Method and system for predicting TOC of source rock in sparse well area, electronic device and medium

Abstract

The application discloses a sparse well area hydrocarbon source rock TOC prediction method and system, an electronic device and a medium, relates to the technical field of hydrocarbon source rock carbon content prediction, and comprises the following steps: establishing a virtual well in a well-free area; determining the seismic attribute of the virtual well based on a seismic attribute extraction method; determining the lithology of the virtual well based on the seismic attribute of the virtual well and a sandstone and mudstone lithology prediction model, and extracting the mudstone section of the virtual well; determining the longitudinal continuous TOC prediction value of the hydrocarbon source rock in the mudstone section of the virtual well based on a single-well longitudinal continuous TOC prediction model; and combining the longitudinal continuous TOC prediction value and the measured TOC value of the hydrocarbon source rock in the mudstone section of the virtual well to construct a three-dimensional TOC attribute body of the target sparse well area based on a preset geological condition constraint, so as to predict the TOC of the hydrocarbon source rock in the target sparse well area. The application realizes higher-precision quantitative evaluation and prediction of the TOC of the hydrocarbon source rock in the sparse well area.

Description

Technical Field

[0001] This invention relates to the field of hydrocarbon source rock carbon content prediction technology, and in particular to a method, system, electronic device and medium for predicting the total carbon content (TOC) of hydrocarbon source rocks in sparse well areas. Background Technology

[0002] Total organic carbon (TOC) content is one of the main indicators for evaluating the hydrocarbon generation capacity of source rocks and is an important parameter for hydrocarbon generation research and resource assessment in oil-bearing basins. However, in offshore operations, due to limitations imposed by drilling engineering, the drilling coverage area is small and the amount of core samples obtained is limited. This means that only a limited number of analytical test samples can represent the entire evaluation interval and region, ignoring the heterogeneity of source rock distribution. Consequently, the predicted organic matter abundance results deviate significantly from the actual situation, and sometimes even lead to erroneous conclusions, posing a great challenge to resource potential assessment.

[0003] Furthermore, traditional methods for predicting the TOC of source rocks in rare-well areas rely on inferences from actual well logging data combined with sedimentary facies, resulting in low accuracy and reliability that cannot meet exploration needs. Summary of the Invention

[0004] The purpose of this invention is to provide a method, system, electronic equipment and medium for predicting the TOC of source rocks in sparse well areas, so as to achieve a more accurate and scientifically based quantitative evaluation and prediction of the TOC of source rocks in sparse well areas.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] In a first aspect, the present invention provides a method for predicting the total chemical oxygen demand (TOC) of source rocks in rare-well areas, comprising:

[0007] Establish virtual wells in areas without existing wells;

[0008] Based on the seismic attribute extraction method, the seismic attributes above the virtual well are determined;

[0009] Based on the sandstone and mudstone lithology prediction model, the lithology of the virtual well is determined according to the seismic attributes of the virtual well, and the mudstone interval of the virtual well is extracted; the sandstone and mudstone lithology prediction model is determined based on the measured wells in the target sparse well area;

[0010] For the mudstone section of the virtual well, the predicted value of the longitudinal continuous TOC of the source rock in the mudstone section of the virtual well is determined based on the single-well longitudinal continuous TOC prediction model; the single-well longitudinal continuous TOC prediction model is determined based on the actual well measurements in the target sparse well area;

[0011] Based on preset geological constraints, and combining the predicted and measured TOC values ​​of the longitudinal continuous source rocks in the mudstone section of the virtual well, a three-dimensional TOC attribute body of the target sparse well area is constructed; the measured TOC value is the measured TOC value corresponding to the measured well in the target sparse well area.

[0012] Based on the three-dimensional TOC attribute volume of the target sparse well area, the TOC of the source rocks in the target sparse well area is predicted.

[0013] Optionally, the construction process of the sandstone and mudstone lithology prediction model specifically includes:

[0014] Acquire measured 3D seismic data in the target sparse well area;

[0015] Based on the measured 3D seismic data, determine the root mean square amplitude data and variance seismic attribute data of each rock layer in the target sparse well area.

[0016] The root mean square amplitude data and the variance seismic attribute data are normalized.

[0017] The threshold for the proportion of sandstone and mudstone is determined based on the normalized root mean square amplitude data and the variance seismic attribute data.

[0018] Optionally, the virtual wellbore seismic attributes include root mean square amplitude attributes and variance seismic attributes;

[0019] The sandstone and mudstone lithology prediction model, based on the seismic attributes of the virtual well, determines the lithology of the virtual well and extracts the mudstone interval of the virtual well, specifically including:

[0020] Based on the sandstone-mudstone ratio threshold, the distribution of mudstone strata in the virtual well is determined according to the root mean square amplitude attribute and variance seismic attribute of each stratum in the virtual well.

[0021] Optionally, the construction process of the single-well vertical continuous TOC prediction model specifically includes:

[0022] Core cuttings testing and analysis were performed on the measured wells in the target sparse well area to obtain discrete TOC data of the measured wells;

[0023] Based on preset response characteristics, the logging curve to be used for the actual logging well is determined; the preset response characteristics include electrical resistance characteristics, acoustic characteristics, and mudstone lithology characteristics;

[0024] Based on the logging curve to be used, the ΔLogR method is used to establish a single-well longitudinal continuous TOC prediction model based on the discrete TOC data of the actual logging well.

[0025] Optionally, the construction of a three-dimensional TOC attribute body for the target sparse well area, based on preset geological constraints and combining the predicted and measured longitudinal continuous TOC values ​​of the source rocks in the mudstone section of the virtual well, specifically includes:

[0026] Geochemical tests and analyses were conducted on the target sparse well area to determine the terrigenous organic matter input data, paleoproductivity data, and preservation condition data of the target sparse well area.

[0027] Based on the location of the measured wells in the target sparse well area, the terrigenous organic matter input data, the paleoproductivity data, and the preservation condition data are weighted respectively; the weighting is used to optimize the vertical continuous TOC prediction value corresponding to the virtual well, so that the difference between the vertical continuous TOC prediction value corresponding to the virtual well and the vertical continuous TOC value corresponding to the measured well in the target sparse well area is within the set difference range.

[0028] Based on the weighted input data of terrigenous organic matter, the paleoproductivity data, and the preservation condition data, the virtual wells are interpolated to the measured wells of the target sparse well area according to the preset variation function, preset planar distribution probability, and preset vertical distribution probability, so as to obtain the three-dimensional TOC attribute body of the target sparse well area.

[0029] Optionally, the method for determining the seismic attributes above the virtual wellbore based on seismic attribute extraction specifically includes:

[0030] A multi-attribute seismic attribute prediction TOC model is constructed. The seismic attributes include root mean square amplitude, average instantaneous frequency, average instantaneous phase, seismic trace integral, and variance seismic attribute. These multiple attributes include lithology and at least one of the aforementioned seismic attributes. The lithology includes mudstone lithological characteristics and sandstone lithological characteristics. The seismic attribute prediction TOC model is trained using a training set and a neural network. The training set includes historical well 3D seismic data and the corresponding multiple attributes of the 3D seismic data.

[0031] Based on a multi-attribute earthquake attribute prediction TOC model, the optimal earthquake attribute prediction TOC model is determined.

[0032] Based on the optimal seismic attribute prediction TOC model, the seismic attributes of the virtual well are determined.

[0033] Optionally, before the step of predicting the TOC of the source rocks in the target sparse well area based on the three-dimensional TOC attribute volume of the target sparse well area, the method further includes:

[0034] Obtain the set of verification wells in the target sparse well region and the measured TOC value corresponding to each verification well in the set; the verification well is any measured well in the target sparse well region.

[0035] For each verification well, the predicted TOC value of the verification well is determined based on the three-dimensional TOC attribute volume of the target sparse well area.

[0036] The difference between the predicted TOC value of the verification well and the measured TOC value of the corresponding verification well is calculated to obtain the verification difference.

[0037] When the verification difference is within the preset verification difference range, the step of predicting the source rock TOC of the target sparse well area based on the three-dimensional TOC attribute body of the target sparse well area is entered.

[0038] If the verification difference is not within the preset verification difference range, return to the step of establishing a virtual well in the wellless area.

[0039] Secondly, the present invention provides a TOC prediction system for hydrocarbon source rocks in rare-well areas, comprising:

[0040] The virtual well creation module is used to create virtual wells in areas without existing wells.

[0041] The virtual well seismic attribute determination module is used to determine the seismic attributes of the virtual well based on the seismic attribute extraction method.

[0042] The virtual well mudstone section determination module is used to determine the lithology of the virtual well based on the sandstone and mudstone lithology prediction model and the seismic attributes above the virtual well, and to extract the mudstone section of the virtual well; the sandstone and mudstone lithology prediction model is determined based on the measured wells in the target sparse well area;

[0043] The virtual well TOC prediction module is used to determine the longitudinal continuous TOC prediction value of the source rock in the mudstone section of the virtual well based on the single-well longitudinal continuous TOC prediction model; the single-well longitudinal continuous TOC prediction model is determined based on the measured wells in the target sparse well area;

[0044] The three-dimensional TOC attribute body construction module is used to construct a three-dimensional TOC attribute body for the target sparse well area based on preset geological conditions and by combining the vertical continuous TOC prediction value and the measured TOC value of the source rock in the mudstone section of the virtual well; the measured TOC value is the measured TOC value corresponding to the measured well in the target sparse well area.

[0045] The prediction module is used to predict the TOC of the source rocks in the target sparse well area based on the three-dimensional TOC attribute volume of the target sparse well area.

[0046] Thirdly, the present invention provides an electronic device, including a memory and a processor;

[0047] The memory is used to store computer programs, and the processor is used to run the computer programs to execute a method for predicting the TOC of source rocks in rare well areas.

[0048] A computer-readable storage medium storing a computer program;

[0049] When the computer program is executed by the processor, it implements the steps of the TOC prediction method for hydrocarbon source rocks in rare well areas.

[0050] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0051] This invention provides a method, system, electronic equipment, and medium for predicting the TOC (Total Organic Carbon) of source rocks in sparsely-well areas. First, a virtual well is established. Based on a seismic attribute extraction method, the seismic attributes of the virtual well are determined. Based on a sandstone-mudstone lithology prediction model, the lithology of the virtual well is determined, and the corresponding mudstone intervals are extracted. For the mudstone intervals of the virtual well, based on a single-well longitudinal continuous TOC prediction model, the longitudinal continuous TOC prediction values ​​of the source rocks within those mudstone intervals are determined. Finally, based on preset geological constraints, and combining the longitudinal continuous TOC prediction values ​​of the source rocks in the mudstone intervals of the virtual well with the corresponding measured TOC values ​​from actual wells in the target sparsely-well area, a three-dimensional TOC attribute body for the target sparsely-well area is constructed, thereby achieving the integration of virtual well data and actual well data. Compared to existing technologies that only calculate TOC contour maps for each group using average values, this invention uses a method combining virtual and actual well data to obtain a three-dimensional TOC attribute body, which can more intuitively and accurately represent the location of dominant source rock stoves.

[0052] Furthermore, compared to existing technologies where TOC prediction of source rocks in rare-well areas is derived solely from inferences based on actual well data and sedimentary facies, this application establishes relevant sandstone and mudstone lithology prediction models and single-well vertical continuous TOC prediction models based on actual well data. Virtual wells are then established based on these two models, resulting in a three-dimensional TOC attribute volume that integrates actual and virtual well data, which has a high lateral resolution and significantly improves scientific rigor and reliability. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a schematic flowchart of the method for predicting the TOC of hydrocarbon source rocks in rare well areas according to the present invention.

[0055] Figure 2 This is a schematic diagram of the TOC prediction system for hydrocarbon source rocks in rare well areas according to the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] To make the objectives, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] Example 1

[0059] like Figure 1 As shown in the figure, this embodiment provides a method for predicting the TOC of hydrocarbon source rocks in rare well areas, including:

[0060] Step 100: Establish virtual wells in the well-free area. Specifically, determine that the virtual wells penetrate the geological strata to be studied in the depth direction, and determine that the initial distribution of the virtual wells on the plane is uniformly distributed within the study area of ​​the target sparse well zone. The actual density of the virtual wells depends on the actual range of the study area of ​​the target sparse well zone. In principle, the higher the density, the more accurate the results.

[0061] Step 200: Based on the seismic attribute extraction method, determine the seismic attributes above the virtual well; specifically, step 200 includes:

[0062] 1) For the virtual well established in step 100, a seismic attribute prediction TOC model with multiple attributes is constructed in Petrel software. The seismic attributes include root mean square amplitude, average instantaneous frequency, average instantaneous phase, seismic trace integral, and variance seismic attribute. These multiple attributes include lithology and at least one of the aforementioned seismic attributes. The lithology includes mudstone lithological characteristics and sandstone lithological characteristics. The seismic attribute prediction TOC model is trained using a training set and a neural network. The training set includes 3D seismic data from historical wells and the multiple attributes corresponding to the 3D seismic data.

[0063] Specifically, Petrel software was used to extract root mean square amplitude, average instantaneous frequency, average instantaneous phase, seismic trace integral, variance seismic attribute, and lithological data from historical well 3D seismic data. In Matlab software, linear regression, Gaussian process regression, decision tree regression, support vector machine, kernel function approximation, and neural network were used to model the TOC (Transmission of Cost) prediction for historical wells. A comparative study of the TOC results obtained from these various regression methods showed that the neural network method avoids overfitting while maintaining a certain level of prediction accuracy; therefore, neural networks were chosen for training the seismic attribute extraction model.

[0064] Then, seismic attributes were selected, and correlation analysis was performed with the TOC data output by the neural network obtained through Matlab software modeling, according to combinations of single attribute plus lithology, two attributes plus lithology, three attributes plus lithology, four attributes plus lithology, and five attributes plus lithology. The details are as follows:

[0065] For the two attributes, two attribute data are randomly selected from the root mean square amplitude attribute data, average instantaneous frequency attribute data, average instantaneous phase attribute data, seismic trace integral attribute data, and variance seismic attribute data, and combined with lithological attributes, as specific attributes of historical wells; based on the specific attributes of the historical wells, the corresponding TOC value is determined, and then the correlation between the specific attributes of the historical wells (two attributes + lithology) and the corresponding TOC value is analyzed.

[0066] For the three attributes, three attribute data are randomly selected from the root mean square amplitude attribute data, average instantaneous frequency attribute data, average instantaneous phase attribute data, seismic trace integral attribute data, and variance seismic attribute data, and combined with lithology attributes as specific attributes of historical wells; based on the specific attributes of the historical wells, the corresponding TOC values ​​are determined, and then the correlation between the specific attributes of the historical wells (three attributes + lithology) and the corresponding TOC values ​​is analyzed.

[0067] Similarly, we can derive the correlation analysis method for four attributes + lithology and five attributes + lithology.

[0068] Based on the correlation analysis results of single attribute + lithology, two attributes + lithology, three attributes + lithology, four attributes + lithology, and five attributes + lithology with TOC values, the attribute with the highest correlation was determined to be five attributes + lithology. Therefore, the optimal neural network is one whose input values ​​include root mean square amplitude data, average instantaneous frequency data, average instantaneous phase data, seismic trace integral data, variance seismic data, and lithology data. This leads to 2) the optimal seismic attribute prediction TOC model based on multiple attributes. The input seismic attributes of the optimal seismic attribute prediction TOC model include root mean square amplitude attribute, average instantaneous frequency attribute, average instantaneous phase attribute, seismic trace integral attribute, variance seismic attribute, and lithology.

[0069] 3) Based on the optimal seismic attribute prediction TOC model, determine the virtual wellbore seismic attributes. The virtual wellbore seismic attributes include root mean square amplitude attribute, average instantaneous frequency attribute, average instantaneous phase attribute, seismic trace integral attribute, variance seismic attribute, and lithology.

[0070] Step 300: Based on the sandstone and mudstone lithology prediction model, determine the lithology of the virtual well according to the seismic attributes above the virtual well, and extract the mudstone interval of the virtual well; the sandstone and mudstone lithology prediction model is determined based on the measured wells in the target sparse well area.

[0071] Preferably, the construction process of the sandstone and mudstone lithology prediction model specifically includes:

[0072] 1) Obtain measured 3D seismic data in the target sparse well area.

[0073] 2) Based on the measured 3D seismic data, determine the root mean square amplitude data and variance seismic attribute data of each rock layer in the target sparse well area; specifically, use Petrel software to extract the root mean square amplitude data and variance seismic attribute data from the measured 3D seismic data.

[0074] 3) Normalize the root mean square amplitude data and the variance seismic attribute data. Assume that the normalized root mean square amplitude data is A and the variance seismic attribute data is B.

[0075] 4) Determine the sandstone-mudstone ratio threshold based on the normalized root mean square amplitude data and the variance seismic attribute data. Specifically, obtain planar mudstone ratio data within the target sparse well area, where the planar mudstone ratio data is a preset value; determine the mudstone ratio percentage X of the actual well based on the planar mudstone ratio data, find the mudstone ratio percentage X that exactly satisfies X = 1 - A + B, and use the mudstone ratio percentage X as the sandstone-mudstone ratio threshold.

[0076] Based on the sandstone and mudstone lithology prediction model constructed above, step 300 specifically includes: based on the sandstone and mudstone proportion threshold, determining the mudstone layer distribution of each rock layer in the virtual well according to the root mean square amplitude attribute and variance seismic attribute of each rock layer in the virtual well.

[0077] Step 400: For the mudstone section of the virtual well, based on the single-well longitudinal continuous TOC prediction model, determine the predicted longitudinal continuous TOC value of the source rock in the mudstone section of the virtual well; the single-well longitudinal continuous TOC prediction model is determined based on the actual well logging in the target sparse well area. Specifically, a specific logging curve is selected, and the predicted longitudinal continuous TOC value of the mudstone section of the virtual well is calculated according to the ΔLogR method.

[0078] Preferably, the construction process of the single-well vertical continuous TOC prediction model specifically includes:

[0079] 1) Core cuttings testing and analysis were performed on the measured wells in the target sparse well area to obtain discrete TOC data of the measured wells.

[0080] 2) Based on preset response characteristics, determine the logging curve to be used for the actual logging well; the preset response characteristics include resistivity characteristics, acoustic characteristics, and mudstone lithology characteristics. Specifically, the preset response characteristics also include the characteristics of the actual logging well in the plane and the characteristics of the actual logging well in the vertical direction. The characteristics of the actual logging well in the plane include: the geological environment conditions and TOC maximum / minimum values ​​of the actual logging well throughout the target sparse well area; the characteristics of the actual logging well in the vertical direction include the stratum interval in which the actual logging well is located. Different preset response characteristics correspond to different modeling wells, and different modeling wells correspond to different logging curves.

[0081] 3) Based on the logging curve to be used, the ΔLogR method is used to establish a single-well longitudinal continuous TOC prediction model based on the discrete TOC data of the actual logging well.

[0082] Furthermore, using the constructed single-well longitudinal continuous TOC prediction model, longitudinal continuous TOC predictions for mudstone sections were performed on wells under different modeling conditions, and the prediction results were verified based on actual test and analysis results. When the verification results were not satisfactory to the preset accuracy conditions, the single-well longitudinal continuous TOC prediction model was reconstructed.

[0083] Step 500: Based on preset geological conditions, and combining the predicted and measured TOC values ​​of the longitudinal continuous source rocks in the mudstone section of the virtual well, a three-dimensional TOC attribute body for the target sparse well area is constructed; the measured TOC value is the measured TOC value corresponding to the measured well in the target sparse well area.

[0084] Step 500 specifically includes:

[0085] 1) Conduct geochemical testing and analysis on the target sparse well area to determine the terrigenous organic matter input data, paleoproductivity data, and preservation condition data of the target sparse well area. Generally, it is necessary to determine the corresponding levels of the terrigenous organic matter input data, paleoproductivity data, and preservation condition data of the target sparse well area; for example, paleoproductivity is divided into high, medium, and low levels, and preservation conditions are also divided into high, medium, and low levels. Different levels correspond to different geochemical environments, and different geochemical environments will affect the actual magnitude of the well TOC value.

[0086] 2) Based on the location of the measured wells in the target sparse well area, weights are assigned to the terrigenous organic matter input data, the paleoproductivity data, and the preservation condition data, respectively; the weight assignment is used to optimize the vertical continuous TOC prediction value corresponding to the virtual well, so that the difference between the vertical continuous TOC prediction value corresponding to the virtual well and the vertical continuous TOC value corresponding to the measured well in the target sparse well area is within the set difference range.

[0087] Specifically, based on the location of the measured wells, values ​​are assigned to the terrigenous organic matter input data, paleoproductivity data, and preservation condition data, and the weights of the assigned values ​​are adjusted to ensure that the predicted values ​​of the virtual wells in different zones are as close as possible to the TOC values ​​of the measured wells after the assignment.

[0088] 3) Based on the weighted input data of terrigenous organic matter, the paleoproductivity data and the preservation condition data, the virtual well is interpolated to the measured wells of the target sparse well area according to the preset variation function, the preset planar distribution probability and the preset vertical distribution probability, so as to obtain the three-dimensional TOC attribute body of the target sparse well area.

[0089] Specifically, in the Petrel software, the virtual wells, after the aforementioned weighting optimization, are interpolated to the measured wells in the target sparse well area. Among them, the preset planar distribution probability is used to adjust the distribution of virtual wells and measured wells in the plane, the preset vertical distribution probability is used to adjust the distribution of virtual wells and measured wells in the vertical direction, and the variation function is used to adjust the degree of change of virtual wells and measured wells in the horizontal or vertical direction within a certain distance range.

[0090] Step 600: Based on the three-dimensional TOC attribute volume of the target sparse well area, predict the TOC of the source rocks in the target sparse well area. In practical applications, the three-dimensional TOC attribute volume is used to predict high-value areas of source rocks in the target sparse well area to provide an accurate data foundation for subsequent exploration.

[0091] In a specific application, the TOC prediction method for hydrocarbon source rocks in rare-well areas also includes:

[0092] Step 5001: Obtain the set of verification wells for the target sparse well region and the measured TOC value corresponding to each verification well in the set; the verification well is any measured well in the target sparse well region, and the verification well is not the same well as the measured well used to construct the model mentioned above. Generally, the set of verification wells includes one or two verification wells.

[0093] Step 5002: For each verification well, determine the predicted TOC value of the verification well based on the three-dimensional TOC attribute volume of the target sparse well area.

[0094] Step 5003: Calculate the difference between the predicted TOC value of the verification well and the measured TOC value corresponding to the verification well to obtain the verification difference.

[0095] Step 5004: When the verification difference is within the preset verification difference range, proceed to step 600; when the verification difference is not within the preset verification difference range, return to step 100 and reconstruct the virtual well related data to obtain a new three-dimensional TOC attribute body for the target sparse well area.

[0096] In addition, the following verifications can be performed based on the three-dimensional TOC attribute body: 1) Verification of the conformity between the TOC prediction value of the verification well and the geological background knowledge corresponding to the verification well; specifically, this includes comparing and verifying the elevation between different rock layers and between different areas of the same rock layer under the geological background knowledge with the corresponding prediction results; 2) Verification of the matching relationship between the prediction results obtained through the three-dimensional TOC attribute body and the existing oil and gas field discovery locations.

[0097] In summary, this embodiment can only obtain discrete TOC content data through core and cuttings testing, especially in exploration blocks with few wells or few core samples. Geochemical analysis alone is insufficient for three-dimensional evaluation of total organic carbon in source rocks. Therefore, geophysical logging is added to obtain higher-precision vertical TOC prediction data for single wells. A seismic attribute extraction model is established using sufficient TOC data and seismic attributes. Finally, geological constraints are used to calculate the results, resulting in a highly accurate and intuitively displayable three-dimensional TOC attribute body, enabling TOC prediction in areas with few wells.

[0098] Example 2

[0099] like Figure 2 As shown, in order to execute the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, this embodiment provides a TOC prediction system for hydrocarbon source rocks in rare-well areas, including:

[0100] Virtual well creation module 101 is used to create virtual wells in areas without wells.

[0101] The virtual well seismic attribute determination module 201 is used to determine the seismic attributes of the virtual well based on the seismic attribute extraction method.

[0102] The virtual well mudstone section determination module 301 is used to determine the lithology of the virtual well based on the sandstone and mudstone lithology prediction model and the seismic attributes of the virtual well, and to extract the mudstone section of the virtual well; the sandstone and mudstone lithology prediction model is determined based on the measured wells in the target sparse well area.

[0103] The virtual well TOC prediction module 401 is used to determine the longitudinal continuous TOC prediction value of the source rock in the mudstone section of the virtual well based on the single-well longitudinal continuous TOC prediction model; the single-well longitudinal continuous TOC prediction model is determined based on the measured wells in the target sparse well area.

[0104] The three-dimensional TOC attribute body construction module 501 is used to construct a three-dimensional TOC attribute body for the target sparse well area based on preset geological conditions and by combining the vertical continuous TOC prediction value and the measured TOC value of the source rock in the mudstone section of the virtual well; the measured TOC value is the measured TOC value corresponding to the measured well in the target sparse well area.

[0105] The prediction module 601 is used to predict the TOC of the source rocks of the target sparse well area based on the three-dimensional TOC attribute volume of the target sparse well area.

[0106] Example 3

[0107] This embodiment provides an electronic device, including a memory and a processor; the memory is used to store a computer program, and the processor is used to run the computer program to execute the TOC prediction method for hydrocarbon source rocks in rare well areas according to Embodiment 1.

[0108] Optionally, the electronic device is a server.

[0109] In addition, this embodiment also provides a computer-readable storage medium storing a computer program; when the computer program is executed by a processor, it implements the steps of the TOC prediction method for hydrocarbon source rocks in rare well areas according to Embodiment 1.

[0110] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0111] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for predicting the total chemical oxygen demand (TOC) of source rocks in sparsely populated well areas, characterized in that, The TOC prediction method for hydrocarbon source rocks in rare well areas includes: Establish virtual wells in areas without existing wells; Based on the seismic attribute extraction method, the seismic attributes above the virtual well are determined; the seismic attributes include root mean square amplitude attribute, average instantaneous frequency attribute, average instantaneous phase attribute, seismic trace integral attribute, and variance seismic attribute. Based on the sandstone and mudstone lithology prediction model, the lithology of the virtual well is determined according to the seismic attributes of the virtual well, and the mudstone interval of the virtual well is extracted; the sandstone and mudstone lithology prediction model is determined based on the measured wells in the target sparse well area; The construction process of the sandstone-mudstone lithology prediction model specifically includes: acquiring measured 3D seismic data in the target sparse well area; determining the root mean square amplitude data and variance seismic attribute data of each rock layer in the measured well in the target sparse well area based on the measured 3D seismic data; normalizing the root mean square amplitude data and variance seismic attribute data; determining the sandstone-mudstone proportion threshold based on the normalized root mean square amplitude data and variance seismic attribute data; specifically, acquiring planar mudstone proportion data in the target sparse well area, where the planar mudstone proportion data is a preset value; determining the mudstone proportion percentage X of the measured well based on the planar mudstone proportion data, finding the mudstone proportion percentage X that exactly satisfies X=1-A+B, and using the mudstone proportion percentage X as the sandstone-mudstone proportion threshold; where A is the root mean square amplitude data and B is the variance seismic attribute data. For the mudstone section of the virtual well, the predicted value of the longitudinal continuous TOC of the source rock in the mudstone section of the virtual well is determined based on the single-well longitudinal continuous TOC prediction model; the single-well longitudinal continuous TOC prediction model is determined based on the actual well measurements in the target sparse well area; Based on preset geological constraints, and combining the predicted and measured TOC values ​​of the longitudinal continuous source rocks in the mudstone section of the virtual well, a three-dimensional TOC attribute body of the target sparse well area is constructed; the measured TOC value is the measured TOC value corresponding to the measured well in the target sparse well area. Based on the three-dimensional TOC attribute volume of the target sparse well area, the TOC of the source rocks in the target sparse well area is predicted.

2. The method for predicting TOC of source rocks in rare-well areas according to claim 1, characterized in that, The virtual wellbore seismic attributes include root mean square amplitude attributes and variance seismic attributes; The sandstone and mudstone lithology prediction model, based on the seismic attributes of the virtual well, determines the lithology of the virtual well and extracts the mudstone interval of the virtual well, specifically including: Based on the sandstone-mudstone ratio threshold, the distribution of mudstone strata in the virtual well is determined according to the root mean square amplitude attribute and variance seismic attribute of each stratum in the virtual well.

3. The method for predicting TOC of hydrocarbon source rocks in rare-well areas according to claim 1, characterized in that, The construction process of the single-well vertical continuous TOC prediction model specifically includes: Core cuttings testing and analysis were performed on the measured wells in the target sparse well area to obtain discrete TOC data of the measured wells; Based on preset response characteristics, the logging curve to be used for the actual logging well is determined; the preset response characteristics include electrical resistance characteristics, acoustic characteristics, and mudstone lithology characteristics; Based on the logging curve to be used, the ΔLogR method is used to establish a single-well longitudinal continuous TOC prediction model based on the discrete TOC data of the actual logging well.

4. The method for predicting TOC of source rocks in rare-well areas according to claim 1, characterized in that, Based on preset geological constraints, and combining the predicted and measured longitudinal continuous TOC values ​​of the source rocks in the mudstone section of the virtual well, a three-dimensional TOC attribute body for the target sparse well area is constructed, specifically including: Geochemical tests and analyses were conducted on the target sparse well area to determine the terrigenous organic matter input data, paleoproductivity data, and preservation condition data of the target sparse well area. Based on the location of the measured wells in the target sparse well area, the terrigenous organic matter input data, the paleoproductivity data, and the preservation condition data are weighted respectively; the weighting is used to optimize the vertical continuous TOC prediction value corresponding to the virtual well, so that the difference between the vertical continuous TOC prediction value corresponding to the virtual well and the vertical continuous TOC value corresponding to the measured well in the target sparse well area is within the set difference range. Based on the weighted input data of terrigenous organic matter, the paleoproductivity data, and the preservation condition data, the virtual wells are interpolated to the measured wells of the target sparse well area according to the preset variation function, preset planar distribution probability, and preset vertical distribution probability, so as to obtain the three-dimensional TOC attribute body of the target sparse well area.

5. The method for predicting TOC of hydrocarbon source rocks in rare-well areas according to claim 1, characterized in that, The method for determining the seismic attributes of a virtual wellbore based on seismic attribute extraction specifically includes: A multi-attribute seismic attribute prediction TOC model is constructed; the seismic attribute prediction TOC model is trained based on a training set and a neural network; the training set includes three-dimensional seismic data from historical wells and multiple attributes corresponding to the three-dimensional seismic data; the multiple attributes include lithology and at least one of the seismic attributes; the lithology includes mudstone lithological characteristics and sandstone lithological characteristics; Based on a multi-attribute earthquake attribute prediction TOC model, the optimal earthquake attribute prediction TOC model is determined. Based on the optimal seismic attribute prediction TOC model, the seismic attributes of the virtual well are determined.

6. The method for predicting TOC of source rocks in rare-well areas according to claim 1, characterized in that, Before the step of predicting the TOC of the source rocks in the target sparse well area based on the three-dimensional TOC attribute volume of the target sparse well area, the method further includes: Obtain the set of verification wells in the target sparse well region and the measured TOC value corresponding to each verification well in the set; the verification well is any measured well in the target sparse well region. For each verification well, the predicted TOC value of the verification well is determined based on the three-dimensional TOC attribute volume of the target sparse well area. The difference between the predicted TOC value of the verification well and the measured TOC value of the corresponding verification well is calculated to obtain the verification difference. When the verification difference is within the preset verification difference range, the step of predicting the source rock TOC of the target sparse well area based on the three-dimensional TOC attribute body of the target sparse well area is entered. If the verification difference is not within the preset verification difference range, return to the step of establishing a virtual well in the wellless area.

7. A TOC prediction system for source rocks in sparsely populated areas, applied to the TOC prediction method for source rocks in sparsely populated areas as described in any one of claims 1-6, characterized in that, The TOC prediction system for source rocks in rare well areas includes: The virtual well creation module is used to create virtual wells in areas without existing wells. The virtual well seismic attribute determination module is used to determine the seismic attributes of the virtual well based on the seismic attribute extraction method. The virtual well mudstone section determination module is used to determine the lithology of the virtual well based on the sandstone and mudstone lithology prediction model and the seismic attributes above the virtual well, and to extract the mudstone section of the virtual well; the sandstone and mudstone lithology prediction model is determined based on the measured wells in the target sparse well area; The virtual well TOC prediction module is used to determine the longitudinal continuous TOC prediction value of the source rock in the mudstone section of the virtual well based on the single-well longitudinal continuous TOC prediction model; the single-well longitudinal continuous TOC prediction model is determined based on the measured wells in the target sparse well area; The three-dimensional TOC attribute body construction module is used to construct a three-dimensional TOC attribute body for the target sparse well area based on preset geological conditions and by combining the vertical continuous TOC prediction value and the measured TOC value of the source rock in the mudstone section of the virtual well; the measured TOC value is the measured TOC value corresponding to the measured well in the target sparse well area. The prediction module is used to predict the TOC of the source rocks in the target sparse well area based on the three-dimensional TOC attribute volume of the target sparse well area.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor; The memory is used to store a computer program, and the processor is used to run the computer program to perform the TOC prediction method for hydrocarbon source rocks in rare well areas according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program; When the computer program is executed by the processor, it implements the steps of the TOC prediction method for hydrocarbon source rocks in rare well areas as described in any one of claims 1-6.

Images