Geological model updating method and system, computer equipment, storage medium and program

Abstract

The invention relates to the field of underground oil and gas reservoir geological modeling, and provides a geological model updating method and system, computer equipment, a storage medium and a program, and the method comprises the following steps: S0, preparing data; s1, establishing a phase model; s2, physical property modeling is conducted, and a three-dimensional oil reservoir physical property model is established according to the facies model, the three-dimensional geological grid model and statistical logging data; s3, a rock physical model is built, the rock physical model is built through an empirical formula, and a three-dimensional oil reservoir elastic model is obtained after calibration; s4, performing three-dimensional synthetic seismic forward modeling; s5, performing correlation analysis, and judging precision; and S6, updating the model, and carrying out iterative updating by adopting a numerical derivative method until the correlation coefficient reaches the preset precision. According to the method, numerical derivative updating is adopted for system accumulated errors generated in the modeling link, and the method is used for any calibrated rock physical model. Through combination of seismic data and logging data, a tectonic model is established by using geostatistics, and an accurate tectonic framework is provided for model updating.

Description

Technical Field

[0001] The present invention relates to the technical field of underground oil and gas reservoir geological modeling, and particularly to a geological model updating method, system, computer device, storage medium and program. Background Art

[0002] Geological modeling is a three-dimensional quantitative stochastic model generated by using computer graphics technology based on the comprehensive analysis of geological, logging, geophysical data and various interpretation results or conceptual models.

[0003] Traditional geological modeling methods have problems such as complex and diverse modeling methods, complex processes, and inconsistent modeling standards. In the process of rock physics inversion of traditional model updating methods, it is necessary to derive the rock physics model, which requires a large amount of time to deduce the results and has low work efficiency.

[0004] Prior Art One

[0005] A Chinese patent with the publication number CN118781282A provides a guiding drilling geological model updating method, device and computer device. The updating method obtains acoustic wave data at the guiding drilling target position, where the acoustic wave data is obtained by logging while drilling at the guiding drilling target position; performs velocity updating and reimaging based on the acoustic wave data and the three-dimensional seismic data in the un-drilled area in front of the drill bit to obtain logging information while drilling; updates the initial guiding drilling geological model through a pre-established intelligent modeling network based on the logging information to obtain an updated guiding drilling geological model. This application is based on a deep learning framework and creates a model rapid establishment and updating method and system suitable for geological guiding drilling. Through the construction of a deep learning big data sample to train a generative adversarial network, logging while drilling data acquisition and seismic imaging update, construction of an initial guiding drilling model and real-time update during the guiding drilling process.

[0006] The defect problems of Prior Art One are as follows:

[0007] The updating method provided by Prior Art One is complex to operate, requires a large amount of geological data to be collected, and requires professional equipment for assistance. The workload is large and the work efficiency is not high. Summary of the Invention

[0008] In order to overcome or alleviate one or more of the above technical problems, the object of the present invention is to provide a geological model updating method, system, computer device, storage medium and program.

[0009] The present invention provides the following technical solutions:

[0010] In the first aspect, the present invention provides a geological model updating method, which includes the following steps:

[0011] S0. Prepare data

[0012] The prepared data includes actual seismic data and logging data;

[0013] S1. Establish a facies model

[0014] Establish a three-dimensional geological grid model by geostatistical methods; for the three-dimensional geological grid model, conduct sedimentary facies and lithofacies modeling to obtain a facies model;

[0015] S2. Physical property modeling

[0016] According to the facies model and the three-dimensional geological grid model, statistically analyze logging data to establish a three-dimensional reservoir physical property model, and the attributes of the three-dimensional reservoir physical property model at least include porosity, shale content, and water saturation;

[0017] S3. Rock physics model modeling

[0018] Through logging data, based on the three-dimensional reservoir physical property model, establish a rock physics model through empirical formulas, and convert the porosity, shale content, and water saturation in the attributes of the three-dimensional reservoir physical property model into elastic parameters and acoustic parameters of the rock, including P-wave velocity, S-wave velocity, and density field data, to obtain a three-dimensional reservoir elastic model;

[0019] S4. Three-dimensional synthetic seismic forward modeling

[0020] Through the three-dimensional reservoir elastic model, combined with density field data, calculate the impedance body of the reservoir medium, select a seismic wavelet for convolutional forward modeling to obtain a three-dimensional synthetic seismic record, and assign it to the three-dimensional geological grid model;

[0021] S5. Correlation analysis and accuracy judgment

[0022] Assign the seismic data to the model grid of the three-dimensional geological grid model, calculate the error between it and the three-dimensional synthetic seismic record, as well as their correlation coefficient, set the required accuracy of the correlation coefficient of the seismic data and the three-dimensional synthetic seismic record as needed, and judge whether the accuracy of the correlation coefficient of the two reaches the preset accuracy. If not, jump to S6 for model update; if so, the update is completed to obtain an updated three-dimensional reservoir physical property model;

[0023] S6. Model update

[0024] Use the numerical derivative method for update, correct the grid points that do not meet the accuracy requirements, update the three-dimensional synthetic seismic record, and then correct and update the three-dimensional reservoir physical property model of the reservoir, and return to step S2 until the correlation coefficient reaches the preset accuracy.

[0025] According to some embodiments, the numerical derivative for update in step S6 includes the following steps:

[0026] S61: Place the observed seismic data and the three-dimensional synthetic seismic record in the same section, calculate the error between the two, and perform data optimization. The calculation method is as follows:

[0027] F = 1 - Corr(S obs , S symth ) + (S obs - S symnth ) 2 (1)

[0028] Where F is the objective function, S obs is the actual observed seismic record, and S symth is the three-dimensional synthetic seismic record. Calculate the correlation between the actual observed seismic record and the three-dimensional synthetic seismic record. The higher the correlation coefficient, the closer the geological model is to the actual geological situation;

[0029] S62: Update the geological model through the numerical derivative method, and the numerical derivative method includes the steepest gradient descent method, the Newton method, and the conjugate gradient method.

[0030] According to some embodiments, the steepest gradient descent method in step S62 is to take the derivative of F with respect to φ k , and the negative direction of the derivative is the model update direction. Update the model through the numerical derivative method, and the updated porosity φ' is equal to:

[0031]

[0032] In the above formula, λ represents the step size of the gradient method;

[0033] The Newton method calculates the porosity φ' through the following formula:

[0034]

[0035] In the formula, H -1 is the second derivative of the objective function;

[0036] The conjugate gradient method calculates the porosity φ' through the following formula:

[0037] φ' = φ k + λd k (4)

[0038] In the formula, d k is the conjugate direction, β k-1 is the conjugate coefficient.

[0039] According to some embodiments, the well logging data at least includes porosity, water saturation, shale content, density field data, post-stack seismic data, seismic interpretation data, and measured depth.

[0040] According to some embodiments, the empirical formula in step S3 includes the Gassmann model, Hudson model, Wyllie model, or Waxman-Smits model.

[0041] According to some embodiments, the calibration in step S3 is to compare the synthetic P-wave and S-wave velocities with the P-wave and S-wave velocities in actual well logging to verify the rock physics model. If there are any unreasonable points, the rock physics model is modified accordingly until the well logging verification and section verification reach the target effect.

[0042] In a second aspect, the present invention provides a geological model update system, which includes:

[0043] A data preparation unit for preparing data, including actual seismic data and well logging data;

[0044] A facies model unit for establishing a three-dimensional geological grid model by geostatistical methods; and performing sedimentary facies and lithofacies modeling on the three-dimensional geological grid model to obtain a facies model;

[0045] A physical property modeling unit for statistically analyzing well logging data to establish a three-dimensional reservoir physical property model according to the facies model and the three-dimensional geological grid model. The attributes of the three-dimensional reservoir physical property model at least include porosity, shale content, and water saturation;

[0046] A rock physics model modeling unit for establishing a rock physics model through empirical formulas on the basis of the three-dimensional reservoir physical property model by well logging data, and converting the porosity, shale content, and water saturation in the attributes of the three-dimensional reservoir physical property model into elastic parameters and acoustic parameters of the rock, including P-wave velocity, S-wave velocity, and density field data, to obtain a three-dimensional reservoir elastic model;

[0047] A three-dimensional synthetic seismic forward modeling unit for calculating the impedance body of the reservoir medium through the three-dimensional reservoir elastic model, combining density field data, selecting a seismic wavelet for convolution forward modeling to obtain a three-dimensional synthetic seismic record, and assigning it to the three-dimensional geological grid model;

[0048] The correlation analysis and judgment accuracy unit is used to assign the seismic data to the model grid of the three-dimensional geological grid model, calculate the error between the data and the three-dimensional synthetic seismic record, as well as their correlation coefficient, set the required accuracy of the correlation coefficient of the seismic data and the three-dimensional synthetic seismic record as needed, and judge whether the accuracy of the correlation coefficient of the two reaches the preset accuracy. If not, jump to S6 for model update; if so, the update is completed, and the updated three-dimensional reservoir physical property model is obtained.

[0049] The model update unit is used to perform updates using the numerical derivative method, correct the grid points that do not meet the accuracy requirements, update the three-dimensional synthetic seismic record, and then correct and update the three-dimensional reservoir physical property model of the reservoir, and return to step S2 until the correlation coefficient reaches the preset accuracy.

[0050] In a third aspect, the present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the above update method are implemented.

[0051] In a fourth aspect, the present invention provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above update method are implemented.

[0052] In a fifth aspect, the present invention provides a computer program. When the computer program is executed by a processor, the steps of the above update method are implemented.

[0053] Compared with the prior art, the defects of the prior art overcome by the present invention and the beneficial effects are as follows:

[0054] 1. Aiming at the problems of complex and diverse modeling methods, complicated processes, and inconsistent modeling standards in traditional geological modeling methods. For the systematic cumulative errors generated in each modeling link of the present invention, taking the actual observed seismic data as the standard, based on the synthetic seismic record obtained by rock physics forward modeling, using the consistency and correlation of the actual observed data and the synthetic seismic record as the objective function, and using the numerical derivative method to update the grid points of the porosity model with low accuracy, and correcting and updating the physical property model. The update method provided by the present invention has a simple process and is easier to operate.

[0055] 2. Aiming at the traditional model update method, in the process of rock physics inversion, it is necessary to derive the rock physics model, which requires a lot of time to deduce the results. The present invention adopts numerical derivatives, can use any calibrated rock physics model, improves the model accuracy, and the obtained updated geological model is more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1An overview diagram of drilling and seismic data of a typical development unit provided in an embodiment of the present invention (Figure a is a spectrum analysis of seismic data; Figure b is the area of ​​the seismic work area and the well spacing).

[0057] Figure 2 An actual seismic profile diagram provided for an embodiment of the present invention.

[0058] Figure 3 A rock physical parameter diagram provided for an embodiment of the present invention.

[0059] Figure 4 A rock physics model calibration diagram provided in an embodiment of the present invention.

[0060] Figure 5 A fitting diagram of a well rock physics model provided in an embodiment of the present invention.

[0061] Figure 6 A cross-sectional diagram of the superposition of synthetic seismic records (black) and observed seismic data (brown) provided in an embodiment of the present invention.

[0062] Figure 7a A three-dimensional display diagram of a position to be updated provided by an embodiment of the present invention.

[0063] Figure 7b A three-dimensional display diagram of the updated position provided by an embodiment of the present invention.

[0064] Figure 8a This is a three-dimensional porosity model before updating provided in an embodiment of the present invention.

[0065] Figure 8b An updated three-dimensional porosity model provided by an embodiment of the present invention.

[0066] Figure 9a A cross section of a porosity model before updating provided in an embodiment of the present invention.

[0067] Figure 9b A cross section of an updated porosity model provided in an embodiment of the present invention.

[0068] Figure 10a A comparison diagram of correlation coefficients between synthetic records and observed data before updating provided in an embodiment of the present invention, synthetic seismic records before updating (black) and actual observed data (brown).

[0069] Figure 10b A comparison diagram of the correlation coefficients of updated synthetic records and observed data provided in an embodiment of the present invention, an updated synthetic seismic record (black) and actual observed data (brown).

[0070] Figure 11a A comparison chart of correlation coefficients before updating provided in an embodiment of the present invention.

[0071] Figure 11b The updated correlation coefficient comparison chart provided by the embodiment of the present invention.

[0072] Figure 12 The flowchart of the geological model updating method provided by the embodiment of the present invention.

[0073] Figure 13 The structural diagram of the geological model updating system provided by the embodiment of the present invention. Detailed implementation manners

[0074] In view of the systematic cumulative error generated in each modeling link during the geological model modeling process, the present invention proposes a geological model updating method. Through multiple links such as structural modeling, facies modeling, and petrophysical modeling, the established three-dimensional model includes information such as the overall framework, internal structure, and petrophysical parameters of the actual geological situation. Certain errors will be generated in all modeling links, thus forming a cumulative error in the system, which directly affects the accuracy of the finally established model. In order to make the established model have a high consistency with the actual geological situation, only using manual local adjustment cannot guarantee the consistency between the two. In addition, the personnel who correct the actual operation have high professional requirements.

[0075] Both drilling data and seismic data are objective expressions of the underground complex geological body. In particular, seismic data can reflect the macroscopic distribution and trend changes of underground media. This three-dimensional geological model automatic updating method takes the actually observed seismic data as the standard, based on the synthetic seismic record obtained by petrophysical forward modeling, takes the consistency and correlation between the actually observed data and the synthetic seismic record as the objective function, and uses the method of numerical derivative to update the grid points of the porosity model with low accuracy and correct the updated physical property model.

[0076] Therefore, the present invention provides a three-dimensional geological model updating method, as Figure 12 shown.

[0077] The present invention will be described in detail below in conjunction with embodiments and drawings. However, it should be understood that the embodiments and drawings are only used for exemplary description of the present invention, and cannot constitute any limitation to the protection scope of the present invention. All reasonable transformations and combinations within the scope of the inventive concept of the present invention fall within the protection scope of the present invention.

[0078] The present invention will be further described below in conjunction with the drawings.

[0079] Embodiment 1

[0080] According to Figure 12 , the geological model updating method provided by this embodiment specifically includes the following steps:

[0081] S0 Prepare data: Prepare seismic data and logging data required for forward modeling of the model (including porosity, water saturation, shale content, density field data, post-stack seismic data, seismic interpretation data, and measured depth).

[0082] S1 Establish a facies model: Establish a three-dimensional geological grid model through geostatistical methods. Based on the established fine three-dimensional geological grid model, perform sedimentary facies and lithofacies modeling to obtain a facies model.

[0083] S2 Physical property modeling: Establish a three-dimensional reservoir physical property model according to the facies model and logging data (its attributes include at least porosity, shale content, and water saturation).

[0084] S3 Rock physics model modeling: Based on the logging data and the three-dimensional reservoir physical property model obtained in step S2, establish a rock physics model through empirical formulas, which can convert the attributes of the three-dimensional reservoir physical property model (porosity, shale content, water saturation, etc.) into elastic parameters and acoustic parameters of rocks, such as P-wave velocity, S-wave velocity, density, etc., to obtain a three-dimensional reservoir elastic model. The rock physics model is different empirical formulas (including but not limited to Gassmann model, Hudson model, Wyllie model, or Waxman-Smits model), and a more suitable empirical formula is selected. The most important thing in this step is to perform rock physics calibration, that is, to test. The process of repeatedly adjusting the parameters in the empirical formula is the rock physics modeling process. Specifically, compare the synthetic P-wave and S-wave velocities with the actual P-wave and S-wave velocities in logging. If there are any unreasonable points, modify the rock physics model accordingly until the logging verification and profile verification have good effects.

[0085] S4 Three-dimensional synthetic seismic forward modeling: After performing rock physics calibration on the rock physics model established in S3 to simulate the actual geological situation, use the three-dimensional reservoir physical property model obtained in step S2 to forward model to obtain data such as seismic wave velocity field and elastic parameters, and then combine the density field data (since there is logging data only at the well locations, so usually there is no need to specifically establish this model. In specific rock physics modeling, a suitable empirical formula can also be selected to calculate the density field data), calculate the impedance body of the reservoir medium, select a seismic wavelet for convolution forward modeling to obtain a three-dimensional synthetic seismic record, and assign it to the model grid of the three-dimensional geological grid model obtained in step S0. The geological model is three-dimensional grid data, for example, composed of many irregular hexahedrons spliced together. The grid size and spatial position are defined by giving the corner point grid data. Usually, in the first modeling process, this corner point grid data will not change, but only the attributes of the grid are changed. Give the attributes of each grid, including porosity, water saturation, and shale content.

[0086] S5 Correlation analysis and judgment accuracy: Calculate the error and correlation coefficient. Assign the seismic data to the model grid of the three-dimensional geological grid model, calculate the error between the data and the three-dimensional synthetic seismic record obtained in step S4, and calculate their correlation coefficient. Set the required accuracy of the correlation coefficient between the seismic data and the three-dimensional synthetic seismic record as needed (i.e., the objective function is greater than the set value), and determine whether the accuracy of the correlation coefficient between the two reaches the preset accuracy. If not, jump to S6 for model update; if so, the update is completed, and the updated three-dimensional reservoir physical property model is obtained.

[0087] S6 Model update: Correct the grid points that do not meet the accuracy requirements, update the three-dimensional synthetic seismic record, and then correct and update the three-dimensional reservoir physical property model of the reservoir (the purpose is to update the three-dimensional reservoir physical property model, but the change in the three-dimensional reservoir physical property model may cause changes in the acoustic parameters and elastic parameters calculated in step S2), that is, return to step S2 until the correlation coefficient reaches the preset required accuracy.

[0088] More specifically, the update process of step S6 is to use the numerical derivative method for update. Specifically, place the observed seismic data and the three-dimensional synthetic seismic record (the three-dimensional synthetic seismic record is calculated by taking a two-dimensional profile during calculation to obtain three-dimensional data) in the same profile, calculate the error between the two, and perform data optimization. The calculation method is as follows:

[0089] F = 1 - Corr(S obs , S symth )+(S obs - S symth ) 2 (1)

[0090] Where F is the objective function, S obs is the actual observed seismic record, and S symth is the three-dimensional synthetic seismic record, and calculate the correlation between the actual observed seismic record and the three-dimensional synthetic seismic record.

[0091] When the correlation coefficient is higher, it indicates that the established geological model is closer to the real geological situation. Take the derivative of F with respect to φ k , and the negative direction of the derivative is the model update direction. Update the model by the method of numerical derivative, then the updated porosity φ′ is equal to:

[0092]

[0093] In the above formula, λ represents the step size of the gradient method.

[0094] Repeat the geostatistical modeling and rock physics modeling processes with the updated porosity to generate a three-dimensional synthetic seismic record, compare it with the observed seismic record, repeat the previous steps, perform multiple iterations using a computer, that is, return to step S2 after solving using the numerical derivative method until the required accuracy is achieved, and obtain the updated three-dimensional reservoir physical property model.

[0095] In this embodiment, by combining seismic data and well logging data, a structural model and an attribute model are established using geostatistics to provide an accurate structural framework for model updating; by analyzing well logging data, a rock physics model is established, and rock physics calibration is performed through comparison between well logging curves and actual data curves to obtain an accurate rock physics model for the model area. Then, forward rock physics modeling is performed on the model, and further seismic forward modeling is carried out to obtain a three-dimensional synthetic seismic record. The three-dimensional synthetic seismic record obtained from forward modeling is compared with the actual observed seismic record, the gradient is obtained using numerical derivatives, and the model is corrected and updated using the method of numerical derivatives to improve the consistency between the seismic data obtained from forward modeling and the actual observed seismic data; finally, by analyzing the matching degree of the synthetic seismic records before and after updating, the effectiveness of seismic-constrained model updating is verified. It is shown that after introducing seismic data constraints to adjust and modify the geological model, the model accuracy can be improved, and the final obtained geological model is more reliable.

[0096] The updating method provided in this embodiment does not require taking the derivative of the forward modeling process formula, reducing the manual calculation cost and reducing the cumulative error due to manual operations in multiple modeling processes.

[0097] Embodiment 2

[0098] This embodiment is an implementation application of the updating method provided in Embodiment 1 based on a representative composite sand body development unit in the research work area.

[0099] 1. Well-seismic work area overview

[0100] This development unit has 11 well logging data, the well spacing is generally greater than 300m, the seismic work area is about 11.7 km 2 , the seismic data bandwidth is about 8 - 55 Hz, and the main frequency is about 47 Hz, as Figure 1 shown. Calculated according to the sandstone velocity of 2350 m / s in the work area, the vertical resolution of the seismic data is about 12.5 m. Obviously, the too large well spacing leads to uncertainty in predicting the sand body between wells, and the number of wells is not sufficient to support the fine characterization of the configuration of the composite sand body. In this case, making full use of seismic data is the key to sand body configuration characterization. The actual seismic profile is as Figure 2 shown.

[0101] 2. Model update

[0102] Select a rock physics model for forward modeling.Figure 3 The selected rock physics model and its parameters are shown as follows. Figure 5 The fitting comparison between the actual acoustic curve and the synthetic acoustic curve of Well A15 is shown as follows. Among them, the black curve is the actual observed data, and the red curve is the synthetic forward modeling data. It can be seen that the forward modeling result of the selected rock physics model is basically consistent with the measured curve in the well. If the curve fitting effect is poor, then through the rock physics calibration method such as Figure 4 changing a certain parameter for debugging.

[0103] Select the dominant frequency of the wavelet based on the observed seismic data, and conduct rock physics forward modeling on the established porosity model. Figure 6 The stacked section of the synthetic seismic record and the observed seismic record is shown as follows. The black waveform is the observed seismic data, and the brown waveform is the synthetic seismic record. Then, by calculating the error between the synthetic record and the observed seismic data and setting the matching threshold, the position to be updated of the model is obtained. Figure 7a The 3D display of the position to be updated is shown as follows. Figure 7b The 3D display of the updated position is shown as follows.

[0104] Automatically update the model for the area below the accuracy. The two-dimensional image after the update is shown in Figure 9, where Figure 9a is the profile of the porosity model before the update, Figure 9b is the profile of the porosity model after the update. The three-dimensional image after the update is shown in Figure 8, where Figure 8a is the three-dimensional porosity model before the update, Figure 8b is the three-dimensional porosity model after the update. The synthetic seismic records before and after the update are compared and shown in Figure 10, where Figure 10a is the synthetic seismic record before the update, Figure 10b is the synthetic seismic record after the update. Both are shown as black waveforms, and the brown background is the actual observed data. It can be seen that the forward modeling synthetic record after the update is more in line with the trend expressed by the observed seismic data, which indicates that the automatic model update improves the consistency between the porosity model and the actual observed geological conditions. Figure 11 is the comparison chart of the correlation coefficients before and after the update.

[0105] To sum up, the key to 3D geological modeling lies in the prediction between wells. In step S2 of establishing the physical property model, geostatistics is used to obtain the data between wells. Its theoretical basis is the geological law between wells, that is, the geological correlation of reservoir parameters in three-dimensional space. By fully mining seismic data to predict the actual geological conditions and participating in the modeling process, and using geological laws to supervise the process of estimation between wells. Considering the systematic cumulative error in the modeling process and taking the actual observed seismic data as the standard to conduct automatic update of the geological model can improve the consistency between the established model and the actual geological conditions, making the modeling results conform to both well point information and seismic information, and also conform to geological laws.

[0106] Example 3

[0107] For the calculation of numerical derivatives, optimization methods such as the steepest gradient descent method, Newton's method, and conjugate gradient method can be used. The method provided in Example 1 is the steepest descent method, which is simple and easy to implement but has a slow convergence rate.

[0108] Newton's method: Equation (2) can be changed to Equation (3):

[0109]

[0110] where H -1 is the second derivative of the objective function. This method has a relatively fast convergence rate, but it requires calculating the Hessian matrix and its inverse matrix, which involves a large amount of computation.

[0111] Conjugate gradient method: Equation (2) can be changed to Equation (4):

[0112] φ′ = φ k + λd k (4)

[0113] where d k is the conjugate direction, is the conjugate coefficient. This method is applicable to large-scale problems, and its convergence rate is between that of the steepest descent method and Newton's method.

[0114] Example 4

[0115] As Figure 13 , this example provides a geological model updating system, which includes:

[0116] A data preparation unit for preparing data, including actual seismic data and logging data;

[0117] A facies model unit for establishing a three-dimensional geological grid model through geostatistical methods; for the three-dimensional geological grid model, sedimentary facies and lithofacies modeling are performed to obtain a facies model;

[0118] A physical property modeling unit for statistically analyzing logging data to establish a three-dimensional reservoir physical property model based on the facies model and the three-dimensional geological grid model. The attributes of the three-dimensional reservoir physical property model at least include porosity, shale content, and water saturation;

[0119] A rock physics model modeling unit for establishing a rock physics model through logging data on the basis of the three-dimensional reservoir physical property model. By using empirical formulas, the porosity, shale content, and water saturation in the attributes of the three-dimensional reservoir physical property model are converted into elastic parameters and acoustic parameters of the rock, including longitudinal wave velocity, transverse wave velocity, and density field data, to obtain a three-dimensional reservoir elastic model;

[0120] 3D synthetic seismic forward modeling unit, which is used to calculate the impedance of reservoir media through a 3D reservoir elastic model, combine density field data, select a seismic wavelet for convolutional forward modeling, obtain a 3D synthetic seismic record, and assign it to a 3D geological grid model;

[0121] Correlation analysis and judgment accuracy unit, which is used to assign seismic data to the model grid of a 3D geological grid model, calculate the error between it and the 3D synthetic seismic record, as well as their correlation coefficient, set the required accuracy of the correlation coefficient of the seismic data and the 3D synthetic seismic record as needed, and judge whether the accuracy of their correlation coefficient reaches the preset accuracy. If not, jump to S6 for model update; if so, the update is completed, and an updated 3D reservoir physical property model is obtained;

[0122] Model update unit, which is used to perform updates using the numerical derivative method, correct the grid points that do not meet the accuracy requirements, update the 3D synthetic seismic record, and then correct and update the 3D reservoir physical property model of the reservoir, and return to step S2 until the correlation coefficient reaches the preset accuracy.

[0123] Example 5

[0124] This embodiment provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the update method provided in Example 1 are implemented.

[0125] Example 6

[0126] This embodiment provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the update method provided in Example 1 are implemented.

[0127] Example 7

[0128] This embodiment provides a computer program, which when executed by a processor implements the steps of the update method provided in Example 1.

[0129] The above embodiments are only the preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be pointed out that for those of ordinary skill in the art, improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A geological model updating method, characterized in that: The following steps are involved: S0. Prepare data The prepared data include actual seismic data and well logging data; S1. Establishing phase model Establishing a three-dimensional geological grid model by using a geostatistical method; performing sedimentary phase and lithofacies modeling on the three-dimensional geological grid model to obtain a phase model; S2. Physical Property Modeling According to the phase model and the three-dimensional geological grid model, a three-dimensional reservoir physical property model is established by statistically analyzing the well logging data, wherein the attributes of the three-dimensional reservoir physical property model include at least porosity, shale content and water saturation; S3. Rock physics modeling Based on the three-dimensional reservoir physical property model, a rock physical model is established by using an empirical formula through well logging data, and the porosity, shale content and water saturation in the attributes of the three-dimensional reservoir physical property model are converted into elastic parameters and acoustic parameters of the rock, including P-wave velocity, S-wave velocity and density field data, to obtain a three-dimensional reservoir elastic model; S4, 3D synthetic seismic forward modeling By using the three-dimensional reservoir elastic model and density field data, the reservoir medium impedance body is calculated, and seismic wavelets are selected for convolution forward modeling to obtain three-dimensional synthetic seismic records, which are assigned to the three-dimensional geological grid model; S5. Correlation analysis and accuracy judgment Assign the seismic data to the model grid of the three-dimensional geological grid model, calculate the error between the seismic data and the three-dimensional synthetic seismic record, and the correlation coefficient between the two, set the required accuracy of the correlation coefficient between the seismic data and the three-dimensional synthetic seismic record as needed, and judge whether the accuracy of the correlation coefficient between the two reaches the preset accuracy. If not, jump to S6 to update the model; if so, the update is completed to obtain an updated three-dimensional reservoir physical property model; S6. Model update The numerical derivative method is used for updating, the grid points that do not meet the accuracy requirements are corrected, the three-dimensional synthetic seismic records are updated, and then the three-dimensional reservoir physical property model of the reservoir is corrected and updated, and the process returns to step S2 until the correlation coefficient reaches the preset accuracy.

2. The geological model updating method according to claim 1, characterized in that: The updating of the numerical derivative in step S6 includes the following steps: S61: Put the observed seismic data and the three-dimensional synthetic seismic record in the same section, calculate the error between the two, and perform data optimization. The calculation method is as follows: F=1-Corr(S obs ,S symth )+(S obs -S symth ) 2 (1) Where F is the objective function, S obs is the actual observed earthquake record, S symth For a three-dimensional synthetic seismic record, the correlation between the actual observed seismic record and the three-dimensional synthetic seismic record is calculated. The higher the correlation coefficient is, the closer the geological model is to the actual geological situation. S62: updating the geological model by numerical derivative methods, wherein the numerical derivative methods include the steepest gradient descent method, Newton's method and conjugate gradient method.

3. The geological model updating method according to claim 2, characterized in that: Step S62: The fastest gradient descent method is F versus φ k Derivative, the negative direction of the derivative is the model update direction, the model is updated by the method of numerical derivative, and the updated porosity φ′ is equal to: In the above formula, λ represents the step size of the gradient method; The Newton method calculates the porosity φ′ by the following formula: In the formula, H -1 is the second-order derivative of the objective function; The conjugate gradient method calculates the porosity φ′ by the following formula: φ′=φ k +λd k (4) Where, d k is the conjugate direction, β k-1 is the conjugate coefficient.

4. The geological model updating method according to claim 1, characterized in that: The logging data at least include porosity, water saturation, shale content, density field data, post-stack seismic data, seismic interpretation information and measurement depth.

5. The geological model updating method according to claim 1, characterized in that: The empirical formula in step S3 includes a Gassmann model, a Hudson model, a Wyllie model or a Waxman-Smits model.

6. The geological model updating method according to claim 1, characterized in that: The calibration in step S3 is to compare the synthesized P-wave and S-wave velocities with the P-wave and S-wave velocities in actual logging to verify the rock physics model. If any unreasonableness is found, the rock physics model is modified accordingly until the logging verification and profile verification achieve the target effect.

7. A geological model updating system, characterized in that: It includes: A data preparation unit is used to prepare data, including actual seismic data and well logging data; The facies model unit is used to establish a three-dimensional geological grid model using geostatistical methods; Perform sedimentary phase and lithofacies modeling on the three-dimensional geological grid model to obtain a phase model; A physical property modeling unit, for establishing a three-dimensional reservoir physical property model based on the phase model and the three-dimensional geological grid model and statistical logging data, wherein the attributes of the three-dimensional reservoir physical property model include at least porosity, shale content and water saturation; A rock physics model building unit is used to establish a rock physics model through empirical formulas based on the three-dimensional reservoir physical property model using well logging data, and to convert the porosity, shale content and water saturation in the attributes of the three-dimensional reservoir physical property model into elastic parameters and acoustic parameters of the rock, including P-wave velocity, S-wave velocity and density field data, to obtain a three-dimensional reservoir elastic model; A three-dimensional synthetic seismic forward modeling unit is used to calculate the reservoir medium impedance body through the three-dimensional reservoir elastic model in combination with density field data, select seismic wavelets for convolution forward modeling, obtain three-dimensional synthetic seismic records, and assign them to the three-dimensional geological grid model; A correlation analysis and accuracy judgment unit is used to assign the seismic data to the model grid of the three-dimensional geological grid model, calculate the error between the seismic data and the three-dimensional synthetic seismic record, and the correlation coefficient between the two, set the required accuracy of the correlation coefficient between the seismic data and the three-dimensional synthetic seismic record as needed, and judge whether the accuracy of the correlation coefficient between the two reaches the preset accuracy. If not, jump to S6 to update the model; if so, the update is completed to obtain an updated three-dimensional reservoir physical property model; The model updating unit is used to update by using the numerical derivative method, correct the grid points that do not meet the accuracy requirements, update the three-dimensional synthetic seismic records, and then correct and update the three-dimensional reservoir physical property model of the reservoir, and return to step S2 until the correlation coefficient reaches the preset accuracy.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images