Geophysical identification method for middle-deep layer under-compacted low-speed mudstone

Abstract

The invention provides a geophysical recognition method for medium-deep under-compacted low-speed mudstone, which comprises the following steps of: obtaining representative longitudinal wave velocity, transverse wave velocity and density of low-speed mudstone and gas-bearing sandstone in logging information through statistical analysis of data, and designing a theoretical model of a superposition relationship between the low-speed mudstone and the gas-bearing sandstone in combination with geological information, so as to obtain the geophysical recognition result of the low-speed mudstone and the gas-bearing sandstone. Performing convolution forward modeling and pre-stack forward modeling on the model, respectively calculating to obtain post-stack seismic attributes and pre-stack seismic attributes of the model, performing sensitive attribute optimization on the attributes, taking the obtained sensitive attributes as learning samples, and identifying the superposition relationship by using probability self-organizing neural network clustering. According to the method, the pre-stack attributes are added into sensitive attribute optimization, so that the fuzzy self-organizing neural network can more accurately recognize the gas-bearing sandstone and low-speed mudstone stacking relation, the low-speed mudstone and the gas-bearing sandstone can be effectively distinguished, and support is provided for subsequent effective reservoir prediction.

Description

Technical Field

[0001] The invention provides a geophysical identification method for medium-deep under-compacted low-velocity mudstone, belonging to the technical field of geological exploration. Background Art

[0002] The deep and medium layers of the Yingqiong Basin have obvious high temperature (ultra) high pressure characteristics, and undercompacted low-velocity mudstone and complex structures are developed, which leads to complex seismic responses and increases the difficulty of reservoir prediction. The undercompacted low-velocity mudstone formed by rapid deposition under high temperature (ultra) high pressure has drastic velocity changes in some areas and a wide range of changes, which directly affects the seismic reflection characteristics of the top of the gas layer. From the rock physical characteristics of low-velocity mudstone analyzed from known logging data, it can be seen that the longitudinal wave impedance of low-velocity mudstone is basically consistent with that of gas-bearing sandstone, or even lower. This leads to the formation of "false bright spot" reflection when low-velocity mudstone contacts with high-impedance background mudstone on the post-stack section, and "dark spot" reflection when low-velocity mudstone contacts with gas-bearing sandstone, which seriously affects the identification and prediction of effective reservoirs. Therefore, it is necessary to clarify the distribution of low-velocity mudstone under high temperature and high pressure and the characteristic laws of its different superposition patterns with gas-bearing sandstone, so as to provide support for the subsequent selection of favorable target exploration areas and reservoir prediction in development.

[0003] Identifying lithology and thickness by optimizing sensitive attributes is a common method in the field of lithology identification. By optimizing the sensitive attributes of sand body thickness, an intelligent well-seismic sand body prediction method is used for various types of sand bodies with different superposition relationships. The attributes used by this technology to identify the superposition relationship of sand bodies are limited to post-stack attributes, and gas-bearing sandstone and low-velocity mudstone are always difficult to distinguish on the post-stack section, so this method is not suitable for identifying the superposition relationship between low-velocity mudstone and gas-bearing sandstone. Summary of the invention

[0004] The present invention mainly focuses on inventing a geophysical identification method for medium-deep undercompacted low-velocity mudstone. The method selects the post-stack and pre-stack sensitive attributes of different superposition relationships between low-velocity mudstone and gas-bearing sandstone, and uses them as input for clustering using a fuzzy self-organizing neural network, thereby distinguishing the different superposition relationships between low-velocity mudstone and gas-bearing sandstone, and then identifying the distribution and thickness of the low-velocity mudstone.

[0005] A geophysical identification method for medium-deep undercompacted low-velocity mudstone obtains representative P-wave velocity, S-wave velocity and density of low-velocity mudstone and gas-bearing sandstone in logging data by statistically analyzing data. Combined with geological information, a theoretical model of the superposition relationship between low-velocity mudstone and gas-bearing sandstone is designed. Convolution forward modeling and pre-stack forward modeling are performed on the model to calculate the post-stack and pre-stack seismic attributes of the model, respectively. Sensitive attributes of these attributes are then optimized, and the obtained sensitive attributes are used as learning samples to identify the superposition relationship using probabilistic self-organizing neural network clustering.

[0006] The specific steps include:

[0007] Step 1: Analyze the logging data and seismic data to obtain the typical rock physical parameters and contact relationship of low-velocity mudstone, gas-bearing sandstone and background mudstone, and establish a theoretical geological model of the superposition relationship between low-velocity mudstone and gas-bearing sandstone;

[0008] Step 2: Perform convolution forward modeling on the superposition model of low-velocity mudstone and gas-bearing sandstone to analyze whether the forward modeling results can represent the response characteristics of different contact relationships, such as the reservoir "dark spot" reflection phenomenon caused by low-velocity mudstone overlying gas-bearing sandstone and the "false bright spot" reflection phenomenon caused by the contact between low-velocity mudstone and background mudstone. If not, return to step 1 and modify the parameters of the superposition model.

[0009] Step 3: Perform pre-stack forward modeling on the superposition pattern of low-velocity mudstone and gas-bearing sandstone, and then calculate the intercept attribute P and gradient attribute G in the pre-stack attributes, calculate different combinations of P and G, and obtain the pre-stack combination attributes.

[0010] Step 4: Select an appropriate time window to extract various types of seismic attributes, and use the principal component analysis (PCA) method to evaluate the sensitivity of each seismic attribute between different superposition relationships, so as to select sensitive attributes that can distinguish different superposition relationships.

[0011] The steps to implement PCA are as follows:

[0012] 1. Standardize Data

[0013]

[0014] where x ij represents the jth seismic attribute of the ith overlay relation, n is the number of overlay relations, and p is the number of seismic attributes.

[0015]

[0016] where μ j is the mean of the jth earthquake attribute, σ j is its standard deviation.

[0017] 2. Perform principal component analysis on the standardized matrix Z and calculate its covariance matrix C:

[0018]

[0019] Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors. The eigenvector corresponds to the direction of the principal component, while the eigenvalue represents the variance of each principal component. By looking at the principal component load (the component of the eigenvector), we can determine which seismic attributes contribute the most to the principal component, thereby selecting the sensitive attributes.

[0020] Step 5: Input multiple sensitive seismic attributes selected in step 4 into the fuzzy self-organizing neural network at the same time. The network performs self-organizing learning competition according to the characteristics of the seismic attributes to see whether the results can distinguish the different superposition relationships between low-velocity mudstone and gas-bearing sandstone in the model. If they cannot be distinguished, it means that the sensitive attributes selected in step 4 are not suitable and need to be further optimized. If they can be distinguished, it means that the sensitive attributes selected in step 4 are sufficient to distinguish different superposition relationships and can be used for actual data.

[0021] Step 6: Extract the sensitive attributes selected in steps 4 and 5 along the appropriate time window in the actual seismic data, input them into the fuzzy self-organizing neural network after preprocessing to obtain the clustering results, and combine them with the logging data to explain the possible superposition relationship corresponding to each category, and then determine the distribution and thickness of the regional low-velocity mudstone.

[0022] The present invention is a method for optimally identifying low-velocity mudstone using sensitive attributes, in which pre-stack attributes are added to the sensitive attribute optimization, so that the fuzzy self-organizing neural network can more accurately identify the superposition relationship between gas-bearing sandstone and low-velocity mudstone, and can effectively distinguish low-velocity mudstone from gas-bearing sandstone, providing support for subsequent effective reservoir prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The superposition model of the low-velocity mudstone and the gas-bearing sandstone of the present invention;

[0024] Figure 2 The recognition results before and after adding the pre-stack attributes of the present invention;

[0025] Figure 3 Topological structure diagram of the fuzzy self-organizing neural network of the present invention. DETAILED DESCRIPTION

[0026] The present invention is described in detail below in conjunction with the accompanying drawings:

[0027] A geophysical identification method for medium-deep undercompacted low-velocity mudstone comprises the following steps:

[0028] Step 1: Analyze the logging data and seismic data to obtain the typical rock physical parameters and contact relationship of low-velocity mudstone, gas-bearing sandstone and background mudstone, and further establish a theoretical geological model of the superposition relationship between low-velocity mudstone and gas-bearing sandstone, such as Figure 1 There are 8 superposition modes: ① gas-bearing sandstone; ② “thick sand-thin mud-thin sand” superposition mode; ③ “upper sand and lower mud” superposition mode; ④ multi-stage sand-mud superposition mode; ⑤ “upper mud and lower sand” superposition mode; ⑥ “thick mud-thin sand-thin mud” superposition mode; ⑦ low-velocity mudstone; ⑧ background mudstone.

[0029] Step 2: Perform convolution forward modeling on the superposition model of low-velocity mudstone and gas-bearing sandstone to analyze whether the forward modeling results can represent the response characteristics of different contact relationships, such as the reservoir "dark spot" reflection phenomenon caused by low-velocity mudstone overlying gas-bearing sandstone and the "false bright spot" reflection phenomenon caused by the contact between low-velocity mudstone and background mudstone. If not, return to step 1 and modify the parameters of the superposition model.

[0030] Step 3: Perform pre-stack forward modeling on the superposition pattern of low-velocity mudstone and gas-bearing sandstone, and then calculate the intercept attribute P and gradient attribute G in the pre-stack attributes, and calculate different combinations of P and G, such as P*G, P+G, etc., to obtain pre-stack combination attributes.

[0031] Step 4: Select an appropriate time window to extract various types of post-stack seismic attributes, such as amplitude, waveform, and frequency seismic attributes, and pre-stack attributes, such as P, G, P*G, P+G, etc. The principal component analysis (PCA) method is used to evaluate the sensitivity of each seismic attribute between different superposition relationships, so as to select sensitive attributes that can distinguish different superposition relationships.

[0032] The steps to implement PCA are as follows:

[0033] 1. Standardize Data

[0034]

[0035] where x ij represents the jth seismic attribute of the ith overlay relation, n is the number of overlay relations, and p is the number of seismic attributes.

[0036]

[0037] where μ j is the mean of the jth earthquake attribute, σ j is its standard deviation.

[0038] 2. Perform principal component analysis on the standardized matrix Z and calculate its covariance matrix C:

[0039]

[0040] Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors. The eigenvector corresponds to the direction of the principal component, while the eigenvalue represents the variance of each principal component. By looking at the principal component load (the component of the eigenvector), it is possible to determine which seismic attributes contribute the most to the principal component, thereby selecting the sensitive attributes.

[0041] Step 5: Input multiple sensitive seismic attributes selected in step 4 into the fuzzy self-organizing neural network at the same time. The network performs self-organizing learning competition according to the characteristics of the seismic attributes to see whether the results can distinguish the different superposition relationships between low-velocity mudstone and gas-bearing sandstone in the model. If they cannot be distinguished, it means that the sensitive attributes selected in step 4 are not suitable and need to be further optimized. If they can be distinguished, it means that the sensitive attributes selected in step 4 are sufficient to distinguish different superposition relationships and can be used for actual data. Figure 2 These are the recognition results before and after adding pre-stack attributes. It can be seen that the results without adding pre-stack attributes cannot effectively distinguish the overlay relationship.

[0042] Step 6: Extract the sensitive attributes selected in steps 4 and 5 along the appropriate time window in the actual seismic data, input them into the fuzzy self-organizing neural network after preprocessing to obtain the clustering results, and combine them with the logging data to explain the possible superposition relationship corresponding to each category, and then determine the distribution and thickness of the regional low-velocity mudstone.

[0043] The fuzzy self-organizing neural network mentioned in steps 5 and 6 is a neural network model based on the combination of self-organizing learning mechanism and fuzzy logic. Its topological structure includes input layer, fuzzy layer, membership layer, and output layer. Figure 3 .

[0044] Among them, the input layer receives external data; the fuzzy layer fuzzifies the data of the input layer and passes it to the membership layer; each node of the membership layer represents the output of a membership function, that is, the membership of the input data to a fuzzy set (the degree to which each input data belongs to each fuzzy set), and the network measures the classification characteristics of the input attributes through the membership layer; the nodes of the output layer represent different categories, and after fuzzy reasoning, the network will give the prediction result of the category to which the input data belongs. The steps of fuzzy self-organizing neural network clustering are as follows:

[0045] 1. Normalization of input earthquake attributes. The purpose of normalization is to standardize the input earthquake attributes to ensure that the value range of each attribute is similar, so as to avoid some attributes from having too great an impact on the results. The formula is as follows:

[0046]

[0047] Where A represents the normalized seismic attribute value; A f Represents the original seismic attribute data; A max Represents the maximum value of the original seismic attribute data; A min Represents the minimum value of the original seismic attribute data.

[0048] 2. Initialize the neural network. Use a random number to determine the initial weight ω of the network cluster center. ik (t). Where 0<ω ik(t)<1, i=1,2,…N is the number of input data, k=1,2,…,K represents the number of cluster centers.

[0049] 3. Fuzzy input earthquake attributes. The normalized earthquake attributes are fuzzified by Gaussian function to determine the degree of subordination between the sample points and the cluster center in the earthquake attributes, that is:

[0050]

[0051] Where i = 1, 2, ..., N is the number of input data, j = 1, 2, ..., M is the number of sample data points in each earthquake attribute, k = 1, 2, ..., K is the number of cluster centers, σ is the width of the Gaussian window, and t is the number of iterations.

[0052] 4. Calculate the membership degree between each sample point in the input earthquake attribute and the neurons in the membership layer, that is:

[0053]

[0054] 5. Adjust the weights of the network, that is

[0055]

[0056] Among them, α represents the learning factor index.

[0057] 6. Determine whether the network has reached stability. This is determined by checking the change in the neuron weights. If the weight update is less than a certain threshold, the network is considered stable, that is:

[0058] |ω ik (t+1)-ω ik (t)|<ε

[0059] Where ε is the set threshold. If the weight change of all neurons is less than the threshold, the network is considered to have converged, otherwise it returns to 3 to continue training.

[0060] 7. Output the clustering results. When the network reaches stability, output the clustering results, namely:

[0061]

[0062] The present invention designs a theoretical model of different superposition relationships between gas-bearing sandstone and low-velocity mudstone based on logging and seismic data, extracts attributes and optimizes the post-stack and pre-stack forward responses, identifies different superposition relationships through fuzzy self-organizing neural network clustering, and further identifies the distribution and thickness of the low-velocity mudstone.

Claims

1. A geophysical identification method for medium-deep undercompacted low-velocity mudstone, characterized in that: By statistically analyzing the data, the representative P-wave velocity, S-wave velocity and density of low-velocity mudstone and gas-bearing sandstone in the logging data are obtained. Combined with geological information, a theoretical model of the superposition relationship between low-velocity mudstone and gas-bearing sandstone is designed. Convolution forward modeling and pre-stack forward modeling are performed to obtain the post-stack and pre-stack seismic attributes of the model, respectively. Sensitive attributes of these attributes are then optimized and the obtained sensitive attributes are used as learning samples to identify the superposition relationship using probabilistic self-organizing neural network clustering.

2. The geophysical identification method of medium-deep undercompacted low-velocity mudstone according to claim 1 is characterized in that: The following steps are involved: Step 1: Analyze the logging data and seismic data to obtain the typical rock physical parameters and contact relationship of low-velocity mudstone, gas-bearing sandstone and background mudstone, and establish a theoretical geological model of the superposition relationship between low-velocity mudstone and gas-bearing sandstone; Step 2: Perform convolution forward modeling on the superposition model of low-velocity mudstone and gas-bearing sandstone to analyze whether the forward modeling results can represent the response characteristics of different contact relationships, such as the reservoir "dark spot" reflection phenomenon caused by low-velocity mudstone overlying gas-bearing sandstone and the "false bright spot" reflection phenomenon caused by the contact between low-velocity mudstone and background mudstone. If not, return to step 1 and modify the parameters of the superposition model; Step 3, pre-stack forward modeling is performed on the stacking pattern of the low-velocity mudstone and the gas-bearing sandstone, and then the intercept attribute P and the gradient attribute G in the pre-stack attributes are calculated, and different combinations of P and G are calculated to obtain the pre-stack combination attributes; Step 4: Select a suitable time window to extract various types of seismic attributes, and use the principal component analysis (PCA) method to evaluate the sensitivity of each seismic attribute between different superposition relationships, so as to select the sensitive attributes that can distinguish different superposition relationships; Step 5: Input multiple sensitive seismic attributes selected in step 4 into the fuzzy self-organizing neural network at the same time. The network performs self-organizing learning competition according to the characteristics of the seismic attributes to see whether the result can distinguish the different superposition relationships between the low-velocity mudstone and the gas-bearing sandstone in the model. If it cannot be distinguished, it means that the sensitive attributes selected in step 4 are not suitable and need to be further optimized; if it can be distinguished, it means that the sensitive attributes selected in step 4 are sufficient to distinguish different superposition relationships and can be used for actual data; Step 6: Extract the sensitive attributes selected in steps 4 and 5 along the appropriate time window in the actual seismic data, input them into the fuzzy self-organizing neural network after preprocessing to obtain the clustering results, and combine them with the logging data to explain the possible superposition relationship corresponding to each category, and then determine the distribution and thickness of the regional low-velocity mudstone.

3. The geophysical identification method of medium-deep undercompacted low-velocity mudstone according to claim 2 is characterized in that: The steps to implement PCA are as follows: (1) Standardized data where x ij represents the jth seismic attribute of the ith overlay relation, n is the number of overlay relations, and p is the number of seismic attributes; where μ j is the mean of the jth earthquake attribute, σ j is its standard deviation; (2) Perform principal component analysis on the standardized matrix Z and calculate its covariance matrix C: Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors; the eigenvectors correspond to the directions of the principal components, while the eigenvalues ​​represent the variance of each principal component; by looking at the principal component loads, i.e., the components of the eigenvectors, determine which seismic attributes contribute the most to the principal components, and thus select the sensitive attributes.

Images