Methods, equipment, media and products for predicting enhanced oil recovery in fractured gas reservoirs

Abstract

This application provides a method, device, medium, and product for predicting the recovery rate of fractured gas reservoirs. The method includes: dividing the area to be predicted into at least one target evaluation unit based on unit division data; determining the well type category of each target evaluation unit based on well type characteristic parameters, and determining the fracture morphology category of each target evaluation unit based on fracture characteristic parameters; obtaining a target recovery rate prediction model corresponding to the well type category and fracture morphology category from a preset recovery rate model database, and obtaining the predicted recovery rate value of the target evaluation unit using the target recovery rate prediction model based on the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters of the target evaluation unit; and obtaining the predicted recovery rate value of the area to be predicted based on the predicted recovery rate values ​​of each target evaluation unit according to a preset regional aggregation rule. This method aims to improve the accuracy of recovery rate prediction.

Description

Technical Field

[0001] This application relates to the field of petroleum development technology, and in particular to a method, equipment, medium and product for predicting the recovery rate of fractured gas reservoirs. Background Technology

[0002] Fractured gas reservoirs are an important type of natural gas reservoir, and their development directly affects the field's stable production capacity, remaining gas utilization, and ultimate recovery rate. These reservoirs typically feature low matrix permeability, well-developed natural fractures, strong heterogeneity, and significant edge or bottom water response. Natural fractures can improve gas flow channels and increase single-well productivity; however, they can also create channels for rapid water advance, leading to earlier water production, faster pressure decline, and further impacting subsequent development adjustments and recovery rates.

[0003] In existing technologies, gas reservoir recovery rate prediction typically employs empirical statistical methods or mass balance methods. Empirical statistical methods generally rely on production data from developed blocks or similar gas reservoirs, selecting factors such as reservoir properties, development dynamics, well network deployment, or water intrusion intensity. Statistical fitting is used to establish the relationship between recovery rate and these factors, enabling rapid estimation of development effectiveness. Mass balance methods typically use data on formation pressure, cumulative gas production, cumulative water production, fluid properties, and water intrusion to analyze energy changes and reserve consumption during reservoir development, estimating recoverable reserves or recovery rate.

[0004] However, for fractured gas reservoirs with complex fracture development and significant differences in gas-water response, the recovery rate at different locations and under different development conditions is affected by a combination of factors, including the degree of fracture development, well type differences, reservoir properties, and water energy. Existing prediction methods often struggle to differentiate between various development conditions during rapid field assessments, resulting in low accuracy in recovery rate predictions. Summary of the Invention

[0005] This application provides a method, equipment, medium, and product for predicting the recovery rate of fractured gas reservoirs, which aims to improve the accuracy of recovery rate prediction by enhancing the ability to respond to differences in different development conditions.

[0006] In a first aspect, embodiments of this application provide a method for predicting the recovery rate of fractured gas reservoirs, including:

[0007] Obtain unit division data of the area to be predicted, and divide the area to be predicted into at least one target evaluation unit based on the unit division data; wherein, the unit division data includes at least one of geological boundary data, fracture attribute data, seepage connectivity data, and well network control data;

[0008] Based on the well type characteristic parameters of the target evaluation unit, the well type category of each target evaluation unit is determined, and based on the fracture characteristic parameters of the target evaluation unit, the fracture morphology category of each target evaluation unit is determined.

[0009] From the preset recovery rate model database, obtain the target recovery rate prediction model corresponding to the well type and fracture morphology of the target evaluation unit, and obtain the recovery rate prediction value of the target evaluation unit by using the target recovery rate prediction model based on the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters and water body characteristic parameters of the target evaluation unit.

[0010] Based on the recovery rate prediction values ​​of each target evaluation unit, the recovery rate prediction value of the region to be predicted is obtained according to the preset regional aggregation rules.

[0011] In one possible implementation, dividing the region to be predicted into at least one target evaluation unit based on the unit partitioning data includes:

[0012] Based on the geological boundary data, determine the structural zoning boundaries within the area to be predicted;

[0013] Based on the crack attribute data, determine the crack partition boundaries and crack attribute categories of each crack partition within the area to be predicted.

[0014] Based on the seepage connectivity data, determine the seepage zone boundaries and seepage connectivity categories of each seepage zone within the area to be predicted;

[0015] Based on the well network control data, determine the well control zone boundaries and well network control categories of each well control zone within the area to be predicted;

[0016] The structural partition boundary, fracture partition boundary, seepage partition boundary and well control partition boundary are superimposed to obtain at least one initial partitioning unit corresponding to the region to be predicted.

[0017] Based on the fracture zone, seepage zone, and well control zone where the initial partitioning unit is located, the fracture attribute category, seepage connectivity category, and well network control category of the initial partitioning unit are determined respectively.

[0018] Traverse each of the initial partitioning units in the region to be predicted, obtain any pair of adjacent initial partitioning units, and when the fracture attribute category, seepage connectivity category and well network control category of any pair of adjacent initial partitioning units are the same, merge the pair of adjacent initial partitioning units to obtain at least one target evaluation unit.

[0019] In one possible implementation, well profile parameters include well inclination data, well trajectory data, horizontal section length, and completion method;

[0020] Accordingly, determining the well type category of each target evaluation unit based on the well type characteristic parameters of the target evaluation unit includes:

[0021] Obtain the well type characteristic parameters of the target well within the target evaluation unit. The well type characteristic parameters include well inclination data, well trajectory data, horizontal section length, and well completion method.

[0022] Based on the well inclination data and well trajectory data, determine whether there are continuous well sections in the well trajectory of the target well with an inclination angle greater than a preset inclination angle threshold;

[0023] If there is no continuous well section with an inclination angle greater than the preset inclination angle threshold, then the well type category of the target evaluation unit is the vertical well category;

[0024] If there is a continuous well section with an inclination angle greater than a preset inclination angle threshold, then the continuous well section is determined as a candidate horizontal section;

[0025] When the length of the candidate horizontal segment is greater than a preset length threshold, it is determined whether the candidate horizontal segment is a completion segment or a producing segment of the target well, based on the completion method of the target well.

[0026] When the candidate horizontal segment is the completion segment or producing segment of the target well, the well type category of the target evaluation unit is determined to be the horizontal well category.

[0027] In one possible implementation, the fracture characteristic parameters include at least one of fracture density, fracture penetration ratio, fracture orientation dispersion parameter, fracture intersection parameter, fracture connectivity parameter, and fracture conductivity parameter.

[0028] Accordingly, based on the crack characteristic parameters of the target evaluation unit, the crack morphology category of each target evaluation unit is determined, including:

[0029] The fracture density of the target evaluation unit is determined based on at least one of core fracture statistics, imaging logging interpretation data, and discrete fracture network model data.

[0030] Based on at least one of seismic fracture interpretation data, imaging logging fracture orientation interpretation data, well connectivity data, tracer monitoring data, and production dynamic data, candidate fracture connectivity bodies are identified within the target evaluation unit, and the extension length of the candidate fracture connectivity bodies is determined.

[0031] The crack penetration ratio of the candidate crack connection is determined based on the ratio of the extension length of the candidate crack connection to the characteristic length of the target evaluation unit.

[0032] Based on at least one of imaging logging interpretation data, core fracture description data, and seismic fracture interpretation data, determine the azimuth data of the fractures within the target evaluation unit, and determine the fracture azimuth dispersion parameters based on the azimuth data.

[0033] Based on at least one of imaging logging interpretation data, core fracture description data, and discrete fracture network model data, determine the cross-cutting relationship between different fractures within the target evaluation unit, and determine the fracture cross-cutting parameters based on the cross-cutting relationship;

[0034] The fracture connectivity parameters and fracture conductivity parameters are determined based on at least one of the following: well test response data, well connectivity data, tracer monitoring data, lost zone identification data, produced zone identification data, water intrusion channel data, and production dynamic data.

[0035] When the crack penetration ratio of the candidate crack body meets the preset penetration condition, the crack flow capacity parameter meets the preset flow condition, and the crack orientation dispersion parameter meets the preset orientation concentration condition, the candidate crack body is identified as the main crack.

[0036] When the crack density meets a preset density condition, the crack orientation dispersion parameter meets a preset multi-directional condition, and the crack intersection parameter meets a preset intersection condition, a crack network is identified within the target evaluation unit.

[0037] When the main crack is identified in the target evaluation unit but the crack network is not identified, the crack morphology category of the target evaluation unit is determined to be a large crack type.

[0038] Alternatively, when the crack network is identified in the target evaluation unit but the main crack is not identified, the crack morphology category of the target evaluation unit is determined to be a crack network type.

[0039] Alternatively, when the main crack and the crack network are identified in the target evaluation unit, the crack morphology category of the target evaluation unit is determined to be a large crack-crack network coupled type.

[0040] In one possible implementation, obtaining the target recovery rate prediction model corresponding to the well type and fracture morphology category of the target evaluation unit includes:

[0041] When the well type of the target evaluation unit is a horizontal well and the fracture morphology is a large fracture, the first recovery rate prediction model is obtained.

[0042] Alternatively, when the well type of the target evaluation unit is a vertical well and the fracture morphology is a large fracture, a second recovery rate prediction model is obtained.

[0043] Alternatively, when the well type of the target evaluation unit is a horizontal well and the fracture morphology is a fracture network, a third recovery rate prediction model is obtained.

[0044] Alternatively, when the well type of the target evaluation unit is a vertical well and the fracture morphology is a fracture network, a fourth recovery rate prediction model is obtained.

[0045] Alternatively, when the well type of the target evaluation unit is a horizontal well and the fracture morphology is a large fracture network coupled type, the fifth recovery rate prediction model is obtained.

[0046] Alternatively, when the well type of the target evaluation unit is a vertical well and the fracture morphology is a large fracture network coupled type, the sixth recovery rate prediction model can be obtained.

[0047] In one possible implementation, obtaining the predicted recovery rate of the region to be predicted based on the predicted recovery rate of each of the target evaluation units, according to a preset regional aggregation rule, includes:

[0048] When the area to be predicted includes a target evaluation unit, the predicted recovery rate of the target evaluation unit is determined as the predicted recovery rate of the area to be predicted.

[0049] Alternatively, when the area to be predicted includes multiple target evaluation units, the geological reserves, controlled reserves, or effective pore volume of each target evaluation unit are obtained, and the following formula is used:

[0050]

[0051] Calculate and obtain the predicted recovery rate R for the area to be predicted. total ;

[0052] Among them, R i Let G be the predicted recovery rate of the i-th target evaluation unit. i Let n represent the geological reserves, controlled reserves, or effective pore volume of the i-th target evaluation unit, and n be the number of target evaluation units.

[0053] In one possible implementation, establishing the recovery rate model database includes:

[0054] Obtain geological interpretation data and development dynamic data of fractured gas reservoirs, and obtain fracture development patterns based on the geological interpretation data and development dynamic data;

[0055] Based on the fracture development pattern, gas reservoir development models are established under different well types and combinations of different fracture morphology categories;

[0056] By adjusting the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters in the gas reservoir development model, multiple sets of parameter combinations are obtained.

[0057] Numerical simulations were performed on each set of parameter combinations to obtain the recovery rate response samples corresponding to each set of parameter combinations.

[0058] The recovery rate response samples are classified according to well type and fracture morphology, and regression processing is performed on the classified recovery rate response samples to obtain the recovery rate prediction model corresponding to each combination of well type and fracture morphology.

[0059] The preset recovery rate model database is established based on each of the recovery rate prediction models.

[0060] Secondly, embodiments of this application provide an apparatus for predicting the recovery rate of fractured gas reservoirs, comprising:

[0061] The region division module is used to acquire unit division data of the region to be predicted, and divide the region to be predicted into at least one target evaluation unit according to the unit division data; wherein, the unit division data includes at least one of geological boundary data, fracture attribute data, seepage connectivity data and well network control data;

[0062] The category identification module is used to determine the well type category of each target evaluation unit based on the well type feature parameters of the target evaluation unit, and to determine the fracture morphology category of each target evaluation unit based on the fracture feature parameters of the target evaluation unit.

[0063] The model matching module is used to obtain the target recovery rate prediction model corresponding to the well type and fracture morphology category of the target evaluation unit from the preset recovery rate model database, and to obtain the recovery rate prediction value of the target evaluation unit by using the target recovery rate prediction model based on the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters and water body characteristic parameters of the target evaluation unit.

[0064] The results aggregation module is used to obtain the predicted recovery rate of the region to be predicted based on the predicted recovery rate of each target evaluation unit and according to the preset regional aggregation rules.

[0065] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0066] The memory stores computer-executed instructions;

[0067] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0068] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0069] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0070] The recovery rate prediction method, equipment, medium, and product for fractured gas reservoirs provided in this application divide the area to be predicted into at least one target evaluation unit by dividing the data into units. This allows the area to be predicted to be partitioned according to differences in internal geological boundaries, fracture attributes, seepage connectivity, or well network control relationships, preserving the differences between different areas during the prediction process. Next, based on the well type characteristic parameters and fracture characteristic parameters of the target evaluation unit, the well type category and fracture morphology category are determined, and the corresponding target recovery rate prediction model is obtained from a preset recovery rate model database. This ensures that the selection of the recovery rate prediction model no longer uses a single general model, but matches it with the development well type and fracture morphology of the target evaluation unit, improving the targeting of the recovery rate prediction process under different development conditions and shortening the prediction process. The well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters are input into the target recovery rate prediction model to obtain the recovery rate prediction value for each target evaluation unit, and further, the recovery rate prediction value for the area to be predicted is obtained according to preset regional aggregation rules. This processing method realizes the transformation from the prediction results of local target evaluation units to the overall prediction results of the area to be predicted. It enables the prediction results to reflect the differences between different target evaluation units and form the overall recovery rate prediction results of the area to be predicted. This solves the technical problem that it is difficult to achieve differentiated processing of different development conditions in complex fractured gas reservoirs, resulting in low recovery rate prediction accuracy. Attached Figure Description

[0071] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0072] Figure 1 A flowchart illustrating the method for predicting the recovery rate of fractured gas reservoirs provided in this application embodiment;

[0073] Figure 2 This is a schematic flowchart illustrating the method for establishing a recovery rate model database provided in an embodiment of this application.

[0074] Figure 3A schematic diagram of the structure of the fractured gas reservoir recovery prediction device provided in the embodiments of this application;

[0075] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0076] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0077] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0078] In existing technologies, for fractured gas reservoirs with complex fracture development and significant differences in gas-water response, the recovery rate at different locations and under different development conditions is affected by a combination of factors, including well type, fracture morphology, fracture parameters, reservoir properties, and water energy. Current prediction methods, when used for rapid field assessments, often struggle to incorporate well type and fracture morphology differences into a unified classification prediction framework. They also find it difficult to quickly match appropriate recovery rate prediction models to different well type-fracture morphology combinations. Consequently, recovery rate predictions under different development conditions lack specificity, have low efficiency for rapid comparison, and limited prediction accuracy.

[0079] The inventive concept of this application is to provide a method for predicting the recovery rate of fractured gas reservoirs, addressing the problems of insufficient regional variation and inefficient rapid prediction in related technologies. For fractured gas reservoirs with complex fracture development and significant differences in gas-water response, directly using the entire area to be predicted as a unified object for recovery rate prediction easily overlooks differences in geological boundaries, fracture properties, seepage connectivity, and well network control relationships at different locations. Furthermore, relying entirely on detailed numerical simulations results in long modeling and calculation cycles in multi-scheme rapid evaluation scenarios. Therefore, this application first divides the area to be predicted into at least one target evaluation unit based on unit partitioning data, enabling the differences within the area to be predicted to be transformed into separately processable evaluation objects.

[0080] After forming the target evaluation unit, the key differences affecting recovery rate prediction are further summarized into two classification dimensions: well type and fracture morphology. In other words, this application does not use a single universal relationship to predict all areas, but rather determines the well type category based on the well type characteristic parameters of the target evaluation unit, and the fracture morphology category based on the fracture characteristic parameters. Then, the well type category and fracture morphology category jointly determine the corresponding target recovery rate prediction model. This allows recovery rate predictions for different development well types and different fracture development patterns to enter different model calculation paths.

[0081] After determining the target recovery rate prediction model, this application inputs the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters of the target evaluation unit into the corresponding model to obtain the recovery rate prediction value for each target evaluation unit. For the prediction area including multiple target evaluation units, the overall recovery rate prediction value of the prediction area is obtained through regional aggregation rules. Thus, a technical concept is formed for regional difference identification, target evaluation unit division, well type and fracture morphology classification, classification model matching, unit prediction, and regional aggregation, which is used to solve the problem of difficulty in quickly obtaining regional recovery rate prediction results that reflect the differences in different development conditions in complex fractured gas reservoirs.

[0082] In multi-scheme evaluation scenarios, multiple candidate development schemes can be set for the same area to be predicted, and the well type, fracture morphology, and input parameters corresponding to each candidate development scheme can be determined. For each candidate development scheme, the corresponding recovery rate prediction value is obtained by following the process of dividing the target evaluation unit, determining the well type, determining the fracture morphology, matching the target recovery rate prediction model, and calculating the recovery rate prediction value. Then, the recovery rate prediction values ​​of each candidate development scheme are sorted to form the development scheme comparison results. This approach can reduce the workload of repeatedly building detailed numerical simulation models during the rapid on-site evaluation process and improve the efficiency of well type selection, well network adjustment, water control and production stabilization, and remaining gas potential tapping scheme screening.

[0083] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0084] Figure 1 This is a flowchart illustrating the method for predicting the recovery rate of fractured gas reservoirs provided in the embodiments of this application, as shown below. Figure 1 As shown, the method includes:

[0085] S11, Obtain the unit division data of the area to be predicted, and divide the area to be predicted into at least one target evaluation unit according to the unit division data; wherein, the unit division data includes at least one of geological boundary data, fracture attribute data, seepage connectivity data and well network control data.

[0086] In this embodiment, the area to be predicted is the region where the recovery rate of fractured gas reservoirs needs to be predicted. The unit partitioning data is a dataset used to determine the evaluation boundaries within the area to be predicted. It reflects at least one spatial difference among geological boundaries, fracture properties, seepage connectivity, and well network control relationships within the area to be predicted. By acquiring the unit partitioning data, a data foundation can be provided for subsequent division of target evaluation units, enabling recovery rate prediction to no longer be uniformly processed for the entire area to be predicted, but rather to form regional units suitable for independent evaluation based on the differences within the area to be predicted.

[0087] After acquiring the unit segmentation data, the area to be predicted is divided into at least one target evaluation unit based on the regional differences reflected in the unit segmentation data. A target evaluation unit refers to a basic regional unit within the area to be predicted, used for well type determination, fracture morphology determination, recovery rate prediction model matching, and recovery rate prediction calculation. When the differences within the area to be predicted are small and can be used for recovery rate prediction as a whole, the area to be predicted can be identified as one target evaluation unit. When differences exist within the area to be predicted, it can be divided into multiple target evaluation units, and subsequent calculations can be performed on each target evaluation unit separately. Converting the area to be predicted into at least one target evaluation unit through unit segmentation data allows for differentiation of geological conditions, fracture development, seepage connectivity, or well network control relationships at different locations within the area to be predicted in subsequent calculations, providing a foundation for determining well type, fracture morphology, and matching the recovery rate prediction model based on the target evaluation units.

[0088] S12. Based on the well type characteristic parameters of the target evaluation unit, determine the well type category of each target evaluation unit, and based on the fracture characteristic parameters of the target evaluation unit, determine the fracture morphology category of each target evaluation unit.

[0089] In this embodiment, well type characteristic parameters are data reflecting the wellbore morphology, wellbore layout, and development well control status within the target evaluation unit. Fracture characteristic parameters are data reflecting the fracture development status, fracture spatial distribution, fracture connectivity, and fracture impact on fluid migration within the target evaluation unit. By acquiring the well type characteristic parameters and fracture characteristic parameters of the target evaluation unit respectively, a data foundation can be provided for subsequent development type and fracture type identification of the target evaluation unit. When determining the well type category, for each target evaluation unit, its well type characteristic parameters are analyzed to identify the wellbore morphology and wellbore layout characteristics of the development wells within the target evaluation unit, and the corresponding well type category of the target evaluation unit is determined accordingly. For a prediction area including multiple target evaluation units, the well type category of each target evaluation unit is determined separately, so that the differences in development wells between different target evaluation units can be distinguished in the subsequent prediction process.

[0090] When determining fracture morphology categories, the fracture characteristic parameters of each target evaluation unit are processed to identify the development degree, spatial combination relationship, and connectivity of fractures within the target evaluation unit, and the corresponding fracture morphology category is determined accordingly. Well type categories are determined through well type characteristic parameters, and fracture morphology categories are determined through fracture characteristic parameters. This enables a categorized expression of the development conditions and fracture development conditions of the target evaluation unit, providing a basis for adopting appropriate prediction calculation paths for different target evaluation units.

[0091] S13. Obtain the target recovery rate prediction model corresponding to the well type and fracture morphology of the target evaluation unit from the preset recovery rate model database. Based on the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters and water body characteristic parameters of the target evaluation unit, use the target recovery rate prediction model to obtain the recovery rate prediction value of the target evaluation unit.

[0092] In this embodiment, the recovery rate model database is a pre-established set of models used to store recovery rate prediction models corresponding to different combinations of well types and fracture morphology categories. The target recovery rate prediction model refers to the recovery rate prediction model selected from the recovery rate model database that matches the well type and fracture morphology category of the current target evaluation unit. By pre-setting the recovery rate model database, recovery rate prediction calculation methods under different development conditions and fracture development conditions can be categorized and stored, providing a foundation for subsequent model calls. After determining the well type and fracture morphology category of the target evaluation unit, the well type and fracture morphology category are used as model retrieval conditions, and a match is performed in the pre-set recovery rate model database to obtain the target recovery rate prediction model corresponding to the target evaluation unit. This matching process ensures that the target evaluation unit no longer uses a uniform prediction model for calculation, but instead selects the appropriate prediction model based on its own well type and fracture morphology category, enabling the prediction process to reflect the differences between different development conditions and fracture development conditions.

[0093] After obtaining the target recovery rate prediction model, the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters of the target evaluation unit are used as model input data. The reservoir characteristic parameters reflect the basic physical properties and effective reservoir conditions within the target evaluation unit, while the water body characteristic parameters reflect the water energy within the target evaluation unit and its impact on the development process. By inputting these parameters into the target recovery rate prediction model, the model can simultaneously utilize information related to well type, fracture, reservoir, and water body for calculation. The target recovery rate prediction model calculates based on the input well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters, outputting the predicted recovery rate value for the target evaluation unit. This predicted recovery rate value represents the predicted recovery rate of the target evaluation unit under the corresponding well type category, fracture morphology category, and parameter conditions. This achieves the acquisition of the predicted recovery rate value for the target evaluation unit and provides a data foundation for subsequent recovery rate aggregation of the area to be predicted.

[0094] S14. Based on the recovery rate prediction values ​​of each target evaluation unit, obtain the recovery rate prediction values ​​of the area to be predicted according to the preset regional aggregation rules.

[0095] In this embodiment, the regional aggregation rule is a calculation rule used to convert the recovery rate prediction values ​​of each target evaluation unit into the overall recovery rate prediction value of the area to be predicted. Since the area to be predicted can be divided into one or more target evaluation units, and each target evaluation unit has a corresponding recovery rate prediction value, it is necessary to aggregate the prediction results of each target evaluation unit according to the preset regional aggregation rule in order to obtain a recovery rate prediction value that can represent the overall development effect of the area to be predicted.

[0096] When the area to be predicted includes only one target evaluation unit, the predicted recovery rate of that target evaluation unit can be used as the predicted recovery rate of the entire area. When the area includes multiple target evaluation units, the predicted recovery rate values ​​of these units can be aggregated regionally to obtain the predicted recovery rate for the entire area. This aggregation rule can be set based on the proportion or contribution of each target evaluation unit within the area, allowing the predicted recovery rate values ​​of different units to participate in the overall result calculation according to their influence within the area. By aggregating the predicted recovery rate values ​​of each target evaluation unit using preset regional aggregation rules, the transformation from local prediction results to overall prediction results is achieved. This processing method, while preserving the differentiated prediction results of each target evaluation unit, forms the overall predicted recovery rate for the entire area, providing a unified prediction result for the development evaluation and scheme comparison of the area.

[0097] In one embodiment, step S11 is further described below. Based on the above embodiment, the region to be predicted is divided into at least one target evaluation unit according to the unit division data, including:

[0098] S111, Based on geological boundary data, determine the structural zoning boundaries within the area to be predicted.

[0099] Specifically, geological boundary data is used to reflect the geological boundaries within the region to be predicted that constrain regional connectivity and evaluation boundaries. When determining the structural zoning boundaries within the region to be predicted based on geological boundary data, one or more of the following can be used as structural zoning boundaries: fault boundaries, tectonic high boundary lines, stratigraphic pinch-out lines, and geological model grid boundaries. Structural zoning boundaries are used to define the basic boundaries within the region to be predicted that should not be crossed or merged, ensuring that the subsequently formed target evaluation units are constrained by tectonic and stratigraphic boundaries. Determining structural zoning boundaries through geological boundary data ensures that the division of target evaluation units conforms to the basic geological pattern of the region to be predicted.

[0100] S112, Based on the crack attribute data, determine the crack partition boundaries and crack attribute categories of each crack partition within the area to be predicted.

[0101] Specifically, fracture attribute data is used to reflect the development and spatial distribution characteristics of fractures within the area to be predicted. When determining fracture zone boundaries and fracture attribute categories for each fracture zone based on fracture attribute data, the boundaries can be determined according to fracture spread characteristics, fracture connectivity, fracture density distribution, fracture penetration ratio distribution, and fracture conductivity distribution. Fracture spread characteristics can be derived from seismic interpretation, imaging logging interpretation, core fracture description, and discrete fracture network models; fracture connectivity can be determined by combining seismic fracture interpretation, well connectivity data, tracer monitoring data, and production dynamics data. Fracture attribute categories can be divided based on the value ranges of parameters such as fracture density, fracture penetration ratio, and fracture conductivity to distinguish areas with different fracture development levels and different fracture control characteristics. Determining fracture zone boundaries and fracture attribute categories through fracture attribute data allows fracture development differences to be incorporated into the subsequent target evaluation unit division process.

[0102] S113, Based on the seepage connectivity data, determine the seepage zone boundaries and seepage connectivity categories of each seepage zone within the area to be predicted.

[0103] Specifically, seepage connectivity data is used to reflect the direction of fluid migration, connectivity relationships, and water intrusion response characteristics within the area to be predicted. When determining the boundaries of seepage zones and the seepage connectivity categories of each zone based on seepage connectivity data, the boundaries can be determined according to the well-to-well tracer response range, water intrusion channels, water intrusion direction, water intrusion velocity, water intrusion propagation range, and well-to-well connectivity relationships. Tracer monitoring data can be used to identify the strength of well-to-well connectivity and dominant channels; water intrusion channels and water intrusion propagation range can be used to identify the main paths of edge water or bottom water propagating into the gas-bearing area; and production dynamics data can be used to assist in judging pressure response, water production response, and well-to-well interference relationships. Seepage connectivity categories can be determined based on water intrusion direction, water intrusion velocity, strength of well-to-well connectivity, and fluid migration channel type. This step determines the seepage zone boundaries and seepage connectivity categories through seepage connectivity data, ensuring that the division of target evaluation units reflects the consistency of flow units.

[0104] S114. Based on the well network control data, determine the well control zone boundaries and well network control categories of each well control zone within the area to be predicted.

[0105] Specifically, well network control data is used to reflect the control range and production response characteristics of development wells on the reservoir within the area to be predicted. When determining the well control zone boundaries and well network control categories for each well control zone based on the well network control data, the well control zone boundaries can be determined according to the well control range, well network control method, and production dynamic control range. The well control range can be determined based on well location deployment, well spacing, single-well control area, or well group control range; the well network control method can be determined based on vertical wells, horizontal wells, or different well network deployment methods; the production dynamic control range can be determined based on gas production, water production, pressure changes, and production contribution range. The well network control category can be determined based on the well network control method, well control intensity, and production dynamic characteristics. Determining the well control zone boundaries and well network control categories through well network control data ensures that the target evaluation unit division matches the actual development control range.

[0106] S115, the structural partition boundary, fracture partition boundary, seepage partition boundary and well control partition boundary are superimposed to obtain at least one initial partitioning unit corresponding to the area to be predicted.

[0107] Specifically, the boundaries of structural zones, fracture zones, seepage zones, and well-controlled zones are unified within the spatial range of the area to be predicted and then overlaid. Overlay processing refers to combining zones from different sources according to their spatial location, allowing different boundaries to jointly divide the area to be predicted. Through overlay processing, at least one initial partitioning unit corresponding to the area to be predicted can be obtained. The initial partitioning unit is a basic area jointly defined by the boundaries of structural zones, fracture zones, seepage zones, and well-controlled zones, enabling different types of regional control factors to be uniformly expressed in the same spatial partitioning result.

[0108] S116, based on the fracture zone, seepage zone, and well control zone where the initial division unit is located, determine the fracture attribute category, seepage connectivity category, and well network control category of the initial division unit.

[0109] Specifically, if an initial partitioning unit is entirely located within a fracture zone, the fracture attribute category of that fracture zone is assigned to that initial partitioning unit; if an initial partitioning unit is entirely located within a seepage zone, the seepage connectivity category of that seepage zone is assigned to that initial partitioning unit; if an initial partitioning unit is entirely located within a well-controlled zone, the well network control category of that well-controlled zone is assigned to that initial partitioning unit. By assigning categories to the initial partitioning units, fracture attributes, seepage connectivity relationships, and well network control relationships can be transformed into data that can be used for subsequent merging and judgment. When assigning categories, the fracture attribute category can be determined based on at least one of fracture density, fracture penetration ratio, fracture conductivity, fracture orientation combination relationship, and fracture intersection relationship; the seepage connectivity category can be determined based on at least one of water intrusion direction, water intrusion velocity, inter-well connectivity relationship, tracer response, pressure response, and production dynamics; the well network control category can be determined based on at least one of well network density, well spacing, well control range, production well distribution pattern, and development well control relationship within the target evaluation unit. This provides a unified category classification basis for the subsequent merging and judgment of adjacent initial partitioning units.

[0110] S117, traverse each initial division unit in the region to be predicted, obtain any pair of adjacent initial division units, and when the fracture attribute category, seepage connectivity category and well network control category of any pair of adjacent initial division units are the same, merge any pair of adjacent initial division units to obtain at least one target evaluation unit.

[0111] Specifically, the fracture attribute category, seepage connectivity category, and well pattern control category of any pair of adjacent initial partitioning units are compared. When adjacent initial partitioning units have the same fracture attribute category, seepage connectivity category, and well pattern control category, they are merged to obtain the target evaluation unit. This merging process can continue until no adjacent initial partitioning units satisfy the merging conditions. Initial partitioning units that do not satisfy the merging conditions are retained as independent target evaluation units. Through category consistency judgment and adjacent unit merging, the target evaluation units are kept consistent in terms of fracture attributes, seepage connectivity relationships, and well pattern control relationships, providing a stable regional object for subsequent well type category determination, fracture morphology category determination, and recovery rate prediction calculation.

[0112] In one embodiment, well type characteristic parameters include well inclination data, well trajectory data, horizontal section length, and completion method. These parameters describe the well trajectory and completion / production characteristics of the target well within the target evaluation unit. Well inclination data refers to the wellbore inclination recorded along the well depth direction; well trajectory data describes the spatial trajectory of the target well; horizontal section length refers to the length of a continuous well section within the well trajectory that meets the criteria for a horizontal well section; and completion method refers to the completion type and location of the target well within the reservoir section. By acquiring well inclination data, well trajectory data, horizontal section length, and completion method, the well type category of the target evaluation unit can be determined from both the wellbore spatial morphology and the actual production location.

[0113] In one embodiment, determining the well type category of each target evaluation unit based on its well type characteristic parameters includes:

[0114] S1211, Obtain the well type characteristic parameters of the target well within the target evaluation unit. The well type characteristic parameters include well inclination data, well trajectory data, horizontal section length, and well completion method.

[0115] S1212, Based on the well inclination data and well trajectory data, determine whether there is a continuous well section in the well trajectory of the target well with an inclination angle greater than the preset inclination angle threshold;

[0116] S1213, If there is a continuous well section with an inclination angle greater than the preset inclination angle threshold, then the continuous well section is determined as a candidate horizontal section;

[0117] S1214, When the length of the candidate horizontal segment is greater than the preset length threshold, determine whether the candidate horizontal segment is the completion segment or the producing segment of the target well according to the completion method of the target well.

[0118] S1215, when the candidate horizontal segment is the completion segment or producing segment of the target well, the well type category of the target evaluation unit is determined to be the horizontal well category.

[0119] In this embodiment, when acquiring the well type characteristic parameters of the target well within the target evaluation unit, well inclination data, well trajectory data, horizontal section length, and completion method can be extracted from drilling design data, actual drilling trajectory data, directional survey data, completion data, and producing section identification data. Well trajectory data may include well depth, vertical depth, horizontal displacement, and the spatial location of the wellbore trajectory. Well inclination data can be used in conjunction with well trajectory data to determine the wellbore morphology of the target well in different well sections. The completion method can be used to confirm whether the corresponding well section in the wellbore trajectory has actually participated in development and production, avoiding the situation where well type classification is determined solely based on wellbore trajectory morphology while ignoring the completion and production status.

[0120] Based on well inclination and well trajectory data, the system sequentially checks along the target well's trajectory to determine if there are continuous well segments with inclination angles greater than a preset inclination angle threshold. The preset inclination angle threshold distinguishes between high-inclination and non-high-inclination continuous well segments, and can be pre-set according to target gas reservoir well type identification standards or on-site engineering evaluation standards. When a continuous well segment with an inclination angle greater than the preset inclination angle threshold exists in the well trajectory, this continuous well segment is identified as a candidate horizontal segment. A candidate horizontal segment refers to a well segment that meets the preliminary identification conditions for a horizontal well segment in terms of well trajectory morphology; further evaluation is needed based on length conditions and well completion / production conditions.

[0121] After identifying candidate horizontal segments, their lengths are compared with a preset length threshold. The preset length threshold is used to determine if the candidate horizontal segment has sufficient horizontal extension to avoid misclassifying short-distance, high-angle well segments as horizontal segments. When the length of a candidate horizontal segment exceeds the preset length threshold, it is further determined whether the candidate horizontal segment is a completion segment or a producing segment of the target well based on the target well's completion method. A completion segment refers to a well segment in the target well that has been completed and is ready for development; a producing segment refers to a segment identified based on production profiles, test data, or production dynamics data that participates in gas or water production.

[0122] When a candidate horizontal segment simultaneously meets the criteria of having a length greater than a preset length threshold and being a completion segment or producing segment of the target well, the well type of the target evaluation unit is determined to be a horizontal well. This determination method combines wellbore trajectory morphology, horizontal extension length, and actual completion and production location, enabling the well type to reflect the actual development method of the target well within the target evaluation unit. Through the joint determination of well inclination data, well trajectory data, horizontal segment length, and completion method, the well type of the target evaluation unit is identified, providing a classification basis for subsequent matching of the target recovery rate prediction model based on the well type.

[0123] Next, if there is no continuous well section with an inclination angle greater than the preset inclination angle threshold, the well type category of the target evaluation unit is vertical well. That is, when it is determined from the well inclination data and well trajectory data that there is no continuous well section with an inclination angle greater than the preset inclination angle threshold in the well trajectory of the target well, it indicates that the target well has not formed a high-inclination continuous well section that meets the preliminary identification conditions for horizontal well sections. At this time, the well type category of the target evaluation unit is determined to be vertical well.

[0124] In one embodiment, fracture characteristic parameters include at least one of fracture density, fracture penetration ratio, fracture azimuth dispersion parameter, fracture intersection parameter, fracture connectivity parameter, and fracture conductivity parameter. These fracture characteristic parameters describe the number, development level, spatial extension, azimuth combination relationship, intersection relationship, connectivity, and conductivity of fractures within the target evaluation unit. Fracture density reflects the intensity of fracture development within the target evaluation unit; fracture penetration ratio reflects the extension of candidate fracture networks relative to the target evaluation unit; fracture azimuth dispersion parameter reflects whether fracture azimuths are concentrated or whether they have multiple sets of azimuth characteristics; fracture intersection parameter reflects whether different fractures form an interwoven relationship; fracture connectivity parameter reflects the degree of connectivity between fractures and between fractures and the well; and fracture conductivity parameter reflects the contribution of fractures to fluid transport.

[0125] In one embodiment, determining the crack morphology category of each target evaluation unit based on its crack characteristic parameters includes:

[0126] S1201, determine the fracture density of the target evaluation unit based on at least one of core fracture statistics, imaging logging interpretation data, and discrete fracture network model data.

[0127] Specifically, core fracture statistics can be used to identify the number of fractures, fracture width, fracture filling state, and fracture intersection relationships within the cored section; imaging logging interpretation data can be used to identify the number, azimuth, dip angle, and opening degree of fractures near the wellbore; discrete fracture network model data can be used to describe the planar and spatial distribution of fractures within the target evaluation unit. For the wellbore scale, fracture density can be calculated based on the number of fractures per unit well length; for the target evaluation unit scale, fracture density can be determined as the ratio between the total fracture length and the area of ​​the target evaluation unit. When identifying fracture networks or determining fracture network morphology, the fracture density is preferably the fracture density at the target evaluation unit scale, or an equivalent fracture density obtained by weighting multiple fracture development intervals. In other words, when determining whether a fracture network exists within the target evaluation unit, the fracture density reflects the overall fracture development intensity within the target evaluation unit, rather than just the number of local fractures near a single wellbore. This allows the calculation scope of fracture density to match the identification scenario of fracture network morphology.

[0128] S1202, based on at least one of seismic fracture interpretation data, imaging logging fracture orientation interpretation data, well connectivity data, tracer monitoring data, and production dynamic data, determine candidate fracture connectivity within the target evaluation unit, and determine the extension length of the candidate fracture connectivity.

[0129] Specifically, candidate fracture connectivity refers to fracture bodies within the target evaluation unit that exhibit continuous distribution or connectivity and may form dominant flow channels. Seismic fault interpretation data can be used to identify the extension direction and spatial distribution range of faults or fracture zones; imaging logging fracture orientation interpretation data can be used to correct the development orientation of fractures near the wellbore; inter-well connectivity data and tracer monitoring data can be used to determine whether fractures have formed connections between wells; and production dynamics data can be used to assist in assessing the impact of fracture connectivity on pressure response, gas production response, or water production response. After identifying candidate fracture connectivity using the above data, the extension length of the candidate fracture connectivity within the target evaluation unit can be further determined.

[0130] S1203, determine the crack penetration ratio of the candidate crack interconnection based on the ratio of the extension length of the candidate crack interconnection to the characteristic length of the target evaluation unit.

[0131] Specifically, the characteristic length of the target evaluation unit can be determined based on the well spacing, model side length, well control range, or the length of the gas-bearing zone in the main water invasion direction. This ratio can be used to determine whether candidate fracture interconnections possess characteristics of penetrating or nearly penetrating the target evaluation unit, providing a quantitative basis for subsequent identification of the main fracture and characterizing the degree of fracture spatial penetration.

[0132] S1204. Based on at least one of imaging logging interpretation data, core fracture description data, and seismic fracture interpretation data, determine the azimuth data of fractures within the target evaluation unit, and determine the fracture azimuth dispersion parameters based on the azimuth data.

[0133] Specifically, imaging logging interpretation data can provide the orientation of fractures near the wellbore, core fracture description data can provide information on fracture orientation and fracture assemblage, and seismic fracture interpretation data can provide fracture propagation directions on a larger scale. Based on this orientation data, it can be determined whether the fracture orientations within the target evaluation unit are concentrated in one main direction or whether multiple directions are developing together. The fracture orientation dispersion parameter is used to subsequently distinguish between the development of a main fracture with concentrated orientations and the development of multiple fracture groups.

[0134] S1205, based on at least one of imaging logging interpretation data, core fracture description data, and discrete fracture network model data, determine the cross-cutting relationship between different fractures within the target evaluation unit, and determine the fracture cross-cutting parameters based on the cross-cutting relationship.

[0135] Specifically, fracture intersection relationships can reflect whether multiple fractures intertwine and form a networked interconnected structure. Core fracture description data can be used to identify the cutting and associated relationships between fractures, imaging logging interpretation data can be used to identify the intersection of multiple sets of fractures near the wellbore, and discrete fracture network model data can be used to describe the spatial intersection and connectivity between different fractures within the target evaluation unit. Fracture intersection parameters can provide a basis for identifying fracture networks.

[0136] S1206. Based on at least one of the following: well test response data, well connectivity data, tracer monitoring data, lost circulation segment identification data, producing layer identification data, water intrusion channel data, and production dynamic data, determine fracture connectivity parameters and fracture conductivity parameters.

[0137] Specifically, well test response data can be used to reflect pressure propagation velocity, seepage morphology, and connectivity; well connectivity data and tracer monitoring data can be used to reflect the existence of rapid connectivity channels between wells; lost circulation segment identification data can be used to reflect the degree of fracture opening and conductivity during drilling or operations; produced zone identification data can be used to reflect the contribution of fractures to gas or water production; and water intrusion channel data and production dynamics data can be used to determine whether fractures control water migration and water production response. By determining fracture connectivity parameters and fracture conductivity parameters using the above data, fracture morphology classification can be based not only on static fracture descriptions but also on fluid migration and production response.

[0138] S1207, when the crack penetration ratio of the candidate crack body meets the preset penetration condition, the crack flow capacity parameter meets the preset flow condition, and the crack orientation dispersion parameter meets the preset orientation concentration condition, the candidate crack body is identified as the main crack.

[0139] Specifically, the identification of a candidate fracture network as a main fracture is determined based on its fracture penetration ratio, fracture conductivity parameters, and fracture azimuth dispersion parameters. When the fracture penetration ratio of a candidate fracture network meets a preset penetration condition, the fracture conductivity parameters meet a preset conductivity condition, and the fracture azimuth dispersion parameters meet a preset azimuth concentration condition, it indicates that the candidate fracture network has high extension, strong conductivity, and concentrated azimuth, and is therefore identified as a main fracture. The main fracture is used to characterize the fracture body within the target evaluation unit that has a dominant conductivity effect. If no candidate fracture network exists, or if the candidate fracture network does not meet at least one of the preset penetration condition, preset conductivity condition, and preset azimuth concentration condition, it is determined that no main fracture has been identified.

[0140] S1208, when the crack density meets the preset density condition, the crack orientation dispersion parameter meets the preset multi-directional condition, and the crack intersection parameter meets the preset intersection condition, the crack network within the target evaluation unit is identified.

[0141] Specifically, the presence of a crack network within the target evaluation unit is determined based on crack density, crack azimuth dispersion parameters, and crack intersection parameters. When the crack density meets a preset density condition, the crack azimuth dispersion parameters meet a preset multi-azimuth condition, and the crack intersection parameters meet a preset intersection condition, it indicates that the target evaluation unit contains a large number of cracks, multiple azimuths, and intersecting relationships between different cracks; therefore, a crack network is identified within the target evaluation unit. A crack network characterizes the interconnected structure formed by multiple interwoven cracks, which differs from a single dominant flow-conducting crack. If at least one of the crack density, crack azimuth dispersion parameters, and crack intersection parameters does not meet the corresponding conditions, a crack network is determined not to have been identified.

[0142] Fracture morphology is determined by combining multiple dimensions of parameters, including fracture density, penetration ratio, azimuth dispersion, and intersection parameters. For example, a fracture network type requires high fracture density, large azimuth dispersion, and high degree of intersection, while a large fracture type requires high penetration ratio, strong conductivity, and concentrated azimuth. This classification logic ensures that the fracture morphology category is consistent with the actual fracture development pattern, matches the corresponding recovery rate prediction model (such as the model differences between fracture network and large fracture types), and improves the accuracy of the prediction results.

[0143] S1209, when a main crack is identified in the target evaluation unit but no crack network is identified, the crack morphology category of the target evaluation unit is determined as a large crack type.

[0144] S1210, when a crack network is identified in the target evaluation unit but no main crack is identified, the crack morphology category of the target evaluation unit is determined as crack network type.

[0145] S1211, When the main crack and crack network are identified in the target evaluation unit, the crack morphology category of the target evaluation unit is determined as large crack network coupling type.

[0146] Specifically, the fracture morphology category of the target evaluation unit is determined based on the identification results. When a main fracture is identified in the target evaluation unit but no fracture network is identified, the fracture morphology category of the target evaluation unit is determined to be a large fracture type; when a fracture network is identified in the target evaluation unit but no main fracture is identified, the fracture morphology category of the target evaluation unit is determined to be a fracture network type; when both a main fracture and a fracture network are identified in the target evaluation unit, the fracture morphology category of the target evaluation unit is determined to be a large fracture fracture network coupled type. By separately identifying the main fracture and the fracture network, the fracture morphology category of the target evaluation unit is classified and determined, providing a basis for subsequent matching of the target recovery rate prediction model based on the fracture morphology category.

[0147] In one specific embodiment, the preset penetration condition is: the crack penetration ratio of the candidate crack interconnect is greater than or equal to a preset penetration threshold. The crack penetration ratio is the ratio of the extension length of the candidate crack interconnect to the characteristic length of the target evaluation unit. When the crack penetration ratio is greater than or equal to the preset penetration threshold, the candidate crack interconnect is considered to have a high degree of extension within the target evaluation unit, thus satisfying the preset penetration condition.

[0148] The preset conductivity condition is that the fracture conductivity parameter is greater than or equal to the preset conductivity threshold. The fracture conductivity parameter can be fracture permeability, fracture aperture, the product of fracture permeability and fracture aperture, or an equivalent conductivity parameter determined based on well test response data, lost circulation segment identification data, and production dynamic data. When the fracture conductivity parameter is greater than or equal to the preset conductivity threshold, the candidate fracture connectivity is considered to have superior conductivity and meets the preset conductivity condition.

[0149] The preset orientation concentration condition is as follows: the orientation dispersion parameter of the candidate fracture connected body is less than or equal to the preset orientation dispersion threshold, or the proportion of fractures located within the same dominant orientation range in the candidate fracture connected body is greater than or equal to the preset orientation concentration threshold. When either of the above conditions is met, the fracture orientation of the candidate fracture connected body is considered to be relatively concentrated, thus satisfying the preset orientation concentration condition.

[0150] The preset density condition is: the fracture density of the target evaluation unit is greater than or equal to a preset density threshold. Fracture density can be the number of fractures per unit well length, or the ratio of the total fracture length to the area of ​​the target evaluation unit. When the fracture density is greater than or equal to the preset density threshold, the fracture development level within the target evaluation unit is considered high, satisfying the preset density condition.

[0151] The preset multi-directional condition is as follows: the crack orientation dispersion parameter within the target evaluation unit is greater than or equal to the preset multi-directional threshold, or the number of dominant crack orientation groups within the target evaluation unit is greater than or equal to the preset threshold for the number of orientation groups. When either of the above conditions is met, it is considered that there are multiple sets of crack orientations within the target evaluation unit, thus satisfying the preset multi-directional condition.

[0152] The preset cross-cutting condition is that the cross-cutting parameter is greater than or equal to the preset cross-cutting threshold. The cross-cutting parameter can be the proportion of the number of cross-cutting cracks to the total number of cracks, the number of cross-cutting points, the number of cross-cutting points per unit area, or a cross-cutting degree parameter determined according to the discrete crack network model. When the cross-cutting parameter is greater than or equal to the preset cross-cutting threshold, it is considered that there is a significant cross-cutting relationship between different cracks within the target evaluation unit, thus satisfying the preset cross-cutting condition.

[0153] It should be further noted that the above thresholds can be preset or corrected based on the core fracture statistics, imaging logging interpretation data, seismic fracture interpretation data, well test response data, tracer monitoring data, and production dynamic data of the area to be predicted.

[0154] In one embodiment, step S13 is further described below. Based on the above embodiment, obtaining the target recovery rate prediction model corresponding to the well type category and fracture morphology category of the target evaluation unit includes:

[0155] In this embodiment, the preset recovery rate model database includes a first recovery rate prediction model, a second recovery rate prediction model, a third recovery rate prediction model, a fourth recovery rate prediction model, a fifth recovery rate prediction model, and a sixth recovery rate prediction model. Each recovery rate prediction model corresponds to a combination of different well types and different fracture morphology types. The natural logarithm function is denoted as ln, and the natural exponential function is denoted as exp. In each model, R is the predicted recovery rate value of the target evaluation unit, S is the well network density, and k is the recovery rate prediction value. f Let w be the fracture permeability, and k be the fracture aperture. m R is the matrix permeability, h is the effective reservoir thickness, and R is the effective reservoir thickness. f D is the crack penetration ratio. f h is the crack density. f Where W is the height of the crack and W is the multiple of the water volume.

[0156] When the well type of the target evaluation unit is horizontal well and the fracture morphology type is large fracture, the first recovery rate prediction model is obtained. The first recovery rate prediction model is as follows:

[0157]

[0158] When the well type of the target evaluation unit is a vertical well and the fracture morphology is a large fracture, the second recovery rate prediction model is obtained. The second recovery rate prediction model is as follows:

[0159]

[0160] When the well type of the target evaluation unit is horizontal well and the fracture morphology is fracture network, the third recovery rate prediction model is obtained. The third recovery rate prediction model is as follows:

[0161]

[0162] When the well type of the target evaluation unit is vertical well and the fracture morphology is fracture mesh, the fourth recovery rate prediction model is obtained. The fourth recovery rate prediction model is as follows:

[0163]

[0164] When the well type of the target evaluation unit is a horizontal well and the fracture morphology is a large fracture network coupled type, the fifth recovery rate prediction model is obtained. The fifth recovery rate prediction model is as follows:

[0165]

[0166] When the well type of the target evaluation unit is a vertical well and the fracture morphology is a large fracture network coupled type, the sixth recovery rate prediction model is obtained. The sixth recovery rate prediction model is as follows:

[0167]

[0168] Using the six recovery rate prediction models described above, the pre-set recovery rate model database can correspond to six combinations: horizontal well type with large fracture type, vertical well type with large fracture type, horizontal well type with fracture network type, vertical well type with fracture network type, horizontal well type with large fracture network coupled type, and vertical well type with large fracture network coupled type. After the target evaluation unit completes the well type and fracture morphology identification, it can obtain the corresponding target recovery rate prediction model from the pre-set recovery rate model database, and substitute the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters of the target evaluation unit into the corresponding model to obtain the recovery rate prediction value of the target evaluation unit.

[0169] In one embodiment, step S14 is implemented as follows. Based on the above embodiment, the predicted recovery rate of the region to be predicted is obtained according to a preset regional aggregation rule, based on the predicted recovery rate of each target evaluation unit, including:

[0170] S141, when the area to be predicted includes a target evaluation unit, the predicted recovery rate of the target evaluation unit is determined as the predicted recovery rate of the area to be predicted.

[0171] S142, When the area to be predicted includes multiple target evaluation units, obtain the geological reserves, control reserves, or effective pore volume of the target evaluation units, and use the formula:

[0172]

[0173] Calculate and obtain the predicted recovery rate R for the area to be predicted. total Among them, R i Let G be the predicted recovery rate of the i-th target evaluation unit. i Let n represent the geological reserves, controlled reserves, or effective pore volume of the i-th target evaluation unit, and n be the number of target evaluation units.

[0174] In this embodiment, a preset regional aggregation rule is used to convert the recovery rate prediction values ​​of each target evaluation unit into the overall recovery rate prediction value of the area to be predicted. The area to be predicted may contain only one target evaluation unit or multiple target evaluation units. When the area to be predicted contains only one target evaluation unit, that target evaluation unit represents the overall evaluation object of the area to be predicted, so the recovery rate prediction value of that target evaluation unit is directly determined as the recovery rate prediction value of the area to be predicted.

[0175] When the area to be predicted includes multiple target evaluation units, each target evaluation unit has a corresponding predicted recovery rate value. Since the reserve size, control range, or effective pore volume of different target evaluation units within the area to be predicted may differ, directly averaging the predicted recovery rate values ​​of each target evaluation unit may not reflect the differences in their contribution to the overall recovery rate of the area. Therefore, in the case of multiple target evaluation units, the geological reserves, controlled reserves, or effective pore volume of each target evaluation unit are obtained and used as weighting parameters for the regional aggregation.

[0176] When performing regional aggregation, it is preferable to use the same type of weighting parameter for each target evaluation unit. That is, when geological reserves are selected as the weighting parameter, each target evaluation unit uses its own geological reserves; when controlled reserves are selected as the weighting parameter, each target evaluation unit uses its own controlled reserves; and when effective pore volume is selected as the weighting parameter, each target evaluation unit uses its own effective pore volume. By using the same type of weighting parameter, a unified weighting basis can be ensured for the regional aggregation calculation, avoiding distortion of the predicted recovery rate values ​​for the area to be predicted due to the mixing of different weighting types.

[0177] Geological reserves can be determined based on reserve calculation data such as the gas-bearing area, effective reservoir thickness, porosity, gas saturation, and gas volume coefficient of the target evaluation unit. Controlled reserves can be determined based on the well control range, well network deployment, and single-well controlled reserves of development wells within the target evaluation unit. Effective pore volume can be determined based on the effective area, effective reservoir thickness, porosity, and effective gas-bearing range of the target evaluation unit. The aforementioned geological reserves, controlled reserves, or effective pore volume can be derived from geological models, reserve evaluation results, well control reserve calculation results, or databases.

[0178] The calculation first involves multiplying the predicted recovery rate of each target evaluation unit by its corresponding weighting parameter, then summing these products to obtain the weighted recovery rate parameter. Simultaneously, the weighting parameters of each target evaluation unit are summed to obtain the regional aggregate base. Finally, the ratio of the weighted recovery rate parameter to the regional aggregate base is determined as the predicted recovery rate for the region to be predicted. This method is suitable for situations where there are multiple fracture morphologies, multiple well network control areas, or multiple flow-connected regions within the region to be predicted. It can generate an overall recovery rate prediction result for the region to be predicted based on the recovery rate predictions of each target evaluation unit.

[0179] In one specific embodiment, well type characteristic parameters include well pattern density; fracture characteristic parameters include fracture permeability, fracture aperture, fracture penetration ratio, fracture density, and fracture height; reservoir characteristic parameters include matrix permeability and effective reservoir thickness; and water body characteristic parameters include water volume ratio. After acquiring these data, the units of well pattern density, fracture permeability, fracture aperture, fracture penetration ratio, fracture density, fracture height, matrix permeability, effective reservoir thickness, and water volume ratio are standardized, dimensions are checked, and applicability is verified.

[0180] When parameters fail the scope of application verification, these parameters are marked as anomalies and corrected based on at least one of the following: analog data from adjacent wells, well test interpretation data, geological statistics data, and production dynamic fitting data. When missing parameters exist, they are supplemented based on at least one of the following: analog data from adjacent wells, well test interpretation data, geological statistics data, and production dynamic fitting data. When the target evaluation unit contains multiple fractured segments, fracture permeability, fracture aperture, fracture density, and fracture height are weighted according to the effective thickness or production contribution of each fractured segment to obtain equivalent fracture permeability, equivalent fracture aperture, equivalent fracture density, and equivalent fracture height. When the target evaluation unit contains multiple main fractures, the equivalent fracture penetration ratio is calculated using the maximum main fracture length or the conductivity-weighted length. The verified or equivalent well network density, fracture permeability, fracture aperture, fracture penetration ratio, fracture density, fracture height, matrix permeability, effective reservoir thickness, and water volume multiple are input into the target recovery prediction model to calculate the predicted recovery value of the target evaluation unit.

[0181] In this embodiment, unit unification refers to converting parameters from different data sources into a unified engineering unit system used in the recovery rate prediction model. For example, fracture permeability and matrix permeability are unified as Darcy (D), fracture aperture, fracture height, and effective reservoir thickness are unified as meters, and fracture density is unified as fractures per meter. Dimensionality checking is used to determine whether the input parameters are consistent with the meaning of variables in the recovery rate prediction model, avoiding the mixing of area parameters, length parameters, permeability parameters, or dimensionless parameters.

[0182] The scope of application verification is used to determine whether the input parameters are within the applicable range of the recovery prediction model. As one example, the applicable range for well density can be 0.5–3 wells per square kilometer, the applicable range for fracture permeability can be 1–500 Darcy, the applicable range for fracture aperture can be 0.005–0.01 meters, and the applicable range for matrix permeability can be 1 × 10⁻⁶ meters. -5 ~1×10 -3 For Darcy, the applicable ranges for fracture penetration ratio are 0.5–0.9, fracture density is 0.0065–0.1 fractures per meter, water volume multiple is 1–20, and fracture height is 50–150 meters. The effective reservoir thickness can be determined based on the interpretation results of the effective reservoir within the target evaluation unit. By validating the applicable ranges, parameters that significantly exceed the model's applicable conditions can be identified, providing a basis for subsequent parameter correction.

[0183] When parameters fail the scope of application verification, they are marked as anomalies. Anomaly marking indicates that the parameter needs reinterpretation, correction, or review. For marked parameters, correction can be performed based on at least one of the following: adjacent well analogy data, well test interpretation data, geological statistics data, and production dynamics fitting data. Adjacent well analogy data can be used to select reference values ​​for parameters under similar structural locations, reservoir conditions, or fracture development conditions; well test interpretation data can be used to correct parameters related to fracture permeability, matrix permeability, and fracture connectivity; geological statistics data can be used to correct spatial distribution parameters such as fracture density, fracture height, and effective reservoir thickness; production dynamics fitting data can dynamically correct fracture parameters and water volume ratios through changes in pressure, gas production, water production, water breakthrough time, and water-gas ratio. Through anomaly marking and parameter correction, the parameters input to the recovery prediction model are matched with the actual development response of the target evaluation unit.

[0184] When missing parameters exist, they are supplemented using at least one of the following: analogy data from adjacent wells, well test interpretation data, geological statistics, and production dynamic fitting data. For missing well density, it can be determined based on the number of effective producing wells and the evaluation area within the target evaluation unit, such as the ratio of the number of effective producing wells N to the evaluation area A, i.e., well density S = N / A. For missing fracture permeability, it can be determined based on well test interpretation results, fracture aperture estimation results, or production dynamic fitting results. For missing fracture aperture, it can be determined based on imaging logging interpretation, core fracture description, fracture conductivity inversion, or production dynamic fitting results. For missing water volume multiples, they can be determined based on geological models, material balance calculations, water intrusion response, or production dynamic fitting results. Alternatively, they can be inverted based on the relationship between production, pressure changes, and water intrusion, i.e., W = V. a / V gW represents the multiple of water volume, V a V is the effective pore volume of the water body. g This refers to the effective pore volume of the gas-bearing zone. The effective pore volume of a water body can be determined based on the extent of edge or bottom water, the effective thickness of the reservoir, porosity, and the connectivity of the water body; the effective pore volume of a gas-bearing zone can be determined based on the gas-bearing area, the effective thickness of the reservoir, porosity, and gas saturation.

[0185] When a target evaluation unit contains multiple fractured segments, fracture permeability, fracture aperture, fracture density, and fracture height are weighted to obtain equivalent parameters that represent the overall fracture characteristics of the target evaluation unit. Weighting can be performed based on the effective thickness of each fractured segment or on its productivity contribution. When weighted by effective thickness, the equivalent value of a fracture parameter can be obtained by summing the products of the parameter values ​​of each fractured segment and their corresponding effective thicknesses, and then dividing by the sum of the effective thicknesses of all fractured segments. When weighted by productivity contribution, the equivalent value of a fracture parameter can be obtained by summing the products of the parameter values ​​of each fractured segment and their corresponding productivity contributions, and then dividing by the sum of the productivity contributions of all fractured segments. This processing yields equivalent fracture permeability, equivalent fracture aperture, equivalent fracture density, and equivalent fracture height, allowing the fracture development of multiple segments to be input into the recovery prediction model with unified parameters.

[0186] When a target evaluation unit contains multiple main fractures, the fracture penetration ratio is equivalently processed. As one embodiment, the equivalent fracture penetration ratio can be calculated using the maximum length of the main fracture among the multiple main fractures; that is, the ratio of the maximum main fracture length to the characteristic length of the target evaluation unit is used as the equivalent fracture penetration ratio. As another embodiment, the equivalent fracture penetration ratio can be calculated using a conductivity-weighted length; that is, a conductivity-weighted length is determined based on the extension length and corresponding conductivity of each main fracture, and then the ratio of the conductivity-weighted length to the characteristic length of the target evaluation unit is used as the equivalent fracture penetration ratio.

[0187] After completing unit unification, dimension checking, scope verification, abnormal parameter correction, missing parameter supplementation, and equivalent parameter calculation, the verified or equivalent well network density, fracture permeability, fracture aperture, fracture penetration ratio, fracture density, fracture height, matrix permeability, effective reservoir thickness, and water volume multiple are input into the recovery rate prediction model. For target evaluation units without multiple fractured sections or multiple main fractures, the verified original parameters can be input into the recovery rate prediction model; for target evaluation units with multiple fractured sections or multiple main fractures, the corresponding equivalent parameters are input into the recovery rate prediction model. The recovery rate prediction model calculates based on the above parameters to obtain the predicted recovery rate value for the target evaluation unit. Through parameter preprocessing, correction, and equivalence, the data input into the model can reflect the actual fracture development, reservoir conditions, water energy, and well network deployment of the target evaluation unit.

[0188] In one embodiment, Figure 2 This is a schematic flowchart illustrating a method for establishing a recovery rate model database, provided in an embodiment of this application. Based on the above embodiments, as follows... Figure 2 As shown, it includes:

[0189] S21, Obtain geological interpretation data and development dynamic data of fractured gas reservoirs, and obtain fracture development patterns based on the geological interpretation data and development dynamic data;

[0190] S22, Based on fracture development patterns, establish gas reservoir development models under different well types and combinations of different fracture morphology categories;

[0191] S23, adjust the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters and water body characteristic parameters in the gas reservoir development model to obtain multiple sets of parameter combinations;

[0192] S24, Numerical simulations were performed on each set of parameter combinations to obtain the recovery rate response samples corresponding to each set of parameter combinations;

[0193] S25. The recovery rate response samples are classified according to the well type and fracture morphology, and regression processing is performed on the classified recovery rate response samples to obtain the recovery rate prediction model corresponding to each combination of well type and fracture morphology.

[0194] S26. Establish a pre-set recovery rate model database based on each recovery rate prediction model.

[0195] In this embodiment, the recovery rate model database is pre-established before performing recovery rate prediction for the area to be predicted. The recovery rate model database is used to store recovery rate prediction models under different combinations of well types and fracture morphology categories, so that subsequent target evaluation units can directly call the corresponding target recovery rate prediction model after determining the well type and fracture morphology category, without having to carry out complete modeling and regression processing for each target evaluation unit again.

[0196] When establishing a recovery rate model database, geological interpretation data and development dynamic data for fractured gas reservoirs are first acquired. Geological interpretation data can include seismic interpretation data, well logging interpretation data, imaging well logging interpretation data, core fracture description data, reservoir distribution interpretation data, fracture interpretation data, and fracture prediction data. Development dynamic data can include formation pressure data, gas production data, water production data, water-to-gas ratio data, water breakthrough time data, well test response data, well connectivity data, and tracer monitoring data. By comprehensively analyzing the geological interpretation data and development dynamic data, fracture development patterns can be obtained. Fracture development patterns describe the main development morphology and flow control characteristics of fractures in the target gas reservoir, and can include large fracture type, fracture network type, and large fracture network coupled type.

[0197] After obtaining the fracture development pattern, gas reservoir development models are established based on the fracture development pattern under different combinations of well types and fracture morphology categories. These gas reservoir development models are used to simulate the development response of fractured gas reservoirs under different well types, fracture morphologies, reservoir conditions, and water body conditions. Specifically, gas reservoir development models can be established corresponding to horizontal well types with large fractures, vertical well types with large fractures, horizontal well types with fracture mesh types, vertical well types with fracture mesh types, horizontal well types with large fracture mesh coupled types, and vertical well types with large fracture mesh coupled types. By establishing gas reservoir development models according to well type and fracture morphology categories, subsequent sample generation and regression processing can have a clear classification basis.

[0198] After establishing the gas reservoir development model, the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters in the model are adjusted to obtain multiple sets of parameter combinations. By changing these parameters within a preset range, multiple sets of parameter combinations covering different development conditions and different fracture development conditions can be formed, providing an input basis for generating recovery rate response samples. Numerical simulations are performed on each set of parameter combinations to obtain the corresponding recovery rate response samples. Numerical simulations are used to calculate pressure propagation, gas-water migration, gas and water production changes, and the final recovery rate response during gas reservoir development. For each set of parameter combinations, it is input into the corresponding gas reservoir development model, and the corresponding recovery rate result is obtained through simulation calculation. This recovery rate result serves as the recovery rate response sample. Through batch numerical simulations, the recovery rate variation patterns under different well types, different fracture morphologies, and different parameter combinations can be obtained.

[0199] After obtaining the recovery rate response samples, they are classified according to well type and fracture morphology. Recovery rate response samples belonging to the same well type and fracture morphology combination are grouped into the same sample set. Subsequently, regression processing is performed on the classified recovery rate response samples to obtain recovery rate prediction models corresponding to each well type and fracture morphology combination. Regression processing can be performed using least squares regression, robust regression, machine learning regression, or numerical simulation recalibration. Through regression processing, discrete recovery rate response samples obtained from numerical simulation can be transformed into directly callable recovery rate prediction models. By pre-establishing a recovery rate model database, the complex numerical simulation and regression modeling processes are brought forward, enabling direct model calculation during the field prediction stage. This improves the efficiency and accuracy of recovery rate prediction for fractured gas reservoirs and allows the prediction models to reflect development differences under different well types and fracture morphology combinations.

[0200] It should be further explained that well type categories can be expanded according to actual development needs. For example, in addition to vertical and horizontal well categories, directional well categories, highly deviated well categories, branching well categories, or multi-bottom well categories can be added. When a new well type category is added, a recovery rate prediction model corresponding to the new well type category can be added to the recovery rate model database, and model matching can be performed according to the combination relationship between well type category and fracture morphology category. At the same time, fracture morphology categories can also be further subdivided according to the fracture development characteristics of the target gas reservoir. For example, in addition to large fracture type, fracture network type, and large fracture network coupled type, it can be further subdivided into small-scale single fracture type, directional through fracture type, or multi-level fracture network type. The subdivision of fracture morphology categories does not change the processing logic of obtaining the target recovery rate prediction model from the recovery rate model database based on well type category and fracture morphology category in this method. The model coefficients in the recovery rate model database can be updated according to new samples. The update method can adopt least squares regression, robust regression, machine learning regression, or numerical simulation recalibration. As long as the processing flow of dividing the target evaluation unit, determining the well type, determining the fracture morphology, matching the target recovery rate prediction model, and calculating the recovery rate prediction value is still used, the recovery rate prediction objective of this application can be achieved.

[0201] After establishing a recovery rate model database in advance using the above method, during the rapid field prediction phase, it is unnecessary to rebuild a complete numerical simulation model for each prediction area or candidate development scheme, nor is it necessary to recalculate all parameter combinations. During field prediction, it is only necessary to retrieve the corresponding target recovery rate prediction model based on the well type and fracture morphology category of the target evaluation unit, and input the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters of the target evaluation unit into the target recovery rate prediction model to obtain the corresponding recovery rate prediction value. Thus, the recovery rate response patterns formed during the numerical simulation phase can be transformed into quickly invoked classification prediction models, improving prediction efficiency for multiple target evaluation units and multiple candidate development schemes.

[0202] In one specific embodiment, if it is necessary to select the optimal solution in this area, three candidate development schemes are set for the same area to be predicted under the same reservoir conditions, and recovery rate predictions are performed for each scheme. The common parameter values ​​are: well density of 1.6 wells per square kilometer, and matrix permeability of 8 × 10⁻⁶. -5 Darcy has a water volume ratio of 3, a fracture height of 85 meters, and an effective reservoir thickness of 60 meters. Three candidate development schemes are proposed: horizontal wells with large fracture morphology, horizontal wells with fracture network morphology, and horizontal wells with a combination of large fracture and fracture network morphology.

[0203] For horizontal wells with a large fracture morphology, the fracture permeability is 180 Darcy, the fracture aperture is 0.008 meters, the fracture penetration ratio is 0.85, and the fracture density is 0.012 fractures per meter. A first-order recovery prediction model is obtained based on the well type and fracture morphology, and the above parameters are input into the first-order recovery prediction model:

[0204]

[0205] The calculation yields: R A =2.369+9.87×0.3114×2.3478=9.59.

[0206] For horizontal wells with a fracture network morphology, the fracture permeability is 120 Darcy, the fracture aperture is 0.006 meters, the fracture penetration ratio is 0.78, and the fracture density is 0.040 fractures per meter. A tertiary recovery prediction model is obtained based on the well type and fracture morphology, and the above parameters are input into the tertiary recovery prediction model:

[0207]

[0208] Calculated R B =2.369+18.63×0.6267×2.7116=34.03.

[0209] For horizontal wells with a large fracture network coupled type of fracture morphology, the fracture permeability is 150 Darcy, the fracture aperture is 0.007 meters, the fracture penetration ratio is 0.82, and the fracture density is 0.030 fractures per meter. A fifth-order recovery prediction model is obtained based on the well type and fracture morphology, and the above parameters are input into the fifth-order recovery prediction model:

[0210]

[0211] The calculation yields: 𝑅 𝐶 =2.369+13.21×0.5185×1.7780=14.55.

[0212] In this embodiment, the predicted recovery rates of three candidate development schemes can be obtained separately and sorted according to the magnitude of the predicted recovery rate values. If the scheme with a horizontal well type and a fracture morphology type of fracture mesh has the highest predicted recovery rate value, then this scheme can be determined as the recommended development scheme. This embodiment achieves rapid comparison of development schemes under different well types and different combinations of fracture morphology types by performing target recovery rate prediction model matching and recovery rate prediction value calculation on multiple candidate development schemes.

[0213] When outputting the comparison results of candidate development schemes, the schemes can be sorted from highest to lowest according to the predicted recovery rate. The well type, fracture morphology, target recovery rate prediction model, main input parameters, and predicted recovery rate value corresponding to each candidate development scheme are also output. When the difference in the predicted recovery rate values ​​of multiple candidate development schemes is less than a preset difference threshold, a comprehensive recommendation result can be generated by further considering water control risk, implementation cost, well network adjustment difficulty, and field operability. Therefore, this embodiment can not only output the predicted recovery rate value for a single target evaluation unit or the area to be predicted, but also provide quantitative basis for well type selection, fracture development pattern selection, and development adjustment scheme screening.

[0214] Figure 3 This is a schematic diagram of the structure of the fractured gas reservoir recovery prediction device provided in the embodiments of this application, as shown below. Figure 3 As shown, the fractured gas reservoir recovery prediction device 30 provided in this embodiment includes:

[0215] The region division module 301 is used to acquire the unit division data of the region to be predicted, and divide the region to be predicted into at least one target evaluation unit according to the unit division data; wherein, the unit division data includes at least one of geological boundary data, fracture attribute data, seepage connectivity data and well network control data;

[0216] The category identification module 302 is used to determine the well type category of each target evaluation unit based on the well type characteristic parameters of the target evaluation unit, and to determine the fracture morphology category of each target evaluation unit based on the fracture characteristic parameters of the target evaluation unit.

[0217] The model matching module 303 is used to obtain the target recovery rate prediction model corresponding to the well type and fracture morphology of the target evaluation unit from the preset recovery rate model database, and to obtain the recovery rate prediction value of the target evaluation unit by using the target recovery rate prediction model based on the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters and water body characteristic parameters of the target evaluation unit.

[0218] The results aggregation module 304 is used to obtain the predicted recovery rate of the region to be predicted based on the predicted recovery rate of each target evaluation unit and according to the preset regional aggregation rules.

[0219] The fractured gas reservoir recovery rate prediction device 30 provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0220] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the electronic device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0221] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0222] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0223] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0224] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0225] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0226] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0227] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0228] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0229] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0230] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0231] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0232] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0233] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0234] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0235] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for predicting the recovery rate of fractured gas reservoirs, characterized in that, include: Obtain unit division data of the area to be predicted, and divide the area to be predicted into at least one target evaluation unit based on the unit division data; wherein, the unit division data includes at least one of geological boundary data, fracture attribute data, seepage connectivity data, and well network control data; Based on the well type characteristic parameters of the target evaluation unit, the well type category of each target evaluation unit is determined, and based on the fracture characteristic parameters of the target evaluation unit, the fracture morphology category of each target evaluation unit is determined. From the preset recovery rate model database, obtain the target recovery rate prediction model corresponding to the well type and fracture morphology of the target evaluation unit, and obtain the recovery rate prediction value of the target evaluation unit by using the target recovery rate prediction model based on the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters and water body characteristic parameters of the target evaluation unit. Based on the recovery rate prediction values ​​of each target evaluation unit, the recovery rate prediction values ​​of the region to be predicted are obtained according to the preset regional aggregation rules.

2. The method according to claim 1, characterized in that, The step of dividing the region to be predicted into at least one target evaluation unit based on the unit division data includes: Based on the geological boundary data, determine the structural zoning boundaries within the area to be predicted; Based on the crack attribute data, determine the crack partition boundaries and crack attribute categories of each crack partition within the area to be predicted. Based on the seepage connectivity data, determine the seepage zone boundaries and seepage connectivity categories of each seepage zone within the area to be predicted; Based on the well network control data, determine the well control zone boundaries and well network control categories of each well control zone within the area to be predicted; The structural partition boundary, fracture partition boundary, seepage partition boundary and well control partition boundary are superimposed to obtain at least one initial partitioning unit corresponding to the region to be predicted. Based on the fracture zone, seepage zone, and well control zone where the initial partitioning unit is located, the fracture attribute category, seepage connectivity category, and well network control category of the initial partitioning unit are determined respectively. Traverse each of the initial partitioning units in the region to be predicted, obtain any pair of adjacent initial partitioning units, and when the fracture attribute category, seepage connectivity category and well network control category of any pair of adjacent initial partitioning units are the same, merge the pair of adjacent initial partitioning units to obtain at least one target evaluation unit.

3. The method according to claim 1, characterized in that, Well type characteristic parameters include well inclination data, well trajectory data, horizontal section length, and completion method; Accordingly, determining the well type category of each target evaluation unit based on the well type characteristic parameters of the target evaluation unit includes: Obtain the well type characteristic parameters of the target well within the target evaluation unit. The well type characteristic parameters include well inclination data, well trajectory data, horizontal section length, and well completion method. Based on the well inclination data and well trajectory data, determine whether there are continuous well sections in the well trajectory of the target well with an inclination angle greater than a preset inclination angle threshold; If there is no continuous well section with an inclination angle greater than the preset inclination angle threshold, then the well type category of the target evaluation unit is the vertical well category; If there is a continuous well section with an inclination angle greater than a preset inclination angle threshold, then the continuous well section is determined as a candidate horizontal section; When the length of the candidate horizontal segment is greater than a preset length threshold, it is determined whether the candidate horizontal segment is a completion segment or a producing segment of the target well, based on the completion method of the target well. When the candidate horizontal segment is the completion segment or producing segment of the target well, the well type category of the target evaluation unit is determined to be the horizontal well category.

4. The method according to claim 1, characterized in that, The fracture characteristic parameters include at least one of the following: fracture density, fracture penetration ratio, fracture orientation dispersion parameter, fracture intersection parameter, fracture connectivity parameter, and fracture conductivity parameter; Accordingly, based on the crack characteristic parameters of the target evaluation unit, the crack morphology category of each target evaluation unit is determined, including: The fracture density of the target evaluation unit is determined based on at least one of core fracture statistics, imaging logging interpretation data, and discrete fracture network model data. Based on at least one of seismic fracture interpretation data, imaging logging fracture orientation interpretation data, well connectivity data, tracer monitoring data, and production dynamic data, candidate fracture connectivity bodies are identified within the target evaluation unit, and the extension length of the candidate fracture connectivity bodies is determined. The crack penetration ratio of the candidate crack connection is determined based on the ratio of the extension length of the candidate crack connection to the characteristic length of the target evaluation unit. Based on at least one of imaging logging interpretation data, core fracture description data, and seismic fracture interpretation data, determine the azimuth data of the fractures within the target evaluation unit, and determine the fracture azimuth dispersion parameters based on the azimuth data. Based on at least one of imaging logging interpretation data, core fracture description data, and discrete fracture network model data, determine the cross-cutting relationship between different fractures within the target evaluation unit, and determine the fracture cross-cutting parameters based on the cross-cutting relationship; The fracture connectivity parameters and fracture conductivity parameters are determined based on at least one of the following: well test response data, well connectivity data, tracer monitoring data, lost zone identification data, produced zone identification data, water intrusion channel data, and production dynamic data. When the crack penetration ratio of the candidate crack body meets the preset penetration condition, the crack flow capacity parameter meets the preset flow condition, and the crack orientation dispersion parameter meets the preset orientation concentration condition, the candidate crack body is identified as the main crack. When the crack density meets a preset density condition, the crack orientation dispersion parameter meets a preset multi-directional condition, and the crack intersection parameter meets a preset intersection condition, a crack network is identified within the target evaluation unit. When the main crack is identified in the target evaluation unit but the crack network is not identified, the crack morphology category of the target evaluation unit is determined to be a large crack type. Alternatively, when the crack network is identified in the target evaluation unit but the main crack is not identified, the crack morphology category of the target evaluation unit is determined to be a crack network type. Alternatively, when the main crack and the crack network are identified in the target evaluation unit, the crack morphology category of the target evaluation unit is determined to be a large crack-crack network coupled type.

5. The method according to claim 1, characterized in that, The method for obtaining the target recovery rate prediction model corresponding to the well type and fracture morphology category of the target evaluation unit includes: When the well type of the target evaluation unit is a horizontal well and the fracture morphology is a large fracture, the first recovery rate prediction model is obtained. Alternatively, when the well type of the target evaluation unit is a vertical well and the fracture morphology is a large fracture, a second recovery rate prediction model is obtained. Alternatively, when the well type of the target evaluation unit is a horizontal well and the fracture morphology is a fracture network, a third recovery rate prediction model is obtained. Alternatively, when the well type of the target evaluation unit is a vertical well and the fracture morphology is a fracture network, a fourth recovery rate prediction model is obtained. Alternatively, when the well type of the target evaluation unit is a horizontal well and the fracture morphology is a large fracture network coupled type, the fifth recovery rate prediction model is obtained. Alternatively, when the well type of the target evaluation unit is a vertical well and the fracture morphology is a large fracture network coupled type, the sixth recovery rate prediction model can be obtained.

6. The method according to claim 1, characterized in that, The step of obtaining the predicted recovery rate of the region to be predicted based on the predicted recovery rate of each of the target evaluation units and according to a preset regional aggregation rule includes: When the area to be predicted includes a target evaluation unit, the predicted recovery rate of the target evaluation unit is determined as the predicted recovery rate of the area to be predicted. Alternatively, when the area to be predicted includes multiple target evaluation units, the geological reserves, controlled reserves, or effective pore volume of each target evaluation unit are obtained, and the following formula is used: Calculate and obtain the predicted recovery rate R for the area to be predicted. total ; Among them, R i Let G be the predicted recovery rate of the i-th target evaluation unit. i Let n represent the geological reserves, controlled reserves, or effective pore volume of the i-th target evaluation unit, and n be the number of target evaluation units.

7. The method according to claim 1, characterized in that, Establish the aforementioned recovery rate model database, including: Obtain geological interpretation data and development dynamic data of fractured gas reservoirs, and obtain fracture development patterns based on the geological interpretation data and development dynamic data; Based on the fracture development pattern, gas reservoir development models are established under different well types and combinations of different fracture morphology categories; By adjusting the well type characteristic parameters, fracture characteristic parameters, reservoir characteristic parameters, and water body characteristic parameters in the gas reservoir development model, multiple sets of parameter combinations are obtained. Numerical simulations were performed on each set of parameter combinations to obtain the recovery rate response samples corresponding to each set of parameter combinations. The recovery rate response samples are classified according to well type and fracture morphology, and regression processing is performed on the classified recovery rate response samples to obtain the recovery rate prediction model corresponding to each combination of well type and fracture morphology. The preset recovery rate model database is established based on each of the recovery rate prediction models.

8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images