A method and apparatus for recommending profile and drive parameters based on similarity calculation model.

Abstract

This invention discloses a method and apparatus for recommending profile control and displacement parameters based on a similarity calculation model, belonging to the field of oil drilling technology. The method includes: acquiring first data information of several historical intervention wells based on preset analogy system parameters; calculating the analogy parameter weights of the preset analogy system parameters; acquiring second data information of intervention wells based on the preset analogy system parameters; calculating the similarity between each historical intervention well using a similarity calculation method; sorting the similarity between each historical intervention well and the intervention well according to their numerical values, and determining the historical intervention well with the highest similarity as a similar intervention well; acquiring historical profile control schemes for similar intervention wells and pushing them to terminal equipment; and the apparatus itself. This invention enables rapid design of profile control and displacement schemes by referencing historical profile control schemes from historical intervention wells that are most similar to the geological statics and production dynamics of the intervention well block.

Description

Technical Field

[0001] This invention belongs to the field of oil drilling technology, and in particular relates to a method and device for recommending profile control and drive parameters based on a similarity calculation model. Background Technology

[0002] Water drive profile adjustment technology can effectively modify the water intake profile of injection wells and improve water drive efficiency. The design of water drive profile adjustment schemes requires consideration of many factors. For example, the injection method for water drive adjustment is generally a composite slug injection, designed based on the principle that the modulating agent preferentially enters large pores and then small pores and thin, poor-permeability layers. The slugs to be designed include pre-slugs, main slugs, and post-slugs. The injection volume of the water drive system also affects the adjustment effect; if the injection volume is too small, it is difficult to achieve the purpose of sealing high-permeability layers, while if the injection volume is too large, the cost will increase accordingly. In current production practice, the main methods for optimizing slug design in water drive schemes include cross-validation and numerical simulation. However, cross-validation cannot comprehensively cover all possible combinations of schemes, and numerical simulation requires highly skilled technicians, is difficult to prepare data for, and has a long simulation period, which cannot meet the requirements for rapid parameter recommendation of water drive schemes. Summary of the Invention

[0003] This invention addresses the problems existing in the prior art. Its purpose is to provide a parameter analogy recommendation method for profile and drive adjustment based on a similarity calculation model, which can quickly design profile and drive adjustment schemes by referring to historical profile adjustment schemes of historical wells that are most similar to the geological statics and production dynamics of the well block.

[0004] The second objective of this invention is to propose a parameter analogy recommendation device for profile adjustment and driving based on a similarity calculation model.

[0005] To achieve the above objectives, the technical solution of the present invention provides a method for recommending profile and drive parameters by analogy based on a similarity calculation model, including:

[0006] Acquire first data information of several historical wells based on preset analogy system parameters, which include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters.

[0007] Obtain second data information of the well based on preset analog system parameters;

[0008] Based on the first data information and the second data information, the similarity between each historical well and the well to be treated is calculated using a similarity calculation method;

[0009] The similarity scores of each historical well and the current well are sorted by numerical value, and the historical well with the highest similarity score is identified as the similar well.

[0010] The historical profile control schemes of similar wells are obtained and pushed to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters, and drive construction parameters.

[0011] Furthermore, based on the first data information and the second data information, a similarity calculation method is used to calculate the similarity between each historical intervention well and the intervention well. This similarity calculation method includes the Euclidean distance method, which comprises: according to the formula... Calculate the Euclidean distance between two points, object X and object Y, in n-dimensional space. The Euclidean distance is used to characterize the similarity.

[0012] Furthermore, after obtaining the first data information of several historical wells based on preset analogy system parameters, the method further includes: calculating the analogy parameter weights of the preset analogy system parameters according to the preset analogy system parameters;

[0013] The similarity calculation method for each historical well and each applied well, based on the first data information and the second data information, further includes: calculating the similarity between each historical well and each applied well based on the analogy parameter weights; the similarity calculation method includes a comprehensive feature method, which includes: based on formula K i =1-H i Calculate the similarity K between the treated well and the historical treated well. i Where H is based on the formula H i =(O i -O min ) / (O max -O min ) Calculation, O i It is a comprehensive index.

[0014] Furthermore, after acquiring the first data information of several historical wells based on preset analogy system parameters, the method further includes: calculating the analogy parameter weights of the preset analogy system parameters according to the preset analogy system parameters. ;

[0015] The similarity calculation method for each historical well and the applied well, based on the first data information and the second data information, further includes: calculating the similarity between the two wells according to the analogy parameter weights. The similarity between historical wells and current wells is calculated using a similarity calculation method; the similarity calculation method includes a multivariate decision-making method, which includes: according to the formula The similarity K between the g-th historical well and the well that was treated can be calculated. g Where g = 1, 2, ..., m, μ(x) giW represents the fuzzy value of the g-th historical measure well. i The corresponding weights of each parameter in the preset analogy system parameters.

[0016] Furthermore, acquiring historical profile control schemes for similar wells and pushing them to the terminal device also includes: acquiring and pushing the profile control effect data of similar wells to the terminal device, wherein the profile control effect data includes the oil increase, economic benefits, and input-output ratio of the similar wells.

[0017] To achieve the above objectives, a second aspect of the present invention provides a profile and drive parameter analogy recommendation device based on a similarity calculation model, comprising:

[0018] The first acquisition module is used to acquire first data information of several historical wells based on preset analog system parameters. The preset analog system parameters include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters.

[0019] The second acquisition module is used to acquire second data information of the measure well based on preset analog system parameters;

[0020] The second calculation module is used to calculate the similarity between each historical well and the well based on the first data information and the second data information using a similarity calculation method.

[0021] The determination module is used to sort the similarity between each historical well and the current well by numerical value, and determine the historical well with the highest similarity as the similar well.

[0022] The third acquisition module is used to acquire the historical profile control schemes of the similar wells and push them to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters, and drive construction parameters.

[0023] Furthermore, in the second calculation module, the similarity calculation method includes the Euclidean distance method, which includes: according to the formula Calculate the Euclidean distance between two points, object X and object Y, in n-dimensional space. The Euclidean distance is used to characterize the similarity.

[0024] Furthermore, after the first acquisition module, there is also a first calculation module, used to calculate the analogy parameter weights of the preset analogy system parameters according to the preset analogy system parameters;

[0025] The second calculation module further includes: calculating the similarity between each historical well and the well to be treated based on the analogy parameter weights using a similarity calculation method; the similarity calculation method includes a comprehensive feature method, which includes: based on formula Ki =1-H i Calculate the similarity K between the treated well and the historical treated well. i Where H is based on the formula H i =(O i -O min ) / (O max -O min ) Calculation, O i It is a comprehensive index.

[0026] Furthermore, after the first acquisition module, the system further includes: a first calculation module, configured to calculate the analogy parameter weights of the preset analogy system parameters based on the preset analogy system parameters. ;

[0027] The second calculation module further includes: further calculating based on the analogy parameter weights. The similarity between historical wells and current wells is calculated using a similarity calculation method; the similarity calculation method includes a multivariate decision-making method, which includes: according to the formula The similarity K between the g-th historical well and the well that was treated can be calculated. g Where g = 1, 2, ..., m, μ(x) gi W represents the fuzzy value of the g-th historical measure well. i The corresponding weights of each parameter in the preset analogy system parameters.

[0028] Furthermore, the third acquisition module, used to acquire the historical profile control schemes of the similar wells and push them to the terminal device, also includes: acquiring the profile control effect data of the similar wells and pushing it to the terminal device, wherein the profile control effect data includes the oil increase, economic benefits, and input-output ratio of the similar wells.

[0029] By adopting the above scheme, this invention can quickly calculate the similarity between the first data information of several historical intervention wells based on preset analogy system parameters and the second data information of intervention wells based on preset analogy system parameters through a similarity calculation method. By sorting the similarity scores, the historical intervention well with the highest similarity score can be identified as the similar intervention well with the best intervention effect. By obtaining data such as historical profile control schemes and drive control effect data of similar intervention wells, decision-making basis can be quickly provided for the design scheme of intervention wells, and support can be provided for predicting the drive control effect. Attached Figure Description

[0030] Figure 1 This is a flowchart of a method for recommending profile and drive parameters based on a similarity calculation model, provided in Embodiments 1 and 2 of the present invention.

[0031] Figure 2This is a flowchart of a method for recommending profile and drive parameters based on a similarity calculation model, provided in embodiments three and four of this invention.

[0032] Figure 3 This is a structural schematic diagram of step S12 in embodiments three and four of the present invention;

[0033] Figure 4 This is a schematic diagram of step S14 in embodiments two to four of the present invention;

[0034] Figure 5 This is a flowchart of the principal component analysis solution in step 121 of Embodiment 3 of the present invention;

[0035] Figure 6 This is a schematic diagram of the structure of a profile and drive parameter analogy recommendation device based on a similarity calculation model provided in Embodiment 5 of the present invention;

[0036] Figure 7 This is a schematic diagram of another profile and drive parameter analogy recommendation device based on a similarity calculation model provided in Embodiment 5 of the present invention. Detailed Implementation

[0037] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. The described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0038] Example 1: As Figure 1 As shown, embodiments of the present invention provide a method for recommending profile and drive parameters based on a similarity calculation model, including:

[0039] Step S11: Obtain first data information of several historical wells based on preset analog system parameters. The preset analog system parameters include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters.

[0040] The parameters of the pre-set analogy system can be easily obtained through existing methods. Among these, the target reservoir characteristic parameters include well group reserves, minimum porosity, maximum porosity, minimum permeability, maximum permeability, coefficient of variation, and permeability range. These parameters can be obtained from oilfield data tables and sub-layer data tables in historical sample databases. The target reservoir fluid property parameters include crude oil viscosity, formation water salinity, and temperature. These parameters can be obtained from formation data tables in historical sample databases. The well group injection-production dynamic characteristic parameters include average wellhead pressure, maximum wellhead pressure, average daily injection rate, average apparent water absorption index, FD value, and PI value. The average wellhead pressure, maximum wellhead pressure, and average daily injection rate can be obtained from water injection well production records, while the average apparent water absorption index test data, FD value, and PI value can be calculated from water injection wellhead pressure drop test data. The types of dominant channel characteristic parameters can be obtained through dominant channel characteristic analysis software, and can be obtained by establishing a water drive capacitive resistance model of the target reservoir and inverting it using the ensemble Kalman filter method.

[0041] Step S13: Obtain the second data information of the measure well based on the preset analog system parameters.

[0042] Step S14: Based on the first data information and the second data information, calculate the similarity between each historical well and the well that was used for the treatment using a similarity calculation method.

[0043] Based on the first and second data information, an analogy model is established, and a similarity calculation method is used to compare the similarity between historical wells and the wells that have undergone intervention.

[0044] Step S15: Sort the similarity scores of each historical well and the current well by numerical value, and determine the historical well with the highest similarity score as the similar well.

[0045] Furthermore, after calculating the similarity between the intervention well and historical intervention wells using similarity calculation methods, the degree of oilfield similarity can also be quantitatively represented numerically. For example, the similarity values ​​can be adjusted to 0-100, and the parameters of similar intervention wells participating in the analogy and the contribution of each parameter to the similarity can be intuitively expressed through lists, radar charts, etc., thereby quantitatively characterizing the similarity of intervention wells.

[0046] Step S16: Obtain the historical profile control schemes of the similar wells and push them to terminal devices such as computers, laptops, and mobile phones. The historical profile control schemes include profile control design parameters, slug design parameters, and profile control construction parameters.

[0047] Among them, profile control design parameters may include total fluid volume, profile control radius, PV number, etc.; slug design parameters may include system type, system combination, system concentration, single slug injection volume, etc.; profile control construction parameters may include displacement, pump pressure, injection time, etc., as shown in Table 1, which provides examples of historical profile control schemes for similar wells.

[0048] Table 1. Examples of historical profile control schemes for wells with similar measures

[0049]

[0050] Furthermore, as shown in Table 2, the historical profile control schemes of similar wells will be obtained and pushed to the terminal device. The process also includes obtaining the profile control effect data of similar wells, which includes the oil production, economic benefits, and input-output ratio of the similar wells.

[0051] Table 2. Example of similarity data on the driving effect of historical control measures wells.

[0052]

[0053] By adopting the technical solution provided in Embodiment 1, this invention selects reasonable preset analogy system parameters before implementing the regulation and drive measures. Through a similarity calculation method, the similarity between the first data information of several historical wells based on the preset analogy system parameters and the second data information of the wells under regulation based on the preset analogy system parameters can be quickly calculated. By sorting the similarities, the historical well with the highest similarity can be identified as the similar well with the best regulation effect. By obtaining data such as the historical profile regulation schemes and regulation and drive effect data of similar wells, a decision-making basis can be quickly provided for the design scheme of the wells under regulation, and support can be provided for predicting the regulation and drive effect.

[0054] Example 2: Figure 1 , Figure 4 As shown, embodiments of the present invention provide a method for recommending profile and drive parameters based on a similarity calculation model, including:

[0055] Step S11: Obtain first data information of several historical wells based on preset analog system parameters. The preset analog system parameters include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters.

[0056] Step S13: Obtain the second data information of the measure well based on the preset analog system parameters.

[0057] Step S14: Based on the first data information and the second data information, calculate the similarity between each historical well and the well that was used for the treatment using a similarity calculation method.

[0058] Step S141: Based on the first data information and the second data information, calculate the similarity between each historical intervention well and the intervention well using a similarity calculation method. The similarity calculation method includes the Euclidean distance method, which includes: according to the formula... ,

[0059] When comparing the similarity between historical wells and current wells, a common method is the Euclidean distance method. Euclidean distance is the most commonly used distance calculation formula, measuring the absolute distance between points in a multidimensional space. It is suitable for situations where the data is dense and continuous. The Euclidean distance between two points, object X (historical well) and object Y (current well), in n-dimensional space is used to characterize similarity.

[0060] The Euclidean distance method can be used as follows: First, take objects X and Y that both contain n-dimensional feature historical wells and measure wells, where X = (x1, x2, x3, ... x...). n Y = (y1, y2, y3, ... y) n According to the formula Calculate the similarity between X and Y. According to the formula... Calculate the Euclidean distance between two points X and Y in n-dimensional space. In this context, a smaller Euclidean distance indicates a greater similarity, while a larger Euclidean distance indicates a smaller similarity.

[0061] Step S15: Sort the similarity scores of each historical well and the current well by numerical value, and determine the historical well with the highest similarity score as the similar well.

[0062] Step S16: Obtain the historical profile control schemes of the similar wells and push them to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters, and drive construction parameters.

[0063] All other technical measures in this embodiment are the same as those in Embodiment 1, and therefore have all the above-mentioned beneficial effects, which will not be repeated here.

[0064] Example 3: Figures 2 to 5 As shown, embodiments of the present invention provide a method for recommending profile and drive parameters based on a similarity calculation model, including:

[0065] Step S11: Obtain first data information of several historical wells based on preset analog system parameters. The preset analog system parameters include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters.

[0066] Step S12: Calculate the analogy parameter weights of the preset analogy system parameters based on the preset analogy system parameters.

[0067] When calculating the analogy parameter weights of the preset analogy system parameters, the methods for calculating the analogy parameter weights may include principal component analysis (PCA) and analytic hierarchy process (AHP).

[0068] Step S121: Apply PCA to calculate the analogy parameter weights. PCA is a data compression and feature extraction technique suitable for processing high-dimensional data with strong correlations between variables. PCA can reduce the dimensionality of data, transforming numerous correlated indicators X1, X2, ..., X... P (Where p is the number of indicators), they are recombined into a smaller set of uncorrelated composite indicators F. m To replace the original indicators.

[0069] Finding the principal components is the process of determining the eigenvalues ​​and eigenvectors based on the covariance matrix of the data source. Principal components can be represented using the orthogonal eigenvectors corresponding to the eigenvalues ​​of the covariance matrix. According to the formula... The original variables X1, X2, ..., X can be... p Recombining them yields new indicators F1, F2...F p These indicators can fully reflect the main information of the original indicators. The formula satisfies the following three conditions:

[0070] I. aij are the elements of the covariance matrix, and the coefficient matrix A is an orthogonal matrix, i.e.: .

[0071] II. Principal components Fi and Fj are independent of each other, and their Cov(Fj) is independent of each other. i F j The covariance is 0, that is: , where i ≠ j, j = 1, 2, ..., p.

[0072] III. The importance of principal components: Var(F) gradually decreases, satisfying the condition that variance decreases sequentially, i.e.: .

[0073] like Figure 5 As shown, principal component analysis includes the following steps:

[0074] The original data sources are the first data information of historical wells based on preset analog system parameters and the second data information of wells based on preset analog system parameters.

[0075] Data preprocessing includes using a "mean substitution" method to preprocess data with missing values. When the number of missing values ​​is only a small fraction of the total, missing values ​​(rows) can be directly deleted. However, if the proportion of missing values ​​is relatively large, this method of directly deleting missing values ​​will lose important information such as the mean.

[0076] According to the formula After standardizing matrix X, we obtain matrix Y, where X = {x} ij}, Y={y ij}, i=1, 2,...n, j=1, 2,...p, , .

[0077] Perform KMO and Bartlett tests. The criteria for performing KMO and Bartlett's test of sphericity are shown in Table 3.

[0078] Table 3. Criteria for Testing the Applicability of KMO and Bartletti Factor Analysis

[0079]

[0080] According to the formula Find the correlation coefficient matrix R.

[0081] According to the formula Find the eigenvalues ​​of the coefficient matrix. and eigenvector I p .

[0082] The eigenvector a is calculated using the above formula. i and a i =(a i1 a i2 , ...a ip Let i = 1, 2, ..., p, and sort the eigenvalues ​​from largest to smallest to obtain a set of principal components F. i : The principal components can then be determined.

[0083] The contribution rate of the kth principal component is calculated as follows: The cumulative contribution rate of the first k principal components is: .

[0084] The appropriate principal components are selected based on their contribution. Preferably, principal components with eigenvalues ​​greater than 1 and a cumulative variance contribution rate greater than 85% are retained, as they are considered to sufficiently reflect the information of the original variables.

[0085] Step S13: Obtain the second data information of the measure well based on the preset analog system parameters.

[0086] Step S14: Based on the first data information, the second data information, and the analogy parameter weights, calculate the similarity between each historical well and the well that received a treatment using a similarity calculation method.

[0087] Step S142, in calculating the similarity between historical intervention wells and intervention wells using a similarity calculation method, the similarity calculation method includes a comprehensive feature method, which includes: according to formula K i =1-H i Calculate the similarity K between the treated well and the historical treated well. i Where H is based on the formula H i =(O i -O min ) / (O max -O min )Calculation, O i It is a comprehensive index.

[0088] When comparing the similarity between historical wells and newly treated wells, a common method is the comprehensive feature method. When calculating the similarity using the comprehensive feature method, the feature vector λ is first determined using PCA, where λ = λ1, λ2, ... λ. n Input profile control and wellbore parameter value X i X i =X 1i X 2i ...X ni Place the implemented wells in the first column, and arrange the historical implemented wells in order; according to O i =λ Xi T Calculate the composite index O i The comprehensive index corresponding to each well is O, where O = O0, O1, ... O i According to formula H i =(O i -O min ) / (O max -O min Normalization is performed, where H = H0, H1, ... H m According to formula K i =1-H i Calculate the similarity K between the treated well and the historical treated well. i Then, the similarity K i Sort in descending order.

[0089] Step S15: Sort the similarity scores of each historical well and the current well by numerical value, and determine the historical well with the highest similarity score as the similar well.

[0090] Step S16: Obtain the historical profile control schemes of the similar wells and push them to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters, and drive construction parameters.

[0091] All other technical measures in this embodiment are the same as those in Embodiment 1, and therefore have all the above-mentioned beneficial effects, which will not be repeated here.

[0092] Example 4: Figures 2 to 4 As shown, embodiments of the present invention provide a method for recommending profile and drive parameters based on a similarity calculation model, including:

[0093] Step S11: Obtain first data information of several historical wells based on preset analog system parameters. The preset analog system parameters include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters.

[0094] Step S12: Calculate the analogy parameter weights of the preset analogy system parameters based on the preset analogy system parameters. .

[0095] When calculating the analogy parameter weights of the preset analogy system parameters, the methods for calculating the analogy parameter weights may include principal component analysis (PCA) and analytic hierarchy process (AHP).

[0096] Step S122: Apply the Analytic Hierarchy Process (AHP) to calculate the analogy parameter weights. The AHP method is a decision analysis method that combines qualitative and quantitative approaches to solve complex multi-objective problems. This method combines quantitative and qualitative analysis, using the decision-maker's experience to judge the relative importance of parameters related to the achievability of each objective, and reasonably assigns weights to each criterion of each decision option. The weights are then used to determine the order of merit of each option, making it effective for problems that are difficult to solve using quantitative methods. The Analytic Hierarchy Process includes the following steps:

[0097] Step S1221: Determine the target A (i.e., the similarity comprehensive evaluation index A in Table 4) and the evaluation factor set U (i.e., the analogy index parameter list C11~C36 in Table 4), as shown in Table 4.

[0098] Table 4. List of Analogy Indicator Parameters

[0099]

[0100] Step S1222: Construct the judgment matrix. Using the 1-9 scale method proposed by Saaty, compare the influence of each of the 16 factors in the evaluation factor set U on the target A in pairs. ij Indicate u i For u j The relative importance values, where j=1, 2, ..., n, n=16, u ij The values ​​are determined by referring to Table 5 to determine the matrix scale and its meaning.

[0101] Table 5. Judgment Matrix Scale and Its Meaning

[0102]

[0103] Based on the meanings of the symbols mentioned above, we can obtain the judgment matrix P, which is called the AU judgment matrix, i.e.:

[0104]

[0105] Step S1223: Calculate the importance ranking. Based on the AU matrix, find the eigenvector corresponding to the largest eigenvalue. The obtained eigenvector is the importance ranking of each evaluation factor, which is the weight allocation.

[0106] The methods for finding the eigenvectors of the AU determination matrix include the root method and the product method.

[0107] (1) Root method

[0108] I. According to the formula Calculate the product of the elements in each row of the AU judgment matrix. , where i, j = 1, 2, ..., n.

[0109] II. According to the formula ,calculate of Root , where i, j = 1, 2, ..., n.

[0110] III. According to the formula For vectors Perform normalization, where i, j = 1, 2, ..., n, and obtain the following results in sequence. This is the desired eigenvector.

[0111] IV. According to the formula Calculate the maximum eigenvalue λmax of the judgment matrix, where, For vectors The Each element, namely

[0112] (2) Sum-and-product method

[0113] Ⅰ. According to the formula , normalize each column of the A-U judgment matrix, where i, j = 1, 2,....n.

[0114] Ⅱ. According to the formula , sum each row of the judgment matrix after normalization for each column, where i, j = 1, 2,....n.

[0115] Ⅲ. According to the formula , perform normalization processing on the vector W = [w1, w2,..., w n T , where i = 1, 2,....n. The obtained in turn is the required eigenvector.

[0116] Ⅳ. According to the formula , calculate the maximum eigenvalue λmax of the A-U judgment matrix.

[0117] Step S1224, check the weight distribution. To check whether the weight distribution is reasonable, the formula can be used to perform consistency check on the judgment matrix. Where: CR is the random consistency ratio of the judgment matrix, CI is the general consistency index of the judgment matrix, and CI can be obtained from the formula CI = (λ max - n) / (n - 1), RI is the average random consistency index of the judgment matrix. For 1-9 order judgment matrices, the RI values are shown in Table 6:

[0118] Table 6 RI value table

[0119]

[0120] The CR value can use the preset value A as the judgment criterion. When CR < A, it is considered that the judgment matrix has satisfactory consistency, indicating that the weight distribution is reasonable; otherwise, the judgment matrix needs to be adjusted until satisfactory consistency is obtained. Preferably, when A = 0.1, it can make the CR value more suitable for checking the consistency of the judgment matrix.

[0121] Step S13, obtain the second data information of the measure well based on the preset analogy system parameters.

[0122] Step S14, according to the first data information, the second data information and the analogy parameter weights, use the similarity calculation method to calculate the similarity between each historical measure well and the measure well.

[0123] ​Step S143: A similarity calculation method is used to calculate the similarity between the implemented well and historical implemented wells. This similarity calculation method includes a multivariate decision-making method, which includes: [The following is a formula...] The similarity K between the g-th historical well and the well that was treated can be calculated. g Where g = 1, 2, ..., m, μ(x) gi W represents the fuzzy value of the g-th historical measure well. i The corresponding weights of each parameter in the preset analogy system parameters.

[0124] When comparing the similarity between historical profile control wells and current control wells, a common method is the multivariate decision method. Also known as fuzzy evaluation, the multivariate decision method introduces the concept of weights into Euclidean distance, moving beyond the absolute spatial distance between two points represented by Euclidean distance. When calculating the similarity of control wells using the multivariate decision method, it is assumed that there are m historical profile control and displacement wells to be evaluated, denoted as M1, M2...M... m Each historical well has n pieces of first data information based on preset analogy system parameters, namely C 11 C 12 ...C 36 The first data information of n in the data is used to represent the i-th index of the g-th well with the corresponding fuzzy value μ(x). ki The expression is described by ), where k = 1, 2, ..., m, i = 1, 2, ..., n. μ is normalized to be within the range of 0-1. According to the formula... It can calculate m historical measures and n-dimensional multivariate decision R. mn, According to the formula The similarity K between the g-th historical well and the well that was treated can be calculated. g Where g = 1, 2, ..., m, W i To define the weights of each parameter in the preset analogy system parameters, W i The similarity can be calculated using the Analytic Hierarchy Process (AHP). See Table 7 for an example table of similarity between the current measure wells and historical measure wells using the multivariate decision-making method.

[0125] Table 7. Examples of similarity between treated wells and historical treated wells using the multivariate decision-making method.

[0126]

[0127] Step S15: Sort the similarity scores of each historical well and the current well by numerical value, and determine the historical well with the highest similarity score as the similar well.

[0128] Step S16: Obtain the historical profile control schemes of the similar wells and push them to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters, and drive construction parameters.

[0129] All other technical measures in this embodiment are the same as those in Embodiment 1, and therefore have all the above-mentioned beneficial effects, which will not be repeated here.

[0130] Example 5: Corresponding to the analogy recommendation method for profile and drive parameters based on a similarity calculation model provided in Examples 1 to 4, such as... Figure 6 As shown, embodiments of the present invention also provide a profile and drive parameter analogy recommendation device based on a similarity calculation model, comprising:

[0131] The first acquisition module 21 is used to acquire first data information of several historical wells based on preset analog system parameters. The preset analog system parameters include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters.

[0132] The second acquisition module 23 is used to acquire second data information of the measure well based on preset analog system parameters.

[0133] The second calculation module 24 is used to calculate the similarity between each historical well and the well based on the first data information and the second data information using a similarity calculation method.

[0134] The determination module 25 is used to sort the similarity between each historical well and the current well by numerical value, and determine the historical well with the highest similarity as the similar well.

[0135] The third acquisition module 26 is used to acquire the historical profile control schemes of the similar wells and push them to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters and drive construction parameters.

[0136] Preferably, in the second calculation module, the similarity calculation method includes the Euclidean distance method, which includes: according to the formula Calculate the Euclidean distance between two points, object X and object Y, in n-dimensional space. The Euclidean distance is used to characterize the similarity.

[0137] Preferred, such as Figure 7 As shown, after the first acquisition module 21, it further includes: a first calculation module 22, used to calculate the analogy parameter weights of the preset analogy system parameters according to the preset analogy system parameters;

[0138] The second calculation module 24 further includes: calculating the similarity between each historical well and the well using a similarity calculation method based on the analogy parameter weights; the similarity calculation method includes a comprehensive feature method, which includes: according to formula K i =1-Hi Calculate the similarity K between the treated well and the historical treated well. i Where H is based on the formula H i =(O i -O min ) / (O max -O min ) Calculation, O i It is a comprehensive index.

[0139] Preferred, such as Figure 7 As shown, after the first acquisition module 21, the system further includes: a first calculation module 22, used to calculate the analogy parameter weights of the preset analogy system parameters based on the preset analogy system parameters. ;

[0140] The second calculation module 24 further includes: further calculating the analogy parameter weights. The similarity between historical wells and current wells is calculated using a similarity calculation method; the similarity calculation method includes a multivariate decision-making method, which includes: according to the formula The similarity K between the g-th historical well and the well that was treated can be calculated. g Where g = 1, 2, ..., m, μ(x) gi W represents the fuzzy value of the g-th historical measure well. i The corresponding weights of each parameter in the preset analogy system parameters.

[0141] Preferably, the third acquisition module 26 is used to acquire the historical profile control schemes of the similar wells and push them to the terminal device, and further includes: acquiring the profile control effect data of the similar wells and pushing it to the terminal device, wherein the profile control effect data includes the oil increase, economic benefits, and input-output ratio of the similar wells.

[0142] By adopting the technical solution provided in Embodiment 5, this invention can quickly calculate the similarity between the first data information of several historical intervention wells based on preset analogy system parameters and the second data information of intervention wells based on preset analogy system parameters through a similarity calculation method. By sorting the similarity scores, the historical intervention well with the highest similarity score can be identified as the similar intervention well with the best intervention effect. By obtaining data such as historical profile control schemes and driving effect data of similar intervention wells, decision-making basis can be quickly provided for the design scheme of intervention wells, and support can be provided for predicting the driving effect.

[0143] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.

[0144] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for recommending profile and drive parameters based on a similarity calculation model, characterized in that, include: Acquire first data information of several historical wells based on preset analogy system parameters, which include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters. Obtain second data information of the well based on preset analog system parameters; Based on the first data information and the second data information, the similarity between each historical well and the well to be treated is calculated using a similarity calculation method; The similarity scores of each historical well and the current well are sorted by numerical value, and the historical well with the highest similarity score is identified as the similar well. The historical profile control schemes of similar wells are obtained and pushed to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters, and drive construction parameters.

2. The method for recommending profile and drive parameters based on a similarity calculation model as described in claim 1, characterized in that, Based on the first data information and the second data information, a similarity calculation method is used to calculate the similarity between each historical well and the applied well. This similarity calculation method includes the Euclidean distance method, which comprises: according to the formula... Calculate the Euclidean distance between two points, object X and object Y, in n-dimensional space. The Euclidean distance is used to characterize the similarity.

3. The method for recommending profile and drive parameters based on a similarity calculation model as described in claim 1, characterized in that, After obtaining the first data information of several historical wells based on preset analogy system parameters, the method further includes: calculating the analogy parameter weights of the preset analogy system parameters according to the preset analogy system parameters; According to the first data information and the second data information, the similarity between each historical treatment well and the treatment well is calculated by using a similarity calculation method, and the similarity calculation method further comprises a comprehensive characteristic method, and the comprehensive characteristic method comprises: calculating the similarity K between the treatment well and the historical treatment well according to the formula K = 1-H, wherein H is calculated according to the formula H = (O-O) / (O-O), and O is a comprehensive index. i i i i i min max min i ​​​​​​​​​ 4. The method for recommending profile and drive parameters based on a similarity calculation model as described in claim 1, characterized in that, After acquiring the first data information of several historical wells based on preset analogy system parameters, the method further includes: calculating the analogy parameter weights of the preset analogy system parameters according to the preset analogy system parameters. ; The similarity calculation method for each historical well and the applied well, based on the first data information and the second data information, further includes: calculating the similarity between the two wells according to the analogy parameter weights. The similarity between historical wells and current wells is calculated using a similarity calculation method; the similarity calculation method includes a multivariate decision-making method, which includes: according to the formula The similarity K between the g-th historical well and the well that was treated can be calculated. g Where g = 1, 2...m, μ(x) gi W represents the fuzzy value of the g-th historical measure well. i The corresponding weights of each parameter in the preset analogy system parameters.

5. The method for recommending profile and drive parameters based on a similarity calculation model as described in claim 1, characterized in that, The process of acquiring historical profile control schemes for similar wells and pushing them to the terminal device also includes: acquiring and pushing the profile control effect data of similar wells to the terminal device, wherein the profile control effect data includes the oil production, economic benefits, and input-output ratio of the similar wells.

6. A device for recommending profile and drive parameters based on a similarity calculation model, characterized in that, include: The first acquisition module is used to acquire first data information of several historical wells based on preset analog system parameters. The preset analog system parameters include target reservoir characteristic parameters, target fluid property parameters, well group injection and production dynamic characteristic parameters, and dominant channel characteristic parameters. The second acquisition module is used to acquire second data information of the measure well based on preset analog system parameters; The second calculation module is used to calculate the similarity between each historical well and the well based on the first data information and the second data information using a similarity calculation method. The determination module is used to sort the similarity between each historical well and the current well by numerical value, and determine the historical well with the highest similarity as the similar well. The third acquisition module is used to acquire the historical profile control schemes of the similar wells and push them to the terminal device. The historical profile control schemes include profile control design parameters, slug design parameters, and drive construction parameters.

7. The profile adjustment and drive parameter analogy recommendation device based on a similarity calculation model as described in claim 6, characterized in that, In the second calculation module, the similarity calculation method includes the Euclidean distance method, which includes: according to the formula Calculate the Euclidean distance between two points, object X and object Y, in n-dimensional space. The Euclidean distance is used to characterize the similarity.

8. The profile adjustment and drive parameter analogy recommendation device based on a similarity calculation model as described in claim 6, characterized in that, After the first acquisition module, the system further includes: a first calculation module, used to calculate the analogy parameter weights of the preset analogy system parameters based on the preset analogy system parameters; The second calculation module further includes: calculating the similarity between each historical well and the well to be treated based on the analogy parameter weights using a similarity calculation method; the similarity calculation method includes a comprehensive feature method, which includes: based on formula K i =1-H i Calculate the similarity K between the treated well and the historical treated well. i Where H is based on the formula H i =(O i -O min ) / (O max -O min ) Calculation, O i It is a comprehensive index.

9. The profile adjustment and drive parameter analogy recommendation device based on a similarity calculation model as described in claim 6, characterized in that, Following the first acquisition module, the system further includes: a first calculation module, configured to calculate the analogy parameter weights of the preset analogy system parameters based on the preset analogy system parameters. ; The second calculation module further includes: further calculating based on the analogy parameter weights. The similarity between historical wells and current wells is calculated using a similarity calculation method; the similarity calculation method includes a multivariate decision-making method, which includes: according to the formula The similarity K between the g-th historical well and the well that was treated can be calculated. g Where g = 1, 2...m, μ(x) gi W represents the fuzzy value of the g-th historical measure well. i The corresponding weights of each parameter in the preset analogy system parameters.

10. The profile adjustment and drive parameter analogy recommendation device based on a similarity calculation model as described in claim 6, characterized in that, The third acquisition module is used to acquire historical profile control schemes of similar wells and push them to the terminal device. It also includes: acquiring the profile control effect data of similar wells and pushing it to the terminal device. The profile control effect data includes the oil increase, economic benefits, and input-output ratio of the similar wells.

Images