistivity is set to ensure that the electrical characteristics of the resistivity model in the oil and gas reservoir area are consistent with the actual situation. Then, the iterative least squares method is used to adjust the model parameters and gradually optimize the resistivity model to make it more consistent with the geological data and electrical response. This optimization process continuously adjusts the parameters to reduce the error between the model prediction value and the observed data, thereby ensuring that the final three-dimensional resistivity distribution model can accurately reflect the geological characteristics and electrical properties in the carbonate reservoir, and provide more reliable support for the subsequent oil and gas reservoir boundary identification and reserve assessment.

[0104] Further, step S500 includes step S510 to step S540.

[0105] Step S510: Perform model parameter initialization processing according to the three-dimensional resistivity distribution model and the resistivity data of the target area, and obtain a dynamic parameter initialization model based on the dynamic response parameters of the geological-electrical coupling relationship model, the dynamic response parameters including formation pressure and temperature gradient parameters, which are used to adjust the initial resistivity value;

[0106] Specifically, it is first necessary to use the three-dimensional resistivity distribution model as a basis and perform initialization adjustments in combination with the actual resistivity data of the target area. Through the geological-electrical coupling relationship model, the pressure and temperature gradient of the formation have a direct impact on the resistivity, so these dynamic response parameters need to be incorporated into the model. Changes in formation pressure may cause changes in porosity, thereby affecting resistivity; changes in temperature gradients will affect the conductivity of the formation. By introducing these dynamic response parameters, the initial value of the resistivity model can be adjusted more accurately so that the model can adapt to the actual geological environment. This step ensures that the resistivity model can reflect the actual geological conditions, and provides reliable initial input for subsequent model optimization and inversion, helping to improve the accuracy of the final resistivity distribution.

[0107] Step S520, initializing the model according to the dynamic parameters, performing adaptive meshing processing, and generating a high-resolution mesh model matching the carbonate cave-crack structure by combining a quadtree algorithm with a preset resistivity gradient threshold;

[0108] It can be understood that adaptive meshing is a technology that dynamically adjusts the mesh resolution according to the characteristics and requirements of different regions in the model. For carbonate reservoirs, especially in areas with complex structures such as caves and fractures, the changes in resistivity are often local and significant. Therefore, the meshing needs to be refined according to these changes to ensure that sufficient resolution is provided in key areas (such as fractures, caves, etc.) to accurately simulate the changes in resistivity. The quadtree algorithm is an adaptive meshing algorithm commonly used for spatial segmentation, which can automatically adjust the size of the grid according to the resistivity gradient of the region. Through the quadtree algorithm, the entire target area can be divided into multiple four sub-areas, and further subdivision can be determined based on the resistivity changes in each area. Specifically, areas with large resistivity gradients (such as fractures and caves) will be subdivided into smaller grid units, while areas with small resistivity changes will maintain larger grid units. On this basis, the resistivity gradient threshold is used as a criterion for judging mesh refinement, which can be set according to actual needs. When the resistivity change exceeds the preset threshold, the grid will be further refined in that area. This method can ensure that higher resolution grids are used in areas with drastic electrical property changes (such as caves and fractures), while lower resolution grids are used in areas with gentle electrical property changes, thereby improving computational efficiency while ensuring model accuracy. Ultimately, the generated high-resolution grid model can effectively reflect the influence of complex geological structures such as caves and fractures in carbonate reservoirs, providing accurate spatial data support for subsequent resistivity prediction, oil and gas reservoir boundary identification, and reserve assessment.

[0109] Step S530, performing parallel calculation and solution processing according to the high-resolution grid model to obtain a preliminary resistivity prediction result;

[0110] It is understandable that the high-resolution grid model is a detailed division of resistivity data, which usually involves a large number of computing units. As the complexity of the resistivity model increases, the amount of calculation will also increase rapidly. Especially for carbonate reservoirs containing complex geological features such as caves and cracks, the amount of data and processing time required for calculation may be very large. Therefore, it is often difficult to meet the requirements of efficiency and accuracy using traditional single calculation methods. In order to improve computing efficiency and speed up the generation of results, parallel computing technology is introduced in this step. Parallel computing allows tasks to be assigned to multiple computing units (such as multiple processors or computing nodes) and executed simultaneously, thereby greatly reducing computing time. Parallel computing can be implemented in many ways, such as multi-core processing, distributed computing, etc.

[0111] Step S540: Optimize the preliminary resistivity prediction results, use the radial basis function interpolation algorithm to fill in the low-confidence data in the cave area, and remove noise in combination with the resistivity anomaly distribution map to obtain the resistivity spatial distribution prediction results of the target area.

[0112] In the preliminary resistivity prediction results, due to data noise, measurement errors or reservoir complexity, some areas (such as caves, fractures, etc.) may have low confidence. These low-confidence areas may lead to inaccuracy in the resistivity model, so optimization is required to improve the accuracy of the model. By further optimizing the resistivity model, the electrical characteristics of complex geological structures (such as caves) can be better simulated, and the reliability of resistivity distribution in the entire target area can be ensured. In this step, radial basis function interpolation is used to fill the low-confidence data in the cave area. The core idea of ​​radial basis function interpolation is to predict the resistivity value of unknown points by the data value of known points, especially for the cave area, because its resistivity is usually low and varies greatly, so radial basis function interpolation can accurately fill in these areas. At the same time, in the resistivity model, due to measurement errors and data anomalies, there may be some noise data, which will have an adverse effect on the final resistivity distribution prediction results. In order to improve the accuracy of the model, it is necessary to combine the resistivity anomaly distribution map to eliminate abnormal data. The resistivity anomaly distribution map reflects the significant areas of resistivity change, especially the abnormal areas related to reservoirs or oil and gas reservoirs. By setting a suitable threshold, the noise data can be identified based on the resistivity anomaly distribution map and the interference of these areas can be eliminated.

[0113] Further, step S600 includes step S610 to step S640.

[0114] Step S610, performing oil and gas reservoir boundary identification processing according to the resistivity spatial distribution prediction result, performing high-order derivative filtering on the resistivity spatial distribution, using the Laplace operator to enhance the boundary area where the resistivity changes sharply, and obtaining the oil and gas reservoir boundary identification result;

[0115] It is understandable that high-order derivative filtering is a mathematical method commonly used to extract parts of signals that change dramatically. In the results of resistivity spatial distribution prediction, the resistivity changes in boundary areas are usually more dramatic, so these changes can be enhanced by calculating the high-order derivatives of the resistivity distribution. In order to be able to identify areas with large changes related to the reservoir, it is not only necessary to rely on traditional high-order derivatives, but also necessary to combine the fracture direction of the reservoir to enhance the resistivity field in a targeted manner. By defining a formula for anisotropic high-order derivative filtering, it is used to enhance changes in the fracture direction, and the smoothness in the non-fracture direction is controlled by the weight coefficient. The specific formula is as follows:

[0116]

[0117] Where R(r) represents the resistivity at the spatial point r = (r x , r y , r z ) at the value; and ω represent the weight coefficients along the crack direction and perpendicular to the crack direction respectively; p represents the order of the derivative.

[0118] The role of the Laplace operator in the electrical field is to reflect the comprehensive degree of electrical changes at each point in space. For carbonate reservoirs, the influence of cracks and caves needs to be fully reflected in the Laplace operator. In order to enhance the influence of cracks and caves, the Laplace operator is improved to adapt to environments with strong geological heterogeneity. The improved Laplace operator formula is:

[0119]

[0120] Among them, k(r) represents the spatial variation function of electrical conductivity, reflecting the influence of geological features such as fractures and caves on electrical distribution; is the gradient symbol, indicating the spatial variation of resistivity; ΔR(r) represents the resistivity variation calculated by the improved Laplace operator; Represents the resistivity field R(r) at the spatial position R=(r x , r y , r z ) at the gradient;

[0121] Combined with the above-mentioned high-order derivative filtering and the enhancement effect of the Laplace operator, the change of resistivity distribution will be significantly improved in areas with cracks and caves. Next, by setting a resistivity change threshold ΔR thresh , the boundary area of ​​the oil and gas reservoir can be determined based on the drastic change in resistivity. The specific boundary identification method can be expressed as follows:

[0122]

[0123] in, Indicates the boundary area of ​​the oil and gas reservoir; ΔR thresh It is a set change threshold used to distinguish the electrical differences between oil and gas reservoirs and surrounding rock formations.

[0124] By combining high-order derivative filtering and the Laplace operator, this step can significantly improve the accuracy of identifying reservoir boundaries. High-order derivative filtering helps to highlight areas with large resistivity changes, while the Laplace operator further enhances the identification of these areas, thereby ensuring accurate delineation of reservoir boundaries.

[0125] Step S620: extract fracture and cave features based on the reservoir boundary identification result, use the topology optimization method to spatially reconstruct the resistivity boundary area and adjust the spatial distribution of fractures and caves to obtain fracture-cavity feature information;

[0126] It is understandable that in carbonate reservoirs, the distribution of fractures and caves has a significant impact on resistivity variation, so accurately identifying and reconstructing these geological features is crucial for the detection of oil and gas reservoirs. First, based on the results of reservoir boundary identification, the areas with drastic resistivity changes are determined, which are usually closely related to the presence of fractures and caves. Next, the topology optimization method is applied to spatially reconstruct these boundary areas. The topology optimization method adjusts the distribution of fractures and caves in space by combining the geometric characteristics of fractures and caves with resistivity boundary data to ensure that the structure of fractures and caves conforms to the actual geological conditions. Specifically, topology optimization optimizes the resistivity distribution by minimizing the discontinuity of fracture and cave areas in the resistivity model, so that these geological features are more accurately reflected in the model. This process not only improves the sensitivity of the model to complex geological structures, but also effectively reduces the errors in the calculation process, providing accurate fracture-cavity characteristic information for subsequent reservoir identification and reserve assessment.

[0127] Step S630, performing 3D geological modeling processing according to the fracture-cavity characteristic information, converting the resistivity and fracture-cavity data into point cloud data, and using a topological reconstruction algorithm to convert the point cloud data into a 3D grid model, thereby obtaining a 3D structural model of the oil and gas reservoir;

[0128] It should be noted that, first, the spatial features of fractures and caves are extracted and converted into point cloud data through the previous processing steps. Each point represents a key position in the resistivity data and marks the characteristics of fractures and caves. These point cloud data provide the basis for subsequent 3D modeling, ensuring sufficient spatial resolution in complex areas of the reservoir (such as fractures and caves). Then, using the topological reconstruction algorithm, these point cloud data are converted into a 3D mesh model after algorithm processing. The role of the topological reconstruction algorithm here is to construct a reasonable 3D mesh structure from discrete point cloud data through spatial interpolation and boundary condition constraints, ensuring the coherence and accuracy of the mesh, especially in the fracture and cave areas. Finally, the obtained 3D structural model of the oil and gas reservoir provides accurate 3D geological information for subsequent oil and gas reservoir boundary identification, reserve assessment and exploration activities, and can reflect the real structure and electrical characteristics of the reservoir.

[0129] Step S640, perform oil and gas reserve assessment processing based on the three-dimensional structural model of the oil and gas reservoir, generate multiple possible reserve distribution scenarios by randomly sampling the resistivity, porosity and permeability data in the three-dimensional structure, simulate the oil and gas reservoir reserves under each scenario, and integrate to obtain the oil and gas reservoir detection results.

[0130] In this process, first, the resistivity, porosity and permeability data obtained from the three-dimensional structural model of the oil and gas reservoir are statistically analyzed, and these parameters are sampled by random sampling. Since the reserves of oil and gas reservoirs are affected by multiple uncertain factors (such as porosity, permeability and fracture distribution), random sampling can simulate different reservoir conditions and generate multiple reserve distribution scenarios, thereby taking into account different geological changes and uncertainties. Next, using these sampling scenarios, the reserves of the oil and gas reservoirs in each scenario are simulated and evaluated. The reserve calculation model of the oil and gas reservoir is used to estimate the reserve distribution in each scenario, and the results of each scenario are combined to obtain the final reserve estimate. This process not only provides different estimates of oil and gas reservoir reserves, but also quantifies the uncertainty of reserve assessment, providing a more reliable decision-making basis for the planning and development of oil and gas resources.

[0131] Embodiment 2:

[0132] like Figure 2 As shown, this embodiment provides a carbonate oil and gas reservoir detection system using resistivity anomaly inversion, the system comprising:

[0133] Acquisition module 1, used to acquire historical data, the historical data including geological parameters, resistivity data, carbonate rock resistance test data and known oil and gas reservoir information of at least one region;

[0134] Classification module 2 is used to perform multi-scale analysis and classification processing based on resistivity data, extract resistivity anomaly features and perform spatial cluster analysis to obtain resistivity anomaly distribution map;

[0135] Modeling module 3, which performs modeling processing based on geological parameters and carbonate rock resistance test data, and obtains a geological-electrical coupling relationship model by simulating the dynamic response characteristics of resistivity under different geological conditions;

[0136] Construction module 4 is used to construct a three-dimensional resistivity distribution model based on the resistivity anomaly distribution map and the geological-electrical coupling relationship model by introducing regularization optimization and Bayesian inversion framework and combining known oil and gas reservoir information to constrain and calibrate the model;

[0137] Prediction module 5, obtains resistivity data of the target area, and inputs the resistivity data of the target area into the resistivity three-dimensional distribution model for calculation, and obtains the resistivity spatial distribution prediction result of the target area through resistivity prediction processing based on adaptive grid generation and parallel calculation;

[0138] Output module 6 is used to identify the boundaries of oil and gas reservoirs and perform three-dimensional geological modeling according to the predicted results of resistivity spatial distribution, and obtain the detection results of oil and gas reservoirs in the target area by reconstructing the three-dimensional structure of the oil and gas reservoirs in the target area and quantitatively evaluating the reserve distribution.

[0139] In a specific implementation disclosed in the present application, the classification module 2 includes:

[0140] The first decomposition unit 21 is used to perform multi-scale decomposition processing according to the resistivity data, extract the resistivity high-frequency anomaly characteristics related to the carbonate reservoir by designing a wavelet basis function matching the cave-fracture scale, and obtain a multi-scale resistivity characteristic map;

[0141] The first enhancement unit 22 is used to perform abnormal feature enhancement processing according to the multi-scale resistivity characteristic map and the porosity distribution data in the geological parameters, and fuse the multi-scale abnormal features and suppress non-reservoir interference signals through a weighted fusion algorithm based on porosity constraints to obtain an enhanced resistivity abnormal feature map;

[0142] The first classification unit 23 is used to perform classification processing based on the heterogeneous statistical model according to the enhanced resistivity anomaly characteristic map, and to define the anomaly threshold and segment the oil and gas reservoir related anomaly area by constructing the carbonate formation resistivity-porosity joint probability distribution function, so as to obtain the classified resistivity anomaly area;

[0143] The first clustering unit 24 is used to perform fracture-cavern spatial correlation analysis based on the classified resistivity anomaly areas, introduce the spatial autocorrelation of carbonate rock fracture trends, and use anisotropic density clustering algorithm to identify the clustering pattern of abnormal areas to obtain a resistivity anomaly distribution map.

[0144] In a specific implementation disclosed in the present application, the modeling module 3 includes:

[0145] The first modeling unit 31 is used to perform pore structure modeling processing according to the porosity data and carbonate rock resistance test data in the geological parameters, and obtain a pore structure resistivity model by constructing a porosity-resistivity relationship model based on Archie's formula and correcting model parameters in combination with the pore structure characteristics of carbonate rocks;

[0146] The second modeling unit 32 is used to perform fracture network modeling processing according to fracture density data and carbonate rock resistance test data in geological parameters, and obtain a fracture network resistivity model by constructing a fracture-resistivity relationship model based on equivalent medium theory and optimizing model parameters in combination with spatial distribution characteristics of fractures;

[0147] The third modeling unit 33 is used to perform rock physics comprehensive modeling processing according to the pore structure resistivity model and the fracture network resistivity model, and to obtain a rock physics comprehensive resistivity model by combining the resistivity responses of the pore structure and the fracture network by introducing a weighted fusion algorithm;

[0148] The fourth modeling unit 34 is used to perform dynamic response modeling processing based on the rock physics comprehensive resistivity model and the formation pressure and temperature data in the geological parameters, simulate the dynamic changes of resistivity under different geological conditions through the finite element numerical simulation method, and calibrate the model parameters in combination with the carbonate rock resistance test data to obtain the geological-electrical coupling relationship model.

[0149] In a specific implementation disclosed in the present application, the building block 4 includes:

[0150] The first construction unit 41 is used to perform model construction processing according to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, and generate an initial three-dimensional resistivity distribution model by using the Kriging interpolation algorithm combined with geological structure constraints;

[0151] The first optimization unit 42 is used to perform regularization optimization processing according to the initial three-dimensional resistivity distribution model, and obtain a regularized resistivity model by introducing anisotropic regularization terms of fracture direction smoothness constraint and cave area gradient constraint;

[0152] The first inversion unit 43 is used to perform Bayesian inversion processing according to the regularized resistivity model, define the prior distribution by combining the known oil and gas reservoir information with the Markov chain Monte Carlo sampling algorithm, solve the three-dimensional resistivity distribution parameter with the maximum posterior probability, and obtain the Bayesian inversion result;

[0153] The second construction unit 44 is used to perform model constraint calibration processing according to the Bayesian inversion result, set the resistivity threshold range in combination with the location and reserve information of the known oil and gas reservoirs, and use the iterative least square method to adjust the model parameters to construct a three-dimensional resistivity distribution model.

[0154] In a specific implementation disclosed in the present application, the prediction module 5 includes:

[0155] The first processing unit 51 is used to perform model parameter initialization processing according to the three-dimensional resistivity distribution model and the resistivity data of the target area, and to adjust the initial value of the resistivity based on the dynamic response parameters of the geological-electrical coupling relationship model, the dynamic response parameters including the formation pressure and temperature gradient parameters, to obtain the dynamic parameter initialization model;

[0156] The first partitioning unit 52 is used to initialize the model according to the dynamic parameters, perform adaptive mesh partitioning, and generate a high-resolution mesh model matching the carbonate cave-crack structure by combining a quadtree algorithm with a preset resistivity gradient threshold;

[0157] The first calculation unit 53 is used to perform parallel calculation and solution processing according to the high-resolution grid model to obtain a preliminary resistivity prediction result;

[0158] The second optimization unit 54 is used to perform optimization processing based on the preliminary resistivity prediction result, fill the low confidence data of the cave area with the radial basis function interpolation algorithm, and remove noise in combination with the resistivity anomaly distribution map to obtain the resistivity spatial distribution prediction result of the target area.

[0159] In a specific implementation disclosed in the present application, the output module 6 includes:

[0160] The first identification unit 61 is used to perform oil and gas reservoir boundary identification processing according to the resistivity spatial distribution prediction result, and obtain the oil and gas reservoir boundary identification result by performing high-order derivative filtering on the resistivity spatial distribution and using the Laplace operator to enhance the boundary area where the resistivity changes sharply;

[0161] The first extraction unit 62 is used to extract fracture and cave features according to the reservoir boundary identification result, use the topology optimization method to spatially reconstruct the resistivity boundary area and adjust the spatial distribution of fractures and caves to obtain fracture-cavity feature information;

[0162] The fifth modeling unit 63 is used to perform three-dimensional geological modeling according to the fracture-cavity characteristic information, by converting the resistivity and fracture-cavity data into point cloud data, and using a topological reconstruction algorithm to convert the point cloud data into a three-dimensional grid model, thereby obtaining a three-dimensional structural model of the oil and gas reservoir;

[0163] The first simulation unit 64 is used to perform oil and gas reserve assessment processing based on the three-dimensional structural model of the oil and gas reservoir. By randomly sampling the resistivity, porosity and permeability data in the three-dimensional structure, a plurality of possible reserve distribution scenarios are generated, and the oil and gas reservoir reserves under each scenario are simulated to obtain the oil and gas reservoir detection results through integration.

[0164] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for detecting carbonate oil and gas reservoirs by inversion using resistivity anomaly, characterized in that: include: Acquiring historical data, the historical data including geological parameters, resistivity data, carbonate rock resistance test data, and known oil and gas reservoir information of at least one region; Performing multi-scale analysis and classification processing on the resistivity data, extracting resistivity anomaly features and performing spatial cluster analysis to obtain a resistivity anomaly distribution map; Modeling is performed based on the geological parameters and the carbonate rock resistance test data, and a geological-electrical coupling relationship model is obtained by simulating the dynamic response characteristics of resistivity under different geological conditions; According to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, by introducing regularized optimization and Bayesian inversion framework, and combining the known oil and gas reservoir information to constrain and calibrate the model, a three-dimensional resistivity distribution model is constructed; Obtaining resistivity data of the target area, and inputting the resistivity data of the target area into the resistivity three-dimensional distribution model for calculation, and obtaining resistivity spatial distribution prediction results of the target area through resistivity prediction processing based on adaptive grid generation and parallel calculation; The oil and gas reservoir boundary identification and three-dimensional geological modeling are performed according to the resistivity spatial distribution prediction results, and the oil and gas reservoir detection results in the target area are obtained by reconstructing the three-dimensional structure of the oil and gas reservoir in the target area and quantitatively evaluating the reserve distribution.

2. The method for detecting carbonate oil and gas reservoirs by inversion using resistivity anomaly according to claim 1, characterized in that: According to the resistivity data, multi-scale analysis and classification processing are performed, and resistivity anomaly characteristics are extracted and spatial cluster analysis is performed to obtain a resistivity anomaly distribution map, including: Performing multi-scale decomposition processing on the resistivity data, extracting high-frequency resistivity anomaly characteristics related to carbonate reservoirs by designing a wavelet basis function matching the cave-fracture scale, and obtaining a multi-scale resistivity characteristic map; According to the multi-scale resistivity characteristic map and the porosity distribution data in the geological parameters, abnormal feature enhancement processing is performed, and the multi-scale abnormal features are fused and non-reservoir interference signals are suppressed through a weighted fusion algorithm based on porosity constraints to obtain an enhanced resistivity abnormal feature map; According to the enhanced resistivity anomaly characteristic map, a classification process based on a heterogeneous statistical model is performed, and by constructing a carbonate formation resistivity-porosity joint probability distribution function, an anomaly threshold is defined and oil and gas reservoir-related anomaly areas are segmented to obtain a classified resistivity anomaly area; According to the classified resistivity anomaly areas, fracture-cavern spatial correlation analysis is performed, and by introducing the spatial autocorrelation of carbonate rock fracture trends, anisotropic density clustering algorithm is used to identify the clustering pattern of abnormal areas to obtain a resistivity anomaly distribution map.

3. The method for detecting carbonate oil and gas reservoirs by inversion using resistivity anomaly according to claim 1, characterized in that: Modeling is performed based on the geological parameters and the carbonate rock resistance test data, and a geological-electrical coupling relationship model is obtained by simulating the dynamic response characteristics of resistivity under different geological conditions, including: According to the porosity data in the geological parameters and the carbonate rock resistance test data, pore structure modeling is performed, by constructing a porosity-resistivity relationship model based on Archie's formula and correcting model parameters in combination with the pore structure characteristics of carbonate rocks, thereby obtaining a pore structure resistivity model; According to the fracture density data in the geological parameters and the carbonate rock resistance test data, fracture network modeling is performed, by constructing a fracture-resistivity relationship model based on equivalent medium theory and optimizing model parameters in combination with the spatial distribution characteristics of fractures, thereby obtaining a fracture network resistivity model; According to the pore structure resistivity model and the fracture network resistivity model, a rock physics comprehensive modeling process is performed, and the resistivity responses of the pore structure and the fracture network are combined by introducing a weighted fusion algorithm to obtain a rock physics comprehensive resistivity model; Dynamic response modeling is performed based on the rock physics comprehensive resistivity model and the formation pressure and temperature data in the geological parameters. The dynamic changes of resistivity under different geological conditions are simulated by the finite element numerical simulation method, and the model parameters are calibrated in combination with the carbonate rock resistance test data to obtain a geological-electrical coupling relationship model.

4. The method for detecting carbonate oil and gas reservoirs by inversion using resistivity anomaly according to claim 1, characterized in that: According to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, by introducing regularized optimization and Bayesian inversion framework, and combining the known oil and gas reservoir information to constrain and calibrate the model, a three-dimensional resistivity distribution model is constructed, including: According to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, a model construction process is performed, and an initial three-dimensional resistivity distribution model is generated by using a Kriging interpolation algorithm combined with geological structure constraints; Performing regularization optimization processing according to the initial three-dimensional resistivity distribution model, and obtaining a regularized resistivity model by introducing anisotropic regularization terms of fracture direction smoothness constraint and cave region gradient constraint; According to the regularized resistivity model, Bayesian inversion processing is performed, a priori distribution is defined by combining the known oil and gas reservoir information with a Markov chain Monte Carlo sampling algorithm, and three-dimensional resistivity distribution parameters with the maximum posterior probability are solved to obtain a Bayesian inversion result; Model constraint calibration is performed according to the Bayesian inversion results, a resistivity threshold range is set in combination with the location and reserve information of the known oil and gas reservoirs, and the model parameters are adjusted using the iterative least squares method to construct a three-dimensional resistivity distribution model.

5. The method for detecting carbonate oil and gas reservoirs by inversion using resistivity anomaly according to claim 1, characterized in that: Obtaining the resistivity data of the target area, and inputting the resistivity data of the target area into the resistivity three-dimensional distribution model for calculation, and obtaining the resistivity spatial distribution prediction result of the target area through resistivity prediction processing based on adaptive mesh generation and parallel calculation, including: Performing model parameter initialization processing according to the three-dimensional resistivity distribution model and the resistivity data of the target area, and obtaining a dynamic parameter initialization model based on the dynamic response parameters of the geological-electrical coupling relationship model, wherein the dynamic response parameters include formation pressure and temperature gradient parameters, and are used to adjust the initial value of the resistivity; Initializing the model according to the dynamic parameters, performing adaptive grid generation processing, and generating a high-resolution grid model matching the carbonate cave-crack structure by combining a quadtree algorithm with a preset resistivity gradient threshold; Performing parallel computing and solving processing according to the high-resolution grid model to obtain a preliminary resistivity prediction result; The preliminary resistivity prediction results are optimized, the radial basis function interpolation algorithm is used to fill in the low-confidence data of the cave area, and the resistivity anomaly distribution map is combined to eliminate noise, so as to obtain the resistivity spatial distribution prediction results of the target area.

6. A carbonate oil and gas reservoir detection system using resistivity anomaly inversion, characterized in that: include: An acquisition module, used to acquire historical data, wherein the historical data includes geological parameters, resistivity data, carbonate rock resistance test data and known oil and gas reservoir information of at least one region; A classification module is used to perform multi-scale analysis and classification processing according to the resistivity data, and obtain a resistivity anomaly distribution map by extracting resistivity anomaly features and performing spatial cluster analysis; A modeling module, which performs modeling processing according to the geological parameters and the carbonate rock resistance test data, and obtains a geological-electrical coupling relationship model by simulating the dynamic response characteristics of resistivity under different geological conditions; A construction module is used to construct a three-dimensional resistivity distribution model based on the resistivity anomaly distribution map and the geological-electrical coupling relationship model by introducing regularized optimization and Bayesian inversion framework and combining the known oil and gas reservoir information to constrain and calibrate the model; A prediction module, which obtains resistivity data of a target area, and inputs the resistivity data of the target area into the resistivity three-dimensional distribution model for calculation, and obtains a resistivity spatial distribution prediction result of the target area through resistivity prediction processing based on adaptive grid generation and parallel calculation; The output module is used to perform oil and gas reservoir boundary identification and three-dimensional geological modeling processing according to the resistivity spatial distribution prediction result, and obtain the oil and gas reservoir detection result of the target area by reconstructing the three-dimensional structure of the oil and gas reservoir in the target area and quantitatively evaluating the reserve distribution.

7. The carbonate oil and gas reservoir detection system using resistivity anomaly inversion according to claim 6 is characterized in that: The classification module comprises: The first decomposition unit is used to perform multi-scale decomposition processing according to the resistivity data, extract the resistivity high-frequency anomaly characteristics related to the carbonate reservoir by designing a wavelet basis function matching the cave-fracture scale, and obtain a multi-scale resistivity characteristic map; A first enhancement unit is used to perform abnormal feature enhancement processing according to the multi-scale resistivity characteristic map and the porosity distribution data in the geological parameters, and to fuse the multi-scale abnormal features and suppress non-reservoir interference signals through a weighted fusion algorithm based on porosity constraints to obtain an enhanced resistivity abnormal feature map; A first classification unit is used to perform classification processing based on a heterogeneous statistical model according to the enhanced resistivity anomaly characteristic map, and to define anomaly thresholds and segment oil and gas reservoir-related abnormal areas by constructing a carbonate formation resistivity-porosity joint probability distribution function to obtain classified resistivity anomaly areas; The first clustering unit is used to perform fracture-cavern spatial correlation analysis according to the classified resistivity anomaly areas, introduce the spatial autocorrelation of carbonate rock fracture trends, adopt anisotropic density clustering algorithm to identify the clustering pattern of abnormal areas, and obtain a resistivity anomaly distribution map.

8. The carbonate oil and gas reservoir detection system using resistivity anomaly inversion according to claim 6 is characterized in that: The modeling module includes: A first modeling unit is used to perform pore structure modeling processing according to the porosity data in the geological parameters and the carbonate rock resistance test data, and obtain a pore structure resistivity model by constructing a porosity-resistivity relationship model based on Archie's formula and correcting model parameters in combination with the pore structure characteristics of the carbonate rock; The second modeling unit is used to perform fracture network modeling processing according to the fracture density data in the geological parameters and the carbonate rock resistance test data, and obtain a fracture network resistivity model by constructing a fracture-resistivity relationship model based on the equivalent medium theory and optimizing the model parameters in combination with the spatial distribution characteristics of the fractures; A third modeling unit is used to perform rock physics comprehensive modeling processing according to the pore structure resistivity model and the fracture network resistivity model, and to obtain a rock physics comprehensive resistivity model by combining the resistivity responses of the pore structure and the fracture network by introducing a weighted fusion algorithm; The fourth modeling unit is used to perform dynamic response modeling processing based on the rock physical comprehensive resistivity model and the formation pressure and temperature data in the geological parameters, simulate the dynamic changes of resistivity under different geological conditions through the finite element numerical simulation method, and calibrate the model parameters in combination with the carbonate rock resistance test data to obtain the geological-electrical coupling relationship model.

9. The carbonate oil and gas reservoir detection system using resistivity anomaly inversion according to claim 6, characterized in that: The building blocks include: A first construction unit is used to perform model construction processing according to the resistivity anomaly distribution map and the geological-electrical coupling relationship model, and generate an initial three-dimensional resistivity distribution model by using a Kriging interpolation algorithm combined with geological structure constraints; A first optimization unit is used to perform regularization optimization processing according to the initial three-dimensional resistivity distribution model, and obtain a regularized resistivity model by introducing anisotropic regularization terms of fracture direction smoothness constraint and cave region gradient constraint; A first inversion unit is used to perform Bayesian inversion processing according to the regularized resistivity model, define a priori distribution by combining the known oil and gas reservoir information with a Markov chain Monte Carlo sampling algorithm, solve the three-dimensional resistivity distribution parameter with the maximum posterior probability, and obtain a Bayesian inversion result; The second construction unit is used to perform model constraint calibration processing according to the Bayesian inversion result, set the resistivity threshold range in combination with the location and reserve information of the known oil and gas reservoirs, and use iterative least squares method to adjust the model parameters to construct a three-dimensional resistivity distribution model.

10. The carbonate oil and gas reservoir detection system using resistivity anomaly inversion according to claim 6, characterized in that: The prediction module comprises: A first processing unit is used to perform model parameter initialization processing according to the three-dimensional resistivity distribution model and the resistivity data of the target area, and to adjust the initial value of the resistivity based on the dynamic response parameters of the geological-electrical coupling relationship model, wherein the dynamic response parameters include formation pressure and temperature gradient parameters, so as to obtain a dynamic parameter initialization model; A first subdivision unit is used to initialize the model according to the dynamic parameters, perform adaptive grid subdivision processing, and generate a high-resolution grid model matching the carbonate cave-crack structure by combining a quadtree algorithm with a preset resistivity gradient threshold; A first calculation unit is used to perform parallel calculation and solution processing according to the high-resolution grid model to obtain a preliminary resistivity prediction result; The second optimization unit is used to perform optimization processing according to the preliminary resistivity prediction result, fill the low confidence data of the cave area with the radial basis function interpolation algorithm, and remove the noise in combination with the resistivity anomaly distribution map to obtain the resistivity spatial distribution prediction result of the target area.

Images