Method, device, medium and equipment for predicting tight thin reservoir

Abstract

The present invention discloses a method, apparatus, medium, and equipment for predicting tight thin reservoirs. The method comprises: obtaining P-wave impedance parameters and lithofacies classification curves based on well logging data; obtaining a probability density function based on the P-wave impedance parameters and lithofacies classification curves; obtaining a P-wave impedance data volume based on seismic data; obtaining a lithofacies probability data volume based on the P-wave impedance data volume and the probability density function; and obtaining a P-wave impedance inversion data volume and a lithofacies inversion data volume through random inversion using the probability density function and the lithofacies probability data volume as trend constraints. The present invention applies seismic lithofacies volumes to random seismic inversion, predicting the probability distribution of different lithofacies volumes based on deterministic inversion and lithofacies probability analysis. The lithofacies probability data volume is added to the random inversion as a constraint, thereby obtaining relatively stable inversion results, reducing the randomness and multi-solution nature of the random inversion, and improving the inversion prediction accuracy for tight thin reservoirs.

Description

Technical Field

[0001] The present invention belongs to the field of seismic reservoir prediction, and in particular relates to a method, device, medium and electronic equipment for predicting tight thin reservoirs based on seismic phase control. Background Art

[0002] With the continuous deepening of seismic exploration, the exploration target has gradually shifted to tight thin reservoirs. Tight thin reservoirs have become an important area of ​​oil and gas exploration. Therefore, tight thin reservoir prediction technology has become an important research topic in oil fields at home and abroad.

[0003] Currently, the most commonly used method for predicting tight thin reservoirs is seismic inversion. Many scholars at home and abroad have proposed corresponding seismic inversion research methods, including sparse pulse inversion, model-constrained inversion, and random seismic inversion. Sparse pulse inversion is based on the wave impedance inversion of seismic data. It is faithful to the seismic data and is less affected by the model, but has low vertical resolution. Model-constrained inversion methods have high vertical resolution, but are seriously affected by the model and have prominent multi-solution characteristics. Random seismic inversion is an inversion method that combines random simulation with seismic inversion. It has high vertical and horizontal resolution, but the lateral trends are often highly random, making it difficult to obtain relatively stable inversion results.

[0004] In summary, the use of seismic data to predict tight thin reservoirs has prominent problems of multi-solution and uncertainty, the vertical resolution of inversion and the horizontal prediction accuracy are low, the prediction accuracy of seismic data is not high, and the application effect is not obvious.

[0005] Therefore, a method for predicting tight thin reservoirs with high vertical resolution and lateral prediction accuracy of inversion and high prediction accuracy of seismic data is needed. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for predicting tight thin reservoirs with high vertical resolution and lateral prediction accuracy of inversion and high prediction accuracy of seismic data.

[0007] In a first aspect, the present invention provides a method for predicting tight thin reservoirs based on seismic phase control, comprising: obtaining P-wave impedance parameters and lithofacies classification curves based on well logging data; obtaining a probability density function based on the P-wave impedance parameters and lithofacies classification curves; obtaining a P-wave impedance data volume based on seismic data; obtaining a lithofacies probability data volume based on the P-wave impedance data volume and the probability density function; and obtaining a P-wave impedance inversion data volume and a lithofacies inversion data volume through random inversion using the probability density function and the lithofacies probability data volume as trend constraints.

[0008] Optionally, the lithofacies classification curve is obtained by the following steps: based on the reservoir type in the study area and according to the well logging data, the category of the reservoir lithofacies is determined; based on the well logging data and the category of the reservoir lithofacies, a lithofacies classification curve is generated.

[0009] Optionally, the probability density function of the longitudinal wave impedance parameter and the lithofacies category curve is obtained by a Bayesian classifier.

[0010] Optionally, the Bayesian classifier is:

[0011]

[0012] Among them, y i represents the i-th type of lithofacies, x is the longitudinal wave impedance parameter, P(y i |x) is the posterior probability of belonging to the i-th category of lithofacies, which means the probability that the sample point of the longitudinal wave impedance parameter belongs to the i-th category of lithofacies, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of the i-th type of lithofacies, and P(x) is the scaling factor.

[0013] Optionally, the P-wave impedance data volume is obtained by using a sparse pulse inversion method using post-stack seismic data, well logging data and seismic interpretation horizon data.

[0014] Optionally, the lithofacies probability data volume is obtained by the following steps: in the probability density function, the longitudinal wave impedance parameter is replaced by the longitudinal wave impedance data volume to obtain the lithofacies probability data volume.

[0015] Optionally, based on seismic data, well logging data and seismic interpretation horizon data, probability density functions and lithofacies probability data volumes are added as trend constraints, and P-wave impedance inversion data volumes and lithofacies inversion data volumes are obtained through random inversion.

[0016] In a second aspect, the present invention further provides an electronic device comprising: a memory storing executable instructions; and a processor executing the executable instructions in the memory to implement the above-mentioned method for predicting tight thin reservoirs based on seismic phase control.

[0017] In a third aspect, the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for predicting tight thin reservoirs based on seismic phase control.

[0018] In a fourth aspect, the present invention also provides a tight thin reservoir prediction device based on seismic phase control, comprising: a lithofacies category acquisition module, which obtains P-wave impedance parameters and lithofacies category curves based on well logging data; a probability density function acquisition module, which obtains a probability density function based on the P-wave impedance parameters and the lithofacies category curve; a P-wave impedance data body acquisition module, which obtains a P-wave impedance data body based on seismic data; a lithofacies probability data body acquisition module, which obtains a lithofacies probability data body based on the P-wave impedance data body and the probability density function; and an inversion data body acquisition module, which uses the probability density function and the lithofacies probability data body as trend constraints to obtain a P-wave impedance inversion data body and a lithofacies inversion data body through random inversion.

[0019] Optionally, the lithofacies classification curve is obtained by the following steps: based on the reservoir type in the study area and according to the well logging data, the category of the reservoir lithofacies is determined; based on the well logging data and the category of the reservoir lithofacies, a lithofacies classification curve is generated.

[0020] Optionally, the probability density function of the longitudinal wave impedance parameter and the lithofacies category curve is obtained by a Bayesian classifier.

[0021] Optionally, the Bayesian classifier is:

[0022]

[0023] Among them, y i represents the i-th type of lithofacies, x is the longitudinal wave impedance parameter, P(y i |x) is the posterior probability of belonging to the i-th category of lithofacies, which means the probability that the sample point of the longitudinal wave impedance parameter belongs to the i-th category of lithofacies, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of the i-th type of lithofacies, and P(x) is the scaling factor.

[0024] Optionally, the P-wave impedance data volume is obtained by using a sparse pulse inversion method using post-stack seismic data, well logging data and seismic interpretation horizon data.

[0025] Optionally, the lithofacies probability data volume is obtained by the following steps: in the probability density function, the longitudinal wave impedance parameter is replaced by the longitudinal wave impedance data volume to obtain the lithofacies probability data volume.

[0026] Optionally, based on seismic data, well logging data and seismic interpretation horizon data, probability density functions and lithofacies probability data volumes are added as trend constraints, and P-wave impedance inversion data volumes and lithofacies inversion data volumes are obtained through random inversion.

[0027] The beneficial effects of the present invention are as follows: the method for predicting tight thin reservoirs based on seismic phase control of the present invention applies seismic lithofacies bodies to random seismic inversion, predicts the probability distribution of different lithofacies bodies based on deterministic inversion and lithofacies probability analysis, and adds lithofacies probability data bodies to random inversion as constraints, thereby obtaining relatively stable inversion results, reducing the randomness and multi-solution characteristics of random inversion, and improving the inversion prediction accuracy of tight thin reservoirs.

[0028] The present invention has other features and advantages that will be apparent from or will be described in detail in the accompanying drawings and the following specific examples incorporated herein, which together serve to explain the specific principles of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present invention.

[0030] Figure 1 A flowchart of a method for predicting tight thin reservoirs based on seismic phase control according to an embodiment of the present invention is shown.

[0031] Figure 2 Another flow chart of a method for predicting tight thin reservoirs based on seismic phase control according to an embodiment of the present invention is shown.

[0032] Figure 3 A cross-sectional diagram of compressional wave impedance parameters of sparse pulse inversion of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown.

[0033] Figure 4 A probability distribution diagram of different lithofacies at well sampling points in a method for predicting tight thin reservoirs based on seismic facies control according to an embodiment of the present invention is shown.

[0034] Figure 5 A sandstone probability volume of a tight thin reservoir prediction method based on seismic facies control according to an embodiment of the present invention is shown.

[0035] Figure 6 A mudstone probability volume of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown.

[0036] Figure 7 The figure shows a compressional wave impedance inversion data volume of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention.

[0037] Figure 8 A sandstone probability attribute diagram of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown.

[0038] Figure 9 A sandstone thickness map obtained by inversion prediction of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown.

[0039] Figure 10 A structural block diagram of a tight thin reservoir prediction device based on seismic phase control according to an embodiment of the present invention is shown.

[0040] Description of Reference Numerals

[0041] 102. Lithofacies category acquisition module; 104. Probability density function acquisition module; 106. P-wave impedance data acquisition module; 108. Lithofacies probability data acquisition module; 110. Inversion data acquisition module. DETAILED DESCRIPTION

[0042] The preferred embodiments of the present invention will be described in more detail below. Although the preferred embodiments of the present invention are described below, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0043] The present invention provides a tight thin reservoir prediction method based on seismic phase control, comprising: obtaining a P-wave impedance parameter and a lithofacies classification curve based on well logging data; obtaining a probability density function based on the P-wave impedance parameter and the lithofacies classification curve; obtaining a P-wave impedance data volume based on seismic data; obtaining a lithofacies probability data volume based on the P-wave impedance data volume and the probability density function; and obtaining a P-wave impedance inversion data volume and a lithofacies inversion data volume through random inversion using the probability density function and the lithofacies probability data volume as trend constraints.

[0044] Specifically, sparse pulse inversion is used to obtain the P-wave impedance data volume, and the P-wave impedance parameters and lithofacies category curves are used to obtain the probability density function. The P-wave impedance data volume and the lithofacies probability density function are used to obtain different lithofacies probability data volumes. The probability density function and the lithofacies probability data volume are added to the random inversion as prior constraints to control the longitudinal and lateral variation trends of different lithofacies, and at the same time, relatively stable and high-resolution P-wave impedance and lithofacies inversion results are obtained.

[0045] According to an exemplary embodiment, the tight thin reservoir prediction method based on seismic phase control applies seismic lithofacies bodies to random seismic inversion, and predicts the probability distribution of different lithofacies bodies based on deterministic inversion and lithofacies probability analysis. The lithofacies probability data body is added to the random inversion as a constraint condition, which can obtain relatively stable inversion results, reduce the randomness and multi-solution of random inversion, and improve the inversion prediction accuracy of tight thin reservoirs.

[0046] As an optional solution, the lithofacies classification curve is obtained by the following steps: based on the reservoir type in the study area and according to the well logging data, the reservoir lithofacies classification is determined; based on the well logging data and the reservoir lithofacies classification, the lithofacies classification curve is generated.

[0047] Specifically, according to the sedimentary characteristics and logging curve features of the study area, the lithofacies division scheme is determined, the reservoir is divided into multiple lithofacies types, and the lithofacies type curve is generated according to the lithologic types and logging data.

[0048] Alternatively, the probability density functions of the P-wave impedance parameters and lithofacies class curves are obtained using a Bayesian classifier.

[0049] As an alternative, a Bayesian classifier is:

[0050]

[0051] Among them, y i represents the i-th type of lithofacies, x is the longitudinal wave impedance parameter, P(y i |x) is the posterior probability of belonging to the i-th category of lithofacies, which means the probability that the sample point of the longitudinal wave impedance parameter belongs to the i-th category of lithofacies, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of the i-th type of lithofacies, and P(x) is the scaling factor.

[0052] Specifically, a probability statistical analysis was performed on the well sample points to establish a probability statistical relationship between different lithofacies and P-wave impedance parameters, namely a probability density function. By adjusting the parameters of the probability density function (such as the mean and variance), the probability distribution of each attribute condition at the sample point was changed to be consistent with the distribution of the well sample points.

[0053] The probability density function represents the probability distribution of elastic parameters corresponding to different rock types. Based on the statistics of well logging data samples, the probability density of different lithofacies and prior geological information can be integrated into the Bayesian classifier. The elastic parameter volume is used to generate a lithofacies classification volume and a lithofacies probability data volume. The Bayesian classification method is based on Bayesian theory. Based on the distribution of samples in each category in the training set, the posterior probability of each sample belonging to each category is calculated. The category of the sample is then determined to be the category corresponding to the maximum posterior probability. The Bayesian classifier can be written as follows:

[0054]

[0055] Among them, x is the variable attribute parameter, y i Indicates different lithofacies categories, P(y i |x) is the posterior probability of belonging to a certain lithofacies, which means the probability that the sample point x belongs to a certain lithofacies under certain attribute conditions, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of a certain lithofacies, which can be obtained based on the statistics of well logging data. P(x) is usually a constant proportional factor that controls the sum of the probabilities of different lithofacies classifications to be 1.

[0056] As an alternative, the P-wave impedance data volume is obtained by sparse pulse inversion method using post-stack seismic data, well logging data and seismic interpretation horizon data.

[0057] Specifically, the P-wave impedance data volume is obtained by using the sparse pulse inversion method using post-stack seismic data, well logging data, and seismic interpretation horizons.

[0058] As an optional solution, the lithofacies probability data volume is obtained by the following steps: in the probability density function, the longitudinal wave impedance parameter is replaced by the longitudinal wave impedance data volume to obtain the lithofacies probability data volume.

[0059] Specifically, the conditional probability distribution of the well sample point attribute statistics is used to convert the seismic inversion wave impedance data volume into the lithofacies probability data volume corresponding to different rocks.

[0060] As an optional solution, based on seismic data, well logging data and seismic interpretation horizon data, probability density functions and lithofacies probability data volumes are added as trend constraints, and the P-wave impedance inversion data volume and lithofacies inversion data volume are obtained through random inversion.

[0061] Specifically, the stochastic seismic inversion method is an inversion algorithm that combines stochastic simulation and seismic inversion. In stochastic inversion based on seismic data, well logging data, and seismic interpretation horizon data, a Markov Chain-Monte Carlo algorithm is used to obtain a set of sample points that meet these two probability conditions based on the probability density function and the lithofacies probability data volume. Each sample point has a P-wave impedance attribute and a lithofacies attribute. The P-wave impedance of each sample point in the sample set is connected to obtain a P-wave impedance inversion data volume, and the lithofacies of each sample point in the sample set are connected to obtain a lithofacies impedance data volume. If the random inversion is constrained by prior information to obtain relatively reliable results and reduce inversion uncertainty, a three-dimensional lithofacies probability volume constraint can be added, using a phase-controlled stochastic inversion method.

[0062] In a second aspect, the present invention further provides an electronic device, comprising: a memory storing executable instructions; and a processor executing the executable instructions in the memory to implement the above-mentioned method for predicting tight thin reservoirs based on seismic phase control.

[0063] In a third aspect, the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for predicting tight thin reservoirs based on seismic phase control.

[0064] In a fourth aspect, the present invention also provides a tight thin reservoir prediction device based on seismic phase control, including: a facies category acquisition module, which obtains P-wave impedance parameters and facies category curves based on logging data; a probability density function acquisition module, which obtains probability density functions based on P-wave impedance parameters and facies category curves; a P-wave impedance data body acquisition module, which obtains P-wave impedance data body based on seismic data; a facies probability data body acquisition module, which obtains facies probability data body based on P-wave impedance data body and probability density function; and an inversion data body acquisition module, which uses probability density functions and facies probability data body as trend constraints to obtain P-wave impedance inversion data body and facies inversion data body through random inversion.

[0065] Specifically, sparse pulse inversion is used to obtain the P-wave impedance data volume, and the P-wave impedance parameters and lithofacies category curves are used to obtain the probability density function. The P-wave impedance data volume and the lithofacies probability density function are used to obtain different lithofacies probability data volumes. The probability density function and the lithofacies probability data volume are added to the random inversion as prior constraints to control the longitudinal and lateral variation trends of different lithofacies, and at the same time, relatively stable and high-resolution P-wave impedance and lithofacies inversion results are obtained.

[0066] According to an exemplary embodiment, the tight thin reservoir prediction method based on seismic phase control applies seismic lithofacies bodies to random seismic inversion, and predicts the probability distribution of different lithofacies bodies based on deterministic inversion and lithofacies probability analysis. The lithofacies probability data body is added to the random inversion as a constraint condition, which can obtain relatively stable inversion results, reduce the randomness and multi-solution of random inversion, and improve the inversion prediction accuracy of tight thin reservoirs.

[0067] As an optional solution, the lithofacies classification curve is obtained by the following steps: based on the reservoir type in the study area and according to the well logging data, the reservoir lithofacies classification is determined; based on the well logging data and the reservoir lithofacies classification, the lithofacies classification curve is generated.

[0068] Specifically, according to the sedimentary characteristics and logging curve features of the study area, the lithofacies division scheme is determined, the reservoir is divided into multiple lithofacies types, and the lithofacies type curve is generated according to the lithologic types and logging data.

[0069] Alternatively, the probability density functions of the P-wave impedance parameters and lithofacies class curves are obtained using a Bayesian classifier.

[0070] As an alternative, a Bayesian classifier is:

[0071]

[0072] Among them, y i represents the i-th type of lithofacies, x is the longitudinal wave impedance parameter, P(y i |x) is the posterior probability of belonging to the i-th category of lithofacies, which means the probability that the sample point of the longitudinal wave impedance parameter belongs to the i-th category of lithofacies, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of the i-th type of lithofacies, and P(x) is the scaling factor.

[0073] Specifically, a probability statistical analysis was performed on the well sample points to establish a probability statistical relationship between different lithofacies and P-wave impedance parameters, namely a probability density function. By adjusting the parameters of the probability density function (such as the mean and variance), the probability distribution of each attribute condition at the sample point was changed to be consistent with the distribution of the well sample points.

[0074] The probability density function represents the probability distribution of elastic parameters corresponding to different rock types. Based on the statistics of well logging data samples, the probability density of different lithofacies and prior geological information can be integrated into the Bayesian classifier. The elastic parameter volume is used to generate a lithofacies classification volume and a lithofacies probability data volume. The Bayesian classification method is based on Bayesian theory. Based on the distribution of samples in each category in the training set, the posterior probability of each sample belonging to each category is calculated. The category of the sample is then determined to be the category corresponding to the maximum posterior probability. The Bayesian classifier can be written as follows:

[0075]

[0076] Among them, x is the variable attribute parameter, y i Indicates different lithofacies categories, P(y i |x) is the posterior probability of belonging to a certain lithofacies, which means the probability that the sample point x belongs to a certain lithofacies under certain attribute conditions, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of a certain lithofacies, which can be obtained based on the statistics of well logging data. P(x) is usually a constant proportional factor that controls the sum of the probabilities of different lithofacies classifications to be 1.

[0077] As an alternative, the P-wave impedance data volume is obtained by sparse pulse inversion method using post-stack seismic data, well logging data and seismic interpretation horizon data.

[0078] Specifically, the P-wave impedance data volume is obtained by using the sparse pulse inversion method using post-stack seismic data, well logging data, and seismic interpretation horizons.

[0079] As an optional solution, the lithofacies probability data volume is obtained by the following steps: in the probability density function, the longitudinal wave impedance parameter is replaced by the longitudinal wave impedance data volume to obtain the lithofacies probability data volume.

[0080] Specifically, the conditional probability distribution of the well sample point attribute statistics is used to convert the seismic inversion wave impedance data volume into the lithofacies probability data volume corresponding to different rocks.

[0081] As an optional solution, based on seismic data, well logging data and seismic interpretation horizon data, probability density functions and lithofacies probability data volumes are added as trend constraints, and the P-wave impedance inversion data volume and lithofacies inversion data volume are obtained through random inversion.

[0082] Specifically, the stochastic seismic inversion method is an inversion algorithm that combines stochastic simulation and seismic inversion. In stochastic inversion based on seismic data, well logging data, and seismic interpretation horizon data, a Markov Chain-Monte Carlo algorithm is used to obtain a set of sample points that meet these two probability conditions based on the probability density function and the lithofacies probability data volume. Each sample point has a P-wave impedance attribute and a lithofacies attribute. The P-wave impedance of each sample point in the sample set is connected to obtain a P-wave impedance inversion data volume, and the lithofacies of each sample point in the sample set are connected to obtain a lithofacies impedance data volume. If the random inversion is constrained by prior information to obtain relatively reliable results and reduce inversion uncertainty, a three-dimensional lithofacies probability volume constraint can be added, using a phase-controlled stochastic inversion method.

[0083] Example 1

[0084] Figure 1 A flowchart of a method for predicting tight thin reservoirs based on seismic phase control according to an embodiment of the present invention is shown. Figure 2 Another flow chart of a method for predicting tight thin reservoirs based on seismic phase control according to an embodiment of the present invention is shown. Figure 3 A cross-sectional diagram of compressional wave impedance parameters of sparse pulse inversion of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown. Figure 4 A probability distribution diagram of different lithofacies at well sampling points in a method for predicting tight thin reservoirs based on seismic facies control according to an embodiment of the present invention is shown. Figure 5 A sandstone probability volume of a tight thin reservoir prediction method based on seismic facies control according to an embodiment of the present invention is shown. Figure 6 A mudstone probability volume of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown. Figure 7 The figure shows a compressional wave impedance inversion data volume of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention. Figure 8 A sandstone probability attribute diagram of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown. Figure 9 A sandstone thickness map obtained by inversion prediction of a tight thin reservoir prediction method based on seismic phase control according to an embodiment of the present invention is shown.

[0085] Combine Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 、 Figure 7 、 Figure 8 and Figure 9 As shown in FIG, the tight thin reservoir prediction method based on seismic phase control includes:

[0086] Step 1: Based on the well logging data, obtain the P-wave impedance parameters and lithofacies classification curve;

[0087] Step 2: Obtain the probability density function based on the P-wave impedance parameters and the lithofacies classification curve;

[0088] Step 3: Based on the seismic data, obtain the P-wave impedance data volume;

[0089] Step 4: Based on the P-wave impedance data volume and the probability density function, obtain the lithofacies probability data volume;

[0090] Step 5: Using the probability density function and lithofacies probability data volume as trend constraints, obtain the P-wave impedance inversion data volume and lithofacies inversion data volume through random inversion.

[0091] The lithofacies classification curve is obtained by the following steps: based on the reservoir type in the study area and according to the well logging data, the reservoir lithofacies classification is determined; based on the well logging data and the reservoir lithofacies classification, the lithofacies classification curve is generated.

[0092] The probability density functions of the longitudinal wave impedance parameters and lithofacies classification curves are obtained through the Bayesian classifier.

[0093] Among them, the Bayesian classifier is:

[0094]

[0095] Among them, y i represents the i-th type of lithofacies, x is the longitudinal wave impedance parameter, P(y i |x) is the posterior probability of belonging to the i-th category of lithofacies, which means the probability that the sample point of the longitudinal wave impedance parameter belongs to the i-th category of lithofacies, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of the i-th type of lithofacies, and P(x) is the scaling factor.

[0096] Among them, the P-wave impedance data volume is obtained by using the sparse pulse inversion method using post-stack seismic data, well logging data and seismic interpretation layer data.

[0097] The lithofacies probability data volume is obtained by the following steps: in the probability density function, the longitudinal wave impedance parameter is replaced by the longitudinal wave impedance data volume to obtain the lithofacies probability data volume.

[0098] Among them, based on seismic data, well logging data and seismic interpretation layer data, probability density function and lithofacies probability data body are added as trend constraints, and the longitudinal wave impedance inversion data body and lithofacies inversion data body are obtained through random inversion.

[0099] According to the reservoir characteristics and types in the study area, different lithofacies are defined using logging data and interpretation results. They are divided into two lithofacies types: sandstone and mudstone. Sandstone corresponds to a natural gamma ray GR ≤ 110 API, while mudstone corresponds to a natural gamma ray GR > 110 API. The lithofacies definition results are required to be basically consistent with the logging interpretation results.

[0100] First, the original seismic data, well logging data and seismic interpretation horizons are used to perform well calibration and wavelet extraction, and the inverted P-wave impedance low-frequency model is established. The P-wave impedance inversion data volume is obtained using the sparse pulse inversion method, such as Figure 3 .

[0101] By comparing the logging response characteristics of different lithofacies and statistically analyzing the differences in P-wave impedance changes, it can be determined that the sandstone in the study area mainly exhibits medium-high P-wave impedance characteristics, and most of the mudstone exhibits low P-wave impedance characteristics, but there is some overlap.

[0102] In order to overcome the multi-solution problem of attribute prediction, the probability statistical analysis method is adopted to establish the probability distribution of different lithofacies using the longitudinal wave impedance parameters. The corresponding function type is selected according to the probability distribution characteristics. This time, the Gaussian kernel function is selected. By adjusting the main parameters of the function including the mean and variance, the interval described by the function is basically consistent with the distribution of well logging points. For example, Figure 4 As shown, the mean P-wave impedance of sandstone is 11,324 g / cm³*m / s, with a standard deviation of 763, accounting for 0.37% of the lithofacies. The mean P-wave impedance of mudstone is 10,687 g / cm³*m / s, with a standard deviation of 676, accounting for 0.63% of the lithofacies.

[0103] use Figure 4 The statistical relationship established is used to convert the P-wave impedance data volume into different lithofacies volumes and corresponding lithofacies probability volumes. From the lithofacies probability prediction results, the well points are basically consistent, and the lateral changes are faithful to the original seismic data, reflecting the heterogeneity change characteristics between wells. However, the vertical resolution of the inversion is relatively low, such as Figure 5 and Figure 6 .

[0104] Using original seismic data, well logging data, seismic interpretation horizons, wavelets and low-frequency models, random inversion is performed, and different lithofacies data bodies are added as inversion trend constraints to obtain high-resolution P-wave impedance data bodies, such as Figure 7 .

[0105] Example 2

[0106] Figure 10 A structural block diagram of a tight thin reservoir prediction device based on seismic phase control according to an embodiment of the present invention is shown.

[0107] like Figure 10As shown, the tight thin reservoir prediction device based on seismic phase control includes:

[0108] The lithofacies classification acquisition module 102 obtains the longitudinal wave impedance parameters and lithofacies classification curve based on the well logging data;

[0109] The probability density function acquisition module 104 obtains the probability density function based on the longitudinal wave impedance parameter and the lithofacies classification curve;

[0110] A longitudinal wave impedance data volume acquisition module 106 is configured to acquire a longitudinal wave impedance data volume based on seismic data;

[0111] The lithofacies probability data volume acquisition module 108 obtains the lithofacies probability data volume based on the longitudinal wave impedance data volume and the probability density function;

[0112] The inversion data volume acquisition module 110 uses the probability density function and the lithofacies probability data volume as trend constraints to obtain the P-wave impedance inversion data volume and the lithofacies inversion data volume through random inversion.

[0113] The lithofacies classification curve is obtained by the following steps: based on the reservoir type in the study area and according to the well logging data, the reservoir lithofacies classification is determined; based on the well logging data and the reservoir lithofacies classification, the lithofacies classification curve is generated.

[0114] The probability density functions of the longitudinal wave impedance parameters and lithofacies classification curves are obtained through the Bayesian classifier.

[0115] Among them, the Bayesian classifier is:

[0116]

[0117] Among them, y i represents the i-th type of lithofacies, x is the longitudinal wave impedance parameter, P(y i |x) is the posterior probability of belonging to the i-th category of lithofacies, which means the probability that the sample point of the longitudinal wave impedance parameter belongs to the i-th category of lithofacies, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of the i-th type of lithofacies, and P(x) is the scaling factor.

[0118] Among them, the P-wave impedance data volume is obtained by using the sparse pulse inversion method using post-stack seismic data, well logging data and seismic interpretation layer data.

[0119] The lithofacies probability data volume is obtained by the following steps: in the probability density function, the longitudinal wave impedance parameter is replaced by the longitudinal wave impedance data volume to obtain the lithofacies probability data volume.

[0120] Among them, based on seismic data, well logging data and seismic interpretation layer data, probability density function and lithofacies probability data body are added as trend constraints, and the longitudinal wave impedance inversion data body and lithofacies inversion data body are obtained through random inversion.

[0121] Example 3

[0122] The present disclosure provides an electronic device comprising: a memory storing executable instructions; and a processor executing the executable instructions in the memory to implement the above-mentioned method for predicting tight thin reservoirs based on seismic phase control.

[0123] An electronic device according to an embodiment of the present disclosure includes a memory and a processor.

[0124] The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc.

[0125] The processor may be a central processing unit (CPU) or other form of processing unit having data processing capability and / or instruction execution capability, and may control other components in the electronic device to perform desired functions. In one embodiment of the present disclosure, the processor is used to execute the computer-readable instructions stored in the memory.

[0126] Those skilled in the art should understand that in order to solve the technical problem of how to obtain a good user experience, this embodiment may also include well-known structures such as a communication bus and an interface, and these well-known structures should also be included in the scope of protection of this disclosure.

[0127] For detailed description of this embodiment, please refer to the corresponding description in the aforementioned embodiments, which will not be repeated here.

[0128] Example 4

[0129] The present disclosure provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method for predicting tight thin reservoirs based on seismic phase control is implemented.

[0130] According to an embodiment of the present disclosure, a computer-readable storage medium stores non-transitory computer-readable instructions, which, when executed by a processor, execute all or part of the steps of the aforementioned methods of the embodiments of the present disclosure.

[0131] The above-mentioned computer-readable storage media include, but are not limited to, optical storage media (e.g., CD-ROMs and DVDs), magneto-optical storage media (e.g., MOs), magnetic storage media (e.g., magnetic tapes or mobile hard disks), media with built-in rewritable non-volatile memory (e.g., memory cards), and media with built-in ROM (e.g., ROM cartridges).

[0132] While various embodiments of the present invention have been described above, the above description is intended to be illustrative, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for predicting tight thin reservoirs based on seismic phase control, characterized in that: include: Based on the well logging data, the P-wave impedance parameters and lithofacies classification curves are obtained; Obtaining a probability density function based on the longitudinal wave impedance parameter and the lithofacies category curve; Based on seismic data, obtain the longitudinal wave impedance data volume; Obtaining a lithofacies probability data volume based on the longitudinal wave impedance data volume and the probability density function; The probability density function and lithofacies probability data volume are used as trend constraints, and a longitudinal wave impedance inversion data volume and a lithofacies inversion data volume are obtained through random inversion.

2. The method for predicting tight thin reservoirs based on seismic phase control according to claim 1, characterized in that: Obtain the lithofacies classification curve by the following steps: Based on the reservoir type in the study area and the well logging data, the category of the reservoir lithofacies is determined, and based on the well logging data and the category of the reservoir lithofacies, a lithofacies category curve is generated.

3. The method for predicting tight thin reservoirs based on seismic phase control according to claim 1, characterized in that: The probability density function of the longitudinal wave impedance parameter and the lithofacies category curve is obtained through a Bayesian classifier.

4. The method for predicting tight thin reservoirs based on seismic phase control according to claim 3, characterized in that: The Bayesian classifier is: Among them, y i represents the i-th type of lithofacies, x is the longitudinal wave impedance parameter, P(y i |x) is the posterior probability of belonging to the i-th category of lithofacies, which means the probability that the sample point of the longitudinal wave impedance parameter belongs to the i-th category of lithofacies, P(x|y i ) is the conditional probability density function, P(y i ) is the prior probability of the i-th type of lithofacies, and P(x) is the scaling factor.

5. The method for predicting tight thin reservoirs based on seismic phase control according to claim 1, characterized in that: The P-wave impedance data volume is obtained by using post-stack seismic data, well logging data and seismic interpretation horizon data through a sparse pulse inversion method.

6. The method for predicting tight thin reservoirs based on seismic phase control according to claim 1, characterized in that: The lithofacies probability data volume is obtained through the following steps: In the probability density function, the longitudinal wave impedance parameter is replaced by the longitudinal wave impedance data volume to obtain a lithofacies probability data volume.

7. The method for predicting tight thin reservoirs based on seismic phase control according to claim 1, characterized in that: Based on seismic data, well logging data and seismic interpretation horizon data, probability density function and lithofacies probability data volume are added as trend constraints, and the P-wave impedance inversion data volume and lithofacies inversion data volume are obtained through random inversion.

8. An electronic device, characterized in that: The electronic device comprises: a memory storing executable instructions; A processor is configured to execute the executable instructions in the memory to implement the method for predicting tight thin reservoirs based on seismic facies control according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for predicting tight thin reservoirs based on seismic phase control according to any one of claims 1 to 7.

10. A device for predicting tight thin reservoirs based on seismic phase control, characterized in that: include: The lithofacies classification acquisition module obtains the longitudinal wave impedance parameters and lithofacies classification curves based on well logging data; A probability density function acquisition module, which obtains a probability density function based on the longitudinal wave impedance parameter and the lithofacies classification curve; The longitudinal wave impedance data volume acquisition module obtains the longitudinal wave impedance data volume based on seismic data; A lithofacies probability data volume acquisition module is configured to obtain a lithofacies probability data volume based on the longitudinal wave impedance data volume and the probability density function; The inversion data volume acquisition module uses the probability density function and the lithofacies probability data volume as trend constraints to obtain the longitudinal wave impedance inversion data volume and the lithofacies inversion data volume through random inversion.

Images