Three-dimensional space distribution optimization prediction method based on reservoir classification

Abstract

The invention discloses a three-dimensional space distribution optimization prediction method based on reservoir classification, and relates to the field of oil-gas exploration and development, and the method comprises the steps: obtaining the physical property data of a core sample in a target region, and classifying the physical property test points through a clustering analysis method; establishing a three-dimensional geologic model of a target area, and determining favorable main control factors and adverse main control factors which influence physical properties through spatial superposition of different types of reservoir bodies and physical property influence factors; a three-dimensional space influence factor superposition three-dimensional body is established through superposition of various physical property influence factor weight coefficients, physical properties are calculated through physical property influence factor homogenization numerical values and the weight coefficients, and three types of reservoir bodies are divided; and continuously verifying and training the precision degree of the prediction model by adding a new drilling well and a new type of influence factors, and predicting the reservoir type of the well-free region by using the corrected and optimized prediction model. The reservoir prediction precision can be improved, the visualization of the prediction result is enhanced, and the exploration and development efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploration and development, and more specifically to a three-dimensional spatial distribution optimization prediction method based on reservoir classification. Background Art

[0002] Oil and gas exploration is the first key link in oil and gas extraction. It is the foundation of oil and gas extraction projects. Its purpose is to find and identify oil and gas resources, use various exploration methods to understand the underground geological conditions, understand the conditions of oil generation, oil storage, oil and gas migration, accumulation, and preservation, comprehensively evaluate the oil and gas prospects, determine favorable areas for oil and gas accumulation, find oil and gas closures, and explore the area of ​​oil and gas fields, and clarify the oil and gas layer conditions and production capacity.

[0003] Deepwater oil and gas, deep-layer oil and gas, and unconventional oil and gas resources have huge potential. With the obvious deterioration of global oil and gas mining objects and the increase in the difficulty of energy exploration, these three types are gradually becoming new areas of oil and gas exploration in the future. However, oil and gas reservoirs are highly heterogeneous, and it is difficult to predict high-quality reservoirs; at the same time, most existing prediction methods are based on the depth segment of a certain area (one-dimensional) or a certain depth area (two-dimensional), ignoring the three-dimensional concept of the reservoir as a geological body.

[0004] Therefore, how to improve the prediction accuracy of reservoirs, enhance the visualization of prediction results, improve exploration and development efficiency, reduce exploration and development costs, and promote the advancement of oil and gas exploration and development technology is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0005] In view of this, the present invention provides a three-dimensional spatial distribution optimization prediction method based on reservoir classification, which solves the problems existing in the background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A three-dimensional spatial distribution optimization prediction method based on reservoir classification includes the following steps:

[0008] Use nuclear magnetic resonance technology to measure core samples in the target area and obtain physical property data;

[0009] Cluster analysis method is used to classify the physical property test points to form specific types of reservoirs;

[0010] Establish a three-dimensional geological model of the target area, load the factors affecting physical properties, and identify the favorable and unfavorable main controlling factors affecting physical properties through the spatial superposition of different types of reservoirs and factors affecting physical properties;

[0011] By superimposing the weight coefficients of various physical property influencing factors, a three-dimensional space influencing factor superposition three-dimensional volume is established. At the same time, the physical properties are calculated by the homogenized values ​​of the physical property influencing factors and the weight coefficients, and the reservoirs are divided into three categories.

[0012] By drilling new wells and adding new types of influencing factors, the accuracy of the prediction model is continuously verified and trained, and the revised and optimized prediction model is used to predict reservoir types in well-free areas.

[0013] Optionally, using nuclear magnetic resonance technology to measure the core samples of the area to be measured includes the following steps:

[0014] Drill, cut, clean, dry and vacuum the cores in the target area to obtain core samples;

[0015] The porosity of standard samples with known porosity is measured by NMR, and the T2 spectra of different echo intervals are inverted to establish the relationship between porosity and NMR response.

[0016] The processed core samples are subjected to nuclear magnetic resonance T2 spectrum signal detection, and the porosity of the core samples is obtained by using the established relationship between porosity and nuclear magnetic resonance response.

[0017] Optionally, the physical property data include one or more of porosity, permeability, oil saturation, pore size distribution and compressibility.

[0018] Optionally, forming a specific type of reservoir comprises the following steps:

[0019] Establish the upper limit and lower limit of normal porosity of the target area, regard the cores with porosity greater than the upper limit as sweet spot type, regard the cores between the upper limit and the lower limit as effective porosity type, and regard the cores with porosity less than the lower limit as invalid porosity type;

[0020] Cluster analysis was performed based on the location and depth of the physical property test points, and each type of physical property test points that were clustered together were combined to form three types of reservoirs.

[0021] Optionally, a three-dimensional geological model of the target area is established, specifically:

[0022] Through simulation methods, the structural model, sedimentary phase model and interlayer model of the target area are established and used as constraints. The three-dimensional geological model of the target area is established using variogram analysis and sequential Gaussian random simulation algorithm.

[0023] Optionally, factors affecting physical properties include: sedimentary factors and diagenetic factors;

[0024] Sedimentary factors are reflected by sedimentary microfacies, which are determined by core facies, logging facies, and seismic facies;

[0025] Diagenetic factors include pressure and temperature, which are comprehensively calibrated through regional thermal burial history, drilling measured temperature and pressure, and fluid inclusion testing and analysis.

[0026] Optionally, clarifying favorable and unfavorable main controlling factors affecting physical properties includes the following steps:

[0027] Through petrel, the factors affecting physical properties are loaded into the established three-dimensional geological model of the target area;

[0028] Based on the sweet spot type, effective porosity type and invalid porosity type, the divided reservoir is loaded into the established three-dimensional geological model of the target area;

[0029] Different types of reservoirs and factors affecting physical properties are spatially superimposed to identify reservoir intervals, single-factor correlation analysis is performed on factors affecting physical properties and different reservoir types, and favorable and unfavorable main controlling factors affecting physical properties are selected.

[0030] Optionally, the weight coefficient is calculated as follows:

[0031] For each physical property influencing factor, the influence coefficient is calculated and normalized as the weight coefficient of this type of physical property influencing factor; where influence coefficient = Σ physical property parameter point * influencing factor point / number of parameters, positive or negative represents favorable or unfavorable, and the numerical value represents the magnitude of the influence.

[0032] Optionally, the training process of the prediction model performs parameter optimization with prediction accuracy as the objective function. When the prediction accuracy meets the preset requirements, the model training is terminated to obtain a corrected and optimized prediction model.

[0033] It can be seen from the above technical solution that compared with the prior art, the present invention discloses a three-dimensional spatial distribution optimization prediction method based on reservoir classification, which has the following beneficial effects: through three-dimensional geological modeling technology, the prediction results are displayed in the form of three-dimensional spatial distribution. The prediction result is a reservoir in an underground three-dimensional space, not a depth segment calculated by the traditional prediction method based on the correlation between two-dimensional depth and porosity / permeability. The present invention has more intuitive applicability, allowing geologists to more intuitively understand the spatial distribution characteristics of the reservoir, which is convenient for decision-making analysis. In addition, the prediction method can improve the prediction accuracy of the reservoir, enhance the visualization of the prediction results, improve the efficiency of exploration and development, reduce the cost of exploration and development, and promote the advancement of oil and gas exploration and development technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0035] Figure 1 A flow chart of a three-dimensional spatial distribution optimization prediction method based on reservoir classification provided by the present invention;

[0036] Figure 2 A diagram of a testing method for core samples of a test area provided by the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] The embodiment of the present invention discloses a three-dimensional spatial distribution optimization prediction method based on reservoir classification, such as Figure 1 As shown, the following steps are included:

[0039] Use nuclear magnetic resonance technology to measure core samples in the target area and obtain physical property data;

[0040] Cluster analysis method is used to classify the physical property test points to form specific types of reservoirs;

[0041] Establish a three-dimensional geological model of the target area, load the factors affecting physical properties, and identify the favorable and unfavorable main controlling factors affecting physical properties through the spatial superposition of different types of reservoirs and factors affecting physical properties;

[0042] By superimposing the weight coefficients of various physical property influencing factors, a three-dimensional space influencing factor superposition three-dimensional volume is established. At the same time, the physical properties are calculated by the homogenized values ​​of the physical property influencing factors and the weight coefficients, and the reservoirs are divided into three categories.

[0043] By drilling new wells and adding new types of influencing factors, the accuracy of the prediction model is continuously verified and trained, and the revised and optimized prediction model is used to predict reservoir types in well-free areas.

[0044] based on Figure 1The three-dimensional spatial distribution optimization prediction method shown in the present embodiment adopts three-dimensional geological modeling technology, and can display the reservoir type prediction results in the form of three-dimensional spatial distribution, which is convenient for geologists to intuitively understand the spatial distribution characteristics of the reservoir, and solves the problem that the existing prediction methods ignore the three-dimensional concept of the reservoir as a geological body. It has the advantage of establishing a favorable reservoir distribution in the predicted three-dimensional space based on the physical property results.

[0045] Next, Figure 1 The specific process of the three-dimensional spatial distribution optimization prediction method shown is elaborated in detail to achieve a deeper understanding.

[0046] 1. Data collection and preprocessing

[0047] As an emerging core analysis method, nuclear magnetic resonance has the advantages of being non-destructive, convenient and fast. In theory, by directly observing the pore fluid signal of rock samples through nuclear magnetic resonance, it can not only reveal the pore structure characteristics of the rock, but also obtain the reservoir parameters such as the physical properties and oil content of the rock, which provides a good method for multi-parameter analysis and comparison of the same sample.

[0048] Therefore, this embodiment proposes to use nuclear magnetic resonance technology to measure the core samples in the test area, such as Figure 2 As shown, the method specifically comprises the following steps:

[0049] Drill, cut, clean, dry and vacuum the cores in the target area to obtain core samples;

[0050] The standard sample with known porosity is measured by NMR, and the T2 spectra with different echo intervals are obtained by inversion. The relationship between porosity and NMR response is established by using the spectrum area method.

[0051] The processed core samples are subjected to nuclear magnetic resonance T2 spectrum signal detection, and the porosity of the core samples is obtained by using the established relationship between porosity and nuclear magnetic resonance response.

[0052] In this embodiment, the raw data directly obtained by using the nuclear magnetic resonance instrument is a spin echo train with signal amplitude decaying over time. By performing multi-exponential inversion fitting on the echo train data, the T2 spectrum can be obtained. Among them, the echo interval is an important factor affecting the accuracy of nuclear magnetic resonance measurement. The echo interval of existing nuclear magnetic resonance instruments is relatively high, and the application ability in complex and conventional oil and gas such as dense and shale is limited. However, this embodiment takes into account the influence of the echo interval on the intervals of different relaxation components, improves the accuracy of nuclear magnetic resonance porosity measurement of dense cores, and is of great significance.

[0053] Specifically, when obtaining core samples, the cores obtained by drilling and cutting need to be cleaned first, and then placed in a constant temperature box for drying for at least 24 hours. The cores are weighed every 8 hours until the mass change is within the preset range, and the residual cleaning agent after cleaning is removed. Then, the cores are continuously vacuumed for 24 hours using a molecular pump to further remove the impurity gas and moisture adsorbed by the cores. After that, the processed core samples are used for nuclear magnetic resonance T2 spectrum signal detection. This method is easy to operate and has a wide range of application value.

[0054] Furthermore, the physical property data obtained by measuring the core sample may also include: one or more of permeability, oil saturation, pore size distribution and compressibility.

[0055] In addition, after obtaining the physical property data of the core samples, the data can be cleaned, screened, and standardized to ensure the quality and consistency of the data before proceeding with subsequent operations.

[0056] 2. Reservoir division

[0057] Any rock formation that can store and permeate fluid is called a reservoir. The stored fluid is determined by the porosity of the rock, and the permeated fluid is determined by the permeability of the rock. Reservoirs can be classified differently for different research purposes. In this embodiment, by establishing the "upper limit of normal porosity" and "lower limit of normal porosity" in a specific research area, three types of reservoirs are distinguished: sweet spots, effective and ineffective.

[0058] For a specific research area, the normal compaction curve is calculated by the empirical formula of particle size compaction simulation, considering the negative transformation of a certain material base sedimentary rock by destructive diagenesis - the influence of compaction and cementation on pore evolution; the cement content is determined by thin section identification, and the compaction reduction is superimposed on the normal compaction curve by the ratio of primary pores to cement in the surface area. On the porosity-depth profile, the smooth connection curve of porosity variation with depth after the superposition of compaction and cementation is used as the "upper limit of normal porosity".

[0059] For a specific research area, considering the relationship between physical properties and production capacity, the "normal porosity lower limit" is defined as the lower limit of the effective reservoir, that is, no oil and gas is shown below this porosity. Therefore, the physical property lower limit of the effective reservoir at different burial depths can be obtained by using the physical property, oil test, pressure pump and other data, respectively using the distribution function curve method, oil test method, bound water saturation method, etc. On this basis, the regression fitting of the physical property lower limit of the effective reservoir and the burial depth is carried out to obtain the quantitative relationship between the physical property lower limit and the burial depth of the beach bar sand body in the study area, which can eliminate the errors caused by the original data error, the number of basic data, the calculation method, etc. in a single method.

[0060] On this basis, this embodiment proposes a specific method for forming a specific type of reservoir, including the following steps:

[0061] Cores with porosity greater than the upper limit of normal porosity are considered as sweet spots, which are caused by abnormally increased porosity and abnormally preserved porosity geological processes, such as fluid overpressure, oil and gas injection, chlorite coating and dissolution, etc. Cores between the upper limit of normal porosity and the lower limit of normal porosity are considered as effective porosity types, which are affected by compaction and cementation, evolve normally with burial depth, and have a certain amount of oil and gas; cores with porosity less than the lower limit of normal porosity are considered as invalid porosity types, which evolve normally with burial depth but do not have a certain amount of oil and gas;

[0062] The physical property test point is a core sample test at a certain depth of a specific well location, with coordinate information such as position and depth. Therefore, cluster analysis is performed based on the position and depth of the physical property test points, and each type of physical property test points clustered together are combined to form three types of reservoirs.

[0063] 3. Establishment of 3D geological model

[0064] The three-dimensional geological model can characterize the three-dimensional spatial distribution and change model of formation parameters. The three-dimensional geological model of a reservoir is the core, which is a specific representation of the distribution and change of reservoir characteristics and its heterogeneity in three-dimensional space.

[0065] Therefore, this embodiment proposes a method for establishing a three-dimensional geological model of a target area, specifically:

[0066] Through simulation methods, the structural model, sedimentary phase model and interlayer model of the target area are established and used as constraints. The three-dimensional geological model of the target area is established using variogram analysis and sequential Gaussian random simulation algorithm.

[0067] Specifically, the logging data and physical property data of the target area are used, the sedimentary microfacies and stratigraphic framework model are used as control conditions, and geostatistics is used as a means to apply advanced conditional simulation technology to predict the changes in reservoir parameters between wells, thereby obtaining a three-dimensional data body of the formation parameter distribution, namely a three-dimensional geological model.

[0068] Among them, the well location coordinate data, well deviation data, porosity data and permeability data of the target area are obtained, the target layer of the target area is divided into multiple layers and the stratification data is obtained; according to the porosity data and permeability data of each layer in the multiple layers, the corresponding flow zone index is calculated; according to the well location coordinate data, well deviation data and stratification data, the structural model of the target area is established; based on the structural model, the sedimentary phase model is established according to the stratification data and logging data.

[0069] In addition, as a basic tool of geostatistics, the variogram can well study regionalized variables that are both random and structural, but the existing method for calculating the variogram only uses the original well data, which causes the loss of spatial information because the seismic information is not considered. Therefore, in this embodiment, the spatial seismic information and the known well data can be comprehensively considered, and the variogram can be obtained by using the anisotropic variogram method. This method can obtain more accurate results and better reflect spatial correlation; the sequential Gaussian simulation method is used for random modeling to improve the calculation speed.

[0070] Furthermore, the factors affecting physical properties include sedimentary factors and diagenetic factors; sedimentary factors are reflected by sedimentary microfacies, which are determined by core phases, logging phases, and seismic phases; diagenetic factors include pressure and temperature, which are comprehensively calibrated through regional thermal burial history, drilling measured temperature and pressure, and fluid inclusion test analysis. Information such as microfacies, temperature, and pressure can be loaded into the above-established three-dimensional geological model through petrel.

[0071] Furthermore, the favorable and unfavorable main controlling factors affecting physical properties are clarified, including the following steps:

[0072] Through petrel, the factors affecting physical properties are loaded into the established three-dimensional geological model of the target area;

[0073] Based on the sweet spot type, effective porosity type and invalid porosity type, the divided reservoir is loaded into the established three-dimensional geological model of the target area;

[0074] Different types of reservoirs and factors affecting physical properties are spatially superimposed to identify reservoir intervals, single-factor correlation analysis is performed on factors affecting physical properties and different reservoir types, and favorable and unfavorable main controlling factors affecting physical properties are selected.

[0075] Specifically, a single factor correlation analysis is performed on each physical property influencing factor to determine the correlation parameters between each physical property influencing factor and the reservoir type classification result, the correlation parameters are sorted from large to small, and the physical property influencing factors corresponding to the first n correlation parameters are obtained as the main controlling factors affecting the physical properties.

[0076] 4. Reserve forecast

[0077] In this embodiment, a weight scoring method for favorable or unfavorable factors is proposed, specifically:

[0078] For each physical property influencing factor, the influence coefficient is calculated and normalized as the weight coefficient of this type of physical property influencing factor; where influence coefficient = Σ physical property parameter point * influencing factor point / number of parameters, positive or negative represents favorable or unfavorable, and the numerical value represents the magnitude of the influence.

[0079] By superimposing the weight coefficients of various physical property influencing factors, a three-dimensional volume of superimposed influencing factors is established. At the same time, the physical properties are calculated by the homogenized values ​​of the physical property influencing factors and the weight coefficients, and the three-dimensional reservoir body is divided.

[0080] 5. Model verification and optimization

[0081] By adding new drilling and new types of influencing factors, the accuracy of the prediction model is continuously verified and trained; based on the verification results, the model is revised and optimized to improve the prediction accuracy.

[0082] Among them, the training process of the prediction model uses prediction accuracy as the objective function to optimize parameters. When the prediction accuracy meets the preset requirements, the model training is ended to obtain a corrected and optimized prediction model.

[0083] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0084] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A three-dimensional spatial distribution optimization prediction method based on reservoir classification, characterized in that: The following steps are involved: Use nuclear magnetic resonance technology to measure core samples in the target area and obtain physical property data; Cluster analysis method is used to classify the physical property test points to form specific types of reservoirs; Establish a three-dimensional geological model of the target area, load the factors affecting physical properties, and identify the favorable and unfavorable main controlling factors affecting physical properties through the spatial superposition of different types of reservoirs and factors affecting physical properties; By superimposing the weight coefficients of various physical property influencing factors, a three-dimensional space influencing factor superposition three-dimensional volume is established. At the same time, the physical properties are calculated by the homogenized values ​​of the physical property influencing factors and the weight coefficients, and the reservoirs are divided into three categories. By drilling new wells and adding new types of influencing factors, the accuracy of the prediction model is continuously verified and trained, and the revised and optimized prediction model is used to predict reservoir types in well-free areas.

2. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 1 is characterized in that: The core samples of the area to be measured are measured using nuclear magnetic resonance technology, including the following steps: Drill, cut, clean, dry and vacuum the cores in the target area to obtain core samples; The porosity of standard samples with known porosity is measured by NMR, and the T2 spectra of different echo intervals are inverted to establish the relationship between porosity and NMR response. The processed core samples are subjected to nuclear magnetic resonance T2 spectrum signal detection, and the porosity of the core samples is obtained by using the established relationship between porosity and nuclear magnetic resonance response.

3. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 1 is characterized in that: The physical property data include one or more of porosity, permeability, oil saturation, pore size distribution and compressibility.

4. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 1 is characterized in that: The formation of a specific type of reservoir includes the following steps: Establish the upper limit and lower limit of normal porosity of the target area, regard the cores with porosity greater than the upper limit as sweet spot type, regard the cores between the upper limit and the lower limit as effective porosity type, and regard the cores with porosity less than the lower limit as invalid porosity type; Cluster analysis was performed based on the location and depth of the physical property test points, and each type of physical property test points that were clustered together were combined to form three types of reservoirs.

5. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 1 is characterized in that: Establish a 3D geological model of the target area, specifically: Through simulation methods, the structural model, sedimentary phase model and interlayer model of the target area are established and used as constraints. The three-dimensional geological model of the target area is established using variogram analysis and sequential Gaussian random simulation algorithm.

6. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 1 is characterized in that: Factors affecting physical properties include: sedimentary factors and diagenetic factors; Sedimentary factors are reflected by sedimentary microfacies, which are determined by core facies, logging facies, and seismic facies; Diagenetic factors include pressure and temperature, which are comprehensively calibrated through regional thermal burial history, drilling measured temperature and pressure, and fluid inclusion testing and analysis.

7. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 4 is characterized in that: Identify the favorable and unfavorable main controlling factors that affect physical properties, including the following steps: Through petrel, the factors affecting physical properties are loaded into the established three-dimensional geological model of the target area; Based on the sweet spot type, effective porosity type and invalid porosity type, the divided reservoir is loaded into the established three-dimensional geological model of the target area; Different types of reservoirs and factors affecting physical properties are spatially superimposed to identify reservoir intervals, single-factor correlation analysis is performed on factors affecting physical properties and different reservoir types, and favorable and unfavorable main controlling factors affecting physical properties are selected.

8. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 1 is characterized in that: The weight coefficient is calculated as follows: For each physical property influencing factor, the influence coefficient is calculated and normalized as the weight coefficient of this type of physical property influencing factor; where influence coefficient = Σ physical property parameter point * influencing factor point / number of parameters, positive or negative represents favorable or unfavorable, and the numerical value represents the magnitude of the influence.

9. The three-dimensional spatial distribution optimization prediction method based on reservoir classification according to claim 1 is characterized in that: The training process of the prediction model uses prediction accuracy as the objective function to optimize parameters. When the prediction accuracy meets the preset requirements, the model training is terminated to obtain a corrected and optimized prediction model.

Images