Tertiary oil recovery layer series combination screening method based on improved LDA method and related device
Through the improved LDA method, the scientificity and accuracy of the three oil layer screening methods were solved, efficient and accurate layer system combination screening was achieved, and the efficiency and resource utilization of the three oil layer production were improved.
Patent Information
- Application Number
- CN202510639454.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-16
- Publication Date
- 2025-08-26
AI Technical Summary
In the prior art, the combined screening method of the three-class oil production layer system of the three-class oil layer lacks scientificity and accuracy, resulting in waste of resources and increased mining costs, making it difficult to meet the needs of refined management of oil field development.
Using the improved LDA method, by obtaining the feature data set, preprocessing and calculating the global mean, obtaining the intra-class and inter-class dispersion matrix, calculating the optimal projection direction, combining feature weights and comprehensive evaluation coefficients, a combination of three oil production layers suitable for the three types of oil layers was selected.
The classification accuracy and scientificity of the three-time oil recovery layer combinations have been improved, the calculation complexity has been reduced, the accuracy and efficiency of the screening results have been ensured, and resource waste and mining costs have been reduced.
Smart Images
Figure CN120541422A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of oil and gas field development engineering and tertiary oil recovery technology, and in particular to a tertiary oil recovery stratum combination screening method based on an improved LDA method and related devices. Background Art
[0002] With the continued development of global oil and gas resources and the growing demand for energy, the development of low-permeability reservoirs faces increasingly severe technical challenges. Tertiary oil recovery (TER), a key means of improving oil recovery, plays an irreplaceable role in continuing to tap remaining oil resources after waterflooding. It is of great significance for improving the economic benefits of oil fields and extending production cycles. In the actual process of oilfield development, tertiary oil recovery (TER) operations are usually carried out in the main oil layers after the completion of waterflooding. However, as the development process progresses, waterflooding development in non-main oil layers (Type III oil layers) is gradually terminated, and there is an urgent need to introduce TER to improve recovery.
[0003] Class III oil layers are characterized by their large number, poor reservoir development, and complex development conditions, significantly different from the main oil layers. Currently, research on the boundaries of tertiary oil recovery (TER) formation combinations for Class III oil layers is severely insufficient. A comprehensive and scientific screening and evaluation method has yet to be developed, and systematic quantitative analysis is lacking. Existing technologies primarily rely on expert experience and simple parameter-based screening methods. These subjective judgments are often limited by experience, and standards vary between experts, leading to biased screening results and failing to meet the demands of refined oilfield development management. Furthermore, simple parameter-based screening methods struggle to comprehensively assess formation suitability, which in turn impacts oil recovery effectiveness. These issues lead to resource waste in actual development. Some formation combinations that haven't been scientifically screened may be prematurely utilized for oil recovery, resulting in low tertiary oil recovery efficiency and underutilization of remaining oil resources. Furthermore, formation combinations unsuitable for TER are incorporated into development, increasing production costs and reducing reservoir development efficiency. Therefore, accurately and efficiently identifying formation combinations suitable for TER in Class III oil layers remains a major challenge facing current technologies. Summary of the Invention
[0004] The purpose of this application is to provide a tertiary oil recovery layer combination screening method and related devices based on the improved LDA method, which can screen and obtain layer combinations suitable for tertiary oil recovery of three types of oil layers.
[0005] To achieve the above objectives, this application provides the following solutions:
[0006] In a first aspect, the present application provides a tertiary oil recovery stratum combination screening method based on an improved LDA method, comprising:
[0007] Acquire multiple feature data sets and pre-process the original values of evaluation parameters in all feature data sets to obtain an evaluation matrix; the feature data sets are data sets related to tertiary oil recovery stratum combinations; the feature data sets include multiple evaluation parameters and a cumulative oil production; the evaluation matrix includes a multidimensional feature vector for each tertiary oil recovery stratum combination;
[0008] The evaluation matrix is calculated using a global mean method to obtain an overall mean vector, and based on the evaluation matrix and the overall mean vector, an intra-class scatter matrix and an inter-class scatter matrix are calculated;
[0009] Based on the intra-class scatter matrix and the inter-class scatter matrix, a plurality of generalized eigenvalues and an eigenvector corresponding to each of the generalized eigenvalues are calculated, and based on the plurality of generalized eigenvalues and the eigenvector corresponding to each of the generalized eigenvalues, an optimal projection direction is obtained; the optimal projection direction is the eigenvector corresponding to the maximum generalized eigenvalue;
[0010] Based on the evaluation matrix and the optimal projection direction, a comprehensive projection score corresponding to each tertiary oil recovery layer combination is calculated;
[0011] Based on all feature data sets, the correlation between each evaluation parameter and cumulative oil production is calculated to obtain the feature weight corresponding to each evaluation parameter;
[0012] Based on the comprehensive projection score, evaluation matrix and characteristic weight, the comprehensive evaluation coefficient corresponding to each tertiary oil recovery layer combination is calculated, and based on the comprehensive evaluation coefficient, the layer combination suitable for tertiary oil recovery of the three types of oil layers is screened.
[0013] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned tertiary oil recovery stratum combination screening method based on the improved LDA method.
[0014] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned tertiary oil recovery stratum combination screening method based on the improved LDA method.
[0015] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned tertiary oil recovery stratum combination screening method based on the improved LDA method.
[0016] According to the specific embodiments provided in this application, this application discloses the following technical effects:
[0017] The present application provides a tertiary oil recovery layer combination screening method and related devices based on the improved LDA method, which integrates multiple evaluation parameters to screen layer combinations suitable for tertiary oil recovery in three types of oil layers, avoiding the deviation caused by the traditional method relying on expert experience and screening based on simple parameters. Moreover, the present application improves the traditional LDA method, and overcomes the problem of lack of prior classification labels for the three types of oil layers by calculating the intra-class / inter-class scatter matrix through the global mean, and is suitable for unsupervised learning scenarios. At the same time, the present application can improve the classification accuracy of the layer combination by calculating the eigenvector corresponding to the maximum generalized eigenvalue and using it as the optimal projection direction, and by using the evaluation matrix and the optimal projection direction, calculate the comprehensive projection score corresponding to each tertiary oil recovery layer combination, thereby projecting the high-dimensional evaluation parameters to a one-dimensional comprehensive score, reducing the computational complexity, while retaining key discriminant information. In addition, the present application can improve the scientific nature of tertiary oil recovery layer combination screening by combining weights to calculate the comprehensive evaluation coefficient. In summary, the present application achieves accurate and efficient screening of layer combinations suitable for tertiary oil recovery in three types of oil layers. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0019] Figure 1 This is an application environment diagram of a tertiary oil recovery stratum combination screening method based on an improved LDA method in Example 1 of the present application;
[0020] Figure 2 A schematic flow chart of a tertiary oil recovery stratum combination screening method based on an improved LDA method provided in Example 1 of the present application;
[0021] Figure 3 A schematic flow chart of the tertiary oil recovery stratum combination screening method provided in Example 2 of the present application;
[0022] Figure 4 A schematic diagram of the structure of a computer device provided in Example 3 of the present application;
[0023] Figure 5 Schematic diagram of the configuration environment and system requirements provided for Example 3 of this application. DETAILED DESCRIPTION
[0024] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.
[0025] The applicant's research found that the screening methods based on simple parameters in the existing technology failed to comprehensively consider multiple characteristic parameters and could not effectively integrate various influencing factors of the stratum combination, resulting in difficulty in comprehensively evaluating the adaptability of the stratum combination, which in turn affected the oil production effect. For example, the existing method may only perform preliminary screening based on reserve abundance or permeability breakthrough coefficient, while ignoring the interaction of other key parameters such as thickness and span, resulting in unreasonable selection of stratum combinations. In addition, the subjective judgment of experts is easily limited by experience, and the standards of different experts are different, resulting in deviations in the screening results, which is difficult to meet the needs of refined management of oil field development. In this regard, the present application provides a tertiary oil production stratum combination screening method and related devices based on an improved LDA (Linear Discriminant Analysis) method.
[0026] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0027] Example 1
[0028] The tertiary oil recovery stratum combination screening method based on the improved LDA method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send all feature data sets to the server 104, pre-process the evaluation parameters in all feature data sets, and obtain an evaluation matrix; calculate the evaluation matrix using the global mean method to obtain the overall mean vector, and calculate the intra-class and inter-class scatter matrices based on the evaluation matrix and the overall mean vector; calculate the optimal projection direction based on the intra-class and inter-class scatter matrices; calculate the comprehensive projection score corresponding to each tertiary oil production layer combination based on the evaluation matrix and the optimal projection direction; obtain the feature weight corresponding to each evaluation parameter based on the correlation between each evaluation parameter in the feature data set and the cumulative oil production; calculate the comprehensive evaluation coefficient corresponding to each tertiary oil production layer combination based on the comprehensive projection score, the evaluation matrix and the feature weight, and screen the layer combination suitable for tertiary oil production of the three types of oil layers. The server 104 may feed back the obtained layer combination suitable for tertiary oil recovery of the three types of oil layers to the terminal 102 .
[0029] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, and IoT devices. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or a cloud server.
[0030] In an exemplary embodiment, Figure 2 As shown, a tertiary oil recovery stratum combination screening method based on the improved LDA method is provided. The method is executed by a computer device, specifically, it can be executed by a computer device such as a terminal or a server alone, or it can be executed by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate the process, including the following steps 201 to 206.
[0031] Step 201, obtain multiple feature data sets, and preprocess the original values of evaluation parameters in all feature data sets to obtain an evaluation matrix; the feature data sets are data sets related to tertiary oil recovery layer combinations; the feature data sets include multiple evaluation parameters and a cumulative oil production; the evaluation matrix includes a multidimensional feature vector for each tertiary oil recovery layer combination.
[0032] Step 202 : Calculate the evaluation matrix using the global mean method to obtain the overall mean vector, and calculate the intra-class scatter matrix and the inter-class scatter matrix based on the evaluation matrix and the overall mean vector.
[0033] Step 203: Based on the intra-class scatter matrix and the inter-class scatter matrix, multiple generalized eigenvalues and the eigenvectors corresponding to each generalized eigenvalue are calculated, and based on the multiple generalized eigenvalues and the eigenvectors corresponding to each generalized eigenvalue, an optimal projection direction is obtained; the optimal projection direction is the eigenvector corresponding to the maximum generalized eigenvalue.
[0034] Step 204 : Based on the evaluation matrix and the optimal projection direction, a comprehensive projection score corresponding to each tertiary oil recovery layer combination is calculated.
[0035] Step 205 : Based on all feature data sets, the correlation between each evaluation parameter and the cumulative oil production is calculated to obtain a feature weight corresponding to each evaluation parameter.
[0036] Step 206 , based on the comprehensive projection score, the evaluation matrix and the characteristic weight, the comprehensive evaluation coefficient corresponding to each tertiary oil recovery layer combination is calculated, and based on the comprehensive evaluation coefficient, the layer combination suitable for tertiary oil recovery of the three types of oil layers is screened.
[0037] Implementing the above-mentioned steps 201 to 206, by integrating multiple evaluation parameters to screen the stratum combination suitable for tertiary oil recovery of the three types of oil layers, avoids the deviation caused by the traditional method relying on expert experience and screening based on simple parameters. Moreover, the present application improves the traditional LDA method, and calculates the intra-class / inter-class scatter matrix by global mean, overcomes the problem of the lack of prior classification labels for the three types of oil layers, and makes it suitable for unsupervised learning scenarios. At the same time, the present application can improve the classification accuracy of the stratum combination by calculating the eigenvector corresponding to the maximum generalized eigenvalue and using it as the optimal projection direction, and by using the evaluation matrix and the optimal projection direction, calculates the comprehensive projection score corresponding to each tertiary oil recovery stratum combination, thereby projecting the high-dimensional evaluation parameters to a one-dimensional comprehensive score, reducing the computational complexity, and retaining key discriminant information. In addition, the present application can improve the scientific nature of the tertiary oil recovery stratum combination screening by combining weights to calculate the comprehensive evaluation coefficient. In summary, the present application achieves accurate and efficient screening of stratum combinations suitable for tertiary oil recovery of the three types of oil layers.
[0038] Furthermore, in step 201, the original values of the evaluation parameters in all feature data sets are preprocessed to obtain an evaluation matrix, which specifically includes:
[0039] In step 2011, first, the original values of the evaluation parameters in all feature data sets are detected for outliers using the box plot method to obtain outliers (including abnormally high values or abnormally low values). Specifically, the method is to calculate the first quartile Q1 (i.e., the 25% quantile, indicating that 25% of the data is less than or equal to this value) and the third quartile Q3 (i.e., the 75% quantile, indicating that 75% of the data is less than or equal to this value). If a data point (the original value of the evaluation parameter) exceeds the upper and lower limit ranges, it is considered an outlier. The calculation formulas for the lower and upper limits are:
[0040] Lower limit=Q1-1.5×(Q3-Q1);
[0041] Upper limit=Q3+1.5×(Q3-Q1).
[0042] Secondly, the original values of the evaluation parameters in all feature data sets are tested for missing values to check whether the data is empty and obtain missing values; finally, the experts check whether the outliers and missing values are reasonable to prevent the accidental deletion of valid data, and select appropriate filling methods to replace the outliers and missing values according to the data distribution of the feature data set to obtain a complete first data set; the specific replacement process is: if the overall distribution of the feature data set is relatively symmetrical, use the mean to fill; if the feature data set has a skewed distribution, using the median to fill is more robust; if the feature data set is highly discrete or has serious outliers, model interpolation or sample removal can be considered; this helps to reduce calculation errors and ensure that the LDA method is not interfered with by abnormal data during dimensionality reduction. The sorted data are numbered to ensure that the data is suitable for subsequent LDA analysis.
[0043] In step 2012, since different features (i.e., evaluation parameters) have different dimensions and value ranges, the first data set is normalized using the minimum-maximum normalization method. The original value of each evaluation parameter is compressed to between 0 and 1, eliminating the dimension differences between different features, and obtaining a normalized second data set. The specific process of normalization is as follows:
[0044]
[0045] Among them, x ij is the original value of the i-th tertiary oil recovery layer combination on the j-th evaluation parameter, x′ ij is the value of the i-th tertiary oil recovery layer combination normalized on the j-th evaluation parameter, max(x j ) and min(x j ) are the minimum and maximum values of the j-th evaluation parameter, respectively.
[0046] In step 2013, the normalized multidimensional feature data (i.e., the normalized second data set) are organized into a standardized matrix to obtain an evaluation matrix, which provides basic data for subsequent LDA dimensionality reduction and classification decisions. Assuming that a total of m samples (i.e., m tertiary oil recovery stratum combinations) are collected, each sample has n evaluation parameters (which correspond to n-dimensional feature vectors), the established matrix is as follows:
[0047]
[0048] During data normalization, only the evaluation parameters within the characteristic parameters are processed, not the cumulative oil production within the characteristic parameters. This is because this application uses cumulative oil production as an important reference value to represent the linear relationship between the evaluation parameters and the cumulative oil production, avoiding normalization, preserving the physical meaning of the cumulative oil production, and improving the practicality of the calculation. Furthermore, the cumulative oil production in this application refers to the cumulative oil production of the primary oil recovery (natural energy development self-flowing) and secondary oil recovery (conventional water flooding) stages before the tertiary oil recovery. This oil production provides a very intuitive view of the reservoir's properties (the quality of the oil layer), providing a potential reference for subsequent tertiary oil recovery.
[0049] Furthermore, in step 202, the evaluation matrix is calculated using the global mean method to obtain the overall mean vector. Based on the evaluation matrix and the overall mean vector, the intra-class scatter matrix and the inter-class scatter matrix are calculated. This can quantify the internal scatter characteristics of the data and provide a solid statistical basis for LDA to solve the optimal projection direction, thereby achieving efficient dimensionality reduction and accurate classification of hierarchical combinations. The specific process is:
[0050] In step 2021, since there is no classification in the initial stage, the mean of the global data is used as the benchmark to calculate the dispersion. The overall mean vector μ is:
[0051]
[0052] Among them, X′ i is the n-dimensional eigenvector of the i-th tertiary oil recovery stratum combination in the evaluation matrix, and m is the total number of samples, that is, the total number of tertiary oil recovery stratum combinations.
[0053] In step 2022, in the traditional LDA algorithm, data is pre-classified. This application uses the final comprehensive evaluation coefficient for discrimination and classification. Therefore, based on the traditional LDA algorithm, this application uses the global data mean to calculate the intra-class and inter-class scatter matrices to avoid errors caused by human classification and improve the objectivity of the calculation. The method is as follows:
[0054] According to the formula Calculate the intra-class scatter matrix;
[0055] According to the formula Calculate the inter-class scatter matrix;
[0056] Among them, S W is the intra-class scatter matrix; S B is the between-class scatter matrix; X i ′ is the n-dimensional eigenvector of the i-th tertiary oil recovery layer combination in the evaluation matrix; m is the total number of tertiary oil recovery layer combinations; μ is the overall mean vector.
[0057] Furthermore, in step 203, based on the intra-class scatter matrix and the inter-class scatter matrix, multiple generalized eigenvalues and the eigenvectors corresponding to each of the generalized eigenvalues are calculated, and the eigenvector corresponding to the maximum eigenvalue is selected as the optimal projection direction v * , the method is as follows:
[0058]
[0059] Where v is the candidate projection direction and λ is the generalized eigenvalue.
[0060] Furthermore, in step 204, based on the evaluation matrix and the optimal projection direction, the comprehensive projection score corresponding to each tertiary oil recovery layer combination is calculated. Specifically, based on the optimal projection direction obtained in step 203, the multidimensional feature vector of each tertiary oil recovery layer combination is projected onto the optimal projection direction to calculate the comprehensive projection score corresponding to each tertiary oil recovery layer combination. The calculation formula is:
[0061] y i =v *T X' i ;
[0062] Among them, y i is the comprehensive projection score corresponding to the i-th tertiary oil recovery layer combination; v * is the optimal projection direction; X i ′ is the n-dimensional characteristic vector of the i-th tertiary oil recovery layer combination in the evaluation matrix.
[0063] Furthermore, in step 205, the correlation between each evaluation parameter and the cumulative oil production is calculated based on the characteristic data set to obtain the characteristic weight corresponding to each evaluation parameter. Specifically, the original value x of the i-th tertiary oil production layer combination on the j-th evaluation parameter is calculated. ij and cumulative oil production Y prod , calculate the Pearson correlation coefficient r between the jth evaluation parameter and the cumulative oil production j , the method is as follows:
[0064]
[0065] Where m is the total number of tertiary oil recovery strata combinations; x ij is the original value of the jth evaluation parameter of the i-th tertiary oil recovery layer combination; is the average value of the jth evaluation parameter; is the cumulative oil production of the i-th tertiary oil production layer combination; is the average value of cumulative oil production.
[0066] Then, based on the correlation between the original value of each evaluation parameter and the cumulative oil production, the weight w corresponding to each evaluation parameter is calculated. j , the method is as follows:
[0067]
[0068] Among them, w j is the feature weight corresponding to the jth evaluation parameter; r j is the Pearson correlation coefficient between the jth evaluation parameter and the cumulative oil production.
[0069] In addition, when some features have low relevance to the cumulative oil production, their feature weights w j It may still affect the final classification results, so the weight setting can be dynamically adjusted based on historical data.
[0070] Furthermore, in step 206, based on the comprehensive projection score, the evaluation matrix and the feature weight, the comprehensive evaluation coefficient corresponding to each tertiary oil recovery layer combination is calculated. The specific process is as follows:
[0071] according to Calculate and obtain the comprehensive evaluation coefficient corresponding to each tertiary oil recovery stratum combination;
[0072] Among them, C i is the comprehensive evaluation coefficient corresponding to the i-th tertiary oil recovery layer combination; α is the weighting factor used to adjust the comprehensive projection score. Based on the actual experience of oilfield development and tertiary oil recovery, experts can pre-set a value that suits the study area to ensure that the comprehensive evaluation model has sufficient flexibility and applicability; y i is the comprehensive projection score corresponding to the i-th tertiary oil production layer combination; w j is the feature weight corresponding to the j-th evaluation parameter; is the normalized value of the i-th tertiary oil recovery layer combination on the j-th evaluation parameter in the evaluation matrix; n is the number of evaluation parameters or the dimension of the eigenvector.
[0073] In addition, in step 206, based on the comprehensive evaluation coefficient, a tertiary oil recovery layer combination suitable for tertiary oil recovery of the three types of oil layers is screened and obtained, specifically including:
[0074] Determine the decision value; determine a reasonable decision value T based on historical data, expert experience and actual production statistical data analysis; at the same time, use the minimum error rate analysis to optimize the selection of T to ensure classification accuracy, which is used to distinguish the comprehensive evaluation coefficient C i height.
[0075] Determine whether the comprehensive evaluation coefficient is greater than the decision value;
[0076] If so, the corresponding tertiary oil recovery layer combination is output and used as the layer combination suitable for tertiary oil recovery of the three types of oil layers; if not, it means that the corresponding layer combination is not suitable for tertiary oil recovery of the three types of oil layers.
[0077] In response to the existing deficiencies and gaps in this research field, this application proposes a comprehensive evaluation method for tertiary oil recovery stratum combination based on the improved LDA method. Linear discriminant analysis has good multi-parameter linear discriminant ability, and can construct the optimal projection direction in the high-dimensional feature space to maximize the difference between classes and minimize the difference within classes, thereby effectively revealing the essential discriminant information between multi-dimensional parameters. By utilizing the characteristics of linear discriminant analysis, the high-dimensional reservoir data is projected into the low-dimensional space, and the data after dimensionality reduction is used for classification to determine the stratum combination suitable for tertiary oil recovery; and the accuracy of the classification results is verified by actual data, and the model parameters are adjusted to optimize the effect. At the same time, considering that the reagents for tertiary oil recovery (chemical flooding) are expensive and costly, some poorly developed oil layers cannot be used for tertiary oil recovery at all, so tertiary oil recovery tests are generally not carried out, and sample data with classification results cannot be obtained. Therefore, this application introduces the total mean vector, weight factor and projection score fusion mechanism. This method can be combined with actual production data for objective, data-driven stratum optimization without relying on prior classification. Practical results show that this method exhibits higher classification accuracy and promotion adaptability under the complex conditions of three types of oil layers, effectively improving the scientific nature of tertiary oil recovery layer screening and development decision-making efficiency.
[0078] The innovation of this application is that it breaks through the limitation of traditional LDA relying on category information, applies it to the field of reservoir engineering, and successfully applies it to unsupervised learning scenarios. On the one hand, LDA dimensionality reduction can effectively retain the discriminant information of the data, ensuring that the classification of the adaptability of the tertiary oil recovery stratum combination is more accurate; on the other hand, LDA is used to calculate the projection score and combine the weight to calculate the comprehensive evaluation coefficient to improve the scientific nature of the tertiary oil recovery stratum combination screening, especially for the reorganization problem of the three types of oil layers, effectively process high-dimensional data, and improve classification accuracy and interpretability. It has been proven in practice that this method is superior to traditional methods in both classification accuracy and efficiency, and can significantly improve the evaluation effect of the tertiary oil recovery stratum combination, reduce development costs, and increase recovery rates.
[0079] The present application also provides an application scenario, which applies the above-mentioned tertiary oil recovery layer combination screening method based on the improved LDA method. Specifically: the tertiary oil recovery layer combination screening method provided in this embodiment can be applied in the tertiary oil recovery layer combination screening scenario. The tertiary oil recovery layer combination screening scenario includes a layer combination screening link and a screening result display link; the layer combination screening link is used to screen layer combinations suitable for tertiary oil recovery of three types of oil layers by integrating multiple evaluation parameters; the screening result display link is used to display the layer combinations suitable for tertiary oil recovery of three types of oil layers obtained by screening; the method of this embodiment belongs to the layer combination screening link and the screening result display link.
[0080] Example 2
[0081] This embodiment takes the actual data of a domestic oil field as an example (see Table 1) and analyzes in detail the tertiary oil recovery stratum combination screening method of this application. The main process is as follows: Figure 3 As shown, the following steps are included in sequence:
[0082] Step A: Based on the characteristic parameters of the tertiary oil recovery stratum combination provided in Table 1 (including evaluation parameters and cumulative oil production), when preliminary statistics of the data were performed and box plots were drawn, it was found that all samples had no missing values and the data distribution was within a reasonable range, with no obvious outliers. There was no need to replace or eliminate outliers in the data.
[0083] The main characteristic parameters are recorded as follows: effective thickness Tcff, converted thickness Tadj, span Sspan, original reserves Rorig, remaining reserves Rres, remaining reserves abundance Rres_den, average permeability Kavg, permeability surge coefficient Krush, permeability variation coefficient Kvar and cumulative oil production Yprod.
[0084] At the same time, according to the formula in step 2012, the normalized value in the example is calculated, and the normalized evaluation matrix is established:
[0085]
[0086] Taking the effective thickness Tcff as an example, its x′ 11 The value is 0.5679, x′ 21 The value is 0.6020, x′ 31 The value is 0.1980.
[0087] Table 1 Characteristic parameters of the tertiary oil recovery stratum combination in an oil field
[0088]
[0089]
[0090] Note: In order to ensure the effectiveness of the examples and protect the confidentiality of the real data, the data listed in this application are not accompanied by units, but this does not affect the applicability of the method and the reliability of the conclusions.
[0091] Step B: Based on the normalized evaluation matrix, the formula in step 202 of the invention is used to obtain the overall mean vector, the intra-class scatter matrix, and the inter-class spread matrix values as follows:
[0092] μ=[0.4135 0.4746 0.4800 0.2992 0.2992 0.3954 0.4832 0.5971 0.6615] T ;
[0093]
[0094] Step C: Solve the generalized eigenvalues based on the intra-class scatter matrix and inter-class scatter matrix values obtained in step B. The results are as follows:
[0095] λ=[0.1111 0.0851 0.0749 0.0613 0.0487 0.0342 0.0201 0.0108 0.0072] T ;
[0096] The optimal projection direction is obtained based on the maximum generalized eigenvalue λ = 0.1111, and the results are as follows:
[0097] v * =[0.7527 -0.4546 -0.1757 -0.0185 0.2291 -0.1078 0.1464 -0.23020.3469] T ;
[0098] Step D: Based on the normalized evaluation matrix and optimal projection direction obtained in steps A and C, the comprehensive projection score corresponding to each tertiary oil recovery layer combination is calculated as follows:
[0099] y i =[0.2882 0.2476 0.0122 0.4150 -0.0322 0.0985 0.3573 -0.08930.0279] T ;
[0100] Taking effective thickness as an example, the Pearson correlation coefficient is calculated. According to the mean of effective thickness: 4.63 and the mean of cumulative oil production: 265.48, the covariance numerator is:
[0101]
[0102] The product of standard deviations is:
[0103]
[0104] The effective thickness Pearson correlation coefficient is obtained:
[0105]
[0106] Taking this as an example, the calculation results are as follows:
[0107] r j =[0.8541 0.7895 0.6123 0.9328 0.9147 0.6782 0.8435 0.7152 0.6034] T ;
[0108] Further get the feature weight w corresponding to each evaluation parameter j :
[0109] w j =[0.1205 0.1114 0.0865 0.1655 0.1622 0.1203 0.1123 0.0974 0.0810] T ;
[0110] Finally, the comprehensive evaluation coefficient corresponding to each tertiary oil recovery stratum combination is calculated:
[0111] C i =[0.4721 0.4387 0.3215 0.5762 0.2984 0.3556 0.5293 0.2714 0.3128] T ;
[0112] In this embodiment, the threshold value T set by the experts based on actual field production data is 0.4124. It can be concluded through judgment that the stratum combinations of Schemes 1, 2, 4, and 7 are suitable for tertiary oil recovery.
[0113] Example 3
[0114] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store processed data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it realizes a comprehensive evaluation method for tertiary oil recovery strata based on the improved LDA method. Since the improved linear discriminant analysis adopted has a large amount of calculation, the configured environment and system requirements are as follows. Figure 5 shown.
[0115] Those skilled in the art will understand that Figure 4 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.
[0116] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.
[0117] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0118] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0119] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.
[0120] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0121] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A tertiary oil recovery stratum combination screening method based on an improved LDA method, characterized in that: include: Acquire multiple feature data sets, and preprocess the original values of the evaluation parameters in all feature data sets to obtain an evaluation matrix; The characteristic data set is a data set related to the tertiary oil recovery stratum combination; The characteristic data set includes multiple evaluation parameters and a cumulative oil production; the evaluation matrix includes a multidimensional characteristic vector of each tertiary oil recovery layer combination; The evaluation matrix is calculated using a global mean method to obtain an overall mean vector, and based on the evaluation matrix and the overall mean vector, an intra-class scatter matrix and an inter-class scatter matrix are calculated; Based on the intra-class scatter matrix and the inter-class scatter matrix, a plurality of generalized eigenvalues and an eigenvector corresponding to each of the generalized eigenvalues are calculated, and based on the plurality of generalized eigenvalues and the eigenvector corresponding to each of the generalized eigenvalues, an optimal projection direction is obtained; the optimal projection direction is the eigenvector corresponding to the maximum generalized eigenvalue; Based on the evaluation matrix and the optimal projection direction, a comprehensive projection score corresponding to each tertiary oil recovery stratum combination is calculated; Based on all feature data sets, the correlation between each evaluation parameter and cumulative oil production is calculated to obtain the feature weight corresponding to each evaluation parameter; Based on the comprehensive projection score, evaluation matrix and characteristic weight, the comprehensive evaluation coefficient corresponding to each tertiary oil recovery layer combination is calculated, and based on the comprehensive evaluation coefficient, the layer combination suitable for tertiary oil recovery of the three types of oil layers is screened.
2. The tertiary oil recovery stratum combination screening method based on the improved LDA method according to claim 1, characterized in that: The preprocessing includes outlier detection, missing value detection and normalization; The original values of the evaluation parameters in all feature data sets are preprocessed to obtain the evaluation matrix, which specifically includes: Performing outlier detection and missing value detection on the original values of the evaluation parameters in all feature data sets to obtain outliers and missing values, and replacing the outliers and missing values using a mean filling method or a median filling method to obtain a first data set; The first data set is normalized using a minimum-maximum normalization method to obtain a second data set, and the second data set is constructed into a standardized matrix to obtain an evaluation matrix.
3. The tertiary oil recovery stratum combination screening method based on the improved LDA method according to claim 2, characterized in that: The evaluation parameters include effective thickness, converted thickness, span, original reserves, remaining reserves, remaining reserve abundance, average permeability, permeability breakthrough coefficient and permeability variation coefficient.
4. The tertiary oil recovery stratum combination screening method based on the improved LDA method according to claim 1, characterized in that: Based on the comprehensive evaluation coefficient, the formation combinations suitable for tertiary oil recovery of three types of oil layers are screened, specifically including: Determine a decision value; the decision value is obtained by analyzing historical data, expert experience, actual production statistics, and a minimum error rate; Determining whether the comprehensive evaluation coefficient is greater than the decision value; If so, the corresponding tertiary oil recovery layer combination is output and used as the layer combination suitable for tertiary oil recovery of the three types of oil layers.
5. The tertiary oil recovery stratum combination screening method based on the improved LDA method according to claim 1, characterized in that: The calculation processes of the intra-class scatter matrix and the inter-class scatter matrix are respectively: According to the formula Calculate the intra-class scatter matrix; According to the formula Calculate the inter-class scatter matrix; Among them, S W is the intra-class scatter matrix; S B is the between-class scatter matrix; X i ′ is the n-dimensional eigenvector of the i-th tertiary oil recovery layer combination in the evaluation matrix; m is the total number of tertiary oil recovery layer combinations; μ is the overall mean vector.
6. The tertiary oil recovery stratum combination screening method based on the improved LDA method according to claim 1, characterized in that: The calculation process of the comprehensive projection score is: According to the formula y i =v *T X' i The comprehensive projection score corresponding to each tertiary oil recovery layer combination is calculated; Among them, y i is the comprehensive projection score corresponding to the i-th tertiary oil recovery layer combination; v * is the optimal projection direction; X i ′ is the n-dimensional characteristic vector of the i-th tertiary oil recovery layer combination in the evaluation matrix.
7. The tertiary oil recovery stratum combination screening method based on the improved LDA method according to claim 1, characterized in that: The calculation process of the comprehensive evaluation coefficient is: according to Calculate and obtain the comprehensive evaluation coefficient corresponding to each tertiary oil recovery stratum combination; Among them, C i is the comprehensive evaluation coefficient corresponding to the i-th tertiary oil production layer combination; α is the weighting factor used to adjust the comprehensive projection score; y i is the comprehensive projection score corresponding to the i-th tertiary oil production layer combination; w j is the feature weight corresponding to the jth evaluation parameter; x′ ij is the normalized value of the i-th tertiary oil recovery layer combination on the j-th evaluation parameter in the evaluation matrix; n is the number of evaluation parameters.
8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the tertiary oil recovery stratum combination screening method based on the improved LDA method as described in any one of claims 1 to 7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the tertiary oil recovery stratum combination screening method based on the improved LDA method according to any one of claims 1 to 7 is realized.
10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the tertiary oil recovery stratum combination screening method based on the improved LDA method according to any one of claims 1 to 7 is realized.