A method for constructing a relationship model for identifying lithology of clastic rocks

Abstract

The present invention discloses a method for constructing a relationship model for identifying clastic rock lithology, which relates to the technical field of lithology identification. The method determines the target lithology type after the initial lithology type is simplified and classified by the logging characteristics corresponding to the sample clastic rock classified by the clastic rock core; and constructs a target relationship model according to the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock. The target relationship model is used to classify and identify at least one mudstone type and sandstone types of different particle sizes. In this way, the complex initial lithology type is simplified to the target lithology type with similar logging characteristics, and the target relationship model is constructed, so that different lithology types can be classified and identified, reducing the dependence on traditional core sampling, helping to simplify the classification process of lithology identification, and being able to adapt to different geological conditions and environments.

Description

Technical Field

[0001] The present invention relates to the technical field of lithology identification, and in particular to a method for constructing a relationship model for identifying clastic rock lithology. Background Art

[0002] The logging classification and identification of clastic rock lithology is the basis for the fine evaluation of reservoir logging. Only on the basis of fine classification and identification of lithology can the precision and accuracy of reservoir parameters, fluid types, production capacity evaluation, etc. be improved, and the breadth and depth of application of logging data in geological evaluation can be expanded. Summary of the invention

[0003] In order to overcome the problems of lithology identification in related technologies, such as the lithology description of clastic rock cores is too complex, and different lithologies have similar logging characteristics, the present invention provides a method for constructing a relationship model for identifying clastic rock lithology. It includes:

[0004] Obtaining well logging characteristics of sample clastic rocks classified by initial lithology types, wherein the initial lithology types are classified by clastic rock cores;

[0005] Determine, according to the logging characteristics of the sample clastic rock, a target lithology type after simplified classification of the initial lithology type, wherein the target lithology type includes at least one mudstone type and sandstone types of different grain sizes;

[0006] A target relationship model is constructed according to the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, and the target relationship model is used to classify and identify the at least one mudstone type and the sandstone types with different grain sizes.

[0007] Optionally, a target relationship model is constructed according to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, including:

[0008] According to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, a first relationship model is constructed, wherein the first relationship model is used to classify and identify the clastic rocks of the mudstone type and the sandstone type;

[0009] A second relationship model is constructed according to the logging characteristics of the sample clastic rocks corresponding to the sandstone type, and the second relationship model is used to classify and identify the clastic rocks of the sandstone type with different particle sizes.

[0010] Optionally, the logging characteristics include an amplitude difference between density and neutron and a compensated density logging, and the amplitude difference is determined based on a compensated neutron logging value and a compensated density logging value; and according to the logging characteristics of the sample clastic rock corresponding to the sandstone type, a second relationship model is constructed, including:

[0011] Analyze the logging characteristics of the sample clastic rocks corresponding to the sandstone types of different grain sizes, and establish a cross-plot between the amplitude difference of the sample clastic rocks and the compensated density logging value corresponding to the sandstone types of different grain sizes;

[0012] The second relationship model is constructed according to the intersection plate.

[0013] Optionally, constructing the second relationship model according to the intersection plate includes:

[0014] Using the intersection plate as the second relationship model; or,

[0015] Based on the distribution of the data points of the cross-plot, the objective function relationship between the amplitude difference corresponding to the clastic rocks of sandstone types with different grain sizes and the compensated density logging value is obtained;

[0016] The objective function relationship is used as the second relationship model.

[0017] Optionally, the sandstone types with different particle sizes include a first sandstone type, a second sandstone type, a third sandstone type and a fourth sandstone type, and the particle sizes of the clastic rocks corresponding to the first sandstone type, the second sandstone type, the third sandstone type and the fourth sandstone type decrease in sequence;

[0018] When the objective function relationship is used as the second relationship model, the second relationship model is:

[0019] The first sandstone type: Diff≥-0.7×Den+2.04, Den<2.527g / cm 3 ;

[0020] Second sandstone type: Diff≥-0.7×Den+2.04, Den≥2.527g / cm 3 ; and / or Diff<-2.1×Den+5.5277,2.4912g / cm 3 ≤Den≤2.5973g / cm 3 ; and / or Diff<-0.7×Den+2.04, Den≤2.4912g / cm 3 ; and / or Diff ≥ 1.5552 × Den - 3.9658, Den ≤ 2.5973 g / cm 3 ;

[0021] The third sandstone type: Diff<-2.1×Den+5.5277, Den>2.5973g / cm 3 ; and / or Diff ≥ -2.1 × Den + 5.5277, 2.4912 g / cm 3≤Den≤2.5973g / cm 3 ; and / or Diff≥-1.3726×Den+3.7403, Den≥2.632g / cm3; and / or Diff≥1.5552×Den-3.9658, 2.5973≤Den≤2.632g / cm 3 ;

[0022] The fourth sandstone type: Diff<-1.3726×Den+3.7403, Den≥2.632g / cm 3 ; and / or Diff<1.5552×Den-3.9658, Den≤2.632g / cm 3 ;

[0023] Where Diff is the amplitude difference between the density and neutron of sandstone-type clastic rocks; Den is the compensated density logging value.

[0024] Optionally, the difference in magnitude between density and neutrons is determined by the following calculation:

[0025] ,

[0026] Where Diff is the amplitude difference between the density and neutron of sandstone-type clastic rocks; Den is the compensated density logging value; CNL 校 compensated neutron logging values ​​after excavation correction;

[0027] CNL 校 Determined by the following calculation formula:

[0028] ,

[0029] Where CNL is the compensated neutron logging value; E is the compensated neutron curve mining correction factor; S w is water saturation; ρ g is the density of natural gas.

[0030] Optionally, the sensitive parameters of clastic rock include natural gamma, compensated neutron and P parameter, wherein the P parameter is used to characterize the change law of the compensated neutron logging value response corresponding to the unit acoustic wave time difference change;

[0031] The first relational model is:

[0032] Sandstone type: meets the target judgment function, wherein the target judgment function is: GR<140API and P>0.587 and CNL<21% or CNL<15%, wherein GR is the natural gamma logging value, P is the P parameter, and CNL is the compensated neutron logging value;

[0033] Mudstone type: does not satisfy the objective judgment function.

[0034] Optionally, the logging characteristics include sensitive parameter data, and determining the target lithology type after simplified classification of the initial lithology type according to the logging characteristics of the sample clastic rock includes:

[0035] Based on the logging characteristics of the sample clastic rocks classified according to the initial lithology type, statistical analysis is performed on the sensitive parameter data of the logging characteristics, and by analyzing the flow stratification indicators, porosity and permeability relationship and particle size characteristics of the sample clastic rocks, the initial lithology type is simplified and classified to obtain the target lithology type.

[0036] Optionally, the initial lithology types include conglomerate, fine conglomerate, gravelly coarse sandstone, coarse sandstone, gravelly sandstone, gravelly medium sandstone, medium sandstone, fine sandstone, argillaceous fine sandstone, argillaceous siltstone and mudstone;

[0037] The sandstone types with different grain sizes include a first sandstone type, a second sandstone type, a third sandstone type and a fourth sandstone type, wherein the first sandstone type includes sandy conglomerate and fine conglomerate, the second sandstone type includes gravel-containing coarse sandstone, coarse sandstone and gravel sandstone, the third sandstone type includes gravel-containing medium sandstone and medium sandstone, and the fourth sandstone type includes fine sandstone;

[0038] The mudstone types include mud fine sandstone, mud siltstone and mudstone.

[0039] Optionally, after constructing the target relationship model according to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, the method further includes:

[0040] Using the target relationship model to identify the lithology of a preset number of stone box samples, and obtaining identification results;

[0041] Based on the recognition results and the actual lithology of the stone box sample, the recognition compliance rate of the target relationship model is verified.

[0042] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0043] The present invention determines the target lithology type after the initial lithology type is simplified and classified by the logging characteristics corresponding to the sample clastic rock classified by the clastic rock core; and constructs a target relationship model according to the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock. The target relationship model is used to classify and identify at least one mudstone type and sandstone types of different particle sizes. In this way, the complex initial lithology type is simplified to the target lithology type with similar logging characteristics, and the target relationship model is constructed, so that different lithology types can be classified and identified, reducing the dependence on traditional core sampling, helping to simplify the classification process of lithology identification, and being able to adapt to different geological conditions and environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0045] Figure 1 The present invention is a flowchart showing a method for constructing a relationship model for identifying lithology of clastic rocks according to an exemplary embodiment.

[0046] Figure 2 The flowchart is another method for constructing a relationship model for identifying lithology of clastic rocks according to an exemplary embodiment.

[0047] Figure 3 It is a schematic diagram showing the difference characteristics of the density and neutron amplitude of sandstones with different grain sizes according to an exemplary embodiment.

[0048] Figure 4 is a schematic diagram of a cross-plot according to an exemplary embodiment.

[0049] Figure 5 is a frequency histogram of P parameter values ​​of mudstone types according to an exemplary embodiment.

[0050] Figure 6 is a schematic diagram showing the relationship between permeability and porosity according to an exemplary embodiment.

[0051] Figure 7 is a schematic diagram showing the consistency analysis of FZI, porosity-permeability relationship and logging characteristics of different lithologies according to an exemplary embodiment. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the specification and drawings in the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0053] Lithology identification is a basic task in geology, which is used to determine the type and characteristics of rocks. Traditional lithology identification methods rely on experience and expertise, are time-consuming and susceptible to subjective factors.

[0054] For the lithology classification and logging identification of clastic rocks, multi-parameters can be comprehensively applied, and mathematical algorithms such as fuzzy clustering, principal component analysis, neural network, artificial intelligence, multivariate linear discriminant and other methods can be used to identify lithology. Crossplot technology can be used to identify lithology, and parameters sensitive to lithology, such as natural gamma, neutron, density, acoustic wave, resistivity, lithology density and other curves, can be selected, and the lithology can be identified by multi-parameter intersection method. Lithology can also be identified based on a combination of element and imaging logging methods.

[0055] It can be seen that the current logging identification of clastic rocks still focuses on identification as the starting point, without considering classification and identification comprehensively. It is all based on the lithology described by the geology for identification, but this is the important reason for the low accuracy of logging identification of lithology. Among them, there are many types of rocks described by the core, but extremely similar logging characteristics correspond to different lithologies, and logging cannot identify the lithology that corresponds to the core description one by one. In addition, some rock types are relatively rare in the formation, not representative, and the logging characteristics are not obvious and cannot be effectively identified.

[0056] In addition, the porosity-permeability relationship based on the rock classification described by the core still shows a complex relationship and does not have a good correlation. In general, the correlation of the porosity-permeability relationship established by rock type should be better, which can improve the accuracy of permeability calculation. However, actual statistics show that the data points are still scattered when the porosity-permeability relationship is established directly by rock type according to the lithology described by the core. The subjectivity of different core describers leads to the fact that the description results between wells are not completely comparable, that is, the lithology described by different people may be different for the same lithology.

[0057] To this end, the present invention provides a method for constructing a relationship model for identifying lithology of clastic rocks, see Figure 1 ,The method of constructing a relationship model for identifying the lithology of clastic rocks includes the following steps.

[0058] S101, obtaining logging characteristics of sample clastic rocks classified by initial lithology types, wherein the initial lithology types are classified by clastic rock cores;

[0059] S102, determining a target lithology type after simplified classification of the initial lithology type according to the logging characteristics of the sample clastic rock, wherein the target lithology type includes at least one mudstone type and sandstone types of different grain sizes;

[0060] S103: construct a target relationship model according to the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, wherein the target relationship model is used to classify and identify the at least one mudstone type and the sandstone types with different grain sizes.

[0061] For example, clastic rock is a sedimentary rock mainly composed of rock fragments (such as sand, silt, gravel, etc.), which is formed by transportation, deposition and compaction. Well logging characteristics are rock physical property data obtained by logging instruments in boreholes, such as resistivity, acoustic wave velocity, density, neutrons, porosity, etc. The core is a rock sample taken from the borehole, which can be used to directly observe the mineral composition, structure and structure of the rock. The initial lithology type is a rock type obtained by direct observation and analysis of the core sample. The target lithology type is a rock type that is further simplified and classified according to the logging characteristics based on the initial lithology type. In the present invention, the target lithology type specifically refers to mudstone and sandstone of different particle sizes.

[0062] For example, a relational model refers to a model that can describe the relationship between different variables, wherein the relational model can identify and classify lithology according to well logging characteristics.

[0063] For example, physical property data of rocks are obtained in the borehole through logging instruments, including but not limited to resistivity, acoustic wave velocity, density, neutron porosity, etc. According to geological and geophysical conditions, comprehensive logging methods are reasonably selected. There are many logging methods, including basic methods such as electrical, acoustic, and radioactive methods, as well as special methods such as cable formation testing, formation dip logging, imaging logging, nuclear magnetic resonance logging, etc. Then, the collected logging data can be processed to ensure the accuracy and availability of the data. Including correction methods such as borehole environment and mud quality, as well as removal of noise and outliers. Using logging data, combined with core analysis results, the initial lithology type classification of clastic rock samples is carried out. These types may include mudstone, sandstone of different particle sizes, etc.

[0064] For example, based on the logging data of the clastic rock samples corresponding to the initial lithology type described by the core, the initial lithology type can be further simplified and classified through these logging data to determine the target lithology type. Among them, data analysis and statistics can be performed according to preset methods, such as principal component analysis (PCA), convolutional neural network (CNN), etc., to find out the similarities and distinctions of different lithology types in logging characteristics. Based on the analysis results obtained in steps S101 and S102, a target relationship model can be constructed to identify and classify at least one mudstone type and sandstone types of different grain sizes according to logging characteristics. The target relationship model can be a machine learning algorithm that learns how to identify different lithologies through a training data set (known lithology and corresponding logging characteristics). Features that can represent different lithologies are extracted from the logging data. The target relationship model can also be a mathematical statistical model, which is not limited here.

[0065] The present invention determines the target lithology type after the initial lithology type is simplified and classified by the logging characteristics corresponding to the sample clastic rock classified by the clastic rock core; and constructs a target relationship model according to the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock. The target relationship model is used to classify and identify at least one mudstone type and sandstone types of different particle sizes. In this way, the complex initial lithology type is simplified to the target lithology type with similar logging characteristics, and the target relationship model is constructed, so that different lithology types can be classified and identified, reducing the dependence on traditional core sampling, helping to simplify the classification process of lithology identification, and being able to adapt to different geological conditions and environments.

[0066] In addition, the present invention may have better applicability and interpretation ability for highly heterogeneous complex lithology formations such as igneous rocks, and is helpful for effective lithology identification in complex formations. In some embodiments, a target relationship model is constructed based on the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, including:

[0067] According to the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, a first relationship model is constructed, and the first relationship model is used to classify and identify the clastic rocks of the mudstone type and the sandstone type; according to the logging characteristics of the sample clastic rock corresponding to the sandstone type, a second relationship model is constructed, and the second relationship model is used to classify and identify the clastic rocks of the sandstone type with different particle sizes.

[0068] For example, based on the initial lithology type, further simplified and classified according to the logging characteristics, after determining the target lithology type, a target relationship model for identifying at least one mudstone type and clastic rocks of sandstone types of different grain sizes can be established. Among them, the first relationship model is a relationship model obtained based on the logging characteristics and the target lithology type, which can be used to classify and identify clastic rocks of mudstone and sandstone types. According to the first relationship model, as well as the logging characteristics of the clastic rock to be tested, such as natural gamma, acoustic time difference, compensated density logging, compensated neutron logging, and a specific range of acoustic velocity, it can be roughly distinguished whether the clastic rock to be tested is a mudstone type or a sandstone type. The second relationship model is obtained based on the logging characteristics of the sample clastic rock corresponding to the sandstone type, and can be used to further classify and identify sandstone types of different grain sizes. For example, the second relationship model can take into account the logging response of clastic rocks of different grain sizes in sandstone, such as the influence of different grain sizes on acoustic time difference, density logging, and neutron logging.

[0069] For example, a first relationship model can be constructed based on the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock. Specifically, a neural network model can be used for training based on the logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, and a neural network model for classifying and identifying the mudstone type and the sandstone type clastic rock can be fitted as the first relationship model. Statistical analysis can also be used to determine the most important distinguishing features of the mudstone type and the sandstone type in the logging characteristics of the sample clastic rock, so as to construct a mathematical model for distinguishing the clastic rocks of the mudstone type and the sandstone type as the first relationship model, which is not limited here.

[0070] For example, the second relationship model can be constructed based on the logging characteristics of the sample clastic rocks corresponding to the sandstone type. Specifically, a neural network model can be used for training based on the logging characteristics of the sample clastic rocks corresponding to the sandstone type, and a neural network model for further subclassifying clastic rocks of sandstone types with different grain sizes can be fitted as the second relationship model. Statistical analysis can also be used to determine the most important distinguishing features of sandstone types with different grain sizes in the logging characteristics of the sample clastic rocks corresponding to the sandstone type, so as to construct a mathematical model for distinguishing clastic rocks of the sandstone type with different grain sizes as the second relationship model, which is not limited here.

[0071] The present invention trains the model through sample data of known lithology, and evaluates the performance and accuracy of the model through methods such as cross-validation, etc. It can more accurately identify and classify the lithology of clastic rocks, providing important technical support for geological exploration and resource assessment.

[0072] In some embodiments, the sensitive parameters of clastic rock include natural gamma, compensated neutron and P parameter, and the P parameter is used to characterize the change law of the compensated neutron logging value response corresponding to the unit acoustic wave time difference change; the first relationship model is: sandstone type: meet the target judgment function, wherein the target judgment function is: GR<140API and P>0.587and CNL<21% or CNL<15%, wherein GR is the natural gamma logging value, P is the P parameter, and CNL is the compensated neutron logging value;

[0073] Mudstone type: does not satisfy the objective judgment function.

[0074] For example, sensitive parameters refer to parameters that can significantly reflect the physical properties of rocks during logging. In this scheme, sensitive parameters include natural gamma, compensated neutron and P parameters. Among them, natural gamma logging is a method of identifying rock types by measuring the intensity of gamma rays emitted by natural radioactivity in the formation. Different rocks have different natural gamma intensities because they contain different amounts of radioactive elements. For example, rocks with higher shale content usually show stronger natural gamma ray intensities. Compensated neutron logging is a method of determining the porosity and fluid type of the formation by measuring the deceleration ability of the formation to neutrons. This method can distinguish rocks with different hydrogen contents, such as water-saturated rocks and gas-saturated rocks. The P parameter is a specific logging parameter used to characterize the changing law of the compensated neutron logging value response corresponding to the unit acoustic wave time difference change. The P parameter eliminates the influence of porosity to a certain extent and directly reflects the characteristics of lithology, so the P parameter helps to further distinguish rock types, especially between different grain sizes of sandstone.

[0075] For example, the target judgment function combines the natural gamma logging value (GR), the compensated neutron logging value (CNL) and the P parameter to distinguish between mudstone and sandstone types in rock types. Specifically, when GR < 140API, P > 0.587, and CNL < 21%, or when GR < 140API, P > 0.587, and CNL < 15%, it can be determined that the clastic rock is a sandstone type. However, if the above target judgment function is not met, it can be determined that the clastic rock is a mudstone type. In other words, this function determines that other types except sandstone types are mudstone types.

[0076] In the present invention, the natural gamma, compensated neutron and acoustic time difference and other sensitive parameter data of the clastic rock samples can be collected by the logging instrument, and the first relationship model can be constructed, and the target judgment function can be used to classify the rock. For example, if the logging data shows that GR is less than 140API, P is greater than 0.587, and CNL is less than 21% or CNL is less than 15%, the rock type is judged to be sandstone. If the logging data does not meet the target judgment function, the rock type is judged to be mudstone. In this way, the lithology of clastic rocks can be more accurately identified and classified, providing important technical support for geological exploration and resource assessment.

[0077] In some embodiments, the logging characteristics include sensitive parameter data, and determining the target lithology type after the simplified classification of the initial lithology type according to the logging characteristics of the sample clastic rock includes:

[0078] Based on the logging characteristics of the sample clastic rocks classified according to the initial lithology type, statistical analysis is performed on the sensitive parameter data of the logging characteristics, and by analyzing the flow stratification indicators, porosity and permeability relationship and particle size characteristics of the sample clastic rocks, the initial lithology type is simplified and classified to obtain the target lithology type.

[0079] For example, sensitive parameters are parameters that have a significant impact on lithology identification during logging. Sensitive parameters include natural gamma, compensated neutron and P parameters. Natural gamma logging value (GR) reflects the radioactivity of the formation, compensated neutron logging value (CNL) reflects the porosity and fluid type of the formation, and P parameter characterizes the change law of the compensated neutron logging value response corresponding to the unit acoustic wave time difference change, which is used to further distinguish lithology.

[0080] For example, by performing statistical analysis on the sensitive parameter data corresponding to the sample clastic rocks to find out the relationships, patterns and trends between the data, the lithology types with similar logging characteristics can be classified. Flow Zone Indicator (FZI) is a variable related to the reservoir porosity and effective hydraulic radius, which can be defined by the ratio of the reservoir quality index to the normalized porosity. The porosity-permeability relationship is the relationship between the porosity and permeability of the rock. Porosity is the ratio of the volume of pores in the rock to the total volume, while permeability is the ability of fluid to pass through the rock. The relationship between the two is crucial to understanding the flow of fluids in rocks. Grain size characteristics are the size distribution of particles in the rock. Grain size characteristics can affect the physical properties of the rock, such as porosity, permeability and strength of the rock.

[0081] For example, based on the sensitive parameter data corresponding to the sample clastic rock, including sensitive parameters such as natural gamma, compensated neutron and acoustic time difference, the characteristic patterns of different lithological types are identified. Then, the flow stratification index, porosity and permeability relationship and grain size characteristics of the sample clastic rock are analyzed. Based on the above analysis, the initial lithological type is simplified and corrected to the target lithological type, that is, the complex lithological type is classified into categories that are easier to manage and interpret, such as mudstone and sandstone of different grain sizes.

[0082] The present invention can realize the classification of rock types with the same FZI and porosity-permeability relationship and similar logging characteristics as one rock type for research, for example, sandstone and fine conglomerate with the same FZI and porosity-permeability relationship and similar logging characteristics can be classified into one type. At the same time, it can also realize the correction of the lithology described by the core. Some well sections have the same FZI and porosity-permeability relationship, similar logging characteristics, and similar particle size characteristics. Because the core description has a strong personal subjectivity, the described lithology is different, so the core described by the core is corrected.

[0083] In some embodiments, the initial lithology type includes conglomerate, fine conglomerate, gravel-containing coarse sandstone, coarse sandstone, gravelly sandstone, gravel-containing medium sandstone, medium sandstone, fine sandstone, muddy fine sandstone, muddy siltstone and mudstone; the sandstone types with different grain sizes include a first sandstone type, a second sandstone type, a third sandstone type and a fourth sandstone type, wherein the first sandstone type includes conglomerate and fine conglomerate, the second sandstone type includes gravel-containing coarse sandstone, coarse sandstone and gravel sandstone, the third sandstone type includes gravel-containing medium sandstone and medium sandstone, and the fourth sandstone type includes fine sandstone;

[0084] The mudstone types include mud fine sandstone, mud siltstone and mudstone.

[0085] In some embodiments, the logging characteristics include an amplitude difference between density and neutron and a compensated density logging, wherein the amplitude difference is determined based on a compensated neutron logging value and a compensated density logging value; and a second relationship model is constructed according to the logging characteristics of the sample clastic rock corresponding to the sandstone type, including:

[0086] Analyze the logging characteristics of the sample clastic rocks corresponding to the sandstone types of different grain sizes, and establish a cross-plot between the amplitude difference of the sample clastic rocks and the compensated density logging value corresponding to the sandstone types of different grain sizes;

[0087] The second relationship model is constructed according to the intersection plate.

[0088] For example, in well logging, amplitude difference refers to the difference between compensated neutron logging value and compensated density logging value. This difference can reflect the porosity and fluid type of the rock, and helps to distinguish different lithologies. Compensated density logging is a density logging method that corrects the influence of the well wall and mud cake, and can more accurately reflect the density characteristics of the formation. In well logging, the intersection plate is a graphical tool used to analyze and interpret well logging data. By intersecting the values ​​of different logging parameters on the plate, the lithology and porosity of the rock and other characteristics can be determined. The inventors compared a large amount of well logging data and found that the most obvious difference between sandstones of different particle sizes is the amplitude difference between density and neutron. For details, please refer to Figure 3 The finer the sandstone grain size, the smaller the difference in density and neutron amplitude. This is because the finer the rock grain size, the larger the rock specific surface area, the higher the bound water saturation, and the more obvious the increase in neutron porosity.

[0089] In addition, the amplitude difference between density and neutron is also affected by factors such as gas content and physical properties. Affected by gas content, the amplitude difference between neutron and density is significantly larger, and gas correction is required.

[0090] In some embodiments, the difference in magnitude between density and neutrons is determined by the following calculation:

[0091] ,

[0092] Where Diff is the amplitude difference between the density and neutron of sandstone-type clastic rock; Den is the compensated density logging value; CNL 校 compensated neutron logging values ​​after correction for mining;

[0093] CNL 校 Determined by the following calculation formula:

[0094] ,

[0095] Where CNL is the compensated neutron logging value; E is the compensated neutron curve mining correction factor; S w is water saturation; ρ g is the density of natural gas.

[0096] For sandstone with relatively poor physical properties, the amplitude difference between neutron and density is relatively small, which reflects that the amplitude difference of neutron density changes with the porosity. The change of porosity should be considered when identifying lithology. Therefore, the cross plot of the amplitude difference between neutron density and the compensated density logging value can be used to identify lithology, which describes both the amplitude difference between neutron and density and the change of porosity. Figure 4, the X-axis is the compensated density logging value Den, and the Y-axis is the amplitude difference value Diff between neutron densities. Each sandstone type occupies a corresponding plate of the cross-plot.

[0097] In some embodiments, constructing the second relationship model according to the intersection diagram includes:

[0098] Using the intersection plate as the second relationship model; or,

[0099] Based on the distribution of the data points of the cross-plot, the objective function relationship between the amplitude difference corresponding to the clastic rocks of sandstone types with different grain sizes and the compensated density logging value is obtained;

[0100] The objective function relationship is used as the second relationship model.

[0101] For example, the intersection chart can directly reflect the relationship between the amplitude difference and the compensated density logging value corresponding to the clastic rocks of sandstone types with different particle sizes. The intersection chart production can use the logging data to draw data points on the intersection chart and analyze the distribution of the data points. For example, based on the amplitude difference parameters and compensated density logging values ​​of the clastic rocks of sandstone types with different particle sizes, its coordinate position in the intersection chart can be determined, and the specific lithology type of the clastic rocks can be determined based on the coordinate position. In this way, it can be more intuitive to find the specific lithology type of clastic rocks based on the intersection chart.

[0102] For example, the objective function relationship refers to a mathematical expression that describes the relationship between the amplitude difference and the compensated density logging value. Some computer applications may not be able to recognize the data of the intersection plate, or may not be able to directly use the intersection plate as a relationship model. In mathematical relationship modeling, because it can be implemented in code, the functional relationship can be used as a more convenient way for computer input and reading. Therefore, based on the intersection plate, the functional relationship between the amplitude difference and the compensated density logging value corresponding to the clastic rock of each category of sandstone type with different grain sizes can be statistically calculated, and the statistical functional relationship can be used as the second relationship model. In this way, the second relationship model can be directly used for user calculations, or it can be installed in a computer to achieve the purpose of obtaining the sandstone type corresponding to the clastic rock after the user inputs the amplitude difference and the compensated density logging value corresponding to the clastic rock, which is conducive to the automatic identification of lithology, making the identification of lithology more convenient and faster.

[0103] In some embodiments, the sandstone types of different particle sizes include a first sandstone type, a second sandstone type, a third sandstone type and a fourth sandstone type, and the particle sizes of the clastic rocks corresponding to the first sandstone type, the second sandstone type, the third sandstone type and the fourth sandstone type decrease in sequence.

[0104] For example, see Figure 4 , the first sandstone type may include Figure 4 The sandstone type corresponding to category I, the second sandstone type may include Figure 4 The sandstone type corresponding to category II-a and the sandstone type corresponding to category II-b, the third sandstone type can include Figure 4 The sandstone types corresponding to category III, the fourth sandstone type can include Figure 4 The sandstone type corresponding to Category IV. Figure 4 By statistically analyzing the data points in the intersection diagram, the second relationship model corresponding to the clastic rocks of sandstone types with different grain sizes can be determined.

[0105] Specifically, when the objective function relationship is used as the second relationship model, the second relationship model is:

[0106] The first sandstone type: Diff≥-0.7×Den+2.04, Den<2.527g / cm 3 ;

[0107] Second sandstone type: Diff≥-0.7×Den+2.04, Den≥2.527g / cm 3 ; and / or Diff<-2.1×Den+5.5277,2.4912g / cm 3 ≤Den≤2.5973g / cm 3 ; and / or Diff<-0.7×Den+2.04, Den≤2.4912g / cm 3 ; and / or Diff ≥ 1.5552 × Den - 3.9658, Den ≤ 2.5973 g / cm 3 ;

[0108] The third sandstone type: Diff<-2.1×Den+5.5277, Den>2.5973g / cm 3 ; and / or Diff ≥ -2.1 × Den + 5.5277, 2.4912 g / cm 3 ≤Den≤2.5973g / cm 3 ; and / or Diff≥-1.3726×Den+3.7403, Den≥2.632g / cm3; and / or Diff≥1.5552×Den-3.9658, 2.5973≤Den≤2.632g / cm 3 ;

[0109] The fourth sandstone type: Diff<-1.3726×Den+3.7403, Den≥2.632g / cm 3 ; and / or Diff<1.5552×Den-3.9658, Den≤2.632g / cm 3 .

[0110] Here, determining the objective function relationship of the category I region corresponding to the sandstone type I may include:

[0111] Through the Diff and Den corresponding to the scattered points of sandstone type I in the cross-plot, a linear fit was performed to determine the initial function Diff = -0.7 × Den + 2.08;

[0112] After changing the intercept of the initial function and the value of the initial function Den, the percentage C1 of the scattered points of sandstone type I in the category I area, the percentage C2 of the scattered points of sandstone type II in the category II-a area, and the percentage C3 of the scattered points of sandstone types III and IV in the category I and category II-a areas are calculated;

[0113] when When the value of is the largest, determine the corresponding objective function relationship.

[0114] That is, the objective function relationship of the category I area corresponding to sandstone type I is: Diff ≥ -0.7 × Den + 2.04, where Den < 2.527 g / cm 3 .

[0115] Here, the objective function relationship for determining the category II-a region corresponding to sandstone type II may include:

[0116] Through the Diff and Den corresponding to the scattered points of sandstone type I in the cross-plot, a linear fit is performed to determine the initial function Diff=-0.7×Den+2.08; after changing the intercept of the initial function and the value of the initial function Den, the percentage C4 of the scattered points of sandstone type I distributed in the category I area and the percentage C5 of the sandstone type II distributed in the category II-a area are statistically calculated; when When the value of is the largest, determine the corresponding objective function relationship.

[0117] That is, the objective function relationship of the category II-a area corresponding to sandstone type II is: Diff ≥ -0.7 × Den + 2.04, where Den ≥ 2.527 g / cm 3 .

[0118] Here, the objective function relationship for determining the category IV region of sandstone type IV may include:

[0119] Determine the initial functions Diff = -a1 × Den + b1 and Diff = a2 × Den + b2. These two linear functions can determine a quadrilateral area (see Appendix Figure 4Category IV in the ); changing the slopes a1 and a2, and the intercepts b1 and b2 of the two initial functions, statistically analyzing the percentage of scattered points of sandstone type IV in the category IV area C6, and statistically analyzing the percentage of scattered points of sandstone types II and III in the category IV area C7; when When the value is the largest, the corresponding objective function relationship is determined.

[0120] That is, the objective function relationship of the category IV area of ​​sandstone type IV is: Diff<-1.3726×Den+3.7403, where Den≥2.632g / cm 3 , and Diff<1.5552×Den-3.9658, where Den≤2.632g / cm 3 .

[0121] Here, the objective function relationship for determining the category III region corresponding to sandstone type III may include:

[0122] On the basis that category I, category II-a, and category IV have been determined, the area between the two is further distinguished, and the initial function Diff = -a3 × Den + b3 is determined;

[0123] By changing the slope a3 and intercept b3 in the initial function, the percentage C8 of the scattered points of sandstone type III in the category III area and the percentage C9 of the scattered points of sandstone type II in the category II-b area are calculated respectively; when the average values ​​of the percentages C8 and C9 are the largest, the corresponding objective function relationship is determined.

[0124] That is, the objective function relationship of the category III area corresponding to sandstone type III is: Diff<-2.1×Den+5.5277, where Den>2.5973g / cm 3 ; and Diff ≥ -2.1 × Den + 5.5277, where 2.4912 g / cm 3 ≤Den≤2.5973g / cm 3 ; and Diff ≥ -1.3726 × Den + 3.7403, where Den ≥ 2.632 g / cm 3 ; and Diff ≥ 1.5552 × Den-3.9658, where 2.5973 g / cm 3 ≤Den≤2.632g / cm 3 .

[0125] Here, the objective function relationship for determining the category II-b region corresponding to sandstone type II may include:

[0126] On the basis that category I, category II-a, and category IV have been determined, the area between the two is further distinguished, and the initial function Diff = -a4 × Den + b4 is determined;

[0127] By changing the slope a4 and intercept b4 in the initial function, the percentage C10 of the scattered points of sandstone type III in the category III area and the percentage C11 of the scattered points of sandstone type II in the category II-b area are calculated respectively. When the average values ​​of the percentages C10 and C11 are the largest, the corresponding objective function relationship is determined.

[0128] That is, the objective function relationship of the category II-b area corresponding to sandstone type II is: Diff<-2.1×Den+5.5277, where 2.4912g / cm 3 ≤Den≤2.5973g / cm 3 ; and Diff<-0.7×Den+2.04, where Den≤2.4912g / cm 3 ; and Diff ≥ 1.5552 × Den - 3.9658, where Den ≤ 2.5973 g / cm 3 .

[0129] In some embodiments, after constructing the target relationship model according to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, the method further includes:

[0130] Using the target relationship model to identify the lithology of a preset number of stone box samples, and obtaining identification results;

[0131] Based on the recognition results and the actual lithology of the stone box sample, the recognition compliance rate of the target relationship model is verified.

[0132] For example, a stone box sample is a rock sample collected from a specific geological layer (such as the stone box formation) and can be used for geological analysis and lithology identification. The recognition conformity rate refers to the degree of consistency between the recognition result of the lithology type of the stone box sample by the target relationship model and the actual lithology of the stone box sample, which can be an important indicator for evaluating the performance of the target relationship model.

[0133] For example, after the relationship model for identifying the lithology of clastic rocks is constructed as above, the recognition compliance rate of the target relationship model can also be verified based on the stone box samples. The target relationship model is used to identify the lithology of a preset number of stone box samples, and then the recognition results are compared with the actual lithology of the samples to calculate the recognition compliance rate to verify the accuracy and reliability of the model. According to the verification results, the model is adjusted and optimized to improve its performance in practical applications.

[0134] In a specific implementation manner, to facilitate understanding of the present invention, this embodiment takes an exemplary sample clastic rock box set as an example to further illustrate the method for constructing a relationship model for identifying clastic rock lithology involved in the present invention.

[0135] In this embodiment, see Figure 2 , a method for constructing a relationship model for identifying lithology of clastic rocks, specifically comprising the following steps:

[0136] Step 401: Lithological geological classification and sensitivity analysis of well logging curves to different rock types

[0137] According to the lithology described by the core, 11 initial lithology types are determined, and the initial lithology types are determined subjectively based on the core description. Among them, the initial lithology types include sandy conglomerate, fine conglomerate, gravel-bearing coarse sandstone, coarse sandstone, gravelly sandstone, gravelly medium sandstone, medium sandstone, fine sandstone, argillaceous fine sandstone, argillaceous siltstone and mudstone. As shown in Table 1 below, the quantity of each lithology type in the sample clastic rock box group is determined based on the initial lithology type.

[0138] Table 1 Lithology classification of sample clastic rock box group

[0139]

[0140] According to Table 1, the sensitivity characteristics of lithology sensitive parameters to different lithology types are analyzed. The parameters sensitive to lithology types include natural gamma, compensated neutron, and P parameter. The frequency histogram analysis of the natural gamma characteristics of clastic rocks of different particle sizes shows that:

[0141] Sandstone and conglomerate: The natural gamma of typical layers is generally lower than 50API, but also distributed between 50 and 70API;

[0142] Gravel-bearing coarse sandstone, coarse sandstone and gravelly sandstone: natural gamma is mainly distributed between 48 and 85PI;

[0143] Gravel-bearing medium sandstone and medium sandstone: natural gamma is mainly distributed between 65 and 105PI;

[0144] Fine sandstone: natural gamma is generally greater than 100API;

[0145] Mudstone: Natural gamma is generally greater than 140API.

[0146] The natural gamma characteristics of the above-mentioned clastic rocks of different particle sizes are summarized in Table 2. The study of the natural gamma characteristics of clastic rocks of different particle sizes can be used to verify the automatic identification results of well logging lithology.

[0147] Table 2 Well logging classification of different lithologies of the Shihezi Formation and their corresponding natural gamma characteristics

[0148]

[0149] The statistics of the natural gamma frequency histogram of sandstone and mudstone types show that among the sandstone types corresponding to No. 1 to No. 8 in Table 2, the data with GR < 140API account for 95%. The high natural gamma sandstone with a gamma greater than 140API may be mainly caused by rock fragments and potassium feldspar. For this type of sandstone, neutrons can be used to identify sandstones with high GR. Among the mudstone types corresponding to the lithology of No. 9 to No. 12 in Table 2, the data with GR < 140API are mainly in the expansion section and interface, accounting for about 24% of the data, see Appendix. Figure 3 ; GR<140API can identify 95% of sand, but 24% of mud in mudstone is also identified as sandstone.

[0150] Among sandstone types, 98.5% of the data with CNL < 21% are sandstone. Among mudstone types, there are also many mudstones with CNL < 21%. 30% of mudstones have CNL < 21%, but the natural gamma is high, which can be further identified in combination with the natural gamma.

[0151] A frequency histogram of P values ​​was drawn for sandstone and mudstone types. The sandstone types had P values ​​greater than 0.587 accounting for 99.5%, and the mudstone types had P values ​​less than 0.587 accounting for 14.2%.

[0152] Among them, the most obvious difference between sandstones of different particle sizes is the difference in density and neutron amplitude. The finer the particle size, the smaller the difference in density and neutron amplitude. This is because the finer the particle size of the rock, the larger the specific surface area of ​​the rock, the higher the bound water saturation, and the more obvious the increase in neutron porosity.

[0153] Step 402: Analysis of FZI, porosity and permeability relationship and particle size characteristics of different lithologies

[0154] The FZI method is the Kozeny-Carman semi-empirical formula, which is used to express the relationship between rock permeability (K) and porosity (ψ) and rock specific surface area (S). The Kozeny-Carman formula is as follows (4)-(9):

[0155] = (4)

[0156] = (5)

[0157] (6)

[0158] (7)

[0159] = (8)

[0160] (9)

[0161] Where K is the permeability (μm2); φ e is the effective porosity, a decimal number, which can be obtained from core analysis data; F S τ 2 ——Kozeny constant, which can vary from 5 to 100; S gv is the surface area per unit particle volume (μm -1 );I FZ I is the flow zone index; RQ It is the reservoir quality index, in μm, which characterizes the reservoir seepage capacity. The larger the value, the stronger the reservoir seepage capacity.

[0162] According to the transformed formula (9) and Figure 5 It can be seen that on the double logarithmic coordinate graph, the same pore structure has the same I FZ , different pore structures are located on different parallel lines, and the slope of different parallel lines is 1. Figure 6 As shown, the relationship between permeability and porosity of different lithologies can be established.

[0163] Step 403: perform logging classification on lithology types based on FZI, porosity-permeability relationship, and grain size data.

[0164] like Figure 7 As shown in the figure, the A-1 and A-3 sections are gravel-bearing coarse sandstone, and the core of the A-2 section is described as coarse sandstone. From the perspective of FZI and porosity-permeability relationship, the FZI and porosity-permeability relationship of the A-1, A-2 and A-3 sections are all on the same trend line, indicating that the rocks corresponding to the A-1, A-2 and A-3 sections have the same seepage characteristics; from the perspective of well logging characteristics, the well logging characteristics corresponding to the A-2 and A-3 sections are similar; from the perspective of grain size characteristics, the grain size characteristics of A-1 and A-2 are basically the same, indicating that the lithology types of A-1 and A-2 are the same. Based on this, coarse sandstone and gravel-bearing coarse sandstone can be classified as one lithology type. This is one level of logging lithology classification, that is, the rock types with the same seepage characteristics and similar logging characteristics are classified as one type.

[0165] For the B-1 and B-2 sections, the upper section of B-1 is medium sandstone, the lower section is coarse sandstone, and the B-2 section is medium sandstone. The FZI and porosity-permeability relationship of the B-1 and B-2 sections are consistent, and the logging characteristics are similar. Therefore, the B-1 and B-2 sections belong to the same lithology type. Combined with the core description, the lithology of the two sections is mainly medium sandstone. After comprehensive consideration, the coarse sandstone lithology of the lower section of B-1 is corrected to medium sandstone. This is the second level of logging lithology classification of lithology. The lithology described by the core is corrected. The FZI and porosity-permeability relationship of some sections are the same, and the logging characteristics and grain size characteristics are similar. Because the core description has a strong personal subjectivity, the described lithology is different, so the core description of the core is corrected.

[0166] Based on this research idea, the initial lithology types described by cores were simplified into five categories, namely Type I, including sandy conglomerate and fine conglomerate; Type II, gravel-bearing coarse sandstone, coarse sandstone and gravelly sandstone; Type III, composed of gravel-bearing medium sandstone and medium sandstone; Type IV, fine sandstone; Type V, non-reservoir, lithology includes muddy fine sandstone, muddy siltstone, muddy medium sandstone and mudstone.

[0167] Step 404, distinguish between sandstone and mudstone. Natural gamma (GR), compensated neutron (CNL) and P parameters can be used for distinction. Among them, sandstone includes type I, type II, type III and type IV. Based on the frequency histogram analysis and sensitive feature analysis of natural gamma, compensated neutron and P value, a comprehensive three-parameter lithology identification method is established, and the discriminant function is:

[0168] Sandstone: GR<140 and P>0.587 and CNL<21 or CNL<15;

[0169] Mudstone: Mudstone is other than sandstone.

[0170] The lithology of the core wells was identified based on the above function. The identification results showed that the above method had a 95% identification accuracy for sandstone and a 90% identification accuracy for mudstone.

[0171] Step 405: Further identify the lithology of sandstone. Identify the lithology by using the difference between neutron and density (Diff) and the intersection of the density curve (Den), and identify the lithology based on clastic rocks of different particle sizes.

[0172] Among them, through the comparison of a large number of logging data, it is found that the most obvious difference between sandstones of different particle sizes is the difference in density and neutron amplitude. The finer the particle size, the smaller the difference in density and neutron amplitude. This is because the finer the particle size of the rock, the larger the specific surface area of ​​the rock, the higher the bound water saturation, and the more obvious the increase in neutron porosity.

[0173] The amplitude difference between density and neutron is also affected by factors such as gas content and physical properties. Affected by gas content, the amplitude difference between neutron and density is significantly larger, and gas correction is required. At the same time, for sandstone with relatively poor physical properties, the amplitude difference between neutron and density is relatively small, reflecting that the amplitude difference between neutron and density changes with porosity. The change of porosity should be considered when identifying lithology. The intersection chart of the amplitude difference between neutron and density Diff and Den can be used to identify lithology, such as Figure 4 As shown, both the amplitude difference between neutrons and density and the change in porosity are described.

[0174] Determining the functional relationship corresponding to different lithologies in the cross-plot may include the following process: Here, determining the target functional relationship of the category I region corresponding to the sandstone type I may include: performing linear fitting through the Diff and Den corresponding to the scattered points of the sandstone type I in the cross-plot to determine the initial function Diff=-0.7×Den+2.08;

[0175] After changing the intercept of the initial function and the value of the initial function Den, the percentage C1 of the scattered points of sandstone type I in the category I area, the percentage C2 of the scattered points of sandstone type II in the category II-a area, and the percentage C3 of the scattered points of sandstone types III and IV in the category I and category II-a areas are calculated;

[0176] when When the value of is the largest, determine the corresponding objective function relationship.

[0177] That is, the objective function relationship of the category I area corresponding to sandstone type I is: Diff ≥ -0.7 × Den + 2.04, where Den < 2.527 g / cm 3 .

[0178] Here, the objective function relationship for determining the category II-a region corresponding to sandstone type II may include:

[0179] Through the Diff and Den corresponding to the scattered points of sandstone type I in the cross-plot, a linear fit was performed to determine the initial function Diff = -0.7 × Den + 2.08;

[0180] After changing the intercept of the initial function and the value of the initial function Den, the percentage of the scattered points of sandstone type I in the category I area C4 is calculated, and the percentage of the scattered points of sandstone type II in the category II-a area C5 is calculated;

[0181] when When the value of is the largest, determine the corresponding objective function relationship.

[0182] That is, the objective function relationship of the category II-a area corresponding to sandstone type II is: Diff ≥ -0.7 × Den + 2.04, where Den ≥ 2.527 g / cm 3 .

[0183] Here, the objective function relationship for determining the category IV region of sandstone type IV may include:

[0184] Determine the initial functions Diff = -a1 × Den + b1 and Diff = a2 × Den + b2. These two linear functions can determine a quadrilateral area (see Appendix Figure 4 Category IV in the

[0185] By changing the slopes a1 and a2, and the intercepts b1 and b2 of the two initial functions, the percentage C6 of the scattered points of sandstone type IV in the category IV area is calculated, and the percentage C7 of the scattered points of sandstone types II and III in the category IV area is calculated;

[0186] when When the value is the largest, the corresponding objective function relationship is determined.

[0187] That is, the objective function relationship of the category IV area of ​​sandstone type IV is: Diff<-1.3726×Den+3.7403, where Den≥2.632g / cm 3 , and Diff<1.5552×Den-3.9658, where Den≤2.632g / cm 3 .

[0188] Here, the objective function relationship for determining the category III region corresponding to sandstone type III may include:

[0189] On the basis that category I, category II-a, and category IV have been determined, the area between the two is further distinguished, and the initial function Diff = -a3 × Den + b3 is determined;

[0190] By changing the slope a3 and intercept b3 in the initial function, the percentage C8 of the scattered points of sandstone type III in the category III area and the percentage C9 of the scattered points of sandstone type II in the category II-b area are calculated respectively; when the average values ​​of the percentages C8 and C9 are the largest, the corresponding objective function relationship is determined.

[0191] That is, the objective function relationship of the category III area corresponding to sandstone type III is: Diff<-2.1×Den+5.5277, where Den>2.5973g / cm 3 ; and Diff ≥ -2.1 × Den + 5.5277, where 2.4912 g / cm 3 ≤Den≤2.5973g / cm 3; and Diff ≥ -1.3726 × Den + 3.7403, where Den ≥ 2.632 g / cm 3 ; and Diff ≥ 1.5552 × Den-3.9658, where 2.5973 g / cm 3 ≤Den≤2.632g / cm 3 .

[0192] Here, determining the objective function relationship of the category II-b area corresponding to the sandstone type II may include: on the basis of the categories I, II-a, and IV having been determined, further distinguishing the area between the two, and determining the initial function Diff = -a4 × Den + b4;

[0193] By changing the slope a4 and intercept b4 in the initial function, the percentage C10 of the scattered points of sandstone type III in the category III area and the percentage C11 of the scattered points of sandstone type II in the category II-b area are calculated respectively. When the average values ​​of the percentages C10 and C11 are the largest, the corresponding objective function relationship is determined. That is, the objective function relationship of the category II-b area corresponding to sandstone type II is: Diff<-2.1×Den+5.5277, where 2.4912g / cm 3 ≤Den≤2.5973g / cm 3 ; and Diff<-0.7×Den+2.04, where Den≤2.4912g / cm 3 ; and Diff ≥ 1.5552 × Den - 3.9658, where Den ≤ 2.5973 g / cm 3 .

[0194] The objective function relationship of lithology identification established based on the intersection chart shows that the identification results are consistent with the lithology types described by the core, and the average comprehensive coincidence rate of lithology identification in Shihezi is 90.2%.

[0195] Table 3 Logging identification coincidence rate of different lithologies of Shihezi Formation

[0196]

[0197] In the related art, the core description is too complicated, and it is impossible for logging to identify the lithology that corresponds to the core description one by one. The present invention simplifies the lithology described by the core by analyzing similar logging characteristics, consistent porosity and permeability relationships, etc., and classifying rock types that transition to each other, effectively solving the problem of poor consistency of some porosity and permeability relationship points in the same lithology section.

[0198] The present invention is applicable to all clastic oil and gas reservoirs, and mainly improves the existing technology from two aspects: first, the identification of lithology is not completely based on the lithology described by the core, but the lithology described by the core should be simplified and corrected so that the logging data can more effectively identify the lithology; second, the basis and method for simplifying the lithology are proposed, that is, the porosity-permeability relationship, FZI division of flow units, logging characteristics, grain size and other data are combined to simplify the rock type. In the present invention, because the lithology described by the core cannot be directly used as the final lithology to establish the corresponding lithology discrimination standard, the core description, logging characteristics, and grain size analysis data are combined, and the divided lithology is finely adjusted in combination with FZI and porosity-permeability, breaking through the previous technical ideas for lithology identification.

[0199] Finally, it should be noted that the embodiments of the present invention disclose only the preferred embodiments of the present invention, which are only used to illustrate the technical solutions of the present invention rather than to limit them. A person skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a relationship model for identifying lithology of clastic rocks, characterized in that: include: Obtaining well logging characteristics of sample clastic rocks classified by initial lithology types, wherein the initial lithology types are classified by clastic rock cores; Determine, according to the logging characteristics of the sample clastic rock, a target lithology type after simplified classification of the initial lithology type, wherein the target lithology type includes at least one mudstone type and sandstone types of different grain sizes; According to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, a target relationship model is constructed, wherein the target relationship model is used to classify and identify the at least one mudstone type and the sandstone types with different grain sizes; According to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, a target relationship model is constructed, including: According to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, a first relationship model is constructed, wherein the first relationship model is used to classify and identify the clastic rocks of the mudstone type and the sandstone type; A second relationship model is constructed according to the logging characteristics of the sample clastic rocks corresponding to the sandstone type, and the second relationship model is used to classify and identify the clastic rocks of the sandstone type with different particle sizes.

2. The method according to claim 1, characterized in that The logging characteristics include the amplitude difference between density and neutron and the compensated density logging, wherein the amplitude difference is determined based on the compensated neutron logging value and the compensated density logging value; and according to the logging characteristics of the sample clastic rock corresponding to the sandstone type, a second relationship model is constructed, including: Analyze the logging characteristics of the sample clastic rocks corresponding to the sandstone types of different grain sizes, and establish a cross-plot between the amplitude difference of the sample clastic rocks and the compensated density logging value corresponding to the sandstone types of different grain sizes; The second relationship model is constructed according to the intersection plate.

3. The method according to claim 2, characterized in that Constructing the second relationship model according to the intersection plate includes: Using the intersection plate as the second relationship model; or, Based on the distribution of the data points of the cross-plot, the objective function relationship between the amplitude difference corresponding to the clastic rocks of sandstone types with different grain sizes and the compensated density logging value is obtained; The objective function relationship is used as the second relationship model.

4. The method according to claim 3, characterized in that: The sandstone types with different particle sizes include a first sandstone type, a second sandstone type, a third sandstone type and a fourth sandstone type, and the particle sizes of the clastic rocks corresponding to the first sandstone type, the second sandstone type, the third sandstone type and the fourth sandstone type decrease in sequence; When the objective function relationship is used as the second relationship model, the second relationship model is: The first sandstone type: Diff≥-0.7×Den+2.04, Den<2.527g / cm 3 ; Second sandstone type: Diff≥-0.7×Den+2.04, Den≥2.527g / cm 3 ; and / or Diff<-2.1×Den+5.5277,2.4912g / cm 3 ≤Den≤2.5973g / cm 3 ; and / or Diff<-0.7×Den+2.04, Den≤2.4912g / cm 3 ; and / or Diff ≥ 1.5552 × Den - 3.9658, Den ≤ 2.5973 g / cm 3 ; The third sandstone type: Diff<-2.1×Den+5.5277, Den>2.5973g / cm 3 ; and / or Diff ≥ -2.1 × Den + 5.5277, 2.4912 g / cm 3 ≤Den≤2.5973g / cm 3 ; and / or Diff≥-1.3726×Den+3.7403, Den≥2.632g / cm3; and / or Diff≥1.5552×Den-3.9658, 2.5973≤Den≤2.632g / cm 3 ; The fourth sandstone type: Diff<-1.3726×Den+3.7403, Den≥2.632g / cm 3 ; and / or Diff<1.5552×Den-3.9658, Den≤2.632g / cm 3 ; Where Diff is the amplitude difference between the density and neutron of sandstone-type clastic rocks; Den is the compensated density logging value.

5. The method according to claim 2, characterized in that: The difference in magnitude between density and neutrons is determined by the following calculation: ， Where Diff is the amplitude difference between the density and neutron of sandstone-type clastic rock; Den is the compensated density logging value; CNL 校 compensated neutron logging values ​​after correction for mining; CNL 校 Determined by the following calculation formula: ， Where CNL is the compensated neutron logging value; E is the compensated neutron curve mining correction factor; S w is water saturation; ρ g is the density of natural gas.

6. The method according to claim 1, characterized in that The sensitive parameters of clastic rocks include natural gamma, compensated neutron and P parameter, wherein the P parameter is used to characterize the change law of the compensated neutron logging value response corresponding to the unit acoustic wave time difference change; The first relational model is: Sandstone type: meets the target judgment function, wherein the target judgment function is: GR<140API and P>0.587and CNL<21% or CNL<15%, wherein GR is the natural gamma logging value, P is the P parameter, and CNL is the compensated neutron logging value; Mudstone type: does not satisfy the objective judgment function.

7. The method according to any one of claims 1 to 6, characterized in that: The logging characteristics include sensitive parameter data. According to the logging characteristics of the sample clastic rock, the target lithology type after the simplified classification of the initial lithology type is determined, including: Based on the logging characteristics of the sample clastic rocks classified according to the initial lithology type, statistical analysis is performed on the sensitive parameter data of the logging characteristics, and by analyzing the flow stratification indicators, porosity and permeability relationship and particle size characteristics of the sample clastic rocks, the initial lithology type is simplified and classified to obtain the target lithology type.

8. The method according to claim 7, characterized in that The initial lithology types include conglomerate, fine conglomerate, gravelly coarse sandstone, coarse sandstone, gravelly sandstone, gravelly medium sandstone, medium sandstone, fine sandstone, argillaceous fine sandstone, argillaceous siltstone and mudstone; The sandstone types with different grain sizes include a first sandstone type, a second sandstone type, a third sandstone type and a fourth sandstone type, wherein the first sandstone type includes sandy conglomerate and fine conglomerate, the second sandstone type includes gravel-containing coarse sandstone, coarse sandstone and gravel sandstone, the third sandstone type includes gravel-containing medium sandstone and medium sandstone, and the fourth sandstone type includes fine sandstone; The mudstone types include mud fine sandstone, mud siltstone and mudstone.

9. The method according to any one of claims 1 to 6, characterized in that: After constructing the target relationship model according to the well logging characteristics of the sample clastic rock and the target lithology type of the sample clastic rock, the method further includes: Using the target relationship model to identify the lithology of a preset number of stone box samples, and obtaining identification results; Based on the recognition results and the actual lithology of the stone box sample, the recognition compliance rate of the target relationship model is verified.

Images