Combined recognition method and device for complex fractured thermal reservoir

Abstract

The invention relates to the technical field of fractured thermal reservoir identification, in particular to a complex fractured thermal reservoir joint identification method and device, and the method specifically comprises the following steps: 1, obtaining and collection of multi-source data: the obtaining and collection of the multi-source data comprise the collection of geological data, geophysical data and logging data; step 2, data preprocessing, wherein the data preprocessing comprises de-noising processing of seismic data and normalization processing of logging data; step 3, establishing a geologic model; step 4, carrying out geophysical inversion; 5, performing logging data feature extraction; step 6, establishing different fracture type probability models for fracture type thermal reservoirs involving different lithology; step 7, fracture connectivity analysis is carried out; 8, heat storage parameter calculation is carried out; step 9, carrying out model verification; and step 10, result visualization processing is carried out. According to the invention, the problems of limited geophysical logging information acquisition range and low seismic exploration resolution are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of fracture type thermal reservoir identification methods, and in particular to a combined identification method and device for complex fracture type thermal reservoirs. Background Art

[0002] Complex fractured thermal reservoirs occupy a significant position in the global distribution of geothermal resources. Their unique geological structure and fluid migration characteristics make them a key research target for geothermal development. Due to the high complexity of the fracture system, complex fractured thermal reservoirs are typically composed of multiple rock types, with significant variation in the scale, orientation, density, and connectivity of the internal fractures. The presence of fractures not only alters the physical properties of the rock, such as permeability and porosity, but also affects the flow of fluids and heat transfer within the thermal reservoir. For example, in granite thermal reservoirs, tectonic fractures and secondary fractures intertwine to form a complex network structure, which enhances the heterogeneity of the thermal reservoir.

[0003] Geophysical logging technologies, such as resistivity logging and sonic logging, can obtain physical parameters near the wellbore, but they have limited information on the distribution and connectivity of fractures far from the wellbore. Seismic exploration can provide underground structural information over a large area, but its resolution for small-scale fractures is low. Summary of the Invention

[0004] In response to the technical problems of limited information acquisition and low resolution in the prior art, the present invention provides a method and device for joint identification of complex fracture-type thermal reservoirs.

[0005] The technical solution adopted by the present invention is: a method for joint identification of complex fractured thermal reservoirs, which specifically includes the following steps:

[0006] Step 1: Acquisition and collection of multi-source data: The acquisition and collection of multi-source data includes the collection of geological data, geophysical data, and well logging data;

[0007] Step 2: Data preprocessing: The data preprocessing includes denoising of seismic data and normalization of well logging data;

[0008] Step three, establish the geological model;

[0009] Step 4: perform geophysical inversion;

[0010] Step 5: Extract logging data features;

[0011] Step 6: Establish different fracture type probability models for fractured thermal reservoirs with different lithologies;

[0012] Step seven, conduct crack connectivity analysis;

[0013] Step 8: Calculate heat storage parameters;

[0014] Step nine, perform model verification;

[0015] Step 10: Visualize the results.

[0016] In one embodiment, in step 1, the specific method for collecting geological data and geophysical data is as follows:

[0017] The geological data collection method is as follows:

[0018] Collect detailed stratigraphic lithology information, including rock type and mineral composition;

[0019] Assume that the number of lithologic categories is n lith ,Different lithologies have different physical and chemical properties, ,which affect the development and distribution of fractures;

[0020] The geophysical data collection method is as follows:

[0021] Collect seismic wave velocity v s and density ρ;

[0022] Collect geophysical data such as gravity and magnetism;

[0023] The logging data collection method is as follows:

[0024] Obtaining the resistivity R, the resistivity difference between the fluid in the fracture and the rock matrix is ​​obvious and is used to detect the presence of fractures;

[0025] Obtain the acoustic time difference Δt. The acoustic time difference can reflect the pore structure of the rock. Cracks will increase the acoustic wave propagation path, resulting in an increase in the acoustic time difference.

[0026] Collect natural gamma ray, neutron porosity and other logging data.

[0027] In one embodiment, in step 2, the method for denoising the seismic data and normalizing the well logging data is as follows:

[0028] The denoising method for seismic data is as follows:

[0029] For the seismic wave velocity v s , using wavelet transform filtering algorithm, the filtered seismic wave velocity is obtained For well logging data, resistivity R, median filtering is used to remove outliers;

[0030] The normalization processing method of logging data is as follows:

[0031] The normalization formula is where ρ m in and ρ maxare the minimum and maximum values ​​in the density data, respectively;

[0032] The normalized data is in the range [0,1].

[0033] In one embodiment, in step 3, the method for establishing the geological model is as follows:

[0034] Assume that the sampling point x is known i The attribute value at is z(x i ), the value at the point x0 to be estimated is z(x0);

[0035] By constructing the semivariogram Where h is the lag distance, N(h) is the number of sample pairs with a spacing of h;

[0036] Determine the weight coefficient based on the semivariogram, perform spatial interpolation on the lithology and structural characteristics of the formation, and obtain the lithology distribution function L(x,y,z) and the structural distribution function T(x,y,z), (x,y,z) is the spatial coordinate;

[0037] Use part of the known data as the validation set and calculate the error between the model prediction value and the true value. If the error is too large, adjust the semivariogram model parameters and re-interpolate until the model accuracy meets the requirements.

[0038] In one embodiment, in step 4, the method for performing geophysical inversion is as follows:

[0039] Wave equation Based on the equation, ρ is density, μ and λ are Lame constants, u is displacement, and t is time. The wave equation is numerically discretized using the finite difference method. The elastic parameters of the underground medium are continuously adjusted through optimization algorithms such as the conjugate gradient method to match the simulated seismic wave propagation with the actual observation data. The velocity model V(x, y, z) of the underground medium is obtained by inversion.

[0040] Combine gravity, magnetic and other data for joint inversion;

[0041] By constructing a joint objective function, the relationship between different geophysical data and the physical properties of underground media is comprehensively considered.

[0042] In one embodiment, in step 5, the method for extracting features from well logging data is as follows:

[0043] Using Archie formula Calculate fracture porosity φ f , where a is a constant related to lithology, and its value range is generally between 0.6 and 1.5, and R w is the formation water resistivity, R t is the true resistivity of the formation, m is the cementation index;

[0044] Other crack characteristic parameters are extracted from the sonic time difference data to extract the crack indication factor, crack apparent porosity Where Δt is the measured sound wave time difference, Δt m a is the acoustic time difference of the rock skeleton, Δt f is the fluid acoustic time difference.

[0045] In one embodiment, in step 6, the method for establishing different fracture type probability models for fractured thermal reservoirs of different lithologies is as follows:

[0046] Assume the probability of crack existence is P f (x,y,z), through the multiple linear regression model P f (x,y,z)=α1L(x,y,z)+α2T(x,y,z)+α3V(x,y,z)+α4φ f (x,y,z)+…+∈ is calculated; where α i is the regression coefficient, determined by the least squares method, and ∈ is the error term;

[0047] Monte Carlo simulation is used to analyze the uncertainty of the model. By randomly generating input data multiple times, the statistical distribution of crack probability is calculated and the reliability of the model results is evaluated.

[0048] In one embodiment, in step seven, the crack connectivity analysis method is as follows:

[0049] Inject tracers into the thermal reservoir and calculate the response time t of the tracers at different monitoring points. i and concentration C i , using the mass conservation equation Calculate the fracture connectivity index C conn ; where v is the fluid velocity, D is the diffusion coefficient, q is the source and sink term, C s is the source concentration;

[0050] A numerical model of underground fluid flow is constructed, and a discrete fracture network (DFN) model is used to simulate the flow path of the fluid in the fracture network. By setting boundary conditions and initial conditions, the fluid flow equation is solved and the connectivity relationship of the fractures is determined.

[0051] In one embodiment, in step eight, the heat storage parameter calculation method is as follows:

[0052] Combining the identified crack distribution and connectivity, according to the heat conduction equation The finite element method is used to solve the heat conduction equation and calculate the effective thermal conductivity k of the heat reservoir. eff; where ρ is the rock density, c is the specific heat capacity, T is the temperature, k is the thermal conductivity, and Q is the heat source term;

[0053] Thermal diffusivity The calculated effective thermal conductivity k e Calculate thermal diffusivity using ff, rock density ρ and specific heat capacity c;

[0054] In step nine, the method for model verification is as follows:

[0055] Use known well data or actual production data to verify the constructed fracture identification model and heat storage parameter calculation results;

[0056] Calculate the root mean square error, Where N is the number of samples, is the predicted value, is the true value;

[0057] The mean absolute error (MAE), determination coefficient (R 2 ) and other indicators to comprehensively evaluate the model performance;

[0058] In step 10, the result visualization processing method is as follows:

[0059] Use professional 3D modeling software, such as Petrel, Surfer, etc. to convert the crack probability model P f (x,y,z), fracture connectivity index C comn (x,y,z) and thermal storage parameters such as k e ff(x,y,z) etc. are presented in intuitive three-dimensional graphics;

[0060] By setting different color mapping and transparency, the spatial distribution of different attributes can be distinguished.

[0061] In one embodiment, a device for joint identification of complex fracture-type thermal reservoirs is provided, which includes a memory and a processor, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the method described above is implemented.

[0062] The present invention has the following beneficial effects: Compared with existing technologies, multi-source data fusion overcomes the limitations of single-source data in data acquisition. Geological data provides macroscopic context, geophysical data reflects differences in physical properties, and well logging data provides detailed information near the wellbore, comprehensively covering information at different scales and locations within the thermal reservoir. This addresses the limited range of geophysical logging information and the low resolution of seismic exploration. Advanced modeling and analysis techniques enable accurate visualization of the internal structure of the thermal reservoir. Fracture probability models and connectivity analysis methods quantify the probability of fractures and their connectivity, overcoming the inaccuracy of fracture network simulations in geological modeling. Furthermore, accurate calculation of thermal reservoir parameters, combined with visualization technology, provides a comprehensive, accurate, and intuitive reference for geothermal resource development, significantly improving development efficiency and reducing development costs and risks associated with inaccurate information. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0064] In the description of the present invention, it should be noted that the terms "front", "up", "down", "left", "right", "vertical", "horizontal", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they cannot be understood as limiting the present invention.

[0065] Reference Attachment Figure 1 In order to solve the problems existing in the background technology, this application proposes the following technical solution: a method for joint identification of complex fracture-type thermal reservoirs, which specifically includes the following steps:

[0066] Step 1: Acquisition and collection of multi-source data: The acquisition and collection of multi-source data includes the collection of geological data, geophysical data, and well logging data;

[0067] The specific method for collecting geological data and geophysical data in step 1 is as follows:

[0068] The geological data collection method is as follows:

[0069] Collect detailed stratigraphic lithology information, including rock type and mineral composition;

[0070] Assume that the number of lithologic categories is n lithDifferent rock types have different physical and chemical properties, which affect the development and distribution of cracks. For example, brittle rocks (such as limestone and sandstone) are more likely to form cracks than plastic rocks. In terms of structural characteristics, the location and scale of folds and faults are recorded. The number of structural types is n. t Stress concentration is often found near fault steps, making them favorable areas for crack development.

[0071] The geophysical data collection method is as follows:

[0072] Collect seismic wave velocity v s and density ρ;

[0073] Collect geophysical data such as gravity and magnetism;

[0074] Collect seismic wave velocity v s It is related to the elastic properties of the rock. The presence of cracks changes the rock's elastic modulus, which in turn affects the propagation velocity of seismic waves. Density ρ is also a key parameter. Density differences between different lithologies and fracture fillings can be used to identify changes in underground structures. Furthermore, geophysical data such as gravity and magnetism can be collected to reveal the characteristics of the underground geological body from different perspectives.

[0075] The logging data collection method is as follows:

[0076] Obtaining the resistivity R, the resistivity difference between the fluid in the fracture and the rock matrix is ​​obvious and is used to detect the presence of fractures;

[0077] Obtain the acoustic time difference Δt. The acoustic time difference can reflect the pore structure of the rock. Cracks will increase the acoustic wave propagation path, resulting in an increase in the acoustic time difference.

[0078] Collect natural gamma ray, neutron porosity and other logging data to provide multi-dimensional information for subsequent analysis.

[0079] Step 2: Data preprocessing: The data preprocessing includes denoising of seismic data and normalization of well logging data;

[0080] Wherein, in said step 2, the method for denoising the seismic data and normalizing the well logging data is as follows;

[0081] The denoising method for seismic data is as follows:

[0082] For the seismic wave velocity v s , using wavelet transform filtering algorithm. Wavelet transform can decompose the signal into different frequency bands, remove the noise frequency components by setting the threshold, and obtain the filtered seismic wave velocity For well logging data, resistivity R, median filtering is used to remove outliers to improve data stability;

[0083] The normalization processing method of logging data is as follows:

[0084] In order to eliminate the influence of different data dimensions, all data are normalized. Taking density ρ as an example,

[0085] The normalization formula is where ρ m in and ρ max are the minimum and maximum values ​​in the density data, respectively;

[0086] The normalized data is in the range of [0,1], which is convenient for subsequent data fusion and analysis.

[0087] Step three, establish the geological model;

[0088] In step 3, the geological model is established as follows:

[0089] Kriging interpolation method based on geostatistics takes into account the spatial correlation of data.

[0090] Assume that the sampling point x is known i The attribute value at is z(x i ), the value at the point x0 to be estimated is z(x0);

[0091] By constructing the semivariogram Where h is the lag distance, N(h) is the number of sample pairs with a spacing of h;

[0092] Determine the weight coefficient based on the semivariogram, perform spatial interpolation on the lithology and structural characteristics of the formation, and obtain the lithology distribution function L(x,y,z) and the structural distribution function T(x,y,z), (x,y,z) is the spatial coordinate;

[0093] Use part of the known data as the validation set and calculate the error between the model prediction value and the true value. If the error is too large, adjust the semivariogram model parameters and re-interpolate until the model accuracy meets the requirements.

[0094] The optimized geological model can more accurately display the spatial structure of the stratum and provide a basic framework for subsequent fracture analysis.

[0095] Step 4: perform geophysical inversion;

[0096] In step 4, the method for performing geophysical inversion is as follows:

[0097] Wave equation Based on the equation, ρ is density, μ and λ are Lame constants, u is displacement, and t is time. The wave equation is numerically discretized using the finite difference method. The elastic parameters of the underground medium are continuously adjusted through optimization algorithms such as the conjugate gradient method to match the simulated seismic wave propagation with the actual observation data. The velocity model V(x, y, z) of the underground medium is obtained by inversion.

[0098] The presence of cracks will cause abnormal areas to appear in the velocity model, providing clues for crack identification.

[0099] Joint inversion of multiple geophysical data In addition to seismic wave velocity, gravity, magnetic and other data are combined for joint inversion;

[0100] By constructing a joint objective function that comprehensively considers the relationship between different geophysical data and the physical properties of the subsurface medium, the resolution and reliability of the inversion results can be improved. For example, gravity data can reflect differences in subsurface density, and magnetic data can indicate the magnetic characteristics of rocks. These data complement the seismic velocity inversion results to more comprehensively identify the subsurface structure.

[0101] Step 5: Extract logging data features;

[0102] Among them, in step 5, the method for extracting the features of the well logging data is as follows:

[0103] Using Archie formula Calculate fracture porosity φ f , where a is a constant related to lithology, and its value range is generally between 0.6 and 1.5, and R w is the formation water resistivity, R t is the true resistivity of the formation, and m is the cementation index. For fractured formations, m is typically between 1.8 and 2.2. Fracture porosity is an important indicator of fracture development and provides a key parameter for subsequent comprehensive fracture identification.

[0104] Other crack characteristic parameters are extracted from the sonic time difference data to extract the crack indication factor, crack apparent porosity Where Δt is the measured sound wave time difference, Δt m a is the acoustic time difference of the rock skeleton, Δt f is the fluid acoustic wave time difference. In addition, imaging logging data can be used to extract information such as the orientation and length of the fracture, describing the fracture characteristics from multiple dimensions.

[0105] Step 6: Establish different fracture type probability models for fractured thermal reservoirs with different lithologies;

[0106] In step 6, the method for establishing different fracture type probability models for fractured thermal reservoirs of different lithologies is as follows:

[0107] Assume the probability of crack existence is P f (x,y,z), through the multiple linear regression model P f (x,y,z)=α1L(x,y,z)+α2T(x,y,z)+α3V(x,y,z)+α4φ f (x,y,z)+…+∈ is calculated; where α i is the regression coefficient, determined by the least squares method, and ∈ is the error term. This model comprehensively considers the relationship between geological, geophysical, and logging data and the probability of fracture existence, and can quantify the possibility of fracture existence in three-dimensional space.

[0108] Model Improvement and Uncertainty Analysis To improve model accuracy, regularization terms are introduced to prevent overfitting. At the same time, Monte Carlo simulation is used to analyze the uncertainty of the model. By randomly generating input data multiple times, the statistical distribution of crack probabilities is calculated to evaluate the reliability of the model results. For example, by analyzing the standard deviation of the probability distribution, the credibility range of the crack identification results is determined.

[0109] Step seven, conduct crack connectivity analysis;

[0110] In step 7, the crack connectivity analysis method is as follows:

[0111] Inject tracers into the thermal reservoir and calculate the response time t of the tracers at different monitoring points. i and concentration C i , using the mass conservation equation Calculate the fracture connectivity index C conn ; where v is the fluid velocity, D is the diffusion coefficient, q is the source and sink term, C s is the source concentration; tracer testing can intuitively reflect the connectivity between fractures, but its application is limited by testing cost and time.

[0112] A numerical model of underground fluid flow was constructed, using a discrete fracture network (DFN) model to simulate the flow paths of fluids within a fracture network. By setting boundary and initial conditions, the fluid flow equations were solved to determine the connectivity of the fractures. Numerical simulations can overcome the shortcomings of tracer testing, comprehensively analyze the connectivity of fractures within thermal reservoirs, and provide an important basis for the design of thermal reservoir development plans.

[0113] Step 8: Calculate heat storage parameters;

[0114] In step eight, the heat storage parameter calculation method is as follows:

[0115] Combining the identified crack distribution and connectivity, according to the heat conduction equation The finite element method is used to solve the heat conduction equation and calculate the effective thermal conductivity k of the heat reservoir. eff; where ρ is the rock density, c is the specific heat capacity, T is the temperature, k is the thermal conductivity, and Q is the heat source term. The presence of cracks will change the heat conduction path and affect the thermal conductivity. Accurately calculating k e ff is crucial for evaluating the heat storage capacity of a thermal reservoir.

[0116] Thermal diffusivity The calculated effective thermal conductivity k e Thermal diffusivity is calculated using ff, rock density ρ, and specific heat capacity c. Thermal diffusivity reflects the rate of heat transfer within a thermal reservoir and is a key parameter for evaluating the heat transfer efficiency of a thermal reservoir.

[0117] Step nine, perform model verification;

[0118] In step nine, the method for model verification is as follows:

[0119] Use known well data or actual production data to verify the constructed fracture identification model and heat storage parameter calculation results; calculate the root mean square error, Where N is the number of samples, is the predicted value, is the true value;

[0120] The mean absolute error (MAE), determination coefficient (R 2 ) and other indicators to comprehensively evaluate the model performance; if the model error exceeds the allowable range, adjust the model parameters, such as the regression coefficient α i , finite element mesh accuracy, etc. For crack probability models, optimization can be achieved by increasing training data, adjusting feature selection, etc. For numerical simulation models, boundary condition settings can be improved, algorithm parameters can be optimized, and model construction and calculations can be re-performed until the model accuracy meets the requirements.

[0121] Step 10: Visualize the results.

[0122] In step 10, the result visualization processing method is as follows:

[0123] Use professional 3D modeling software, such as Petrel, Surfer, etc. to convert the crack probability model P f (x,y,z), fracture connectivity index C comn (x,y,z) and thermal storage parameters such as k e ff(x,y,z) etc. are presented in intuitive three-dimensional graphics;

[0124] By setting different color mapping and transparency, the spatial distribution of different attributes can be distinguished; for example, red represents areas with high crack probability, and blue represents areas with low crack probability.

[0125] Interactive visualization allows users to observe the internal structure of the thermal reservoir through rotation, zooming, and slicing. In combination with virtual reality (VR) and augmented reality (AR) technologies, this provides an immersive visualization experience, enabling a more intuitive understanding of thermal reservoir characteristics and providing a clear and accurate reference for formulating geothermal resource development plans.

[0126] In summary, the present invention adopts a method combining multi-source data joint inversion with multivariate linear regression to achieve deep fusion of geological, geophysical, and logging data. Unlike the traditional simple data superposition method, by constructing a unified mathematical model, the intrinsic connection between different data is fully explored, thereby improving the accuracy and reliability of fracture identification.

[0127] This approach integrates tracer testing and numerical simulation, leveraging the intuitive nature of tracer testing and the comprehensiveness of numerical simulation to mutually validate and complement each other. Furthermore, an advanced discrete fracture network model is incorporated into the numerical simulation to more realistically simulate fluid flow within the fracture network, improving the accuracy of fracture connectivity analysis. Virtual reality and augmented reality technologies are incorporated into the visualization of thermal reservoir identification results, providing users with an immersive experience. Compared to traditional 2D or static 3D visualizations, users can observe the internal structure of the thermal reservoir more intuitively and deeply, improving decision-making efficiency and accuracy.

[0128] This invention, through systematic multi-source data acquisition and collection, coupled with advanced data processing and modeling techniques, achieves comprehensive and accurate identification of reservoir fracture distribution, connectivity, and thermal reservoir parameters. Innovative data fusion strategies, fracture analysis methods, and visualization techniques surpass traditional identification methods in performance. The application of this method will provide more reliable technical support for geothermal resource development.

[0129] In this paper, multi-source data fusion overcomes the limitations of single-source data in data acquisition. Geological data provides macroscopic context, geophysical data reflects differences in physical properties, and well logging data provides detailed information near the wellbore. This comprehensive coverage of geothermal reservoirs at different scales and locations addresses the limited range of geophysical logging information and the low resolution of seismic exploration. Advanced modeling and analysis techniques enable precise visualization of the internal structure of geothermal reservoirs. Fracture probability models and connectivity analysis methods quantify the probability of fractures and their connectivity, overcoming the inaccuracy of fracture network simulations in geological modeling. Furthermore, accurate calculation of geothermal reservoir parameters, combined with visualization technology, provides a comprehensive, accurate, and intuitive reference for geothermal resource development, significantly improving development efficiency and reducing the costs and risks associated with inaccurate information.

[0130] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A joint identification method for complex fractured thermal reservoirs, characterized in that: The specific steps include: Step 1: Acquisition and collection of multi-source data: The acquisition and collection of multi-source data includes the collection of geological data, geophysical data, and well logging data; Step 2: Data preprocessing: The data preprocessing includes denoising of seismic data and normalization of well logging data; Step three, establish the geological model; Step 4: perform geophysical inversion; Step 5: Extract logging data features; Step 6: Establish different fracture type probability models for fractured thermal reservoirs with different lithologies; Step seven, conduct crack connectivity analysis; Step 8: Calculate heat storage parameters; Step nine, perform model verification; Step 10: Visualize the results.

2. A combined identification method for complex fractured thermal reservoirs according to claim 1, characterized in that: In step 1, the specific method for collecting geological data and geophysical data is as follows: The geological data collection method is as follows: Collect detailed stratigraphic lithology information, including rock type and mineral composition; Assume that the number of lithologic categories is n lith ,Different lithologies have different physical and chemical properties, ,which affect the development and distribution of fractures; The geophysical data collection method is as follows: Collect seismic wave velocity v s and density ρ; Collect geophysical data such as gravity and magnetism; The logging data collection method is as follows: Obtaining the resistivity R, the resistivity difference between the fluid in the fracture and the rock matrix is ​​obvious and is used to detect the presence of fractures; Obtain the acoustic time difference Δt. The acoustic time difference can reflect the pore structure of the rock. Cracks will increase the acoustic wave propagation path, resulting in an increase in the acoustic time difference. Collect natural gamma ray, neutron porosity and other logging data.

3. A combined identification method for complex fractured thermal reservoirs according to claim 2, characterized in that: In step 2, the method for denoising the seismic data and normalizing the well logging data is as follows: The denoising method for seismic data is as follows: For the seismic wave velocity v s , using wavelet transform filtering algorithm, the filtered seismic wave velocity is obtained For well logging data, resistivity R, median filtering is used to remove outliers; The normalization processing method of logging data is as follows: The normalization formula is where ρ m in and ρ max are the minimum and maximum values ​​in the density data, respectively; The normalized data is in the range [0,1].

4. A combined identification method for complex fractured thermal reservoirs according to claim 3, characterized in that: In step 3, the method for establishing the geological model is as follows: Assume that the sampling point x is known i The attribute value at is z(x i ), the value at the point x0 to be estimated is z(x0); By constructing the semivariogram Where h is the lag distance, N(h) is the number of sample pairs with a spacing of h; Determine the weight coefficient based on the semivariogram, perform spatial interpolation on the lithology and structural characteristics of the formation, and obtain the lithology distribution function L(x,y,z) and the structural distribution function T(x,y,z), (x,y,z) is the spatial coordinate; Use part of the known data as the validation set and calculate the error between the model prediction value and the true value. If the error is too large, adjust the semivariogram model parameters and re-interpolate until the model accuracy meets the requirements.

5. A combined identification method for complex fractured thermal reservoirs according to claim 4, characterized in that: In step 4, the method for geophysical inversion is as follows: Wave equation Based on the equation, ρ is density, μ and λ are Lame constants, u is displacement, and t is time. The wave equation is numerically discretized using the finite difference method. The elastic parameters of the underground medium are continuously adjusted through optimization algorithms such as the conjugate gradient method to match the simulated seismic wave propagation with the actual observation data. The velocity model V(x, y, z) of the underground medium is obtained by inversion. Combine gravity, magnetic and other data for joint inversion; By constructing a joint objective function, the relationship between different geophysical data and the physical properties of underground media is comprehensively considered.

6. A combined identification method for complex fractured thermal reservoirs according to claim 5, characterized in that: In step 5, the method for extracting features from well logging data is as follows: Using Archie formula Calculate fracture porosity φ f , where a is a constant related to lithology, and its value range is generally between 0.6 and 1.5, and R w is the formation water resistivity, R t is the true resistivity of the formation, m is the cementation index; Other crack characteristic parameters are extracted from the sonic time difference data to extract the crack indication factor, crack apparent porosity Where Δt is the measured sound wave time difference, Δt m a is the acoustic time difference of the rock skeleton, Δt f is the fluid acoustic time difference.

7. A combined identification method for complex fractured thermal reservoirs according to claim 6, characterized in that: In step 6, the method for establishing different fracture type probability models for fractured thermal reservoirs with different lithologies is as follows: Assume the probability of crack existence is P f (x,y,z), through the multiple linear regression model P f (x,y,z)=α1L(x,y,z)+α2T(x,y,z)+α3V(x,y,z)+α4φ f (x,y,z)+…+∈ is calculated; where α i is the regression coefficient, determined by the least squares method, and ∈ is the error term; Monte Carlo simulation is used to analyze the uncertainty of the model. By randomly generating input data multiple times, the statistical distribution of crack probability is calculated and the reliability of the model results is evaluated.

8. A combined identification method for complex fractured thermal reservoirs according to claim 7, characterized in that: In step seven, the crack connectivity analysis method is as follows: Inject tracers into the thermal reservoir and calculate the response time t of the tracers at different monitoring points. i and concentration C i , using the mass conservation equation Calculate the fracture connectivity index C conn ; where v is the fluid velocity, D is the diffusion coefficient, q is the source and sink term, C s is the source concentration; A numerical model of underground fluid flow is constructed, and a discrete fracture network (DFN) model is used to simulate the flow path of the fluid in the fracture network. By setting boundary conditions and initial conditions, the fluid flow equation is solved and the connectivity relationship of the fractures is determined.

9. A combined identification method for complex fractured thermal reservoirs according to claim 7, characterized in that: In step eight, the heat storage parameter calculation method is as follows: Combining the identified crack distribution and connectivity, according to the heat conduction equation The finite element method is used to solve the heat conduction equation and calculate the effective thermal conductivity k of the heat reservoir. eff ; where ρ is the rock density, c is the specific heat capacity, T is the temperature, k is the thermal conductivity, and Q is the heat source term; Thermal diffusivity The calculated effective thermal conductivity k e Calculate thermal diffusivity using ff, rock density ρ and specific heat capacity c; In step nine, the method for model verification is as follows: Use known well data or actual production data to verify the constructed fracture identification model and heat storage parameter calculation results; Calculate the root mean square error, Where N is the number of samples, is the predicted value, is the true value; The mean absolute error (MAE), determination coefficient (R 2 ) and other indicators to comprehensively evaluate the model performance; In step 10, the result visualization processing method is as follows: Use professional 3D modeling software, such as Petrel, Surfer, etc. to convert the crack probability model P f (x,y,z), fracture connectivity index C comn (x,y,z) and thermal storage parameters such as k e ff(x,y,z) etc. are presented in intuitive three-dimensional graphics; By setting different color mapping and transparency, the spatial distribution of different attributes can be distinguished.

10. A device for joint identification of complex fractured thermal reservoirs, characterized in that: The complex fracture type thermal reservoir joint identification device includes a memory and a processor, the memory is used to store a computer program, and when the computer program is executed by the processor, it implements the method according to any one of claims 1 to 9.

Images