Method for predicting reservoir pressure of fractured body and storage medium

Abstract

The application provides a fractured reservoir pressure prediction method, which comprises the following steps: based on the effective stress principle, pressure prediction is carried out based on the Bowers model to obtain macroscopic formation pressure distribution; the heterogeneity of the fractured reservoir is described; the fluid characteristics of the reservoir space are described; based on the heterogeneity and fluid properties of the fractured oil and gas reservoir, formation pressure correction is carried out to obtain the pore fluid pressure P of the fractured oil and gas reservoir f The application is based on the effective stress principle and the Bowers model to obtain macroscopic pressure distribution. Further, the spatial heterogeneity of the fractured reservoir is quantitatively characterized by seismic multi-attribute extraction and dimension reduction and classification; then, the fluid characteristics of the reservoir space are described based on the post-stack attenuation attribute and the pre-stack elastic inversion and AVO attribute. Finally, based on the actual drilling and measured pressure data, a formation pressure correction model based on the heterogeneity and fluid properties of the fractured oil and gas reservoir is constructed, so as to improve the prediction accuracy of the formation pressure of the reservoir.

Description

Technical Field

[0001] This invention relates to the field of oil and gas geophysics, specifically to a method for predicting reservoir pressure in fractured bodies and a storage medium. Background Technology

[0002] The Shunbei Block is located in the northern part of the Shuntogole Low Uplift, at the junction of the Awati, Manjiaer Depression and the Shaya Uplift. Its favorable structural location and abundant oil resources make it a prime area for long-term oil and gas migration and accumulation. Oil and gas exploration in the Shunbei area has yielded promising results.

[0003] However, drilling practice in the Shunbei oil and gas field has shown that the formations at depths deeper than 6000.0m are extremely complex, presenting a series of drilling challenges: the Silurian formations are fractured, highly pressure sensitive, and pose a significant risk of lost circulation; the Ordovician formations are highly fractured and poorly cemented, prone to collapse and blockage, leading to stuck pipe and other downhole failures. Accurate formation pressure prediction is crucial for safe and economical drilling; during the development phase, formation pressure prediction results help determine the reservoir's drive and connectivity.

[0004] By studying the geological conditions of the work area and the main controlling factors of abnormal pressure in carbonate rocks, the abnormal high pressure in the Middle and Lower Ordovician carbonate strata is mainly affected by two factors: first, changes in the volume of formation fluids caused by oil and gas charging and thermal evolution; and second, changes in fluid storage space caused by dissolution, with gas rising to the upper part of the storage space, and gas reservoirs rising to the upper part of the closed space, which also leads to the generation of abnormal high pressure. Although the Shunbei area has been subjected to compressive stress during geological history, the measured pressures at different locations (compression section / pull-out section / slip section) of Zone 5 show no significant differences.

[0005] In summary, the abnormal pressure in the fractured hydrocarbon reservoirs of the Shunbei Block mainly originates from changes in fluid storage space caused by dissolution and the hydrocarbon charging state. Conventional formation pressure prediction models are mostly based on undercompaction, and therefore are completely unsuitable for predicting the pressure of complex fractured hydrocarbon reservoirs. Summary of the Invention

[0006] The technical problem to be solved by this invention is to overcome the shortcomings of the prior art and provide a method for predicting reservoir pressure in fractured bodies and a storage medium. Starting from the causes of abnormal pressure in fractured oil and gas reservoirs, this invention grasps the two key points of reservoir heterogeneity and oil and gas charging, improves the accuracy of fractured body pressure prediction, provides important support for the design of ultra-deep wells, and thus reduces the drilling risk of ultra-deep fractured bodies.

[0007] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows:

[0008] A method for predicting reservoir pressure in fractured bodies includes the following steps: S01, based on the effective stress principle, pressure prediction is performed using the Bowers model to obtain the macroscopic formation pressure distribution; S02, the heterogeneity of the fractured reservoir is described; S03, the fluid characteristics of the reservoir space are described; S04, formation pressure correction is performed based on the heterogeneity and fluid properties of the fractured oil and gas reservoir to obtain the pore fluid pressure P of the oil and gas fractured reservoir. f .

[0009] The method for predicting reservoir pressure in fractured bodies according to the present invention is based on the effective stress principle and the Bowers model to obtain the macroscopic formation pressure distribution. Further, through seismic multi-attribute extraction, dimensionality reduction, and classification, the spatial heterogeneity of fractured reservoirs is quantitatively characterized. Then, based on post-stack attenuation attributes, pre-stack elastic inversion, and AVO attributes, the fluid characteristics of the reservoir space are described. Finally, based on actual drilling and measured pressure data, a formation pressure correction model based on the heterogeneity and fluid properties of fractured oil and gas reservoirs is constructed, thereby improving the formation pressure prediction accuracy of this type of reservoir.

[0010] The above technical solution can be further improved as described below.

[0011] According to the method for predicting reservoir pressure in fractured bodies of the present invention, in a preferred embodiment, in step S01:

[0012] Under the condition of a sealed formation, the pressure of the overlying formation is expressed as:

[0013] P ov =P p +P e

[0014] In the formula, P ov For the overlying formation pressure, P p P is the formation pore pressure. e For effective pressure;

[0015] Bowers' stress prediction model is expressed as follows:

[0016] P p =P OV -((v p -v0) / A) 1 / B

[0017] In the formula, v0 is the longitudinal wave velocity of the mudstone line, v p Let A represent the P-wave velocity at a given depth, and B be parameters used to correct the relationship between P-wave velocity and effective stress data. B is typically defined as a constant less than 1, indicating that the velocity-effective stress relationship is concave downwards.

[0018] Specifically, under sealed formation conditions, the pressure of the overlying strata is borne jointly by the particles that make up the rock and the fluid in the pores, P e The effective pressure is the pressure borne by the rock skeleton. The effective stress principle still applies to calculating the macroscopic formation pressure of the hydrocarbon reservoir in the Shunbei fault zone. First, we still need to calculate the overlying formation pressure P. ov and effective stress P e Regarding the overlying formation pressure P ov The calculations are all based on formation density integrals, consistent with the methods used to calculate overlying formation pressure in general formations. The effective stress and velocity in carbonate formations are less sensitive than in clastic formations, and the effects of tectonic unloading are taken into account. Therefore, Bowers' pressure prediction model is more suitable for carbonate reservoirs.

[0019] Specifically, in a preferred embodiment, step S02 includes the following sub-steps: S021: extracting and optimizing seismic attributes; S022: determining the fractured reservoir space type.

[0020] Specifically, in a preferred embodiment, the seismic attributes include amplitude attributes, frequency attributes, geometric curvature attributes, and morphological edge detection attributes.

[0021] Specifically, in a preferred embodiment, step S021 includes the following sub-steps: S0211: Extracting amplitude change gradient and coherent amplitude energy using a characteristic amplitude extraction method to highlight the sudden changes in amplitude when the reservoir is composed of internally developed structural fracture zones or interconnected small pore aggregates; S0212: Extracting frequency domain amplitude energy and conducting analysis and research on the main amplitude, main frequency and frequency band characteristics to achieve hierarchical description of geological targets at different scales; S0213: Extracting dip angle, azimuth angle and curvature attributes to reflect the geological phenomena of fault development at a certain scale with different development angles of the target layer by detecting regular and irregular wavefield spatial changes at abrupt changes in seismic data; S0214: Extracting discontinuity, variational PDE and multi-scale edge detection attributes to highlight the discontinuous characteristics of fault boundaries.

[0022] Specifically, amplitude-related attributes include: Extracting various amplitude energies through different methods, such as amplitude variation gradients and coherent amplitude energy. Characteristic amplitude extraction methods highlight abrupt amplitude changes when a reservoir is composed of internally developed structural fracture zones or a series of well-connected small pores, effectively enabling fracture identification. Frequency-related attributes involve extracting various frequency domain amplitude energies through different methods, analyzing dominant amplitude, dominant frequency, and frequency band characteristics to achieve hierarchical descriptions of geological targets at different scales. Geometric curvature-related attributes include dip angle, azimuth angle, and curvature attributes. These reflect the geological phenomena of fracture development at different scales within the target layer by detecting regular and irregular wavefield spatial changes at abrupt changes in seismic data. Morphological edge detection attributes include discontinuities, variational PDEs, and multi-scale edge detection attributes. Edge detection attributes highlight the discontinuous characteristics of fracture boundaries, resulting in more refined and reliable fracture edge detection attribute volumes.

[0023] Specifically, in a preferred embodiment, in step S022, the seismic attribute set obtained through step S021 is defined as...

[0024] {Attri 1 Attri 2 ,L,Attri n}

[0025] The fused attribute volume Attri is obtained through attribute dimensionality reduction algorithm, and the reservoir space type data volume L is obtained through actual drilling analysis. The reservoir space types are defined as L1, L2, L3, and L4. Among them, L1 is bedrock and limestone, L2 is large-scale fracture, L3 is small-scale crack, and L4 is karst cave with a large space.

[0026] Specifically, through attribute dimensionality reduction algorithms (LLE, PCA, fuzzy logic, etc.), we can obtain the fused attribute volume Attri. Through actual drilling analysis, we set reasonable attribute threshold values ​​to obtain the reservoir space type data volume L. The values ​​of L are discrete variables, not continuous variables, representing different reservoir space types. In Shunbei, we define the reservoir space types as L1, L2, L3, and L4. L1 is bedrock and limestone, characterized by extremely low porosity, density, poor pore connectivity, and is mainly under normal pressure; L2 is large-scale faults; L3 is small-scale fractures; and L4 is karst caves with relatively large spaces.

[0027] Specifically, in another preferred embodiment, in step S022, the reservoir space type data volume L is directly obtained based on Bayesian classification or other machine learning algorithms through actual drilling constraints, which can ensure that the required reservoir space type data volume L is obtained quickly and accurately. The reservoir space types are defined as L1, L2, L3, and L4; where L1 is bedrock and limestone, L2 is large-scale fracture, L3 is small-scale crack, and L4 is a karst cave with a large space.

[0028] Specifically, in a preferred embodiment, step S03 includes the following sub-steps: S031: Based on the scattering of seismic waves and the attenuation of seismic energy, extract attributes such as absorption coefficient and attenuation factor to reflect the hydrocarbon-bearing characteristics of the reservoir space; S032: Based on rock physics analysis, qualitatively identify reservoir fluid characteristics by obtaining reservoir fluid-sensitive elastic parameters through pre-stack seismic inversion; S033: Utilize CDP gather data of seismic reflections to analyze the variation law of reflected wave amplitude with shot-receiver distance at the reservoir interface, extract AVO intercept P and AVO slope, Poisson's ratio and fluid factor, and further infer the lithology and hydrocarbon-bearing properties of the reservoir; S034: Clarify the hydrocarbon-water properties and saturation Sw.

[0029] Specifically, the description of reservoir fluid characteristics includes: energy attenuation attributes: based on seismic wave scattering and seismic energy attenuation, attributes such as absorption coefficient and attenuation factor are extracted to reflect the hydrocarbon-bearing characteristics of the reservoir space; pre-stack seismic inversion of fluid-sensitive elastic parameters: based on rock physical analysis, the method of obtaining reservoir fluid-sensitive elastic parameters through pre-stack seismic inversion is used to qualitatively identify reservoir fluid characteristics; fluid detection based on AVO analysis: using CDP gather data of seismic reflections, the variation law of reflected wave amplitude with shot-receiver distance at the reservoir interface is analyzed, and AVO attribute parameters (AVO intercept P and AVO slope G), Poisson's ratio, and fluid factor are extracted to further infer the lithology and hydrocarbon-bearing properties of the reservoir. Through the above methods, the fluid properties of the reservoir are described, and the hydrocarbon-water properties and saturation Sw are clarified.

[0030] Specifically, in a preferred embodiment, in step S04:

[0031] When L = L1 (bedrock), because the bedrock is dense, has low porosity, and high effective stress, the bedrock is usually under normal pressure, and P is defined as... f =P p The parameters of the formation pressure prediction model were calibrated using bedrock as the normal pressure.

[0032] When L = L² (large-scale fracture), large-scale fractures typically cause pressure relief due to low leakage pressure. Simultaneously, fluid injection conditions also affect formation pressure within the reservoir. When the reservoir is a gas-water mixture, increased gas saturation will positively impact pore pressure to some extent. However, when the reservoir is an oil-water mixture, the incompressibility of oil and water means its impact on formation pressure is negligible. Therefore, when the reservoir is an oil-water mixture... c>1, where dis is the distance of the vertical fault in space;

[0033] When the storage space is a mixture of gas and water c>1, where dis is the distance between the vertical faults in space, Sw is the water saturation, and a and b are constants used to calibrate the accuracy of the model; when Sw=1, a=1, b=0, it is independent of the fluid and only related to the fault.

[0034] When L = L3 (small-scale fracture), the leakage pressure from small-scale fractures is high, and oil and gas accumulate in the confined space, which usually leads to an increase in formation pressure. At this time, The adjustment coefficient 'a' is larger when the reservoir fluid is a gas-water mixture than when it is an oil-water mixture. When L = L4 (a larger karst cavern), the low leakage pressure of the karst reservoir usually causes a depressurization effect. Simultaneously, the fluid filling situation also affects the formation pressure within the reservoir. When the reservoir is a gas-water mixture, the increase in gas saturation will have a positive effect on increasing pore pressure to some extent. However, when the reservoir is an oil-water mixture, due to the incompressibility of oil and water, the impact on formation pressure within the reservoir can be ignored. Therefore:

[0035] d>1, where a and b are constants used to calibrate the accuracy of the model.

[0036] The storage medium of the second aspect of the present invention stores a computer program, which, when run by a processor, executes the aforementioned method for predicting reservoir pressure in fractured bodies.

[0037] Compared to existing technologies, the advantages of this invention are as follows: Based on the effective stress principle and the Bowers model, a macroscopic pressure distribution is obtained. Furthermore, through seismic multi-attribute extraction, dimensionality reduction, and classification, the spatial heterogeneity of fractured reservoirs is quantitatively characterized. Then, based on post-stack attenuation properties and pre-stack elastic inversion and AVO properties, the fluid characteristics of the reservoir space are described. Finally, based on actual drilling and measured pressure data, a formation pressure correction model is constructed based on the heterogeneity and fluid properties of fractured hydrocarbon reservoirs, thereby improving the formation pressure prediction accuracy for this type of reservoir. Attached Figure Description

[0038] The invention will now be described in more detail with reference to embodiments and the accompanying drawings.

[0039] Figure 1 The flowchart of the fractured reservoir formation pressure prediction method according to an embodiment of the present invention is illustrated.

[0040] Figure 2 The illustration schematically shows a macroscopic pressure field profile through a well based on the effective stress principle in an embodiment of the present invention.

[0041] Figure 3 The diagram schematically illustrates a multi-scale fracture crack prediction planar view in an embodiment of the present invention.

[0042] Figure 4 The diagram schematically illustrates a plan view of the fluid characteristics of the storage space in an embodiment of the present invention;

[0043] Figure 5 The illustration schematically shows the formation pressure cross-well profile based on the heterogeneity and fluid properties correction of fractured oil and gas reservoirs in an embodiment of the present invention.

[0044] In the accompanying drawings, the same parts use the same reference numerals. The drawings are not drawn to scale. Detailed Implementation

[0045] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments, but this does not limit the scope of protection of the present invention.

[0046] Figure 1 The flowchart of the fault-bounded reservoir formation pressure prediction method of Embodiment 1 of the present invention is illustrated. Figure 2 The illustration schematically shows a macroscopic pressure field profile through a well based on the effective stress principle in an embodiment of the present invention. Figure 3 The diagram illustrates a multi-scale fracture crack prediction planar view in an embodiment of the present invention. Figure 4 The diagram schematically illustrates a plan view of the fluid characteristics of the storage space in an embodiment of the present invention. Figure 5 The illustration schematically shows the formation pressure cross-well profile based on the heterogeneity and fluid properties correction of fractured oil and gas reservoirs in an embodiment of the present invention.

[0047] Example 1

[0048] This embodiment takes the high-precision prediction of formation pressure in the Shunbei No. 5 fractured reservoir as an example.

[0049] Starting from the causes of anomalous pressure in hydrocarbon reservoirs along the Shunbei fault, and focusing on the two key aspects of reservoir heterogeneity and hydrocarbon charging, a high-precision method for predicting reservoir pressure in faulted bodies is proposed, such as... Figure 1 As shown, the specific process includes:

[0050] S01. Based on the effective stress principle, pressure prediction is carried out using the Bowers model to obtain the macroscopic formation pressure distribution.

[0051] Under closed formation conditions, the pressure of the overlying formation is borne jointly by the rock particles and the fluid in the pores, and can be expressed as:

[0052] P ov =P p +P e

[0053] In the formula, P ov P is the overlying formation pressure (static rock pressure). p P is the formation pore pressure (formation pressure). e The effective pressure is the pressure borne by the rock skeleton. The effective stress principle still applies to the Shunbei fault fracture body. Therefore, to calculate the formation pressure of the hydrocarbon reservoir in the Shunbei fault fracture body, we still need to first calculate the overlying formation pressure P. ov and effective stress P e .

[0054] Regarding the overlying formation pressure P ov The calculations are all based on the density integral of the formation, which is consistent with the calculation method of the overlying formation pressure of general formations.

[0055] However, carbonate formations are less sensitive to effective stress and velocity than clastic formations, and the unloading effect caused by tectonic movement needs to be considered. Therefore, Bowers' pressure prediction model is more suitable for carbonate reservoirs. It can be expressed as:

[0056] P p =P OV -((v p -v0) / A) 1 / B

[0057] In the formula, v0 is the P-wave velocity of the mudstone line (i.e., the seabed or the surface), v p For a given depth, the P-wave velocity is denoted by A and B, which are parameters used to correct the relationship between P-wave velocity and effective stress data. B is typically defined as a constant less than 1, indicating that the velocity-effective stress relationship is concave downwards. For example... Figure 2 As shown, conventional models can only provide relatively macroscopic pressure prediction results, and the abnormal pressure distribution cannot correspond one-to-one with drilling leakage and overflow situations on a spatial scale. The accuracy of formation pressure prediction needs to be improved.

[0058] S02, Description of Heterogeneity of Fractured Reservoirs

[0059] Attributes such as coherence, multi-scale coherence of curved waves, edge detection, structural curvature, dip angle, and lateral variation of similarity are extracted. Through attribute optimization and analysis, sensitive attributes and attribute combinations of fractures are determined. Furthermore, through attribute dimensionality reduction and classification, the spatial type of fracture reservoir is determined.

[0060] ① Seismic attribute extraction and optimization

[0061] Amplitude-related attributes: Various amplitude energies are extracted through different methods, including amplitude change gradient and coherent amplitude energy. The characteristic amplitude extraction method highlights the sudden changes in amplitude when the reservoir is composed of internally developed structural fracture zones or a series of well-connected small pores, which can effectively carry out fracture identification.

[0062] Frequency-related attributes: Various frequency domain amplitude energy extraction methods are used to conduct analysis and research on the main amplitude, main frequency and frequency band characteristics, and to achieve hierarchical description of geological targets at different scales.

[0063] Geometric curvature attributes include dip angle, azimuth angle, and curvature attributes. By detecting regular and irregular wavefield spatial changes at abrupt changes in seismic data, these attributes reflect geological phenomena of fault development at a certain scale with different development angles in the target layer.

[0064] Morphological edge detection attributes include discontinuity, variational PDE, and multi-scale edge detection attributes. Edge detection attributes can highlight the discontinuous features of fracture boundaries, resulting in a more refined and reliable fracture edge detection attribute volume.

[0065] ② Determine the spatial type of fractured storage based on attribute dimensionality reduction and classification.

[0066] Define the set of seismic attributes obtained through the above attribute extraction as follows:

[0067] {Attri 1 Attri 2 ,L,Attri n}

[0068] Reservoir space types are obtained through attribute dimensionality reduction or classification algorithms. Using attribute dimensionality reduction algorithms (LLE, PCA, fuzzy logic, etc.), we can obtain a fused attribute volume Attri. Through actual drilling analysis, we set reasonable attribute threshold values ​​to obtain the reservoir space type data volume L. The values ​​of L are discrete variables, not continuous variables, representing different reservoir space types. In Shunbei, we define reservoir space types as L1, L2, L3, and L4. L1 represents bedrock and limestone, characterized by extremely low porosity, density, poor pore connectivity, and is primarily at normal pressure; L2 represents large-scale fractures; L3 represents small-scale fissures; and L4 represents karst caves with relatively large spaces.

[0069] Alternatively, the reservoir space type data volume L can be directly obtained through actual drilling constraints, based on classification algorithms such as Bayesian classification or other machine learning algorithms. In this embodiment, multiple attributes such as amplitude attribute, coherence attribute, curvature attribute, ant body attribute, maximum likelihood attribute, and ant body are used as inputs to obtain the reservoir space type data volume L through a Bayesian classification algorithm, such as... Figure 3 As shown.

[0070] S03, Description of Fluid Characteristics in Storage Space

[0071] Energy attenuation attributes: Based on the scattering of seismic waves and the attenuation of seismic energy, attributes such as absorption coefficient and attenuation factor are extracted to reflect the hydrocarbon-bearing characteristics of the reservoir space.

[0072] Pre-stack seismic inversion of fluid-sensitive elastic parameters: Based on rock physics analysis, a method for qualitatively identifying reservoir fluid characteristics is obtained by pre-stack seismic inversion to obtain reservoir fluid-sensitive elastic parameters.

[0073] Fluid detection based on AVO analysis: Using CDP gather data of seismic reflections, the variation of reflected wave amplitude with shot-receiver distance at the reservoir interface is analyzed, and AVO attribute parameters (AVO intercept P and AVO slope G), Poisson's ratio and fluid factor are extracted to further infer the lithology and hydrocarbon properties of the reservoir.

[0074] The above methods are used to describe the fluid properties of the reservoir, clarifying the oil, gas, and water properties and the saturation level Sw. In this embodiment, through comparison of various fluid detection methods with actual drilling conditions, the fluid detection method based on the post-stack high-frequency attenuation gradient attribute is ultimately selected as the best. Furthermore, based on regression relationships, it is transformed into a quantitative prediction data volume for water saturation, such as... Figure 4 As shown.

[0075] S04. Based on the heterogeneity and fluid properties of fractured oil and gas reservoirs, formation pressure correction is performed to obtain the pore fluid pressure P of fractured oil and gas reservoirs. f

[0076] ① When L = L1 (bedrock), because the bedrock is dense, has low porosity, and high effective stress, the bedrock is usually under normal pressure. Let P be defined as... f =P p The parameters of the formation pressure prediction model were calibrated using bedrock as the normal pressure.

[0077] ② When L = L2 (large-scale fracture), large-scale fractures typically cause pressure relief due to low leakage pressure. Simultaneously, fluid injection conditions also affect formation pressure within the reservoir. When the reservoir is a gas-water mixture, increased gas saturation will positively impact pore pressure to some extent. However, when the reservoir is an oil-water mixture, the incompressibility of oil and water means its impact on formation pressure is negligible. Therefore:

[0078] When the storage space contains a mixture of oil and water c>1, where dis is the distance of the vertical fault in space.

[0079] When the storage space is a mixture of gas and water c>1, where dis is the distance to the vertical fault in space, Sw is the water saturation, and a and b are constants used to calibrate the accuracy of the model. When Sw=1, a=1, b=0, it is independent of the fluid and only related to the fault.

[0080] ③ When L = L3 (small-scale fracture), the leakage pressure from small-scale fractures is high, and oil and gas accumulate in the confined space, which usually leads to an increase in formation pressure. At this time, The adjustment coefficient 'a' is larger when the fluid in the storage space is a mixture of gas and water than when it is a mixture of oil and water.

[0081] ④ When L = L4 (a relatively large karst cavern), the low leakage pressure of the karst reservoir usually causes pressure relief. Simultaneously, the fluid filling situation also affects the formation pressure within the reservoir space. When the reservoir space is a gas-water mixture, the increase in gas saturation will have a positive effect on increasing pore pressure to some extent. However, when the reservoir space is an oil-water mixture, due to the incompressibility of oil and water, the impact on formation pressure within the reservoir space can be ignored. Therefore:

[0082] d>1, where a and b are constants used to calibrate the accuracy of the model.

[0083] like Figure 5 As shown, under the constraints of spatial constraints and fluid filling conditions in fractured reservoirs, the abnormal pressure distribution in fractured reservoirs also exhibits strong heterogeneity. Furthermore, the pressure prediction results match the actual overflow and leakage conditions observed during drilling better, indicating that the method proposed in this patent can better adapt to the formation pressure prediction of fractured oil and gas reservoirs.

[0084] Example 2

[0085] The storage medium of this invention stores a computer program, which, when run by a processor, executes the fractured reservoir pressure prediction method of Embodiment 1.

[0086] As can be seen from the above embodiments, the fault-bounded reservoir pressure prediction and storage method of the present invention is based on the effective stress principle and uses the Bowers model to obtain the macroscopic pressure distribution. Further, through seismic multi-attribute extraction, dimensionality reduction, and classification, the spatial heterogeneity of the fault-bounded reservoir is quantitatively characterized. Then, based on post-stack attenuation attributes and pre-stack elastic inversion and AVO attributes, the fluid characteristics of the reservoir space are described. Finally, based on actual drilling and measured pressure data, a formation pressure correction model based on the heterogeneity and fluid properties of fault-bounded hydrocarbon reservoirs is constructed, thereby improving the formation pressure prediction accuracy of this type of reservoir.

[0087] Although the invention has been described with reference to preferred embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, the technical features mentioned in the various embodiments can be combined in any manner as long as there is no structural conflict. The invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for predicting reservoir pressure in fractured bodies, characterized in that, Includes the following steps: S01. Based on the effective stress principle, pressure prediction is carried out using the Bowers model to obtain the macroscopic formation pressure distribution. S02, Describe the heterogeneity of fractured reservoirs; S03, Describe the fluid characteristics of the storage space; S04. Formation pressure correction based on the heterogeneity and fluid properties of fractured oil and gas reservoirs to obtain the pore fluid pressure of fractured oil and gas reservoirs. ; In step S01: Under the condition of a sealed formation, the pressure of the overlying formation is expressed as: ； In the formula, Due to the pressure of the overlying strata, For formation pore pressure, For effective pressure; Bowers' stress prediction model is expressed as follows: ； In the formula, The longitudinal wave velocity of the mudstone line. Given the formation P-wave velocity at a given depth, A and B are parameters for correcting the relationship between P-wave velocity and effective stress data; The data volume L of the storage space type is obtained, and the storage space types are defined as L1, L2, L3, and L4; where L1 is bedrock and limestone, L2 is large-scale faults, L3 is small-scale fissures, and L4 is a karst cave with a large space. In step S04: When L=L1, define The parameters of the formation pressure prediction model were calibrated using bedrock as the normal pressure. When L=L2, and the storage space is a mixture of oil and water, , >1, where The distance between vertical faults in space; When the storage space is a mixture of gas and water , >1, where The distance between vertical faults in space. Water saturation , It is a constant used to calibrate the accuracy of the model; when =1, =1, When =0, it is unrelated to fluids and only related to faults; When L=L3 adjustment coefficient The size is greater when the fluid in the storage space is a mixture of gas and water than when it is a mixture of oil and water. When L=L4 , >1, , It is a constant used to calibrate the accuracy of the model.

2. The method for predicting reservoir pressure in fractured bodies according to claim 1, characterized in that, Step S02 includes the following sub-steps: S021: Extract and optimize seismic attributes; S022: Determine the type of storage space for the fractured joint.

3. The method for predicting reservoir pressure in fractured bodies according to claim 2, characterized in that, Earthquake attributes include amplitude attributes, frequency attributes, geometric curvature attributes, and morphological edge detection attributes.

4. The method for predicting reservoir pressure in fractured bodies according to claim 3, characterized in that, Step S021 includes the following sub-steps: S0211: The characteristic amplitude extraction method is used to extract the amplitude change gradient and coherent amplitude energy, highlighting the sudden change in amplitude when the reservoir is composed of internally developed structural fracture zones or interconnected small pore aggregates; S0212: Extract frequency domain amplitude energy, conduct analysis and research on the main amplitude, main frequency and frequency band characteristics, and realize hierarchical description of geological targets at different scales; S0213: Extract dip angle, azimuth angle, and curvature attributes to reflect the geological phenomena of fault development at a certain scale with different development angles in the target layer by detecting regular and irregular wavefield spatial changes at abrupt changes in seismic data. S0214: Extract discontinuity, variational PDE, and multi-scale edge detection attributes to highlight the discontinuity features of fracture boundaries.

5. The method for predicting reservoir pressure in fractured bodies according to claim 4, characterized in that, In step S022, the seismic attribute set obtained through step S021 is defined as follows: ； The fused attribute body is obtained through an attribute dimensionality reduction algorithm. Through actual drilling analysis, the reservoir space type data volume L was obtained.

6. The method for predicting reservoir pressure in fractured bodies according to claim 4, characterized in that, In step S022, the reservoir space type data volume L is directly obtained based on Bayesian classification or other machine learning algorithms by means of actual drilling constraints.

7. The method for predicting reservoir pressure in fractured bodies according to claim 5 or 6, characterized in that, Step S03 includes the following sub-steps: S031: Based on the scattering of seismic waves and the attenuation of seismic energy, the absorption coefficient and attenuation factor attributes are extracted to reflect the hydrocarbon-bearing characteristics of the reservoir space. S032: Based on rock physics analysis, a method for qualitatively identifying reservoir fluid characteristics by obtaining reservoir fluid-sensitive elastic parameters through pre-stack seismic inversion; S033: Using CDP gather data of seismic reflection, analyze the variation law of reflected wave amplitude with shot-receiver distance on reservoir interface, extract AVO intercept P, AVO slope, Poisson's ratio and fluid factor, and further infer the lithology and hydrocarbon properties of reservoir. S034: Determine the properties of oil, gas, and water, and water saturation. .

8. A storage medium, characterized in that, It stores a computer program, which, when run by a processor, executes the method for predicting reservoir pressure in fractured bodies as described in any one of claims 1 to 7.

Images