Carbonate fractured-vuggy oil and gas reservoir communication attribute construction method and device

Abstract

The invention discloses a carbonate fractured-vuggy oil and gas reservoir communication attribute construction method and device. The method comprises the steps that well drilling geophysical data, well logging leakage point position drilling fluid leakage data and well group dynamic communication data are obtained; screening geophysical attributes related to the drilling fluid leakage speed of the logging leakage point from the drilling geophysical data, and taking the geophysical attributes as basic attributes for representing inter-well connectivity; based on the logging leakage data and the well group dynamic connection data, constructing a connection index at a logging leakage point; adopting a feedforward deep neural network algorithm to construct a complex nonlinear correlation between the basic attribute and the connectivity index, and obtaining a neural network prediction model conforming to the prediction precision; and inputting the basic attribute of the well logging in the research area into the neural network prediction model, and predicting the connectivity index of the well logging in the research area to obtain a connectivity fusion attribute which effectively characterizes the seepage capacity of the reservoir. According to the invention, semi-quantitative prediction of the inter-well communication mode and the communication path can be realized.

Description

Technical Field

[0001] This invention relates to the field of geophysical comprehensive interpretation, and more specifically, to a method and apparatus for constructing connectivity properties of carbonate fracture-vuggy oil and gas reservoirs. Background Technology

[0002] Currently, connectivity prediction for fractured-vuggy oil and gas reservoirs largely relies on production dynamic data, such as tracer concentration, energy indicator curves, and water injection indicator curves. However, this often only qualitatively identifies inter-well connectivity relationships, lacking a clear understanding of specific connectivity paths and strengths. This invention, from a geophysical perspective, selects multiple fundamental attributes related to reservoir connectivity and combines them with well logging and dynamic information through deep learning attribute fusion to construct a comprehensive connectivity attribute that effectively matches actual production data. This achieves the goal of semi-quantitative prediction of inter-well connectivity modes and paths. Summary of the Invention

[0003] The purpose of this invention is to propose a method and apparatus for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs, so as to achieve semi-quantitative prediction of inter-well connectivity modes and connectivity paths.

[0004] To achieve the above objectives, this invention proposes a method for constructing connectivity attributes of fractured-vuggy hydrocarbon reservoirs in carbonate rocks, comprising:

[0005] Acquire drilling geophysical data, well logging loss point locations, drilling fluid loss data, and well group dynamic connectivity data;

[0006] Geophysical attributes related to drilling fluid leakage rate at well logging loss points are selected from the drilling geophysical data to serve as basic attributes characterizing inter-well connectivity.

[0007] Based on well logging loss data and well group dynamic connectivity data, a connectivity index at well logging loss points is constructed to characterize reservoir seepage capacity, i.e., inter-well connectivity performance.

[0008] A feedforward deep neural network algorithm is used to construct the complex nonlinear correlation between the basic attributes and the connectivity index, and a neural network prediction model that meets the prediction accuracy is obtained.

[0009] The basic attributes of well logging in the study area are input into the neural network prediction model to predict the connectivity index of well logging in the study area, thereby obtaining the connectivity fusion attribute that effectively characterizes the reservoir seepage capacity. Based on the connectivity fusion attribute, the connectivity structure and inter-well connectivity mode in the study area are analyzed and predicted in detail.

[0010] Optionally, the drilling geophysical data includes mobility, porosity, fracture density, and fracture pressure;

[0011] The drilling fluid loss data at the location of the well logging loss point includes the leakage rate of the drilling fluid at the loss point;

[0012] The dynamic connectivity data of the well group includes the well spacing between each well in the well group and the water injection response time of each well.

[0013] Optionally, the basic properties include flowability, porosity, and fracture density.

[0014] Optionally, the construction of a connectivity index at the well logging loss point based on well logging loss data and well group dynamic connectivity data includes:

[0015] Define the connectivity coefficient X of well A in the well group as:

[0016]

[0017] Where m is the number of wells connected to well A, and d i Let t be the distance between well i and well A. i Let be the water injection response time of the i-th well;

[0018] Define the connectivity index C at the leakage point of well A as:

[0019] C = S·X

[0020] Where S is the leakage rate at the leakage point of well A.

[0021] Optionally, the step of constructing the complex nonlinear correlation between the basic attributes and the connectivity index using a feedforward deep neural network algorithm includes:

[0022] Construct a feedforward deep neural network, which includes an input layer, a hidden layer, and an output layer;

[0023] Construct training sample data, which includes three basic attributes: porosity, fracture density, and flowability, as well as the corresponding connectivity index;

[0024] A feedforward deep neural network is trained, with porosity, fracture density, and flowability as input data and connectivity index as output data. The trained neural network prediction model can output the corresponding connectivity index based on the input porosity, fracture density, and flowability.

[0025] Optionally, during training, the backpropagation algorithm is used to iteratively optimize the weights of each neuron in each layer of the neural network.

[0026] Optionally, the number of hidden layers is 3, and the number of neuron nodes in the 3 hidden layers are 8, 6, and 4, respectively.

[0027] Secondly, the present invention provides an electronic device, the electronic device comprising:

[0028] At least one processor; and,

[0029] A memory communicatively connected to the at least one processor; wherein,

[0030] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs as described in the first aspect.

[0031] Thirdly, the present invention proposes a non-transitory computer-readable storage medium that stores computer instructions for causing a computer to execute the method for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs as described in the first aspect.

[0032] Fourthly, this invention proposes a device for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs, comprising:

[0033] The data input module is used to input drilling geophysical data, well logging loss point locations, drilling fluid loss data, and well group dynamic connectivity data;

[0034] The basic attribute filtering module is used to filter geophysical attributes related to drilling fluid leakage rate at well logging loss points from the drilling geophysical data, as basic attributes characterizing inter-well connectivity.

[0035] The connectivity index construction module is used to construct the connectivity index at the well leakage point based on well logging leakage data and well group dynamic connectivity data, in order to characterize the reservoir seepage capacity, i.e., the inter-well connectivity performance.

[0036] The connectivity fusion attribute construction module is used to construct the complex nonlinear correlation between the basic attributes and the connectivity index using a feedforward deep neural network algorithm, and to obtain a neural network prediction model that meets the prediction accuracy.

[0037] The connectivity fusion attribute prediction module is used to input the basic attributes of well logging in the study area into the neural network prediction model, predict the connectivity index of well logging in the study area, obtain the connectivity fusion attribute that effectively characterizes the reservoir seepage capacity, and perform fine analysis and prediction of the connectivity structure and inter-well connectivity mode in the study area based on the connectivity fusion attribute.

[0038] The beneficial effects of this invention are as follows:

[0039] (1) Compared with the qualitative judgment of inter-well connectivity based on production dynamic data, the method of the present invention can effectively describe the differences in the strength of connectivity within the reservoir and perform semi-quantitative judgment and prediction of inter-well connectivity.

[0040] (2) The method of the present invention screens and integrates basic attributes of multiple connectivity representations, and adds drilling leakage data and known well group connectivity information as reference indicators, which can obtain comprehensive and accurate connectivity fusion attributes that are consistent with actual production data.

[0041] The system of the present invention has other features and advantages that will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0042] The above and other objects, features and advantages of the present invention will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.

[0043] Figure 1 This is a step diagram illustrating a method for constructing connectivity attributes of a carbonate fracture-vuggy oil and gas reservoir according to an embodiment of the present invention.

[0044] Figure 2 This is a flowchart illustrating a method for constructing connectivity attributes of a carbonate fracture-vuggy oil and gas reservoir according to an embodiment of the present invention.

[0045] Figure 3a This is a scatter plot showing the relationship between porosity and leakage rate in one embodiment of the present invention.

[0046] Figure 3b This is a scatter plot showing the relationship between crack density and leakage rate in one embodiment of the present invention.

[0047] Figure 3c This is a scatter plot showing the relationship between flow rate and leakage rate in one embodiment of the present invention.

[0048] Figure 4 This is a schematic diagram of a deep neural network model in one embodiment of the present invention.

[0049] Figure 5 This is a schematic diagram of the well group connectivity in one embodiment of the present invention.

[0050] Figure 6 This is a graph showing the root mean square error (RMSE) curve of a deep neural network model prediction in one embodiment of the present invention.

[0051] Figure 7 This is a cross-sectional view of the porosity, fracture density, flowability, and connectivity properties of well A in one embodiment of the present invention.

[0052] Figure 8This is an example diagram of well group connectivity analysis in one embodiment of the present invention. Detailed Implementation

[0053] Currently, predicting connectivity in fractured-vuggy oil and gas reservoirs presents challenges, including unclear connectivity patterns within injection-production well groups in carbonate reservoirs and the inability to effectively predict inter-well connectivity. Furthermore, relying on production dynamic data only allows for qualitative assessments of inter-well connectivity and is limited to the existing injection-production well groups, failing to effectively predict connectivity for wells outside the injection-production well network.

[0054] This invention proposes a method for constructing connectivity attributes in carbonate fractured-vuggy oil and gas reservoirs to address the aforementioned problems. Specifically, it is a technical method that, under the constraints of well logging and production dynamic data, integrates multiple attributes that can effectively characterize seepage capacity through a deep neural network to construct a comprehensive connectivity characterization attribute. Within carbonate fractured-vuggy reservoirs, drilling fluid loss often occurs when wells encounter cavities or fractured bodies. The magnitude of the loss rate directly reflects the fluid seepage capacity. Analysis of well logging loss data in the study area shows a good correlation between porosity, fracture density, and mobility and the loss rate; therefore, these are considered three basic geophysical attributes. This method, based on actual production conditions, combines well logging loss rate data and well group connectivity dynamic data to construct a connectivity index as a parameter for connectivity strength. A deep neural network model predicts the correlation between the three attributes and the connectivity index, constructing a comprehensive connectivity characterization attribute to represent the connectivity relationships within the well group. This solves the current problem of lacking effective means to characterize seepage strength and unclear inter-well connectivity relationships in carbonate reservoirs. Verified in actual work areas, this method can effectively determine the inter-well connectivity mode and semi-quantitatively predict the inter-well connectivity path, providing a good geophysical basis for well group injection and production development and well network construction.

[0055] The invention will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0056] Example 1

[0057] like Figure 1 As shown in the figure, this embodiment provides a method for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs, including:

[0058] S1: Acquire drilling geophysical data, well logging loss point locations, drilling fluid loss data, and well group dynamic connectivity data;

[0059] Among them, drilling geophysical data includes mobility, porosity, fracture density, and fracture pressure; well logging data on the location of lost circulation points and drilling fluid loss includes the leakage rate of the drilling fluid at the loss point; and well group dynamic connectivity data includes the well spacing between wells in the well group and the water injection response time of each well.

[0060] S2: Screen geophysical attributes related to drilling fluid leakage rate at well logging loss points from drilling geophysical data as basic attributes characterizing inter-well connectivity;

[0061] In this step, the basic properties include flowability, porosity, and fracture density.

[0062] S3: Based on well logging loss data and well group dynamic connectivity data, construct the connectivity index at well logging loss points to characterize reservoir seepage capacity, i.e., inter-well connectivity performance.

[0063] In this step, based on well logging loss data and well group dynamic connectivity data, a connectivity index at the well logging loss point is constructed, including:

[0064] Define the connectivity coefficient X of well A in the well group as:

[0065]

[0066] Where m is the number of wells connected to well A, and d i Let t be the distance between well i and well A. i Let be the water injection response time of the i-th well;

[0067] Define the connectivity index C at the leakage point of well A as:

[0068] C = S·X

[0069] Where S is the leakage rate at the leakage point of well A.

[0070] S4: A feedforward deep neural network algorithm is used to construct a complex nonlinear correlation between basic attributes and connectivity index, and a neural network prediction model that meets the prediction accuracy is obtained.

[0071] In this step, a feedforward deep neural network algorithm is used to construct a complex nonlinear correlation between basic properties and connectivity indices, including:

[0072] Construct a feedforward deep neural network, which includes an input layer, hidden layers, and an output layer; preferably, the number of hidden layers is 3, and the number of neurons in the 3 hidden layers are 8, 6, and 4, respectively.

[0073] Construct training sample data, which includes three basic attributes: porosity, crack density, and flowability, as well as the corresponding connectivity index.

[0074] A feedforward deep neural network is trained, with porosity, fracture density, and flowability as input data and connectivity index as output data. The trained neural network prediction model can output the corresponding connectivity index based on the input porosity, fracture density, and flowability.

[0075] Preferably, during the training process, the backpropagation algorithm is used to iteratively optimize the weights of each neuron in each layer of the neural network.

[0076] S5: Input the basic attributes of well logging in the study area into the neural network prediction model to predict the connectivity index of well logging in the study area, obtain the connectivity fusion attribute that effectively characterizes the reservoir seepage capacity, and conduct fine analysis and prediction of the connectivity structure and inter-well connectivity mode in the study area based on the connectivity fusion attribute.

[0077] This method starts from actual drilling data, statistically analyzes drilling fluid leakage information at wellbore leakage points, and constructs a connectivity index parameter to characterize reservoir permeability, i.e., connectivity performance, in conjunction with well group connectivity information. It analyzes and selects three basic attributes related to leakage rate: mobility, porosity, and fracture density, and constructs a complex nonlinear relationship between the three attributes and the connectivity index through a feedforward deep neural network algorithm, thereby effectively predicting the connectivity within the reservoir.

[0078] Example 2

[0079] This embodiment provides a method for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs, such as... Figure 2 As shown, the specific process is as follows:

[0080] (1) Preferred connectivity characterization of basic properties

[0081] Porosity characterizes the size of the pore space within carbonate reservoirs. Well logging statistics confirm a positive correlation between porosity and permeability in this type of reservoir. Porosity can be predicted through pre-stack and post-stack inversion. Fracture density indicates the degree of fracture development within the reservoir. Higher fracture development generally indicates stronger seepage capacity. This attribute can be obtained through pre-stack and post-stack fracture detection. Mobility, defined in permeability mechanics as the ratio of the permeability of porous framework rocks to the viscosity coefficient of reservoir fluids, indicates higher mobility and lower viscosity coefficients of the fluids contained in the reservoir, resulting in stronger fluid seepage capacity. This attribute can also be obtained through seismic data analysis.

[0082] This embodiment statistically analyzes the leakage velocity and three types of attribute values ​​of well logging leakage points in the study area, as shown in the attached figure. Figures 3a-3c The figure shows a scatter plot of the relationship between porosity, fracture density, and flow rate and leakage rate. It can be seen that there is a non-linear correlation between these three attributes and leakage rate. Porosity, fracture density, and flow rate can be used as basic attributes for connectivity characterization.

[0083] (2) Creating the connectivity index parameter

[0084] This invention combines well logging loss data with well group dynamic connectivity data to comprehensively consider connectivity performance. Assume a well has a network of m other wells connected to it, with well distances d1, d2, d3…d… m The water injection response times are t1, t2, t3…t m The well connectivity coefficient X is then defined as:

[0085]

[0086] The connectivity index C at the well leakage point is defined as:

[0087] C = S·X(2)

[0088] The connectivity index C is a comprehensive reflection of the leakage velocity at wellbore loss points and the connectivity between the well and surrounding wells. This parameter can well describe the interconnected seepage situation within the reservoir, so it is used as a predictive indicator to judge and analyze connectivity.

[0089] (3) Construction of connectivity and fusion attributes

[0090] A feedforward deep neural network algorithm is used to construct a complex nonlinear correlation between porosity, flowability, fracture density properties and connectivity index (e.g., Figure 4 (As shown). A feedforward deep neural network consists of an input layer, hidden layers, and an output layer. Each layer consists of several neurons, and the weights of each neuron are iteratively optimized using the backpropagation algorithm.

[0091] In this embodiment, the feedforward neural network propagates information using the following formula:

[0092] z (l) =W (l) ·a (l-1) +b (l) ,a (l) =f l (z (l) (3)

[0093] By transmitting information layer by layer, the final output 'a' of the network is obtained. (L) Where L represents the number of layers in the neural network, f l W represents the activation function of neurons in layer l. (l) Let b represent the weight matrix from layer l-1 to layer l. (l) z represents the offset from layer l-1 to layer l. (l) a represents the net input to neurons in layer l. (l) This represents the output of a neuron in layer l. The entire network can be viewed as a composite function, taking vector x as the input a of layer 1.(0) The output a of layer L (L) As the output of the function.

[0094] The training process of a feedforward neural network based on the backpropagation algorithm can be divided into the following three steps:

[0095] 1. Calculate the net input z for each layer during forward propagation. (l) and activation value a (l) until the last floor;

[0096] 2. Calculate the error term δ for each layer using backpropagation. (l) ;

[0097] δ (l) =f l '(z (l) )⊙(W (l+1) ) T δ (l+1) (4)

[0098] 3. Calculate the partial derivatives of the parameters for each layer and update the parameters.

[0099]

[0100] Porosity, fracture density, and flowability are used as input data, and connectivity index is used as output data for neural network prediction. After testing various parameters, a neural network prediction model meeting the required accuracy is obtained. Through model prediction, a connectivity fusion attribute (predicted connectivity index) that effectively characterizes seepage capacity can be obtained, enabling detailed analysis and prediction of connectivity structures and inter-well connectivity methods within the study area.

[0101] Example 3

[0102] In this embodiment, the study area is located in the Shuntogole low uplift zone of the Tarim Basin. The reservoir type is a carbonate fault-controlled fracture-vuggy reservoir. The area has undergone multiple phases of fault activity, resulting in highly developed and complex fractures and fissures. Vuggy and cavernous reservoirs are accompanied by well-developed main faults. The area is currently in the late stage of development, with reservoir pressure declining. Stable production is maintained by constructing an injection-production well network and supplementing energy through water and gas injection. Some inter-well connectivity relationships have been qualitatively determined using production dynamic data such as water injection indicator curves and tracer analysis. However, the connection methods and strengths between wells are not understood, making it impossible to accurately predict the exact connectivity structure within the well group, thus hindering efficient reserve development and well location design.

[0103] like Figure 5The diagram shows the connection relationship of a group of wells in the work area. The dashed lines represent the connection between wells. Taking well A2 as an example, it is connected to wells A1, A3 and A4. The distances between the wells are 0.767km, 0.9km and 1.558km respectively. When well A2 is injected with water, the response days of wells A1, A3 and A4 are 9 days, 8 days and 65 days respectively. According to formula (1), the connection coefficient of well A2 is 0.074. According to formula (2), the connection index at the leakage sample point above well A2 can be obtained.

[0104] Porosity, crack density, and flowability are used as input data, and connectivity index is used as output data. A deep feedforward neural network model is used for prediction, and their correlations are established. A total of 24 sample points are used. 18 sample points are randomly selected as the training set for neural network training, including a validation set to verify model errors. The remaining 6 sample points are used as the test set to evaluate the effectiveness of the model's prediction accuracy. The number of hidden layers and hidden node counts significantly affect prediction accuracy. Therefore, different parameters for the number of hidden layers and nodes were selected for model accuracy testing. Ultimately, a minimum root mean square error was achieved with 3 hidden layers and 8, 6, and 4 neurons, resulting in the construction of a high-quality neural network model. Figure 6 The figure shows the root mean square error curve of the model prediction. The model was updated and iterated 7 times. The minimum root mean square error on the validation set reached 0.003, and the root mean square error on the final test set was 0.005, which met the accuracy requirements.

[0105] like Figure 7 The figure shows the porosity, fracture density, flow rate, and predicted connectivity profile of well A in the study area. The bottom of the well in the original trajectory and the sidetrack trajectory of well A is the drilling fluid leakage point. It can be seen that the location of the strength and weakness critical point of the connectivity attribute is more consistent with the location of the leakage point than the three basic attributes, and is closer to the actual production data. It can more accurately describe the internal connectivity of the reservoir.

[0106] like Figure 8 As shown, the connectivity of an injection-production well group within the work area was analyzed using the constructed connectivity attributes. Production data shows that when well B was injected with water, wells C and D all responded. Well C had a pressure wave duration of 21 days and a production enhancement effect time of 38 days; well D had a pressure wave duration of 177 days and a production enhancement effect time of 226 days. The well distances between wells C and D and well B are roughly equivalent, and the differences in the rate of effect cannot be effectively explained solely by injection-production indicator curves. The connectivity attribute analysis reveals a strong connectivity between wells B and C, while the area between wells B and D is a weakly connected region. Therefore, after well B was injected with water, well C responded faster than well D.

[0107] As can be seen, the method of this invention, through well profile analysis, shows that this attribute matches well with well logging loss data. Through well group connectivity example analysis, this attribute can intuitively and accurately describe the inter-well connectivity relationship and matches well with the dynamic data of injection-production effectiveness, verifying the effectiveness of the method of this invention. This invention, through the constructed fused connectivity attribute, can effectively verify the injection-production effectiveness of well groups and accurately predict regional connectivity structures. It achieves a semi-quantitative description and prediction of inter-well connectivity in carbonate fractured-vuggy reservoirs, providing geophysical evidence for injection-production well network construction and well trajectory design, and has significant guiding significance for field injection-production work.

[0108] Example 4

[0109] This embodiment provides a device for constructing connectivity properties of carbonate fracture-vuggy oil and gas reservoirs, including:

[0110] The data input module is used to input drilling geophysical data, well logging loss point locations, drilling fluid loss data, and well group dynamic connectivity data;

[0111] The basic attribute filtering module is used to filter geophysical attributes related to drilling fluid leakage rate at well logging loss points from the drilling geophysical data, as basic attributes characterizing inter-well connectivity.

[0112] The connectivity index construction module is used to construct the connectivity index at the well leakage point based on well logging leakage data and well group dynamic connectivity data, in order to characterize the reservoir seepage capacity, i.e., the inter-well connectivity performance.

[0113] The connectivity fusion attribute construction module is used to construct the complex nonlinear correlation between the basic attributes and the connectivity index using a feedforward deep neural network algorithm, and to obtain a neural network prediction model that meets the prediction accuracy.

[0114] The connectivity fusion attribute prediction module is used to input the basic attributes of well logging in the study area into the neural network prediction model, predict the connectivity index of well logging in the study area, obtain the connectivity fusion attribute that effectively characterizes the reservoir seepage capacity, and perform fine analysis and prediction of the connectivity structure and inter-well connectivity mode in the study area based on the connectivity fusion attribute.

[0115] For the specific functions of each functional module in this embodiment, please refer to Embodiments 1 and 2 above, which will not be repeated here.

[0116] Example 5

[0117] This embodiment provides an electronic device, the electronic device comprising:

[0118] At least one processor; and,

[0119] A memory communicatively connected to the at least one processor; wherein,

[0120] The memory stores instructions that can be executed by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the method for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs as described in the above embodiments.

[0121] An electronic device according to embodiments of the present disclosure includes a memory and a processor. The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0122] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of this disclosure, the processor is used to execute computer-readable instructions stored in the memory.

[0123] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0124] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0125] Example 6

[0126] Thirdly, the present invention proposes a non-transitory computer-readable storage medium that stores computer instructions for causing a computer to execute the method for constructing connectivity attributes of carbonate fracture-vuggy oil and gas reservoirs as described in the above embodiments.

[0127] A computer-readable storage medium according to embodiments of the present disclosure stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the methods described in the foregoing embodiments of the present disclosure are performed.

[0128] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0129] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for constructing connectivity attributes of a carbonate fractured-vugular reservoir, characterized in that, include: Acquire drilling geophysical data, well logging loss point locations, drilling fluid loss data, and well group dynamic connectivity data; Geophysical attributes related to drilling fluid leakage rate at well logging loss points are selected from the drilling geophysical data to serve as basic attributes characterizing inter-well connectivity. Based on well logging loss data and well group dynamic connectivity data, a connectivity index at well logging loss points is constructed to characterize reservoir seepage capacity, i.e., inter-well connectivity performance. A feedforward deep neural network algorithm is used to construct the complex nonlinear correlation between the basic attributes and the connectivity index, and a neural network prediction model that meets the prediction accuracy is obtained. The basic attributes of well logging in the study area are input into the neural network prediction model to predict the connectivity index of well logging in the study area, thereby obtaining the connectivity fusion attribute that effectively characterizes the reservoir seepage capacity. Based on the connectivity fusion attribute, the connectivity structure and inter-well connectivity mode in the study area are analyzed and predicted in detail.

2. The method of claim 1, wherein, The drilling geophysical data includes mobility, porosity, fracture density, and fracture pressure; The drilling fluid loss data at the location of the well logging loss point includes the leakage rate of the drilling fluid at the loss point; The dynamic connectivity data of the well group includes the well spacing between each well in the well group and the water injection response time of each well.

3. The method of claim 2, wherein, The basic properties include flowability, porosity, and fracture density.

4. The method of claim 3, wherein, The connectivity index at well loss points is constructed based on well logging loss data and well group dynamic connectivity data, including: Define the connectivity coefficient X of well A in the well group as: wherein m is the number of wells that have a communication relationship with the well A, d i is the well spacing between the i-th well and the well A, t i is the water injection response time of the i-th well; Define the connectivity index C at the leakage point of well A as: C = S·X Where S is the leakage rate at the leakage point of well A.

5. The method of claim 4, wherein, The method of constructing a complex nonlinear correlation between the basic attributes and the connectivity index using a feedforward deep neural network algorithm includes: Construct a feedforward deep neural network, which includes an input layer, a hidden layer, and an output layer; Construct training sample data, which includes three basic attributes: porosity, fracture density, and flowability, as well as the corresponding connectivity index; A feedforward deep neural network is trained, with porosity, fracture density, and flowability as input data and connectivity index as output data. The trained neural network prediction model can output the corresponding connectivity index based on the input porosity, fracture density, and flowability.

6. The method of claim 5, wherein, During training, the backpropagation algorithm is used to iteratively optimize the weights of each neuron in each layer of the neural network.

7. The method of claim 5, wherein, The number of hidden layers is 3, and the number of neuron nodes in the 3 hidden layers are 8, 6 and 4 respectively.

8. An electronic device, comprising: The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method for constructing connectivity properties of carbonate fracture-vuggy oil and gas reservoirs according to any one of claims 1-5.

9. A non-transitory computer-readable storage medium, comprising: The non-transitory computer-readable storage medium stores computer instructions for causing a computer to execute the method for constructing connectivity properties of carbonate fracture-vuggy oil and gas reservoirs as described in any one of claims 1-5.

10. A device for building connectivity attributes of a carbonate fractured-vugular reservoir, characterized in that, include: The data input module is used to input drilling geophysical data, well logging loss point locations, drilling fluid loss data, and well group dynamic connectivity data; The basic attribute filtering module is used to filter geophysical attributes related to drilling fluid leakage rate at well logging loss points from the drilling geophysical data, as basic attributes characterizing inter-well connectivity. The connectivity index construction module is used to construct the connectivity index at the well leakage point based on well logging leakage data and well group dynamic connectivity data, in order to characterize the reservoir seepage capacity, i.e., the inter-well connectivity performance. The connectivity fusion attribute construction module is used to construct the complex nonlinear correlation between the basic attributes and the connectivity index using a feedforward deep neural network algorithm, and to obtain a neural network prediction model that meets the prediction accuracy. The connectivity fusion attribute prediction module is used to input the basic attributes of well logging in the study area into the neural network prediction model, predict the connectivity index of well logging in the study area, obtain the connectivity fusion attribute that effectively characterizes the reservoir seepage capacity, and perform fine analysis and prediction of the connectivity structure and inter-well connectivity mode in the study area based on the connectivity fusion attribute.

Images