A method and system for constructing a three-dimensional geological volume

Abstract

The application discloses a method for constructing a three-dimensional geological body, which comprises the following steps: establishing a three-dimensional seismic attribute body about a work area to be studied; based on the three-dimensional seismic attribute body, combining conventional logging data and imaging logging data, respectively dividing rock types under geological scale, logging scale and imaging scale; according to the rock type division results under different scales, combining core data, establishing a multi-scale correlation model for sequentially performing scale coarsening and attribute prediction from core, imaging, logging to geology; and updating the original three-dimensional geological body according to the multi-scale correlation model. The application can apply fine scale data to the construction of a macroscopic geological body.

Description

Technical Field

[0001] This invention relates to the field of petroleum exploration technology, and in particular to a method and system for constructing three-dimensional geological bodies. Background Technology

[0002] Three-dimensional geological modeling is an important method for predicting, planning, dynamically analyzing, and evaluating the effects of oil, natural gas, and other mineral extraction. Accurately constructing three-dimensional geological data volumes is of great significance for the quantitative description of the drilling geological environment and the three-dimensional characterization of oil reservoirs.

[0003] Seismic exploration is an important technical means for pre-drilling mineral exploration and also the most important data source for 3D geological modeling. Through seismic inversion technology, based on surface observation seismic data and using known geological laws and drilling and logging data as constraints, spatial modeling of properties such as seismic layer velocity, rock physical and mechanical parameters can be performed, and the imaging process of subsurface spatial structure and physical properties can be realized.

[0004] Existing 3D geological data volume construction technologies have the following shortcomings:

[0005] (1) Incomplete utilization of geological data leads to inaccurate geological modeling and reservoir description. Currently, multi-source data can provide a basis for macro-geological modeling, but only seismic data and conventional logging data are currently used, ignoring the reservoir description that fine-scale data such as core samples and imaging logging can provide. This is especially true for complex reservoirs with strong heterogeneity, where small-scale heterogeneity is easily overlooked. Moreover, the large differences in scale and resolution among multi-source data make it difficult to apply fine-scale data in constructing macro-three-dimensional geological bodies.

[0006] (2) The scope of fine description of the wellbore perimeter and the update of parameter volumes based on wellbore data such as core and imaging is limited. Due to the limited coring range and imaging logging interval, it is currently only applied to fine description of a small area around the wellbore, and there is a lack of technical methods to expand the scope of attribute updates based on correlation studies between different scales. Summary of the Invention

[0007] The purpose of this invention is to provide a three-dimensional geological data volume construction scheme based on multi-scale data fusion, so as to accurately construct a three-dimensional geological data volume and effectively guide scientific drilling design and construction, and efficient reservoir development.

[0008] To address the aforementioned technical problems, this invention provides a method for constructing a three-dimensional geological body, comprising: establishing a three-dimensional seismic attribute body for the study area; classifying rock types at geological, well logging, and imaging scales based on the three-dimensional seismic attribute body and in conjunction with conventional well logging data and imaging well logging data; establishing a multi-scale correlation model for scale coarsening and attribute prediction sequentially from core, imaging, well logging to geology based on the rock type classification results at different scales and in conjunction with core data; and updating the original three-dimensional geological body based on the multi-scale correlation model.

[0009] Preferably, the step of classifying rock types at the geological, well logging, and imaging scales based on the three-dimensional seismic attribute volume and in combination with conventional well logging data and imaging well logging data includes: extracting three-dimensional geological grid information based on the three-dimensional seismic attribute volume and classifying rock types within the geological space to obtain rock type classification results at the geological scale; using the three-dimensional geological grid information as a constraint boundary, segmenting the well logging curves and calculating the attribute value of each well logging segment to obtain the rock type classification results at the well logging scale; using the well logging curve segmentation points as constraint boundaries, segmenting the imaging data and calculating the attribute value of each imaging segment to obtain the rock type classification results at the imaging scale.

[0010] Preferably, before the steps of segmenting the imaging data using well logging curve segmentation points as constraint boundaries and obtaining the rock type classification result at the imaging scale by calculating the attribute value of each imaging segment, the two-dimensional imaging data is converted into one-dimensional imaging data so as to classify the imaging rock type based on the one-dimensional imaging data.

[0011] Preferably, the segmentation of imaging data / logging curves includes: using the logging curve segmentation points / the three-dimensional geological grid information as constraint boundaries to perform initial segmentation of the imaging data / logging curves, and calculating the imaging / logging attribute value of each initial segment; using the initial segmentation boundary of the imaging data / logging curves as constraints to perform secondary segmentation of the imaging data / logging curves, and calculating the imaging / logging attribute value of each secondary segment; and based on the average value of the imaging / logging attributes within each secondary segment, dividing different rock types at the imaging / logging scale to obtain imaging / logging attribute data points corresponding to different imaging / logging rock types.

[0012] Preferably, the process of performing secondary segmentation on the logging curve or the imaging data includes: performing smoothing filtering on the imaging data or the logging curve using wavelet transform; obtaining secondary segmentation points of the imaging data or the logging curve, wherein the second derivative of the smoothed imaging data or logging curve is calculated, thereby determining the points where the second derivative is zero as the corresponding secondary segmentation points.

[0013] Preferably, the step of constructing the multi-scale correlation model includes: segmenting the three-dimensional core attribute volume based on the segmentation points of the imaging secondary segmentation to obtain core attribute values ​​for different imaging secondary segmentation segments; constructing the imaging-core correlation model based on the core attribute values ​​of the different imaging secondary segmentation segments and several imaging attribute data points corresponding to different imaging rock types; constructing the logging-imaging correlation model based on the imaging attribute data points and core attribute data points at the imaging scale and several logging attribute data points corresponding to different logging rock types; and constructing the geological-logging correlation model based on the logging attribute data points, imaging attribute data points, and core attribute data points at the logging scale and the seismic layer velocities indicated in the three-dimensional seismic attribute volume corresponding to different geological rock types.

[0014] Preferably, several imaging attribute data points corresponding to each type of imaging rock are used as input, and the core attribute values ​​of the different imaging secondary segmentation segments are used as output. The imaging and core correlation model is established by using machine learning methods or multivariate function fitting methods.

[0015] Preferably, several logging attribute data points corresponding to each logging rock type are used as input, and the imaging attribute values ​​and core attribute values ​​of different logging secondary segments are used as output. The logging and imaging correlation model is established by using machine learning methods or multivariate function fitting methods.

[0016] Preferably, the seismic layer velocity corresponding to each geological rock type is used as input, and the well logging attribute value, imaging attribute value and core attribute value corresponding to the three-dimensional geological grid are used as output. The geological and well logging correlation model is established by using a linear or nonlinear function fitting method.

[0017] Preferably, before constructing the imaging-core correlation model / the logging-imaging correlation model, the method further includes: analyzing the importance of different imaging attributes / logging attributes, so as to establish corresponding correlation models based on the attribute data points corresponding to important imaging / logging attributes.

[0018] Preferably, the step of establishing a three-dimensional seismic attribute body for the work area to be studied includes: collecting seismic data of the work area to be studied, as well as conventional logging data, well logging data, imaging logging data and core data of multiple wells; and using conventional logging data as constraints, performing inversion on the seismic data to obtain the three-dimensional seismic attribute body.

[0019] On the other hand, the present invention also provides a system for constructing a three-dimensional geological body, comprising: a seismic attribute body construction module configured to establish a three-dimensional seismic attribute body for the study area; a multi-scale rock type classification module configured to classify rock types at geological, well logging, and imaging scales based on the three-dimensional seismic attribute body and in combination with conventional well logging data and imaging well logging data; a multi-level correlation model construction and attribute expansion module configured to establish a multi-scale correlation model for scale coarsening and attribute prediction from core, imaging, well logging to geology, based on the rock type classification results at different scales and in combination with core data; and a multi-scale attribute characterization module configured to update the original three-dimensional geological body based on the multi-scale correlation model.

[0020] Compared with the prior art, one or more embodiments of the above solutions may have the following advantages or beneficial effects:

[0021] This invention discloses a method and system for constructing three-dimensional geological volumes. The method and system integrate multi-source data at different scales, including core data, imaging logging, conventional logging, and seismic data. It fully considers the significant differences in scale and resolution among the multi-source data, extending the local range of fine-scale attributes to the spatial scope of the macroscopic geological volume, and obtaining a three-dimensional geological volume updated based on attributes at different scales (updating the three-dimensional geological data volume based on attributes at different scales such as core, imaging, and logging). This invention improves the accuracy of quantitative description of drilling geological factors and three-dimensional reservoir characterization, making it possible to apply fine-scale data in constructing macroscopic geological volumes, and providing effective reference and guidance for drilling design and reservoir development.

[0022] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0023] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0024] Figure 1This is a step diagram of a method for constructing a three-dimensional geological body according to an embodiment of this application.

[0025] Figure 2 This is a flowchart illustrating a method for constructing a three-dimensional geological body according to an embodiment of this application.

[0026] Figure 3 This is an example diagram of the rock type classification results at the well logging scale in the method for constructing a three-dimensional geological body according to an embodiment of this application.

[0027] Figure 4 This is an example diagram of the rock type classification results at the imaging scale in the method for constructing a three-dimensional geological body according to an embodiment of this application.

[0028] Figure 5 This is a structural block diagram of a system for constructing a three-dimensional geological body according to an embodiment of this application. Detailed Implementation

[0029] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples, so that the process of how the present invention uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly. It should be noted that, as long as there is no conflict, the various embodiments and features in the various embodiments of the present invention can be combined with each other, and the resulting technical solutions are all within the protection scope of the present invention.

[0030] Furthermore, the steps illustrated in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than that shown here.

[0031] Three-dimensional geological modeling is an important method for predicting, planning, dynamically analyzing, and evaluating the effects of oil, natural gas, and other mineral extraction. Accurately constructing three-dimensional geological data volumes is of great significance for the quantitative description of the drilling geological environment and the three-dimensional characterization of oil reservoirs.

[0032] Seismic exploration is an important technical means for pre-drilling mineral exploration and also the most important data source for 3D geological modeling. Through seismic inversion technology, based on surface observation seismic data and using known geological laws and drilling and logging data as constraints, spatial modeling of properties such as seismic layer velocity, rock physical and mechanical parameters can be performed, and the imaging process of subsurface spatial structure and physical properties can be realized.

[0033] Existing 3D geological data volume construction technologies have the following shortcomings:

[0034] (1) Incomplete utilization of geological data leads to inaccurate geological modeling and reservoir description. Currently, multi-source data can provide a basis for macro-geological modeling, but only seismic data and conventional logging data are currently used, ignoring the reservoir description that fine-scale data such as core samples and imaging logging can provide. This is especially true for complex reservoirs with strong heterogeneity, where small-scale heterogeneity is easily overlooked. Moreover, the large differences in scale and resolution among multi-source data make it difficult to apply fine-scale data in constructing macro-three-dimensional geological bodies.

[0035] (2) The scope of fine description of the wellbore perimeter and the update of parameter volumes based on wellbore data such as core and imaging is limited. Due to the limited coring range and imaging logging interval, it is currently only applied to fine description of a small area around the wellbore, and there is a lack of technical methods to expand the scope of attribute updates based on correlation studies between different scales.

[0036] Therefore, to address one or more of the aforementioned technical problems, this application proposes a method and system for constructing three-dimensional geological volumes. The method and system perform smoothing filtering and fine segmentation on multi-scale data in the order of geological-logging-imaging scales. Based on the fine segmentation at each scale, an artificial intelligence clustering algorithm is used to classify rock types. Based on rock types at different scales, correlation prediction models for imaging-core, logging-imaging, and geological-logging are established using function fitting or machine learning methods, respectively. Applying these correlation prediction models, core-scale attributes are extended to the imaging interval, imaging-scale attributes are extended to the logging interval, and logging-scale attributes are extended to the three-dimensional geological space, respectively.

[0037] In this way, the present invention integrates multi-source data such as core data, imaging logging, conventional logging, and seismic data, and fully considers the huge differences in scale and resolution of multi-source data, thus providing the possibility for the application of fine-scale data in constructing macroscopic geological bodies.

[0038] Figure 1 This is a step diagram illustrating a method for constructing a three-dimensional geological body according to an embodiment of this application. Figure 1As shown, the method for constructing a three-dimensional geological body (hereinafter referred to as the "three-dimensional geological body construction method") disclosed in this embodiment of the invention includes the following steps: Step S110 establishes a three-dimensional seismic attribute body for the study area; Step S120, based on the three-dimensional seismic attribute body established in Step S110, and combined with conventional well logging data and imaging well logging data, classifies the rock types at the geological scale, well logging scale, and imaging scale respectively; then, Step S130, based on the rock type classification results at different scales obtained in Step S120, and combined with core data, establishes a multi-scale correlation model for scale coarsening and attribute prediction from core, imaging, well logging to geology; finally, Step S140, updates the original three-dimensional geological body based on the multi-scale correlation model constructed in Step S130.

[0039] Figure 2 This is a detailed flowchart of a method for constructing a three-dimensional geological body according to an embodiment of this application. The following is in conjunction with… Figure 1 and Figure 2 The specific process of the three-dimensional geological body construction method described in the embodiments of the present invention will be explained.

[0040] Step S110 involves collecting geological data and wellbore data from multiple wells within the study area to construct a three-dimensional seismic attribute volume, i.e., the original three-dimensional geological volume, for the current study area. In step S110, firstly, seismic data and conventional logging data, well logging data, imaging logging data, and core data (e.g., core CT scan data) from multiple wells within the study area are collected. Then, using the collected conventional logging data as constraints, the seismic data is inverted to obtain the three-dimensional seismic attribute volume.

[0041] After establishing the three-dimensional seismic attribute volume, proceed to step S120, where, based on the three-dimensional seismic attribute volume, conventional logging data and imaging logging data are used to sequentially classify rock types based on geological scale, logging scale, and imaging scale.

[0042] In step S120, firstly, based on the three-dimensional seismic attribute volume, three-dimensional geological grid information is extracted and rock types are classified within the geological space to obtain the rock type classification results at the geological scale; then, using the three-dimensional geological grid information obtained in the previous step as the constraint boundary, the well logging data (well logging curves) is segmented, and the rock type classification results at the well logging scale are obtained by calculating the attribute value of each well logging segment; finally, using the well logging curve segmentation points as the constraint boundary, the imaging data is segmented, and the rock type classification results at the imaging scale are obtained by calculating the attribute value of each imaging segment.

[0043] Specifically, when classifying rock types at the geological scale, it is necessary to extract three-dimensional geological grid information based on the three-dimensional seismic attribute volume, and then apply artificial intelligence clustering algorithms (such as one-dimensional clustering algorithms) to classify rock types in three-dimensional geological space, thereby obtaining geological attribute data points corresponding to different geological rock types (i.e., the classification results of different rock types at the geological scale). Thus, different geological rock types at the geological scale are extracted from the three-dimensional seismic attribute volume, allowing for further classification of rock types at the well logging scale.

[0044] When classifying rock types at the logging scale, step S101 (not shown) uses the aforementioned three-dimensional geological grid information as a constraint boundary to initially segment the logging curves and calculates the average value of logging attributes within each initial segment. Specifically, using the three-dimensional geological grid as a constraint boundary, the logging curves are first coarsely segmented. Since each coarse segment corresponds to multiple data points with logging attribute values, when obtaining the logging attribute values ​​corresponding to each coarse segment, it is necessary to calculate the average value of the same type of logging attribute for all data points within each coarse segment.

[0045] Further, in step S102 (not shown), based on the logging attribute data points within each initial logging segment, and constrained by the initial segmentation boundary of the logging curve, the logging curve is further segmented a second time to obtain the secondary segmentation points. Then, the average attribute value of the logging attributes within each secondary segment is calculated, thereby obtaining the rock type classification result at the logging scale. In step S102, based on the coarse segmentation of the logging curve, the logging data is further refined. Specifically, firstly, the wavelet transform formula defined by convolution is applied to smooth the coarsely segmented logging curve. The wavelet transform formula is expressed by the following expression:

[0046]

[0047] Where t represents depth; S(t) represents the coarsely segmented logging curve; WT represents the logging curve after smoothing and filtering; σ represents the scaling parameter, which represents the degree of smoothing; τ represents the translation amount, used to control the translation of the wavelet function; and g represents the smoothing function, where the smoothing function is a Gaussian smoothing function.

[0048] Second, when forming the secondary segmentation points of the logging curve, the second derivative is calculated for the smoothed logging curve, and the points where the second derivative is zero are determined as the secondary segmentation points of the logging curve, thus forming multiple continuous secondary (fine) segments of the logging curve. Since each logging fine segment corresponds to the logging attribute values ​​of multiple data points, when obtaining the logging attribute values ​​corresponding to each logging fine segment, it is necessary to calculate the average value of the same type of logging attribute for all data points within each logging fine segment.

[0049] Further, in step S103 (not shown), based on the average attribute value of the logging attributes within each secondary logging segment (i.e., the logging attribute value within each fine logging segment), different rock types at the logging scale are divided to obtain logging attribute data points corresponding to different logging rock types. Based on the logging curves after fine segmentation, the average attribute value of each logging fine segment is taken as a data point. According to the magnitude of the logging attribute values ​​corresponding to various logging attributes in each logging fine segment (e.g., using cluster analysis), all logging fine segments are divided into rock types, thereby obtaining various logging rock types at the logging scale. Then, for each logging fine segment, the corresponding logging rock type is obtained. See [link to relevant documentation]. Figure 3 This yields logging attribute data points corresponding to different logging rock types (i.e., the classification results of different rock types at the logging scale).

[0050] It should be noted that when obtaining logging curves corresponding to multiple logging attributes, this embodiment of the invention can process logging curves of different attributes according to steps S101 and S102. Based on the logging attribute values ​​of each secondary segment corresponding to different logging attributes, and fully considering the numerical range of different logging attribute values, a multi-parameter clustering analysis method is used to comprehensively classify logging rock types at the logging scale by combining multiple logging attributes. In other words, logging attribute data points of different logging attributes corresponding to each logging rock type are obtained. In addition, this embodiment of the invention can arbitrarily select logging curves of one logging attribute for primary and secondary segmentation to determine the segmentation boundary. Logging curves of other logging attributes use a unified segmentation boundary, eliminating the need to perform segmentation calculations for logging curves of each logging attribute.

[0051] Next, to improve the accuracy and convenience of rock type classification at the imaging scale, this embodiment of the invention converts the two-dimensional imaging data into one-dimensional imaging data before classifying the rock types at the imaging scale, so as to classify the imaging rock types based on the one-dimensional imaging data. For a certain type of imaging logging curve, the average value of the imaging logging data points along the direction perpendicular to the wellbore is calculated, and the two-dimensional imaging data is converted into one-dimensional imaging data.

[0052] When classifying rock types at the imaging scale, step S201 (not shown) uses the division points of the aforementioned logging curves as boundaries to perform initial segmentation of the (a certain type) imaging data, and calculates the average attribute value of the imaging attributes within each initial segmentation. Specifically, using the secondary division points of the logging curves as boundaries, coarse segmentation is performed on the one-dimensional imaging data. Since each coarse segmentation block corresponds to multiple data points with imaging attribute values, when obtaining the imaging attribute values ​​corresponding to each coarse segmentation block, it is necessary to calculate the average attribute value of the same type of imaging attribute for all data points within each coarse segmentation block along the wellbore direction.

[0053] Further, in step S202 (not shown), based on the imaging attribute data points within each initial imaging segment, and constrained by the initial segmentation boundary of the one-dimensional imaging data, a second segmentation is performed on the one-dimensional imaging data to obtain the second segmentation points. Then, the average value of the imaging attributes within each second segmentation is calculated. In step S202, based on the coarse segmentation of the one-dimensional imaging data, a finer segmentation process is further performed. Specifically, in the first step, the wavelet transform formula defined by convolution is applied to perform smoothing filtering on the coarsely segmented one-dimensional imaging data.

[0054] The second step involves forming secondary segmentation points for the one-dimensional imaging data. For the smoothed one-dimensional imaging data, the second derivative is calculated, and points where the second derivative is zero are identified as secondary segmentation points, thus forming multiple continuous secondary (fine) segments. Since each fine segment corresponds to multiple data points with imaging attribute values, obtaining the imaging attribute values ​​for each fine segment requires calculating the average value of the same type of imaging attribute for all data points within each segment.

[0055] Further, in step S203 (not shown), based on the average value of the imaging attributes within each secondary imaging segment (i.e., the imaging attribute value within each fine imaging segment), different rock types at the imaging scale are divided to obtain imaging attribute data points corresponding to different imaging rock types. Based on the one-dimensional imaging data after fine segmentation, the average attribute value of each fine imaging block is taken as a data point. According to the magnitude of the imaging attribute values ​​corresponding to various imaging attributes in each fine imaging block (e.g., using cluster analysis), all fine imaging blocks are divided into rock types, thereby obtaining multiple imaging rock types at the imaging scale. Then, for each fine imaging block, the corresponding imaging rock type is obtained. See [link to relevant documentation]. Figure 4 This yields imaging attribute data points corresponding to different imaging rock types (i.e., the classification results of different rock types at the imaging scale).

[0056] It should be noted that when obtaining one-dimensional imaging data corresponding to multiple imaging attributes, this embodiment of the invention can process the one-dimensional imaging data of different attributes according to steps S201 and S202. Based on the imaging attribute values ​​of each secondary segment corresponding to different imaging attributes, and fully considering the numerical range of different imaging attribute values, a multi-parameter clustering analysis method is used to comprehensively classify the imaging rock types at the imaging scale by combining multiple imaging attributes. In other words, imaging attribute data points of different imaging attributes corresponding to each imaging rock type are obtained. In addition, this embodiment of the invention can arbitrarily select one type of imaging attribute for primary and secondary segmentation to determine the segmentation boundary. The imaging one-dimensional imaging data of other imaging attributes uses a unified segmentation boundary, without the need to perform segmentation calculations for each type of imaging attribute.

[0057] Furthermore, after completing the rock type classification process based on the imaging scale, the process proceeds to step S130 to establish a multi-level correlation model.

[0058] In step S130, this embodiment of the invention requires segmenting the collected core data as in step S1301 (not shown). Based on the segmentation points of the imaging secondary segmentation, the three-dimensional core attribute volume is segmented to obtain the core attribute value corresponding to each imaging secondary segment. Specifically, the three-dimensional core attribute volume is segmented using the fine segmentation points of the one-dimensional imaging data as boundaries, and the average value of the core attribute of each imaging secondary segment is calculated, that is, a corresponding core attribute value (such as core porosity) is obtained for each imaging secondary segment.

[0059] Then, in step S1302 (not shown), the imaging-core association model is constructed based on the core attribute values ​​of different imaging secondary segmentation segments and combined with the imaging attribute data points corresponding to (several) different types of imaging rock.

[0060] Specifically, several imaging attribute data points corresponding to each type of imaged rock are used as input, and the core attribute values ​​of different imaging secondary segments are used as output. Machine learning methods or multivariate function fitting methods are employed to establish an imaging-core correlation model, thereby constructing an imaging-core correlation model that predicts core attribute values ​​using different important imaging attributes. Furthermore, before constructing the imaging-core correlation model, this embodiment of the invention analyzes the importance of different imaging attributes to use the imaging attribute data points corresponding to several important imaging attributes as input data for establishing the imaging-core correlation model. Preferably, important imaging attributes include: nuclear magnetic resonance imaging porosity and acoustic imaging porosity.

[0061] More specifically, based on the resulting imaging rock type results based on imaging scale, the core attribute values ​​of different imaging secondary segmentation segments are used as output values, and one of the imaging attributes corresponding to each imaging rock type is used as a feature. The importance of all features is analyzed, and the imaging attributes with higher importance are used as input values. Multivariate function fitting or machine learning methods are used to establish an imaging-core correlation model.

[0062] Thus, by applying the established imaging-core correlation model in this embodiment of the invention, core properties within the imaging range can be predicted, thereby extending the prediction of core properties to the imaging range to obtain core property values ​​at the imaging scale. Specifically, using the imaging-core correlation model, core property data points corresponding to different imaging rock types are obtained based on the imaging property data points corresponding to different imaging rock types obtained above.

[0063] Next, step S1303 (not shown) constructs a logging-imaging association model based on imaging attribute data points and core attribute data points at the imaging scale, combined with (several) logging attribute data points corresponding to different logging rock types.

[0064] Specifically, several logging attribute data points corresponding to each logging rock type (corresponding to the attribute average value of each logging sub-segment) are used as input, and the imaging attribute values ​​and core attribute values ​​corresponding to the logging sub-segments are used as output. Machine learning methods or multivariate function fitting methods are employed to establish a logging-imaging correlation model, thereby constructing a logging-imaging correlation model that predicts imaging and core attribute values ​​using different important logging attributes. Furthermore, before constructing the logging-imaging correlation model, this embodiment of the invention analyzes the importance of different logging attributes to use logging attribute data points of several logging attributes as input data required for establishing the logging-imaging correlation model. Among these, important logging attributes preferably include: neutron porosity, density porosity, and acoustic porosity.

[0065] More specifically, based on the well logging rock type results formed at the well logging scale, the imaging attribute values ​​and core attribute values ​​corresponding to different well logging sub-blocks are used as output values, and one well logging attribute corresponding to each well logging rock type is used as a feature. The importance of all features is analyzed, and the well logging attributes with higher importance are used as input values. A well logging and imaging correlation model is established using multivariate function fitting or machine learning methods.

[0066] Thus, by applying the established logging-imaging correlation model in this embodiment of the invention, core properties within a logging interval can be predicted, thereby extending the prediction of core properties to the logging interval to obtain core property values ​​at the logging scale. Specifically, using the logging-imaging correlation model, core property data points corresponding to different logging rock types are obtained based on the logging property data points corresponding to different logging rock types obtained above.

[0067] Further, in step S1304 (not shown), a geological and well logging association model is constructed based on the well logging attribute values, imaging attribute values, and core attribute values ​​at the well logging scale, and in combination with the seismic layer velocities indicated in the three-dimensional seismic attribute volume.

[0068] Specifically, based on the geological rock type results formed at the geological scale, the seismic layer velocity data points corresponding to different geological rock types are used as inputs, and the well logging attribute values, imaging attribute values, and core attribute values ​​corresponding to the geological segmentation blocks (three-dimensional geological grid) are used as outputs. A geological-logging correlation model is established by using linear or nonlinear function fitting methods, thereby constructing a geological-logging correlation model that uses seismic attributes to predict well logging scale attribute values.

[0069] Thus, by applying the established geological and well logging correlation model in this embodiment of the invention, core properties within a geological interval can be predicted, thereby extending the prediction of core properties to the geological interval to obtain core property values ​​at the geological scale. Specifically, using the geological and well logging correlation model, core property data points corresponding to different geological rock types are obtained based on the seismic layer velocities corresponding to different geological rock types obtained above.

[0070] After establishing the three correlation models, the process proceeds to step S140. In step S140, the original three-dimensional geological body is updated using the geological and well logging correlation model, thereby obtaining corresponding well logging attribute data points, imaging attribute data points, and core attribute data points for different geological rock types. Furthermore, step S140 sequentially updates the three-dimensional geological body using the geological and well logging correlation model, the well logging and imaging correlation model, and the imaging and core correlation model, respectively obtaining three-dimensional geological bodies updated with different well logging attributes, three-dimensional geological bodies updated with different imaging attributes, and three-dimensional geological bodies updated with different core attributes.

[0071] Example

[0072] The following describes the construction process and effects of the three-dimensional geological body construction method described in the embodiments of the present invention when applied to the X work area.

[0073] (1) Based on the three-dimensional seismic layer velocity volume obtained by seismic inversion, a one-dimensional clustering algorithm was used to divide three rock types in three-dimensional geological space, namely geological rock type 1, geological rock type 2 and geological rock type 3.

[0074] (2) The acoustic logging curves were coarsely segmented using a three-dimensional geological grid as the boundary. Based on the coarse segmentation, the logging data were further segmented using a smoothing filter. Four rock types were identified based on the average logging attribute points of each segment: logging rock type 1, logging rock type 2, logging rock type 3, and logging rock type 4. Figure 3 As shown.

[0075] (3) Based on the fine segmentation boundary of the well logging curve, coarse segmentation is performed on the one-dimensional nuclear magnetic resonance imaging data. Based on the coarse segmentation, fine segmentation processing based on smoothing filtering is performed on the one-dimensional imaging data, and five rock types are identified according to the average imaging attribute points of each fine segment: Imaging Rock Type 1, Imaging Rock Type 2, Imaging Rock Type 3, Imaging Rock Type 4, and Imaging Rock Type 5. Figure 4 As shown.

[0076] (4) For each rock type corresponding to the imaging scale, core porosity was used as the output value, and two imaging attributes were selected as input values: nuclear magnetic resonance imaging porosity and acoustic imaging porosity. A correlation model for predicting core porosity based on imaging attributes was established. The fitting parameters for different rock types are shown in Table 1.

[0077] Table 1 Fitting parameters for predicting core porosity using imaging attributes

[0078] Imaging scale rock type Parameter a Parameter b 1 0.37 0.61 2 0.28 0.75 3 0.64 0.39 4 0.35 0.61 5 0.45 0.53

[0079] (5) Apply the imaging-core correlation model based on the imaging scale for each rock type to predict the core porosity in the imaging interval and obtain the core porosity at the imaging scale.

[0080] (6) For each rock type based on logging scale, core porosity was used as the output value, and three logging attributes were selected as input values: neutron porosity, density porosity, and sonic porosity. A correlation model for predicting core porosity based on logging attributes was established. The fitting parameters for different rock types are shown in Table 2.

[0081] Table 2 Fitting parameters for predicting core porosity using well logging attributes

[0082] Well logging scale rock type Parameter c Parameter d Parameter e 1 0.17 0.68 0.23 2 0.35 0.26 0.31 3 0.22 0.39 0.47 4 0.16 0.42 0.51

[0083] (7) Apply the logging-imaging correlation model based on logging scale for each rock type to predict the core porosity of the logging interval and obtain the core porosity at the logging scale.

[0084] (8) For each rock type corresponding to the geological scale, core porosity was used as the output value and seismic layer velocity was used as the input value. A linear function fitting was performed to establish a correlation model for predicting core porosity based on seismic attributes. The fitting parameters for different rock types are shown in Table 3.

[0085] Table 3 Fitting parameters for predicting core porosity based on seismic attributes

[0086] Rock types at geological scale Parameter f 1 0.86 2 0.93 3 1.14

[0087] (9) Apply the geological-logging correlation model for each rock type at the geological scale to predict the core porosity of the three-dimensional geological body interval, obtain the core porosity at the geological scale, and thus construct a three-dimensional geological data body based on core porosity updates.

[0088] On the other hand, based on the above-mentioned three-dimensional geological body construction method, this embodiment of the invention also provides a system for constructing three-dimensional geological bodies (hereinafter referred to as "three-dimensional geological body construction system"). Figure 5 This is a structural block diagram of a system for constructing a three-dimensional geological body according to an embodiment of this application.

[0089] like Figure 5 As shown, the three-dimensional geological body construction system of the present invention includes: a seismic attribute body construction module 501, a multi-scale rock type classification module 502, a multi-level correlation model construction and attribute expansion module 503, and a multi-scale attribute characterization module 504. Specifically, the seismic attribute body construction module 501 is implemented according to the method described in step S110 above, and is configured to establish a three-dimensional seismic attribute body for the study area; the multi-scale rock type classification module 502 is implemented according to the method described in step S120 above, and is configured to classify rock types at geological, well logging, and imaging scales based on the three-dimensional seismic attribute body and combined with conventional well logging data and imaging well logging data; the multi-level correlation model construction and attribute expansion module 503 is implemented according to the method described in step S130 above, and is configured to establish a multi-scale correlation model for scale coarsening and attribute prediction from core, imaging, well logging to geology based on the rock type classification results at different scales and combined with core data; the multi-scale attribute characterization module 504 is implemented according to the method described in step S140 above, and is configured to update the original three-dimensional geological body based on the multi-scale correlation model.

[0090] This invention discloses a method and system for constructing three-dimensional geological volumes. The method and system integrate multi-source data at different scales, including core data, imaging logging, conventional logging, and seismic data. It fully considers the significant differences in scale and resolution among the multi-source data, extending the local range of fine-scale attributes to the spatial scope of the macroscopic geological volume, and obtaining a three-dimensional geological volume updated based on attributes at different scales (updating the three-dimensional geological data volume based on attributes at different scales such as core, imaging, and logging). This invention improves the accuracy of quantitative description of drilling geological factors and three-dimensional reservoir characterization, making it possible to apply fine-scale data in constructing macroscopic geological volumes, and providing effective reference and guidance for drilling design and reservoir development.

[0091] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0092] It should be understood that the embodiments disclosed herein are not limited to the specific structures, processing steps, or materials disclosed herein, but should be extended to equivalent substitutions of these features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0093] The phrase "an embodiment" or "an embodiment" used in this specification means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" or "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0094] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A method for constructing a three-dimensional geological body, characterized in that, include: S1, Establish a three-dimensional seismic attribute volume for the work area to be studied; S2, Based on the three-dimensional seismic attribute volume, and combined with conventional logging data and imaging logging data, rock types are classified at the geological scale, logging scale and imaging scale respectively; S3. Based on the rock type classification results at different scales and combined with core data, a multi-scale correlation model is established to perform scale coarsening and attribute prediction sequentially from core, imaging, logging to geology. S4, based on the multi-scale correlation model, update the original three-dimensional geological body. Step S2 includes: Based on the three-dimensional seismic attribute volume, three-dimensional geological grid information is extracted and rock types are classified in geological space to obtain rock type classification results at the geological scale. Using the three-dimensional geological grid information as a constraint boundary, the logging curves are segmented, and the rock type classification results at the logging scale are obtained by calculating the attribute values ​​of each logging segment. Using well logging curve segmentation points as constraint boundaries, the imaging data is segmented, and the rock type classification at the imaging scale is obtained by calculating the attribute values ​​of each imaging segment. When segmenting imaging data / logging curves, the following is included: Using the well logging curve segmentation points / the three-dimensional geological grid information as constraint boundaries, the imaging data / well logging curves are initially segmented, and the imaging / well logging attribute values ​​of each initial segment are calculated. Using the initial segmentation boundary of the imaging data / logging curve as a constraint, the imaging data / logging curve is segmented a second time, and the imaging / logging attribute value of each secondary segment is calculated. This includes: using wavelet transform to perform smoothing filtering on the imaging data or the logging curve, and obtaining the secondary segmentation points of the imaging data or the logging curve. Specifically, the second derivative of the smoothed imaging data or logging curve is calculated, and the points where the second derivative is zero are determined as the corresponding secondary segmentation points. Based on the average value of the imaging / logging attributes within each imaging / logging secondary segment, different rock types at the imaging / logging scale are divided to obtain imaging / logging attribute data points corresponding to different imaging / logging rock types.

2. The method according to claim 1, characterized in that, Before the steps of segmenting the imaging data using well logging curve segmentation points as constraint boundaries and obtaining the rock type classification results at the imaging scale by calculating the attribute values ​​of each imaging segment, the two-dimensional imaging data is converted into one-dimensional imaging data so as to classify the imaging rock types based on the one-dimensional imaging data.

3. The method according to claim 1, characterized in that, The steps of constructing the multi-scale correlation model include: Based on the segmentation points of the imaging secondary segmentation, the three-dimensional core attribute volume is segmented to obtain the core attribute values ​​of different imaging secondary segmentation segments. Based on the core attribute values ​​of the different imaging secondary segmentation segments, and combined with several imaging attribute data points corresponding to different imaging rock types, an imaging-core correlation model is constructed. Based on imaging attribute data points and core attribute data points at the imaging scale, and combined with several logging attribute data points corresponding to different logging rock types, a logging-imaging correlation model is constructed. Based on well logging attribute data points, imaging attribute data points, and core attribute data points at the well logging scale, and combined with the seismic layer velocities indicated in the three-dimensional seismic attribute volume corresponding to different geological rock types, a geological-logging correlation model is constructed.

4. The method according to claim 3, characterized in that, Using several imaging attribute data points corresponding to each type of imaging rock as input and the core attribute values ​​of different imaging secondary segments as output, a model relating imaging and core is established by employing machine learning methods or multivariate function fitting methods.

5. The method according to claim 3, characterized in that, Using several logging attribute data points corresponding to each logging rock type as input, and the imaging attribute values ​​and core attribute values ​​of different logging secondary segments as output, the logging and imaging correlation model is established by using machine learning methods or multivariate function fitting methods.

6. The method according to claim 3, characterized in that, Using the seismic layer velocity corresponding to each geological rock type as input, and the well logging attribute value, imaging attribute value, and core attribute value corresponding to the three-dimensional geological grid as output, the geological and well logging correlation model is established by using linear or nonlinear function fitting methods.

7. The method according to claim 4 or 5, characterized in that, Before constructing the imaging-core correlation model / the logging-imaging correlation model, the following steps are also included: The importance of different imaging / logging attributes is analyzed in order to establish corresponding correlation models based on the attribute data points corresponding to important imaging / logging attributes.

8. The method according to any one of claims 1-6, characterized in that, Step S1 includes: Seismic data, conventional logging data, well logging data, imaging logging data, and core data of multiple wells in the study area were collected. Using conventional logging data as constraints, the three-dimensional seismic attribute volume is obtained by inverting seismic data.

9. A system for constructing three-dimensional geological bodies, characterized in that, include: The seismic attribute volume construction module is configured to create a three-dimensional seismic attribute volume for the work area to be studied. The multi-scale rock type classification module is configured to classify rock types at geological, well logging, and imaging scales based on the three-dimensional seismic attribute volume and in combination with conventional well logging data and imaging well logging data. The multi-level association model construction and attribute expansion module is configured to establish a multi-scale association model based on the rock type classification results at different scales and combined with core data, to perform scale coarsening and attribute prediction sequentially from core, imaging, logging to geology. The multi-scale attribute representation module is configured to update the original three-dimensional geological body based on the multi-scale correlation model, wherein... The multi-scale rock type classification module is also configured as follows: Based on the three-dimensional seismic attribute volume, three-dimensional geological grid information is extracted and rock types are classified in geological space to obtain rock type classification results at the geological scale. Using the three-dimensional geological grid information as a constraint boundary, the logging curves are segmented, and the rock type classification results at the logging scale are obtained by calculating the attribute values ​​of each logging segment. Using well logging curve segmentation points as constraint boundaries, the imaging data is segmented, and the rock type classification at the imaging scale is obtained by calculating the attribute values ​​of each imaging segment. When segmenting imaging data / logging curves, the following is included: Using the well logging curve segmentation points / the three-dimensional geological grid information as constraint boundaries, the imaging data / well logging curves are initially segmented, and the imaging / well logging attribute values ​​of each initial segment are calculated. Using the initial segmentation boundary of the imaging data / logging curve as a constraint, the imaging data / logging curve is segmented a second time, and the imaging / logging attribute value of each secondary segment is calculated. This includes: using wavelet transform to perform smoothing filtering on the imaging data or the logging curve, and obtaining the secondary segmentation points of the imaging data or the logging curve. Specifically, the second derivative of the smoothed imaging data or logging curve is calculated, and the points where the second derivative is zero are determined as the corresponding secondary segmentation points. Based on the average value of the imaging / logging attributes within each imaging / logging secondary segment, different rock types at the imaging / logging scale are divided to obtain imaging / logging attribute data points corresponding to different imaging / logging rock types.

Images