A method and apparatus for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference.

Abstract

This manual relates to a method and apparatus for analyzing the main controlling factors of hydraulic fracturing effects based on causal inference. Using causal inference to study the factors influencing the production increase of oil and gas wells after hydraulic fracturing, and clarifying the causal effect value of each influencing factor on production, helps personnel better understand the factors affecting production, thereby guiding field personnel in optimizing the design of hydraulic fracturing parameters.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of oil and gas extraction technology, and in particular to a method and apparatus for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference. Background Technology

[0002] For the development of tight oil and gas reservoirs, there are many factors that affect the effectiveness of hydraulic fracturing, but at present it is impossible to accurately determine the main controlling factors affecting the effectiveness of hydraulic fracturing, thus making it impossible to improve the production capacity of hydraulic fracturing. Summary of the Invention

[0003] To address the problems existing in the prior art, this specification provides a method and apparatus for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference. The method uses causal inference to study the factors affecting the increase in oil and gas well production after hydraulic fracturing, and clarifies the causal effect value of each influencing factor on production. This helps staff to better understand the factors affecting production, and thus guides on-site staff to optimize the design of hydraulic fracturing parameters.

[0004] The specific technical solutions of the embodiments in this specification are as follows:

[0005] On the one hand, the embodiments of this specification provide a method for analyzing the main controlling factors of hydraulic fracturing effects based on causal inference, the method comprising:

[0006] Obtain numerical values ​​for multiple types of background variables, multiple types of post-intervention variables, and production data from multiple exploited wells;

[0007] Using a type of post-intervention variable as the intervention variable to be analyzed, the values ​​of the intervention variable to be analyzed for the multiple wells to be clustered to obtain multiple categories, and the multiple wells to be analyzed are divided into intervention group and control group according to the multiple categories;

[0008] For each exploited well in the intervention group and the control group, the endogeneity control coefficients of the background variables and post-intervention variables of the exploited well are calculated based on the values ​​of the background variables of the multiple types of the exploited well, the values ​​of the post-intervention variables of the other types besides the intervention variable to be analyzed, and the values ​​of the production data.

[0009] For each exploited well in the intervention group and the control group, the propensity score of the intervention variable to be analyzed for the exploited well is calculated based on the values ​​of the background variables of multiple types for the exploited well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficient of each variable.

[0010] The weighting coefficient of the intervention variable to be analyzed is calculated for each well in the intervention group and the control group based on the propensity score of the intervention variable to be analyzed.

[0011] Based on the propensity score, the exploited wells in the intervention group were matched with the exploited wells in the control group to obtain multiple matching groups;

[0012] The intervention effect of the intervention variable to be analyzed on the production data is calculated based on the production data values ​​of the two wells in each of the matching groups, the weighting coefficient, and the total number of the matching groups, so that the staff can optimize the design of construction parameters based on the intervention effect.

[0013] Furthermore, dividing the multiple exploited wells into intervention and control groups based on the multiple categories further includes:

[0014] The categories belonging to the intervention group and the control group are determined based on the mean values ​​of the intervention variables to be analyzed in each category.

[0015] Furthermore, calculating the endogeneity control coefficients of the various types of background variables and various types of post-intervention variables for the exploited well based on the values ​​of the background variables of the various types, the values ​​of the post-intervention variables other than the intervention variable to be analyzed, and the values ​​of the production data further includes:

[0016] At least one instrumental variable shall be determined from the multiple types of background variables and other types of post-intervention variables besides the intervention variable to be analyzed;

[0017] Using the least squares method to apply the formula The solution is performed to obtain the endogeneity control coefficients for the background variables and post-intervention variables of multiple types, where Y represents the numerical value of the production data, X... i Let βi represent the background variable of the i-th type or the post-intervention variable other than the intervention variable to be analyzed, β0, ..., βi. i ..., β D These are the endogeneity control coefficients, where β is... i Here, η is the endogeneity control coefficient for the background variable or post-intervention variable of the i-th type, where η is a constant term. in Z represents the alternative value for the intervention variable to be analyzed. i Let αi represent the numerical values ​​of the instrumental variable of type i, α0, ..., αi. i Let ε be the coefficient and ε be the constant term; the loss function of the least squares method is: Where Loss is the loss value, and n represents the total number of background variables or post-intervention variables other than the intervention variable to be analyzed.

[0018] Furthermore, the formula for calculating the propensity score of the intervention variable to be analyzed for the well-existing production well, based on the values ​​of the background variables of multiple types, the values ​​of post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficients of each variable, is as follows:

[0019]

[0020] Where e(D) represents the propensity score of the intervention variable D to be analyzed, X1, X2, ..., X... n These represent the numerical values ​​of the background variables of multiple types, and the numerical values ​​of post-intervention variables of other types besides the intervention variable to be analyzed, β1, β2, ..., β... n These are the endogeneity control coefficients for the corresponding background variables or other types of post-intervention variables besides the intervention variables to be analyzed.

[0021] Furthermore, calculating the weighting coefficient of the intervention variable to be analyzed for each well in the intervention group and the control group, based on the propensity score of the intervention variable to be analyzed for each well in the control group, further includes:

[0022] If the wells already in operation belong to the intervention group, then according to the formula... Calculate the weighting coefficients, where w i e(D) represents the weighting coefficient of the intervention variable D to be analyzed in the i-th well of the intervention group. i ) represents the propensity score of the intervention variable D to be analyzed in the i-th well in the intervention group;

[0023] If the wells already in operation belong to the control group, then according to the formula... Calculate the weighting coefficients, where w j e(D) represents the weight coefficient of the intervention variable D to be analyzed in the j-th well of the control group. j ) represents the propensity score of the intervention variable D to be analyzed in the j-th well of the control group.

[0024] Furthermore, based on the propensity score, the exploited wells in the intervention group are matched with the exploited wells in the control group to obtain multiple matching groups, which further include:

[0025] Calculate the difference between the tendency score of each exploited well in the intervention group and the tendency score of each exploited well in the control group, and identify exploited wells in the control group whose difference is less than a threshold, and combine the exploited well in the intervention group and the exploited well in the control group into a matching group;

[0026] If a well in the intervention group has no matching well in the control group, then that well in the intervention group is discarded.

[0027] Furthermore, if the values ​​of the intervention variables to be analyzed from the multiple exploited wells are clustered to obtain two or more categories, dividing the multiple exploited wells into intervention and control groups based on the multiple categories further includes:

[0028] The intervention group and the control group are determined according to the mean values ​​of the intervention variables to be analyzed in each category.

[0029] Based on the propensity score, the exploited wells in the intervention group were matched with the exploited wells in the control group to obtain multiple matching groups, which further included:

[0030] Calculate the difference between the tendency score of each exploited well in each intervention group and the tendency score of each exploited well in the control group, and identify exploited wells in the control group whose difference is less than a threshold, and pair the exploited well in the intervention group and the exploited well in the control group as a matching group;

[0031] If a well in any intervention group has no matching well in the control group, then that well in the intervention group is discarded.

[0032] Furthermore, the formula for calculating the intervention effect of the intervention variable to be analyzed on the production data based on the production data values ​​of the two exploited wells in each of the matching groups, the weighting coefficients, and the total number of the matching groups is as follows:

[0033]

[0034] Where ATE represents the intervention effect of the intervention variable to be analyzed on the production data, k represents the total number of matched groups, T represents the intervention group set, C represents the control group set, and Y represents the control group set. i Y represents the production data of the i-th well belonging to intervention group T. j This represents the production data of the j-th well in the control group C, which belongs to the same matching group as the i-th well in the intervention group T.

[0035] Furthermore, after obtaining numerical values ​​for multiple types of background variables, multiple types of post-intervention variables, and production data from multiple exploited wells, the method further includes:

[0036] The multiple exploited wells are grouped according to the values ​​of the background variables to obtain multiple groups. This allows for the execution of the step of using a type of post-intervention variable as the intervention variable to be analyzed for multiple exploited wells in each group, ultimately obtaining the intervention effect of the intervention variable to be analyzed on the production data in each group.

[0037] On the other hand, embodiments of this specification also provide a device for analyzing the main controlling factors of hydraulic fracturing effects based on causal inference, the device comprising:

[0038] The data acquisition unit is used to acquire numerical values ​​of multiple types of background variables, multiple types of post-intervention variables, and production data from multiple exploited wells.

[0039] Clustering unit is used to cluster the values ​​of the intervention variables to be analyzed for the multiple wells that have been exploited, taking a type of post-intervention variable as the intervention variable to be analyzed, to obtain multiple categories, and to divide the multiple wells into intervention group and control group according to the multiple categories;

[0040] The endogeneity control coefficient calculation unit is used to calculate the endogeneity control coefficients of each type of background variable and each type of post-intervention variable for each well in the intervention group and the control group, based on the values ​​of the background variables of multiple types of the well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the values ​​of the production data.

[0041] The propensity score calculation unit is used to calculate the propensity score of the intervention variable to be analyzed for each well in the intervention group and the control group based on the values ​​of the background variables of multiple types of the well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficient of each variable.

[0042] The weighting coefficient calculation unit is used to calculate the weighting coefficient of the intervention variable to be analyzed for each well in the intervention group and the control group based on the tendency score of the intervention variable to be analyzed for each well in the intervention group and the control group.

[0043] A matching unit is used to match the exploited wells in the intervention group with the exploited wells in the control group based on the propensity score, thereby obtaining multiple matching groups;

[0044] The intervention effect calculation unit is used to calculate the intervention effect of the intervention variable to be analyzed on the production data based on the production data values ​​of the two wells in each of the matching groups, the weighting coefficients, and the total number of the matching groups, so that the staff can optimize the design of construction parameters based on the intervention effect.

[0045] The embodiments in this specification pre-define the types of background variables and post-intervention variables, then obtain the numerical values ​​of multiple types of background variables, post-intervention variables, and production data from multiple exploited wells. Then, a causal inference method is used to form a multi-level and comprehensive causal inference framework, which not only effectively solves the problems of confounding factors, endogeneity, and heterogeneous intervention, but also provides accurate causal effect analysis and optimization guidance, and has broad application prospects. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 The figure shown is a schematic diagram of the implementation system of a method for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference in an embodiment of this specification;

[0048] Figure 2 The diagram shown is a flowchart illustrating a method for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference in an embodiment of this specification.

[0049] Figure 3 The diagram shown is a flowchart illustrating how the multiple exploited wells are divided into intervention and control groups according to the multiple categories in an embodiment of this specification.

[0050] Figure 4 The diagram shown is a flowchart illustrating how, in another embodiment of this specification, the plurality of exploited wells are divided into intervention and control groups according to the plurality of categories.

[0051] Figure 5 The diagram shown is a schematic representation of a device for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference, as described in an embodiment of this specification.

[0052] Figure 6 The diagram shown is a structural schematic of the computer device in an embodiment of this specification.

[0053] [Explanation of Figure Markers]:

[0054] 101. Terminal;

[0055] 102. Server;

[0056] 501. Data Acquisition Unit;

[0057] 502. Clustering unit;

[0058] 503. Endogeneity control coefficient calculation unit;

[0059] 504. Propensity Score Calculation Unit;

[0060] 505. Weighting coefficient calculation unit;

[0061] 506. Matching unit;

[0062] 507: Intervention effect calculation unit;

[0063] 602. Computer equipment;

[0064] 604, Processor;

[0065] 606. Memory;

[0066] 608. Drive mechanism;

[0067] 610. Input / output module;

[0068] 612. Input devices;

[0069] 614. Output devices;

[0070] 616. Presentation equipment;

[0071] 618. Graphical User Interface;

[0072] 620. Network interface;

[0073] 622. Communication link;

[0074] 624. Communication bus. Detailed Implementation

[0075] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the embodiments of this specification.

[0076] It should be noted that the terms "first," "second," etc., in the description, claims, and accompanying drawings of the embodiments herein are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0077] It should be noted that the acquisition, storage, use, and processing of data in the technical solutions of the embodiments of this specification all comply with the relevant provisions of national laws and regulations.

[0078] It should be noted that in the embodiments of this specification, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, they do not mean that the applicant has used or necessarily used the solution.

[0079] like Figure 1 The diagram shown is a schematic representation of an implementation system for a method for analyzing the main controlling factors of hydraulic fracturing effects based on causal inference, as described in this specification. The system includes a terminal 101 and a server 102. The terminal 101 and server 102 can communicate via a network, which may include a local area network (LAN), a wide area network (WAN), the Internet, or a combination thereof, and is connected to a website, user equipment (e.g., computing devices), and a backend system.

[0080] Staff input the values ​​of multiple types of background variables, multiple types of post-intervention variables, and production data from multiple operational wells into server 102 via terminal 101. Server 102 analyzes and calculates the values ​​of these variables using causal inference to obtain the intervention effect of the specified intervention variable on the production data. Server 102 can then provide the staff with the intervention effect of the intervention variable on the production data via terminal 101, enabling them to optimize construction parameters based on the intervention effect.

[0081] Alternatively, server 102 may be a node of a cloud computing system (not shown in the figure), or each server may be a separate cloud computing system comprising multiple computers interconnected by a network and operating as a distributed processing system.

[0082] In addition, it should be noted that, Figure 1 The examples shown are merely one application environment provided by the embodiments in this specification. In practical applications, other application environments may also be included, and this specification does not impose any limitations.

[0083] To address the problems existing in the prior art, this specification provides a method for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference. Figure 2 The diagram shows a flowchart illustrating the method for analyzing the controlling factors of hydraulic fracturing effects based on causal inference in an embodiment of this specification. The diagram depicts the process of analyzing the intervention effect of the intervention variable on the production data. The order of steps listed in the embodiment is merely one possible execution order among many and does not represent the only possible order. In actual system or device products, the methods shown in the embodiment or the accompanying drawings can be executed sequentially or in parallel.

[0084] Specific examples Figure 2 As shown, the method may include:

[0085] Step 201: Obtain numerical values ​​for multiple types of background variables, multiple types of post-intervention variables, and production data for multiple wells that have been exploited;

[0086] Step 202: Using a type of post-intervention variable as the intervention variable to be analyzed, cluster the values ​​of the intervention variable to be analyzed for the multiple exploited wells to obtain multiple categories, and divide the multiple exploited wells into intervention group and control group according to the multiple categories;

[0087] Step 203: For each well in the intervention group and the control group that has been exploited, calculate the endogeneity control coefficients of the background variables and post-intervention variables of the various types for the well based on the values ​​of the background variables of the various types for the well, the values ​​of the post-intervention variables of the various types other than the intervention variable to be analyzed, and the values ​​of the production data.

[0088] Step 204: For each well in the intervention group and the control group, calculate the propensity score of the intervention variable to be analyzed for the well based on the values ​​of the background variables of multiple types of the well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficient of each variable.

[0089] Step 205: Calculate the weighting coefficient of the intervention variable to be analyzed for each well in the intervention group and the control group based on the propensity score of the intervention variable to be analyzed for each well in the intervention group and the control group;

[0090] Step 206: Match the exploited wells in the intervention group with the exploited wells in the control group based on the propensity score to obtain multiple matching groups;

[0091] Step 207: Calculate the intervention effect of the intervention variable to be analyzed on the production data based on the production data values ​​of the two wells in each matching group, the weighting coefficient, and the total number of the matching groups, so that the staff can optimize the design of construction parameters based on the intervention effect.

[0092] The embodiments in this specification pre-define the types of background variables and post-intervention variables, then obtain the numerical values ​​of multiple types of background variables, post-intervention variables, and production data from multiple exploited wells. Then, a causal inference method is used to form a multi-level and comprehensive causal inference framework, which not only effectively solves the problems of confounding factors, endogeneity, and heterogeneous intervention, but also provides accurate causal effect analysis and optimization guidance, and has broad application prospects.

[0093] In the embodiments described in this specification, background variables are variables representing the inherent characteristics of geology and wells, factors that are not affected by other variables but can influence subsequent intervention variables, including but not limited to:

[0094] Oil layer thickness (H), unit: m;

[0095] Porosity (φ), unit: %;

[0096] Oil reservoir permeability (k), unit: mD;

[0097] Oil reservoir encounter rate (R), unit: %;

[0098] Horizontal segment length (L) h ), unit: m.

[0099] Post-intervention variables can be construction parameters, which are influenced by background variables and can be adjusted through intervention, including but not limited to:

[0100] Liquid strength (Q) f (Unit: m) 3 / min;

[0101] Sand strength (Q) s ), unit: kg / min;

[0102] Sand ratio (S), unit: %;

[0103] Segment spacing (D), unit: m;

[0104] Cluster spacing (C), unit: m.

[0105] Production data includes, but is not limited to, the yield increase factor (ΔP), in meters. 3 / d (increase in production after fracturing).

[0106] The embodiments in this specification can first preprocess the aforementioned raw data, performing cleaning, standardization, and normalization to meet the requirements of the causal inference model. The data normalization and continuous variable transformation methods are as follows:

[0107] (1) Data standardization and normalization

[0108] For continuous variables, standardization (Z-score standardization) is performed, as shown in the following formula:

[0109]

[0110] Where X is the original data value, μ is the sample mean, and σ is the sample standard deviation.

[0111] Suppose we have the following data points, each representing data from an exploited well:

[0112] Oil layer thickness (H) = [30, 45, 50, 65, 70, 55, 60, 48, 40, 72];

[0113] Porosity (φ) = [15,18,20,22,23,19,17,21,20,24];

[0114] Oil reservoir permeability (k) = [80, 120, 140, 160, 110, 95, 105, 130, 140, 125];

[0115] Standardize these data to obtain standardized data:

[0116] Standardized data for oil layer thickness:

[0117] H std = [0.25, 0.85, 1.05, 1.65, 1.95, 1.15, 1.45, 0.75, 0.05, 2.05];

[0118] Porosity standardized data:

[0119] φ std = [-0.95, -0.15, 0.15, 0.55, 0.75, 0.25, -0.35, 0.45, 0.15, 0.95];

[0120] Standardized data for oil reservoir permeability:

[0121] k std = [-1.31, 1.07, 1.67, 2.17, 0.57, -0.11, 0.07, 1.17, 1.67, 1.07];

[0122] (2) Continuous variables are transformed into categorical variables to facilitate user viewing.

[0123] For continuous variables (such as liquid usage intensity, sand addition intensity, etc.), unsupervised clustering methods (such as K-means) are used for classification. Assume that the liquid usage intensity (Q) is... f ) and sand strength (Q) s Group them.

[0124] Assuming liquid strength data (unit: m) 3 ( / min) is:

[0125] Q f =[5,7,8,6,7,5,9,6,8,7];

[0126] They were divided into 3 categories using K-means clustering. The following classification results were obtained after clustering:

[0127] Category 1: Q f =[5,6,5];

[0128] Category 2: Q f =[7,6,7,7];

[0129] Category 3: Q f =[8,9,8];

[0130] Next, the sand strength (Q) will be increased. s They are divided into 4 categories. Through similar clustering processing, the following classification results are obtained:

[0131] Category 1: Q s = [100, 120];

[0132] Category 2: Q s =[130,140,145];

[0133] Category 3: Q s =[150,155,160];

[0134] Category 4: Q s = [180, 185];

[0135] At this point, these variables are converted into categorical variables for easier viewing by users.

[0136] Furthermore, in this embodiment, a type of post-intervention variable is used as the intervention variable to be analyzed. This intervention variable can be specified by the staff, indicating the intervention effect of the variable on production. The stronger the intervention effect, the more significant the increase in production will be due to adjusting the construction parameters corresponding to the variable. After clustering the post-intervention variables, multiple exploited wells are divided into intervention groups and control groups based on the clustering results corresponding to the intervention variable to be analyzed among the post-intervention variables.

[0137] Specifically, dividing the multiple exploited wells into intervention and control groups based on the multiple categories further includes:

[0138] The categories belonging to the intervention group and the control group are determined based on the mean values ​​of the intervention variables to be analyzed in each category.

[0139] For example, the clustering result based on liquid strength is: Category 1: Q f =[5,6,5]; Category 2: Q f =[7,6,7,7]; Category 3: Q f =[8,9,8]. The wells corresponding to Category 1 with the lowest mean were assigned to the control group, and the wells corresponding to Categories 2 and 3 were assigned to the two intervention groups with different intervention levels. Alternatively, the wells corresponding to Category 3 with the highest mean were assigned to the control group, and the wells corresponding to Categories 1 and 2 were assigned to the two intervention groups with different intervention levels.

[0140] In addition, staff can divide the exploited wells into one or more intervention groups and a control group based on the clustering results. This specification does not limit the scope of the embodiments.

[0141] According to one embodiment of this specification, since the intervention effect of the same intervention variable to be analyzed on different formation characteristics is unbalanced, after obtaining the values ​​of multiple types of background variables, multiple types of post-intervention variables, and production data of multiple wells already in production, this embodiment of the specification can further group the multiple wells already in production according to the values ​​of the background variables to obtain multiple groups. This allows for the execution of the step of using a type of post-intervention variable as the intervention variable to be analyzed for multiple wells in each group, and finally obtains the intervention effect of the intervention variable to be analyzed on the production data in each group.

[0142] For example, oil and gas wells can be grouped according to formation characteristics (such as porosity and permeability). To simplify, the samples are divided into three groups, representing low porosity (φ<18%), medium porosity (18%≤φ<21%), and high porosity (φ≥21%) oil and gas wells, respectively.

[0143] Group 1 (low porosity): φ = [15, 17];

[0144] Group 2 (Medium porosity): φ=[18,19,20,21];

[0145] Group 3 (high porosity): φ = [22, 23, 24, 24];

[0146] By grouping, it is ensured that background variables such as porosity and permeability of the oil reservoirs within each group have similar distributions. This allows for the application of a single type of post-intervention variable as the intervention variable to be analyzed for multiple exploited wells within each group. Ultimately, the intervention effect of the intervention variable to be analyzed on the production data in each group is obtained. This helps reduce the impact of background variable imbalances on the analysis of intervention effects. In practice, division can also be based solely on blocks, development layers, and development technologies; the embodiments in this specification do not impose such limitations.

[0147] Then, a quantitative description of the intervention effect and a synthesis of multiple methods were performed:

[0148] (1) First layer: control of endogeneity, instrumental variable method (IV) and two-stage least squares method (2SLS) are used to calculate the endogeneity control coefficient.

[0149] Specify an intervention variable to be analyzed. If the intervention variable has endogeneity issues, use the instrumental variable method (IV) to eliminate bias in the estimation of the intervention effect. Endogeneity may arise from omitted variables, measurement errors, or inverse causal relationships. The steps are as follows:

[0150] ① Selecting an instrumental variable: An instrumental variable needs to be selected that is related to the intervention variable to be analyzed, but not directly related to the dependent variable (production enhancement factor). This variable is selected manually, usually from background variables. In the embodiments of this specification, the geological characteristics of the well (such as reservoir thickness, porosity, etc.) can be used as an instrumental variable because they determine the design of hydraulic fracturing, but do not directly affect the production after fracturing.

[0151] ② Instrumental variable method: Instrumental variables are used to replace the intervention variable to be analyzed (such as liquid strength, sand addition strength, etc.). Estimation is performed through the following two-stage process:

[0152] Phase 1: Estimating the predicted values ​​of the intervention variables to be analyzed (such as liquid intensity and sand addition intensity) using instrumental variables. Prediction methods include, but are not limited to, linear regression, machine learning, Bayesian regression, multinomial regression, and other regression prediction methods. For example, linear regression is used as a case study.

[0153]

[0154] in Z represents the alternative value for the intervention variable to be analyzed.i Let αi represent the numerical values ​​of the instrumental variable of type i, α0, ..., αi. i Let α be the coefficient and ε be the constant term. The value of the coefficient α is calculated using the least squares method, yielding... result.

[0155] ③ Endogenous control: using To estimate the causal effect and ensure that the intervention effect is not affected by endogeneity bias, linear regression was used.

[0156]

[0157] Where Y represents the numerical value of the production data, X i Let βi represent the background variable of the i-th type or the post-intervention variable other than the intervention variable to be analyzed, β0, ..., βi. i ..., β D These are the endogeneity control coefficients, where β is... i η is the endogeneity control coefficient of the background variable or the post-intervention variable of the i-th type, where η is a constant term.

[0158] The estimation is performed using two-stage least squares (2SLS), and the loss function for this least squares step is:

[0159]

[0160] Where Loss is the loss value, and n represents the total number of background variables or post-intervention variables other than the intervention variable to be analyzed.

[0161] (2) Second layer: Data preprocessing, propensity score matching (PSM) and inverse probability weighting (IPW) were used to match the exploited wells in the intervention group and the control group.

[0162] To ensure the accuracy of the intervention effect estimation and reduce the influence of confounding factors, propensity score matching (PSM) was first used to handle confounding variables in the sample. The goal of PSM is to balance the experimental and control groups on confounding variables, making the samples as similar as possible in terms of background variables. The basic steps are as follows:

[0163] ① Calculate the propensity score: The propensity score refers to the probability that a sample will accept intervention given background variables (such as porosity, permeability, etc.). A logistic regression model can be used to estimate the propensity score.

[0164]

[0165] Where e(D) represents the propensity score of the intervention variable D to be analyzed, X1, X2, ..., X... nThese represent the numerical values ​​of the background variables of multiple types, and the numerical values ​​of post-intervention variables of other types besides the intervention variable to be analyzed, β1, β2, ..., β... n These are the endogeneity control coefficients for the corresponding background variables or other types of post-intervention variables besides the intervention variables to be analyzed.

[0166] ② Determining weights using the Inverse Probability Weighting (IPW) method: To further reduce intervention score bias, the Inverse Probability Weighting (IPW) method is used for weight adjustment. The weight of each sample is determined by the reciprocal of its propensity score. The weight adjustment formula is as follows:

[0167] If the wells already in operation belong to the intervention group, then according to the formula... Calculate the weighting coefficients, where w i e(D) represents the weighting coefficient of the intervention variable D to be analyzed in the i-th well of the intervention group. i ) represents the propensity score of the intervention variable D to be analyzed in the i-th well in the intervention group;

[0168] If the wells already in operation belong to the control group, then according to the formula... Calculate the weighting coefficients, where w j e(D) represents the weight coefficient of the intervention variable D to be analyzed in the j-th well of the control group. j ) represents the propensity score of the intervention variable D to be analyzed in the j-th well of the control group.

[0169] These weights can be used to make a weighted estimate of the intervention effect, thereby further reducing the bias caused by sample imbalance.

[0170] ③ Sample Matching: Based on the calculated propensity score, samples from the intervention and control groups are matched. Matching methods can include Nearest Neighbor Matching, Caliper Matching, etc. The goal is to make the intervention and control groups as similar as possible in terms of propensity scores, thereby reducing bias caused by confounding variables.

[0171] like Figure 3 As shown, based on the propensity score, the exploited wells in the intervention group are matched with the exploited wells in the control group to obtain multiple matching groups, which further include:

[0172] Step 301: Calculate the difference between the tendency score of each exploited well in the intervention group and the tendency score of each exploited well in the control group, and identify exploited wells in the control group whose difference is less than a threshold, and match the exploited well in the intervention group and the exploited well in the control group as a matching group.

[0173] Step 302: If a well in the intervention group has no matching well in the control group, then the well in the intervention group is discarded.

[0174] For example, samples 1 to 6 and their propensity scores:

[0175] For each sample in the intervention group, find a control group sample with the closest propensity score (difference within 0.05). For example:

[0176] Well 1 in the intervention group (propensity score = 0.014) was the closest to Well 4 in the control group (propensity score = 0.018), and was thus matched as a matched group.

[0177] Well 3 in the intervention group (propensity score = 0.25) was the closest to well 2 in the control group (propensity score = 0.23), and was thus matched as a matched group.

[0178] Well 5 in the intervention group (propensity score = 0.15) was the closest to well 6 in the control group (propensity score = 0.18), and was thus matched as a matched group.

[0179] When calculating the intervention effect analysis, using matched samples (discarding mismatched samples) can effectively reduce the impact of confounding factors on the intervention effect estimation.

[0180] like Figure 4 As shown, if the values ​​of the intervention variables to be analyzed in the multiple exploited wells are clustered to obtain two or more categories, the division of the multiple exploited wells into intervention groups and control groups based on the multiple categories further includes:

[0181] Step 401: Determine multiple categories belonging to the intervention group and one category belonging to the control group based on the mean values ​​of the intervention variables to be analyzed in each category;

[0182] Based on the propensity score, the exploited wells in the intervention group were matched with the exploited wells in the control group to obtain multiple matching groups, which further included:

[0183] Step 402: Calculate the difference between the tendency score of each exploited well in each intervention group and the tendency score of each exploited well in the control group, and identify exploited wells in the control group whose difference is less than a threshold, and match the exploited well in the intervention group and the exploited well in the control group as a matching group.

[0184] Step 403: If a well in any intervention group has no matching well in the control group, then the well in that intervention group is discarded.

[0185] For example, well 7 in the intervention group (propensity score = 0.16, intervention level 2) was the closest to well 6 in the control group (propensity score = 0.18), thus forming a matched group.

[0186] Well 8 in the intervention group (propensity score = 0.99, intervention level 2) was the closest to well 9 in the control group (propensity score = 0.98), and was thus matched as a matched group.

[0187] The matching relationships here are shown in Table 1 below, with a total of 5 matching groups.

[0188] Table 1

[0189]

[0190] ④ Conduct intervention effect analysis: The effect of the intervention is assessed by calculating the weighted average treatment effect (ATE) of the matched experimental and control groups.

[0191] The formula for calculating the intervention effect of the intervention variable to be analyzed on the production data, based on the production data values ​​of the two exploited wells in each of the matching groups, the weighting coefficient, and the total number of the matching groups, is as follows:

[0192]

[0193] Where ATE represents the intervention effect of the intervention variable to be analyzed on the production data, k represents the total number of matched groups, T represents the intervention group set, C represents the control group set, and Y represents the control group set. i Y represents the production data of the i-th well belonging to intervention group T. j This represents the production data of the j-th well in the control group C, which belongs to the same matching group as the i-th well in the intervention group T.

[0194] For example, the relationships between the endogeneity control coefficient β, propensity score e(D), grouping (intervention group and control group), weighting coefficient w, and intervention effect ATE are shown in Table 2 below, where T=0 represents the control group and T=1 represents the intervention group:

[0195] Table 2

[0196]

[0197] In the embodiments of this specification, the influence of the intervention variable on the target parameter is explained based on the results of the intervention effect. For example, the intervention effect of the intervention variable to be analyzed (sand addition intensity) on the yield data is 0.17, which means that adjusting the sand addition intensity can increase the yield by 17%.

[0198] Furthermore, based on the magnitude of the absolute value of the intervention effect, the importance of variables is ranked as follows: sand addition intensity > fluid application intensity > porosity > reservoir thickness. This means that sand addition intensity has the greatest impact on the hydraulic fracturing effect, while porosity and reservoir thickness also play a supporting role to some extent.

[0199] The above analysis reveals how various variables affect the production increase of hydraulic fracturing, further elucidating their causal relationships. Specifically, the causal explanation can be derived from the following aspects:

[0200] ① The causal effect of intervention variables on output increase:

[0201] Propane injection intensity: The results show that proppane injection intensity has the greatest impact on production increase. This conclusion can be explained by causal inference analysis: increasing proppane injection intensity increases the amount of proppant injected during hydraulic fracturing, thereby forming more fractures in the formation, improving oil and gas flowability, and ultimately leading to increased production.

[0202] Fluid strength: Increasing the fluid strength can also promote fracturing effect. Increasing the amount of fluid injected helps the fracture to expand and maintain, thereby increasing production.

[0203] ② The influence of background variables:

[0204] Porosity: Increased porosity means that there are larger voids in the rock, making it easier to store oil and gas. Therefore, porosity also plays a certain role in promoting the effect of hydraulic fracturing.

[0205] Oil layer thickness: The thickness of the oil layer affects the coverage and effect of fracturing. Thicker oil layers may require stronger fracturing force and more refined parameter design, which affects the final effect of hydraulic fracturing.

[0206] After analyzing the intervention effects of multiple intervention variables on production data, staff can optimize construction parameters according to the following steps:

[0207] Optimize sand addition intensity: Given the importance of sand addition intensity to the production enhancement effect, it is recommended to give priority to optimizing sand addition intensity in hydraulic fracturing design, especially in shallow oil reservoirs.

[0208] Adjusting the fluid strength: Optimizing the fluid strength can also effectively improve the fracturing effect, but it should be adjusted flexibly according to the characteristics of the oil reservoir.

[0209] Optimization for deep oil layers: For deep oil layers, it is recommended to increase the inter-segment spacing and sand addition intensity through refined design to compensate for the lack of fracturing effect.

[0210] Model robustness and interpretability: Cross-validation and comparison with different models ensured the robustness of the estimation results. Interpretive analysis of the model (e.g., ranking the importance of decision tree models) enabled staff to clearly understand the mechanisms of action of each factor, thus facilitating better parameter optimization design.

[0211] Based on the same inventive concept, embodiments of this specification also provide a device for analyzing the main controlling factors of hydraulic fracturing effects based on causal inference. For example... Figure 5 As shown, the device includes:

[0212] The data acquisition unit 501 is used to acquire the values ​​of multiple types of background variables, multiple types of post-intervention variables, and production data of multiple exploited wells;

[0213] Clustering unit 502 is used to cluster the values ​​of the intervention variables to be analyzed for the multiple wells, taking a type of post-intervention variable as the intervention variable to be analyzed, to obtain multiple categories, and to divide the multiple wells into intervention group and control group according to the multiple categories;

[0214] Endogeneity control coefficient calculation unit 503 is used to calculate the endogeneity control coefficients of each type of background variable and each type of post-intervention variable for each well in the intervention group and the control group, based on the values ​​of the background variables of multiple types of the well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the values ​​of the production data.

[0215] The propensity score calculation unit 504 is used to calculate the propensity score of the intervention variable to be analyzed for each well in the intervention group and the control group based on the values ​​of the background variables of multiple types of the well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficient of each variable.

[0216] The weighting coefficient calculation unit 505 is used to calculate the weighting coefficient of the intervention variable to be analyzed for each well in the intervention group and the control group based on the tendency score of the intervention variable to be analyzed for each well in the intervention group and the control group.

[0217] Matching unit 506 is used to match the exploited wells in the intervention group with the exploited wells in the control group according to the propensity score, so as to obtain multiple matching groups;

[0218] The intervention effect calculation unit 507 is used to calculate the intervention effect of the intervention variable to be analyzed on the production data based on the production data values ​​of the two wells in each of the matching groups, the weighting coefficients, and the total number of the matching groups, so that the staff can optimize the design of construction parameters based on the intervention effect.

[0219] Since the principle of the above-mentioned device in solving the problem is similar to that of the above-mentioned method, the implementation of the above-mentioned system can refer to the implementation of the above-mentioned method, and the repeated parts will not be described again.

[0220] like Figure 6 The diagram shown is a structural schematic of a computer device according to an embodiment of this specification. The methods described in this specification can be applied to the computer device of this embodiment.

[0221] Computer device 602 may include one or more processors 604, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. Computer device 602 may also include any memory 606 for storing information of any kind, such as code, settings, data, etc. Non-limitingly, for example, memory 606 may include any type of RAM, any type of ROM, flash memory, hard disk, optical disk, etc. More generally, any storage resource can be used to store information using any technology.

[0222] Furthermore, any storage resource can provide volatile or non-volatile retention of information.

[0223] Furthermore, any storage resource can represent a fixed or removable component of the computer device 602. In one case, when the processor 604 executes associated instructions stored in any storage resource or combination of storage resources, the computer device 602 can perform any operation of the associated instructions. The computer device 602 also includes one or more drive mechanisms 608 for interacting with any storage resource, such as a hard disk drive system, an optical disk drive system, etc.

[0224] Computer device 602 may also include an input / output module 610 (I / O) for receiving various inputs (via input device 612) and providing various outputs (via output device 614). A specific output mechanism may include a presentation device 616 and an associated graphical user interface (GUI) 618. In other embodiments, the input / output module 610 (I / O), input device 612, and output device 614 may be omitted, and the device may function solely as a computer device within a network. Computer device 602 may also include one or more network interfaces 620 for exchanging data with other devices via one or more communication links 622. One or more communication buses 624 couple the components described above together.

[0225] Communication link 622 can be implemented in any way, such as via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, or any combination thereof. Communication link 622 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.

[0226] This specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0227] This specification also provides computer-readable instructions, wherein when a processor executes the instructions, the program therein causes the processor to perform the above-described method.

[0228] It should be understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0229] It should also be understood that, in the embodiments of this specification, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the embodiments of this specification, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0230] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this specification can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the embodiments in this specification.

[0231] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0232] In the embodiments provided in this specification, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, devices, or units, or they may be electrical, mechanical, or other forms of connection.

[0233] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments described in this specification, depending on actual needs.

[0234] Furthermore, the functional units in the various embodiments of this specification can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0235] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this specification, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this specification. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0236] This specification describes the principles and implementation methods of the embodiments using specific examples. The above descriptions of the embodiments are only for the purpose of helping to understand the methods and core ideas of the embodiments in this specification. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the embodiments in this specification. Therefore, the content of this specification should not be construed as a limitation on the embodiments in this specification.

Claims

1. A method for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference, characterized in that, The method includes: Obtain numerical values ​​for multiple types of background variables, multiple types of post-intervention variables, and production data from multiple exploited wells; Using a type of post-intervention variable as the intervention variable to be analyzed, the values ​​of the intervention variable to be analyzed for the multiple wells to be clustered to obtain multiple categories, and the multiple wells to be analyzed are divided into intervention group and control group according to the multiple categories; For each exploited well in the intervention group and the control group, the endogeneity control coefficients of the background variables and post-intervention variables of the exploited well are calculated based on the values ​​of the background variables of the multiple types of the exploited well, the values ​​of the post-intervention variables of the other types besides the intervention variable to be analyzed, and the values ​​of the production data. For each exploited well in the intervention group and the control group, the propensity score of the intervention variable to be analyzed for the exploited well is calculated based on the values ​​of the background variables of multiple types for the exploited well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficient of each variable. The weighting coefficient of the intervention variable to be analyzed is calculated for each well in the intervention group and the control group based on the propensity score of the intervention variable to be analyzed. Based on the propensity score, the exploited wells in the intervention group were matched with the exploited wells in the control group to obtain multiple matching groups; The intervention effect of the intervention variable to be analyzed on the production data is calculated based on the production data values ​​of the two wells in each of the matching groups, the weighting coefficient, and the total number of the matching groups, so that the staff can optimize the design of construction parameters based on the intervention effect.

2. The method according to claim 1, characterized in that, Further classifying the multiple exploited wells into intervention and control groups based on the multiple categories includes: The categories belonging to the intervention group and the control group are determined based on the mean values ​​of the intervention variables to be analyzed in each category.

3. The method according to claim 1, characterized in that, The calculation of the endogeneity control coefficients for the various types of background variables and various types of post-intervention variables of the exploited well, based on the values ​​of the background variables of the various types, the values ​​of the post-intervention variables other than the intervention variable to be analyzed, and the production data, further includes: At least one instrumental variable shall be determined from the multiple types of background variables and other types of post-intervention variables besides the intervention variable to be analyzed; Using the least squares method to apply the formula The solution is performed to obtain the endogeneity control coefficients for the background variables and post-intervention variables of multiple types, where Y represents the numerical value of the production data, X... i Let βi represent the background variable of the i-th type or the post-intervention variable other than the intervention variable to be analyzed, β0, ..., βi. i ..., β D These are the endogeneity control coefficients, where β is... i Here, η is the endogeneity control coefficient for the background variable or post-intervention variable of the i-th type, where η is a constant term. in Z represents the alternative value for the intervention variable to be analyzed. i Let αi represent the numerical values ​​of the instrumental variable of type i, α0, ..., αi. i Let ε be the coefficient and ε be the constant term; the loss function of the least squares method is: Where Loss is the loss value, and n represents the total number of background variables or post-intervention variables other than the intervention variable to be analyzed.

4. The method according to claim 3, characterized in that, The formula for calculating the propensity score of the intervention variable to be analyzed for the well-existing facility, based on the values ​​of the background variables of multiple types, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficients of each variable, is as follows: Where e(D) represents the propensity score of the intervention variable D to be analyzed, X1, X2, ..., X... n These represent the numerical values ​​of the background variables of multiple types, and the numerical values ​​of post-intervention variables of other types besides the intervention variable to be analyzed, β1, β2, ..., β... n These are the endogeneity control coefficients for the corresponding background variables or other types of post-intervention variables besides the intervention variables to be analyzed.

5. The method according to claim 4, characterized in that, The calculation of the weighting coefficient of the intervention variable to be analyzed for each well in the intervention group and the control group, based on the propensity score of each well in the control group, further includes: If the wells already in operation belong to the intervention group, then according to the formula... Calculate the weighting coefficients, where w i e(D) represents the weighting coefficient of the intervention variable D to be analyzed in the i-th well of the intervention group. i ) represents the propensity score of the intervention variable D to be analyzed in the i-th well in the intervention group; If the wells already in operation belong to the control group, then according to the formula... Calculate the weighting coefficients, where w j e(D) represents the weight coefficient of the intervention variable D to be analyzed in the j-th well of the control group. j ) represents the propensity score of the intervention variable D to be analyzed in the j-th well of the control group.

6. The method according to claim 5, characterized in that, Based on the propensity score, the exploited wells in the intervention group were matched with the exploited wells in the control group to obtain multiple matching groups, which further included: Calculate the difference between the tendency score of each exploited well in the intervention group and the tendency score of each exploited well in the control group, and identify exploited wells in the control group whose difference is less than a threshold, and combine the exploited well in the intervention group and the exploited well in the control group into a matching group; If a well in the intervention group has no matching well in the control group, then that well in the intervention group is discarded.

7. The method according to claim 6, characterized in that, If the values ​​of the intervention variables to be analyzed from the multiple exploited wells are clustered to obtain two or more categories, further dividing the multiple exploited wells into intervention and control groups based on the multiple categories includes: Based on the mean values ​​of the intervention variables to be analyzed in each category, multiple categories belonging to the intervention group and one category belonging to the control group are determined; Based on the propensity score, the exploited wells in the intervention group were matched with the exploited wells in the control group to obtain multiple matching groups, which further included: Calculate the difference between the tendency score of each exploited well in each intervention group and the tendency score of each exploited well in the control group, and identify exploited wells in the control group whose difference is less than a threshold, and pair the exploited well in the intervention group and the exploited well in the control group as a matching group; If a well in any intervention group has no matching well in the control group, then that well in the intervention group is discarded.

8. The method according to claim 6, characterized in that, The formula for calculating the intervention effect of the intervention variable to be analyzed on the production data, based on the production data values ​​of the two exploited wells in each of the matching groups, the weighting coefficient, and the total number of the matching groups, is as follows: Where ATE represents the intervention effect of the intervention variable to be analyzed on the production data, k represents the total number of matched groups, T represents the intervention group set, C represents the control group set, and Y represents the intervention effect. i Y represents the production data of the i-th well belonging to intervention group T. j This represents the production data of the j-th well in the control group C, which belongs to the same matching group as the i-th well in the intervention group T.

9. The method according to claim 1, characterized in that, After obtaining numerical values ​​for multiple types of background variables, multiple types of post-intervention variables, and production data from multiple exploited wells, the method further includes: The multiple exploited wells are grouped according to the values ​​of the background variables to obtain multiple groups. This allows for the execution of the step of using a type of post-intervention variable as the intervention variable to be analyzed for multiple exploited wells in each group, ultimately obtaining the intervention effect of the intervention variable to be analyzed on the production data in each group.

10. A device for analyzing the main controlling factors of hydraulic fracturing effect based on causal inference, characterized in that, The device includes: The data acquisition unit is used to acquire numerical values ​​of multiple types of background variables, multiple types of post-intervention variables, and production data from multiple exploited wells. Clustering unit is used to cluster the values ​​of the intervention variables to be analyzed for the multiple wells that have been exploited, taking a type of post-intervention variable as the intervention variable to be analyzed, to obtain multiple categories, and to divide the multiple wells into intervention group and control group according to the multiple categories; The endogeneity control coefficient calculation unit is used to calculate the endogeneity control coefficients of each type of background variable and each type of post-intervention variable for each well in the intervention group and the control group, based on the values ​​of the background variables of multiple types of the well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the values ​​of the production data. The propensity score calculation unit is used to calculate the propensity score of the intervention variable to be analyzed for each well in the intervention group and the control group based on the values ​​of the background variables of multiple types of the well, the values ​​of the post-intervention variables of other types besides the intervention variable to be analyzed, and the endogeneity control coefficient of each variable. The weighting coefficient calculation unit is used to calculate the weighting coefficient of the intervention variable to be analyzed for each well in the intervention group and the control group based on the tendency score of the intervention variable to be analyzed for each well in the intervention group and the control group. A matching unit is used to match the exploited wells in the intervention group with the exploited wells in the control group based on the propensity score, thereby obtaining multiple matching groups; The intervention effect calculation unit is used to calculate the intervention effect of the intervention variable to be analyzed on the production data based on the production data values ​​of the two wells in each of the matching groups, the weighting coefficients, and the total number of the matching groups, so that the staff can optimize the design of construction parameters based on the intervention effect.

Images