Method, device, equipment and storage medium for estimating initial production value of oil well

Abstract

The application provides an oil well initial production value estimation method, device, equipment and storage medium. The method comprises the following steps: collecting logging data of an oil well to be exploited; determining oil-bearing property estimation data of the oil well to be exploited based on oil-bearing property estimation characteristic data of an exploited oil well and the logging data of the oil well to be exploited; inputting the oil-bearing property estimation data of the oil well to be exploited into an oil well oil-bearing property estimation model to obtain oil-bearing property distribution of the oil well to be exploited; determining perforation thickness of the oil well to be exploited according to the oil-bearing property distribution of the oil well to be exploited; determining fracturing sanding amount of the oil well to be exploited according to reservoir section fracturing sanding amount data of the exploited oil well and the perforation thickness of the oil well to be exploited; determining initial production estimation data of the oil well to be exploited based on initial production estimation characteristic data of the oil well to be exploited; and inputting the initial production estimation data of the oil well to be exploited into an oil well initial production estimation model to obtain an initial production estimation value of the oil well to be exploited. The method improves the accuracy of oil well initial production value estimation.

Description

Technical Field

[0001] This application relates to the field of petroleum exploration, and in particular to a method, apparatus, equipment and storage medium for estimating the initial production value of an oil well. Background Technology

[0002] Before allocating production to an oil well, it is necessary to know the initial production status of the well to be exploited, so as to select appropriate pumping units, nozzles, and tubing. Oilfield field data is large in volume, diverse in type, and contains a lot of missing and abnormal data. Data processing often consumes a lot of time and effort. However, data-driven and big data-based data processing methods and data modeling algorithms can efficiently and accurately extract feature data from massive amounts of data and establish an oil well initial production prediction model. This is a new method for conducting oil well initial production prediction research.

[0003] Existing technology involves establishing an initial production prediction model for oil wells based on data from other exploited oil wells within the same block, using data-driven and big data processing methods, and then estimating the initial production value of the oil wells to be exploited.

[0004] However, the data from other exploited wells in the same block cannot be completely matched with the situation of the wells to be exploited, which will result in inaccurate initial production value estimates for the wells to be exploited, and further lead to the problem of mismatch between production and production allocation for the wells to be exploited. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, and storage medium for estimating the initial production value of an oil well, in order to solve the problem of inaccurate estimation of the initial production value of an oil well to be exploited.

[0006] In a first aspect, this application provides a method for estimating the initial production value of an oil well, including:

[0007] Collect logging data from oil wells to be exploited;

[0008] Based on the oil-bearing potential characteristic data of the already produced oil wells in the block where the oil well to be produced is located and the well logging data of the oil well to be produced, the oil-bearing potential data of the oil well to be produced is determined.

[0009] The oil-bearing potential prediction data of the oil well to be exploited is input into the oil well oil-bearing potential prediction model to obtain the oil-bearing potential distribution data of the oil well in the depth direction of the logging data.

[0010] Based on the oil-bearing distribution data of the well to be exploited in the depth direction of the logging data, the perforation thickness data of the well to be exploited is determined.

[0011] Based on the fracturing proppant addition data of the reservoir section of the already produced oil well, the perforation thickness data of the already produced oil well, and the perforation thickness data of the oil well to be produced, the fracturing proppant addition data of the reservoir section of the oil well to be produced is determined.

[0012] Based on the initial production prediction characteristic data of the wells already produced in the block where the well to be developed is located, the logging data of the well to be developed, the perforation thickness data of the well to be developed, and the reservoir section fracturing and proppant addition data of the well to be developed, the initial production prediction data of the well to be developed is determined.

[0013] The initial production estimate data of the oil well to be exploited is input into the oil well initial production estimate model to obtain the initial production estimate value of the oil well to be exploited.

[0014] Secondly, this application provides an apparatus for estimating the initial production value of an oil well, comprising:

[0015] The calculation module is used to determine the estimated oil content of the oil well to be developed based on the oil content prediction characteristic data of the oil wells already developed in the block where the oil well to be developed is located and the logging data of the oil well to be developed.

[0016] The calculation module is also used to input the oil-bearing prediction data of the oil well to be exploited into the oil well oil-bearing prediction model to obtain the oil-bearing distribution data of the oil well to be exploited in the depth direction of the logging data.

[0017] The calculation module is also used to determine the perforation thickness data of the oil well to be developed based on the oil-bearing distribution data of the well logging data in the depth direction; and to determine the reservoir fracturing sand addition data of the oil well to be developed based on the reservoir section fracturing sand addition data of the oil well already developed, the perforation thickness data of the oil well already developed, and the perforation thickness data of the oil well to be developed.

[0018] The calculation module is also used to determine the initial production estimate data of the oil well to be developed based on the initial production prediction characteristic data of the oil wells already developed in the block where the oil well to be developed is located, the logging data of the oil well to be developed, the perforation thickness data of the oil well to be developed, and the reservoir section fracturing sand addition data of the oil well to be developed.

[0019] The determination module is used to input the initial production prediction data of the oil well to be exploited into the oil well initial production prediction model to obtain the initial production prediction value of the oil well to be exploited.

[0020] Thirdly, this application provides an apparatus for estimating the initial production value of an oil well, comprising:

[0021] Processor, memory, communication interface;

[0022] The memory is used to store the executable instructions of the processor;

[0023] The processor is configured to execute the method for estimating the initial production value of an oil well as described in the first aspect above by executing the executable instructions.

[0024] Fourthly, this application provides a readable storage medium, comprising:

[0025] When the computer program is executed by the processor, it implements the method for estimating the initial production value of the oil well as described in the first aspect above.

[0026] The method, apparatus, equipment, and storage medium for estimating the initial production value of oil wells provided in this application first estimate the oil-bearing distribution of the oil well to be exploited using logging data, thereby improving the accuracy of oil-bearing distribution estimation. Furthermore, based on the logging data of the oil well to be exploited and the aforementioned estimated oil-bearing distribution, the initial production value of the oil well to be exploited is estimated, achieving an even more accurate estimation of the initial production value. Attached Figure Description

[0027] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0028] Figure 1 A schematic diagram of the method for estimating the initial production value of an oil well provided in this application embodiment;

[0029] Figure 2 A flowchart illustrating the process for determining the oil-bearing potential characteristics of exploited oil wells, as provided in this application embodiment;

[0030] Figure 3 A flowchart illustrating the process of establishing an oil well oil content prediction model provided in this application embodiment;

[0031] Figure 4 A flowchart illustrating the process for determining the initial production prediction characteristics of an exploited oil well, as provided in an embodiment of this application.

[0032] Figure 5 A flowchart illustrating the process of establishing the oil well initial production prediction model provided in this application embodiment;

[0033] Figure 6 A schematic diagram of the structure of the oil well initial production estimation device provided in the embodiments of this application;

[0034] Figure 7 A schematic diagram of the structure of the oil well initial production estimation device provided in the embodiments of this application.

[0035] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0036] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0037] Existing technologies for predicting the initial production of oil wells mainly rely on data from other exploited oil wells within the same block. However, due to the large quantity and variety of data from these other exploited oil wells, which includes many missing and abnormal data, data processing often consumes a lot of time and effort. Therefore, a data-driven and big data-based data processing method is adopted to establish an oil well initial production prediction model and to predict the initial production value of the oil wells to be exploited.

[0038] The method, apparatus, equipment, and storage medium for predicting the initial production of oil wells provided in this application first predict the oil-bearing distribution of the oil well to be developed based on the logging data of the oil well to be developed and the data of already developed oil wells in the block where the oil well to be developed is located, using an oil well oil-bearing prediction model. Then, based on the predicted oil-bearing distribution of the oil well to be developed and the data of already developed oil wells in the block where the oil well to be developed is located, the initial production value of the oil well to be developed is predicted using the same oil well oil-bearing prediction model. This application, in the process of predicting the initial production value of the oil well to be developed, not only refers to the data of already developed oil wells in the block where the oil well to be developed is located, but also incorporates the logging data of the oil well to be developed, solving the problem that the data of other developed oil wells in the same block cannot completely match the situation of the oil well to be developed, thus improving the accuracy of the initial production value prediction of the oil well to be developed. Furthermore, the initial production value estimate in this application is based on the logging data of the oil well to be exploited and the aforementioned estimated oil-bearing distribution, which makes the initial production value estimate of the oil well to be exploited more accurate.

[0039] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0040] Figure 1This is a schematic diagram of the method for estimating the initial production value of an oil well provided in the first embodiment of this application.

[0041] like Figure 1 As shown, the method for estimating the initial production value of an oil well provided in this embodiment may include the following steps:

[0042] Step S101: Collect logging data from the oil well to be exploited.

[0043] Specifically, the first step is to collect logging data from the oil wells to be exploited.

[0044] Optionally, to improve the accuracy of subsequent steps, the logging data of the well to be exploited can be preprocessed. This preprocessing may include: removing outliers from the logging data and filling gaps with the average value of the logging data from the well to be exploited.

[0045] Specifically, the logging data collected from the oil wells to be exploited, after outlier screening and gap filling, constitutes the data for the oil wells to be exploited. This data can be represented in various formats, for example, it can be expressed as... It can be represented in the following form:

[0046]

[0047] Among them, there are n oil wells to be exploited. w One logging data point, p-type logging data It refers to the nth oil well to be exploited w The value of the p-th type of logging data point.

[0048] Step S102: Based on the oil-bearing characteristics of the already exploited oil wells in the block where the oil well to be exploited is located and the logging data of the oil well to be exploited, determine the oil-bearing prediction data of the oil well to be exploited.

[0049] Optionally, before performing step S102, the oil well data to be exploited generated in step S101 can be standardized using, for example, a zero-mean normalization method, to obtain standardized data. The mean and standard deviation in the standardization process are derived from the mean and standard deviation of the oil-bearing potential characteristic data of already produced wells within the block where the oil well to be developed is located. ij This refers to the value of the i-th type of logging data at the j-th logging data point of the oil well to be exploited. This refers to the standardized result of the i-th type of logging data at the j-th logging data point of the oil well to be exploited.

[0050] To obtain the oil-bearing potential feature data of a well to be developed, we can first extract the oil-bearing potential feature data of already developed wells within the block where the well to be developed is located. The principal component weights obtained during this process can be represented, for example, as a vector [a1 a2 … a m The principal components are represented in the form [a1 a2 … a], where the corresponding weights are [a1 a2 … a]. m Combined with the standardized data of the oil wells to be exploited mentioned above, principal component data F of the oil wells to be exploited is generated. n As shown below:

[0051]

[0052] in, It is the principal component data of the j-th logging data point of the oil well to be exploited, and the i-th type of logging data.

[0053] Further, depth data from logging points of the oil well to be exploited can be extracted. This depth data can be combined with the aforementioned principal component data to generate principal component data of the oil well to be exploited, corresponding to the depth of the logging points, i.e., the oil-bearing capacity prediction data F of the oil well to be exploited. n d As shown below:

[0054]

[0055] Where, d i This refers to the depth corresponding to the i-th logging data point of the oil well to be exploited.

[0056] Step S103: Input the oil-bearing prediction data of the oil well to be exploited into the oil well oil-bearing prediction model to obtain the oil-bearing distribution data of the oil well in the depth direction of the logging data.

[0057] Specifically, the oil-bearing potential dataset of the oil well to be exploited can be input into the oil well oil-bearing potential prediction model, which will output the oil-bearing potential corresponding to the depth of the logging data points. The prediction results are as follows:

[0058]

[0059] Wherein, `prediction` is the set of oil-bearing prediction results, i.e., the oil-bearing distribution data of the oil well to be exploited along the depth direction of the logging data; `SVMmodel` is the oil-bearing prediction model for the oil well; and `d`... i This refers to the depth corresponding to the i-th logging point of the oil well to be exploited, y i For the oil-bearing prediction result of the i-th logging data point of the oil well to be exploited, y i The value is 0 (no oil) or 1 (with oil).

[0060] Step S104: Determine the perforation thickness data of the oil well to be exploited based on the oil-bearing distribution data in the depth direction of the logging data.

[0061] The perforation thickness is equal to the total thickness of all oil-bearing layers in the well. Therefore, the thickness of each oil-bearing layer in the well to be exploited can be calculated based on the oil-bearing distribution data in the depth direction of the logging data. Then, the thicknesses of each oil-bearing layer are added together to obtain the total oil-bearing layer thickness, which is the perforation thickness data. For example: when the oil-bearing prediction result y... i When the value is 1, the formation corresponding to the i-th logging data point is an oil-bearing layer. The depth of the formation corresponding to the logging data points where the oil-bearing prediction result is consecutively 1 is extracted, which is the thickness of the oil-bearing layer in the well to be exploited. For example: when the oil-bearing prediction result is from... arrive Both are 1, and n i <n j , hour, arrive This is an oil-bearing layer of an oil well to be developed, and the thickness of this oil-bearing layer is...

[0062] The thickness of all oil-bearing layers in the well to be exploited is calculated based on the oil-bearing distribution data along the depth direction of the logging data. These thicknesses are then summed to obtain the perforation thickness data (hp) of the well. new .

[0063] Step S105: Based on the fracturing sand addition data of the reservoir section of the well that has been produced, the perforation thickness data of the well that has been produced, and the perforation thickness data of the well to be produced, determine the fracturing sand addition data of the reservoir section of the well to be produced.

[0064] Specifically, the fracturing sand addition intensity Sd of the block where the wells are located is first calculated based on the fracturing sand addition data of the reservoir section of the wells and the perforation thickness data of the wells. This is the average sand addition per unit perforation thickness, as shown in the following formula:

[0065]

[0066] Where N is the number of oil wells already in production within the block, and Vs i hp is the amount of fracturing sand added to the i-th produced oil well. i Let be the perforation thickness of the i-th exploited oil well.

[0067] Secondly, the perforation thickness data of the oil well to be developed, determined in step 104, can be multiplied by the fracturing sand injection intensity Sd of the block to obtain the fracturing sand injection amount data Vs of the oil well to be developed. new The calculation formula is as follows:

[0068] Vs new =hp new ×Sd (6)

[0069] Step S106: Based on the initial production prediction characteristic data of the wells already produced in the block where the well to be produced is located, the logging data of the well to be produced, the perforation thickness data of the well to be produced, and the reservoir section fracturing and sand addition data of the well to be produced, determine the initial production prediction data of the well to be produced.

[0070] Specifically, firstly, based on the data of the oil wells to be exploited in step S101... Calculate the average value of logging data from the wells to be developed, column by column; based on the generation process of initial production prediction characteristic data of already produced wells within the block where the wells to be developed are located, algorithms such as the lasso regression model can be used to optimize the average value of logging data from the wells to be developed, thus obtaining the optimized average value of logging data from the wells to be developed. The formula is as follows:

[0071]

[0072] Where t is the number of the optimized average values.

[0073] Then, the average value of the logging data from the selected oil wells to be exploited is calculated. The perforation thickness hp calculated in step 104 new The amount of fracturing sand added to the oil well to be exploited, Vs, calculated in step S105. new Combining these, we obtain the initial production estimate data (newdata) for the oil wells to be exploited, as shown in the following formula:

[0074]

[0075] Step S107: Input the initial production prediction data of the oil well to be exploited into the oil well initial production prediction model to obtain the initial production prediction value of the oil well to be exploited.

[0076] Specifically, before performing step S107, the initial production forecast data of the oil wells to be exploited generated in step S106 can be standardized, wherein the mean value... and variance σ j The mean and variance of the initial production estimate characteristic data of existing wells within the block where the oil well to be developed is used. The standardized initial production estimate data is denoted as newdata′, and its formula is shown below:

[0077]

[0078] Specifically, the standardized initial production prediction dataset of the oil well to be developed can be input into the initial production prediction model, and the initial production prediction value (prod) of the oil well to be developed can be output.new The formula is shown below:

[0079] prod new =RFmodel(newdata′) (10)

[0080] RFmodel is the trained oil well initial production prediction model.

[0081] The method for estimating the initial production value of an oil well provided in this embodiment first collects logging data of the oil well to be developed and combines it with data from already developed oil wells within the block to estimate the oil-bearing distribution of the oil well to be developed. Then, it further uses this estimated oil-bearing distribution data, combined with the data from already developed oil wells within the block, to estimate the initial production value of the oil well to be developed. This embodiment, in estimating the initial production value of the oil well to be developed, not only references data from already developed oil wells within the block but also incorporates logging data from the oil well to be developed. This solves the problem that data from other developed oil wells within the same block cannot perfectly match the situation of the oil well to be developed, thus improving the accuracy of the initial production value estimation.

[0082] Figure 2 This is a flowchart illustrating the process of determining the oil-bearing potential characteristics of an exploited oil well, as provided in the second embodiment of this application.

[0083] like Figure 2 As shown, the process for determining the oil-bearing potential characteristics of exploited oil wells provided in this embodiment may include the following steps:

[0084] Step S201: Collect reservoir data of already exploited oil wells in the block where the oil well to be exploited is located. The reservoir data includes: logging data, perforation thickness data, oil testing data, fracturing sand addition data, and production data.

[0085] Specifically, an oil production area may include multiple blocks, each containing multiple exploited wells and unexploded wells. Due to the similarity in geological characteristics within the same block, when estimating the initial production of unexploded wells within a block, data from the reservoir sections of exploited wells within that block can be referenced. This data includes, but is not limited to: well logging data, perforation thickness data, oil testing data, fracturing and proppant injection data, and production data. Well logging data includes various types of data, including but not limited to: spontaneous potential (SP) data, natural gamma ray (GR) data, acoustic transit time (AC) data, resistivity (RT) data, permeability (PERM) data, porosity (POR) data, clay content (VSH) data, and water saturation (SW) data.

[0086] Specifically, firstly, reservoir data of already exploited oil wells within the block where the oil well to be exploited is located are collected. Then, the logging data of the already exploited oil well reservoir sections are preprocessed. This preprocessing may include, but is not limited to, removing outliers from the logging data of the already exploited oil well reservoir sections and filling gaps with the average value of the logging data of that reservoir section.

[0087] Step S202: Merge the logging data of the exploited oil wells according to the data type.

[0088] In step S201, the preprocessed logging data of the exploited oil well reservoir section is merged according to the logging data type to form vertically merged logging data of the exploited oil well reservoir section.

[0089] Specifically, assuming all exploited oil wells have a total of n logging data points in the reservoir section, the logging data point samples of all wells can be vertically merged according to the logging data type to obtain n samples containing all data points of all wells. Each sample point has p feature indicators, i.e., p types of logging data. The logging data of all exploited oil wells in the reservoir section can be constructed into a sample matrix of size n×p, i.e., the merged logging data x of the exploited oil wells. matrix As shown below:

[0090]

[0091] Where x np The data is the p-th type of well logging data for the nth sample.

[0092] Step S203: Use Principal Component Analysis (PCA) to analyze the logging data of the merged exploited oil wells and obtain the principal component weight values.

[0093] First, the logging data of the merged exploited oil wells is standardized to obtain standardized logging data of the merged exploited oil wells.

[0094] Specifically, the mean value can be calculated column-wise from the sample matrix formed by the merged logging data of the exploited oil wells in step S202. in It is the sample matrix x matrix The average value of column j, i.e., the average value of the j-th type of logging data from all exploited oil well reservoir sections; the standardized data X can be calculated using, for example, a zero-mean normalization method. ij , where X ij For the standardized data in the i-th row and j-th column, σ jThe variance in column j represents the combined logging data of the exploited oil wells, i.e., the standardized combined logging data sample matrix X of the exploited oil wells obtained after standardizing the original sample matrix. matrix :

[0095]

[0096] Secondly, principal component analysis (PCA) is used to optimize the standardized and merged logging data of exploited oil wells to obtain principal component weight values.

[0097] Specifically, the covariance matrix of the standardized merged logging data of the exploited oil wells is calculated, which is the correlation coefficient matrix of the merged logging data of the exploited oil wells.

[0098]

[0099] in It is a standardized sample matrix X matrix The average value of the i-th column, It is the sample matrix X matrix The average value of column j.

[0100] Calculate the eigenvalues ​​and eigenvectors of the correlation coefficient matrix of the merged well logging data of the exploited oil wells.

[0101] Calculate the eigenvalues ​​of the correlation coefficient matrix of the merged logging data of the exploited oil wells, and sort them in descending order: λ1≥λ2≥…≥λ p ≥0.

[0102] Calculate the eigenvectors of the correlation coefficient matrix of the merged well logging data of the exploited oil wells. The eigenvectors are the weight vectors corresponding to each principal component.

[0103] Feature vector:

[0104] Calculate the principal component contribution rate and the cumulative contribution rate;

[0105]

[0106] The weight vector corresponding to each principal component is combined with the cumulative contribution rate of the principal component to generate the principal component weight vector, i.e., the principal component weight value. Based on the cumulative contribution rate, the top m (m≤p) principal components with the largest contribution rates corresponding to features with a cumulative contribution rate exceeding 80% can be selected. The weights corresponding to these principal components are then filtered out, resulting in the principal component weight vector, i.e., the principal component weight value [a1 a2 … a m ].

[0107] Step S204: Obtain the oil-bearing prediction characteristic data of the exploited oil wells based on the principal component weight values ​​and the merged logging data.

[0108] Specifically, the principal component weight values ​​obtained in step S203 can be combined with the standardized and merged logging data X of the exploited oil wells. matrix Combining these components yields the principal components for oil content prediction, i.e., the oil content prediction feature data. In this embodiment, the oil content prediction feature data consists of n×m principal component data. For example: matrix X matrix The j-th principal component calculated from the data in the i-th row is F. i ,

[0109] F ij =a 1j X i1 +a 2j X i2 +…+a pj X ip (i=1,2,…,n; j=1,2,…,m) (14)

[0110] Among them, oil content reflects the possibility of oil-bearing in the test formation of the oil well reservoir section, and the further development of the oil well can be determined based on the oil content prediction.

[0111] The process of determining the oil-bearing characteristics of exploited oil wells provided in this embodiment uses principal component analysis to optimize the reservoir data of exploited oil wells in the block where the oil well to be exploited is located. This process selects more reliable model training data, improves the accuracy of oil-bearing prediction for the oil well to be exploited, and further increases the accuracy of initial production prediction for the oil well to be exploited.

[0112] Figure 3 This is a flowchart illustrating the process of establishing the oil well oil content prediction model provided in the third embodiment of this application.

[0113] like Figure 3 As shown, the process of establishing the oil well oil content prediction model provided in this embodiment may include the following steps:

[0114] Step S301: Based on the logging data, determine the oil-bearing characteristics of the exploited oil wells.

[0115] Specifically, based on well logging data from existing wells within the block where the oil well to be developed is located, the oil-bearing potential characteristics of the existing wells can be determined. For the specific process, please refer to steps S201-S204.

[0116] Step S302: Extract the depth data of the logging data points corresponding to the oil-bearing prediction feature data of the exploited oil wells, and generate oil-bearing prediction feature data of the exploited oil wells with depth labels.

[0117] This involves extracting the depth data of logging data points corresponding to the oil-bearing prediction feature data of already exploited oil wells, combining it with the oil-bearing prediction feature data of already exploited oil wells, and generating oil-bearing prediction feature data of already exploited oil wells with depth labels.

[0118] Specifically, the oil-bearing potential characteristic data of exploited oil wells consists of n×m principal component data. According to formula (14), the oil-bearing potential characteristic data of oil wells, i.e., the principal component data F, can be obtained.

[0119]

[0120] Among them, F nm It is the m-th principal component data in the n-th row of the principal component data F.

[0121] Specifically, each row of oil-bearing potential characteristic data, i.e., principal component data, corresponds to the depth data of well logging points in the reservoir section of an already produced oil well. By combining the depth of the well logging points in the reservoir section of each well with the oil-bearing potential characteristic data of the already produced oil well, i.e., the principal component data F, we obtain the oil-bearing potential characteristic data F of the already produced oil well with depth labels. dep .

[0122]

[0123] Among them, dep n It is the depth of the nth data point.

[0124] Step S303: Mark the oil-bearing characteristics of the extracted oil wells with depth labels by marking them with oil-bearing labels based on the oil testing data of the reservoir section of the extracted oil wells.

[0125] The oil testing data from exploited oil well reservoirs can classify formations into oil-bearing layers, poor-oil-bearing layers, oil-water co-existing layers, water-bearing layers, and dry layers. Based on the oil-bearing characteristics of these formations, they can be divided into two categories: oil-bearing layers and oil-free layers. Oil-bearing layers, poor-oil-bearing layers, and oil-water co-existing layers are oil-bearing layers in the oil testing data from exploited oil well reservoirs, while water-bearing layers and dry layers are oil-free layers. Oil-bearing layers and oil-free layers can be labeled with different tags; for example, oil-bearing layers can be labeled with "1," and oil-free layers can be labeled with "0."

[0126] Table 1 shows the oil-bearing labels corresponding to the formations.

[0127] Table 1

[0128]

[0129] Specifically, based on the above-mentioned stratigraphic classification method, the oil-bearing labels corresponding to the oil-bearing potential characteristic data of the exploited oil wells can be determined using the reservoir testing data of the exploited oil wells. The oil-bearing potential characteristic data of the exploited oil wells with depth labels obtained in step S303 are then labeled with oil-bearing labels.

[0130] However, since not all exploited oil well reservoir sections are testing sections, the process of labeling the oil-bearing potential characteristic data of exploited oil wells with depth labels does not automatically assign an oil-bearing label to all exploited oil wells with depth labels. For example, in this embodiment, in n rows of exploited oil well oil-bearing potential characteristic data F with depth labels... dep There is only n f The row data corresponds to the testing section of the reservoir segment of an already exploited oil well, i.e., it carries an oil-bearing label. Therefore, it is necessary to extract the oil-bearing prediction feature data with the oil-bearing label. This data is then combined with the oil-bearing label corresponding to the formation to generate oil-bearing prediction feature data F of the exploited oil well, labeled with both oil-bearing and depth labels. labeldateset .

[0131]

[0132] in, It is the nth one with an oily label. f The depth of each data point; It is the nth one with an oily label. f Line, the m-th principal component data; It is the nth f The oil content label corresponding to each data point.

[0133] Step S304: Using the oil-bearing prediction feature data of the already exploited oil wells as features, and the corresponding oil-bearing labels and depth data as labels, the support vector machine algorithm is used to train the model to obtain the oil well oil-bearing prediction model.

[0134] This involves establishing an oil well oil-bearing prediction model based on the oil-bearing potential characteristics and testing data of already-exploited oil wells. Specifically, the oil-bearing potential characteristics of already-exploited oil wells can be used as features, and depth data obtained from these characteristics and oil-bearing labels obtained from testing data can be used as labels. A support vector machine (SVM) algorithm, for example, can be used to train the model, resulting in the oil well oil-bearing potential prediction model. By adjusting the training parameters of the SVM, multiple oil well oil-bearing potential prediction models can be trained. The model performance can be evaluated using methods such as the F1 score, and the model with the best prediction performance can be selected.

[0135] First, the oil content prediction feature data labeled with oil content and depth is split into an oil content prediction training set, an oil content prediction validation set, and an oil content prediction test set.

[0136] Specifically, the oil content prediction feature data labeled with oil content and depth can be split into an oil content prediction training set, an oil content prediction validation set, and an oil content prediction test set using a random splitting method, for example. The oil content prediction training set accounts for 80%, the oil content prediction test set accounts for 10%, and the oil content prediction validation set accounts for 10%.

[0137] Secondly, using the oil content prediction training set, multiple oil well oil content prediction models were trained by setting different combinations of training parameters.

[0138] Specifically, a support vector machine (SVM) model can be used to train an oil well oil content prediction model. During the training process, a kernel function is used to project the SVM model onto a high-dimensional plane, achieving linear separability.

[0139] In this embodiment, a Gaussian kernel function can be used as the kernel function of the support vector machine model for training the oil well oil content prediction model. Specifically, before training the oil well oil content prediction model using the Gaussian kernel function as the kernel function of the support vector machine model, training parameters can be set. The training parameters may include: penalty term coefficient C and Gaussian kernel function coefficient γ. The penalty term coefficient is used to prevent overfitting and enhance the generalization ability of the model. The optimized value range is [0.1, 10]. Where σ is the Gaussian kernel bandwidth, and the optimized value range is [0.01, 10]. Seven × eight experimental combinations can be obtained, with C = 0.1, 0.2, 0.5, 1.0, 2.0, 5.0, 10.0 and γ = 0.01, 0.1, 0.2, 0.5, 1.0, 2.0, 5, 10. For each parameter combination, a support vector machine model with a Gaussian kernel function is set. Oil well oil content prediction models are then trained using an oil content training set, generating 56 oil well oil content prediction models.

[0140] Finally, the prediction performance of the oil well oil content prediction model was evaluated using the oil content prediction validation set and the oil content prediction test set, and the oil well oil content prediction model with the best prediction performance was selected.

[0141] Specifically, evaluation methods such as the F1 score can be used to evaluate the prediction effect of the oil well oil content prediction model. F1 refers to the F1 score, which is an indicator used to measure the accuracy of the binary classification model. The larger the F1 value, the better the model prediction effect.

[0142]

[0143] Among them, TP represents the true cases, which are the number of samples where the actual oil content label is "1" and the predicted oil content label is also "1"; n f TN represents the total number of samples; TN represents the true negative examples, which are the number of samples where the actual oil content label is "0" and the predicted oil content label is also "0".

[0144] Input the relevant data of the oil-bearing validation set into the oil-bearing prediction model of the above oil well to obtain the predicted oil-bearing result label; by inputting the predicted oil-bearing result label and the actual oil-bearing label into formula (18), the F1 score values ​​of different oil well oil-bearing prediction models are obtained respectively; select the oil well oil-bearing prediction model with the largest F1 score value as the preferred oil well oil-bearing prediction model; input the relevant data of the oil-bearing test set into the preferred oil well oil-bearing prediction model to obtain the predicted oil-bearing result label; calculate the F1 score value of the preferred oil well oil-bearing prediction model to verify the optimization effect.

[0145] The oil well oil content prediction model provided in this embodiment can be used to predict the distribution of oil content in a well to be exploited along the depth direction of logging data points, i.e., the oil content distribution of the well to be exploited. Furthermore, the perforation location of the well to be exploited can be determined and the preset perforation thickness can be calculated based on the oil content distribution. In the process of establishing this oil well oil content prediction model, not only were optimized oil content prediction feature data selected, but different oil well oil content prediction models were generated by adjusting the parameter combinations used in model training. Finally, the oil well oil content prediction model with the best prediction effect was selected through an optimization method, thereby improving the accuracy of oil well oil content prediction.

[0146] Figure 4 This is a flowchart illustrating the process for determining the initial production prediction characteristic data of an exploited oil well, as provided in the fourth embodiment of this application.

[0147] like Figure 4 As shown, the process for determining the initial production prediction characteristic data of an exploited oil well, as provided in this embodiment, may include the following steps:

[0148] Step S401: Based on the production data, determine the average production rate of the wells that have been exploited for the initial set time.

[0149] Specifically, the production data of the wells in the already produced oilfield is extracted based on the initial production time set for, for example, three months. The average production of the wells in the already produced oilfield over the initial production time is then calculated. This average production of the wells in the initial production time is used as the initial production value data of the wells in the already produced oilfield.

[0150] Step S402: Determine the average value of logging data for the exploited oil well reservoir section based on the logging data type.

[0151] Specifically, logging data for the reservoir section of each exploited oil well is collected. The average logging data for each exploited oil well reservoir section is calculated according to the logging data type, resulting in the average logging data for each exploited oil well reservoir section. The average value of the j-th type of logging data for a single well is... The average logging data of the kth well that has been put into production is: Where, n well This represents the number of logging data points in the reservoir section of this single well.

[0152] The order of this step and step S401 is not limited.

[0153] Step S403: Based on the average production rate of the wells at the initial production set time, a regression model is used to analyze the average logging data of the reservoir section of the wells to obtain the initial production prediction characteristic data of the wells.

[0154] In this process, the average production value of the exploited oil wells at the initial production set time generated in step S401 can be combined with the average logging data value of the exploited oil well reservoir section generated in step S402 using, for example, a Lasso regression model, to obtain the preferred average logging data value of the exploited oil well reservoir section, which is the initial production prediction characteristic data of the exploited oil wells.

[0155] Specifically, the lasso regression model uses the Cost loss function to optimize the vector composed of average well logging data. The formula for the loss function is shown below:

[0156]

[0157] Where N is the sample size, i.e., the total number of oil wells; prod k ω represents the average initial production of the k-th oil well over the first three months, i.e., the estimated initial production label value of the oil well; ω represents the weights of each feature in the regression model, i.e., the weights of the average logging data of the k-th oil well. λ is the average logging data of the kth exploited oil well described in step S402; λ is the penalty term, which is used to prevent model overfitting, and is the penalty term coefficient, which is 0.1.

[0158] Specifically, the key influencing factors and their corresponding weights ω can be obtained by finding the minimum value of formula (19), i.e., the minimum value of the loss function. Since formula (19) incorporates the L1 norm of a vector, which contains absolute values, its cost function is not differentiable everywhere. Therefore, it is impossible to obtain an analytical solution for ω directly by taking the derivative of the function. Thus, the minimum value of the loss function can be found using, for example, the coordinate descent method, to obtain the weight vector ω. TSome of these factors will have a weight of 0. After removing these factors, the remaining features are the optimal results, namely the oil well initial production prediction feature data.

[0159] The process of determining the initial production prediction characteristics of already exploited oil wells provided in this embodiment combines the average production of already exploited oil wells at the initial set time of exploitation with the average logging data of the reservoir section of already exploited oil wells in the block where the oil well to be exploited is located. This optimizes the data by selecting more reliable model training data and improving the accuracy of the initial production prediction of the oil well to be exploited.

[0160] Figure 5 This is a flowchart illustrating the process of establishing the initial production prediction model for oil wells provided in the fifth embodiment of this application.

[0161] like Figure 5 As shown, the process of establishing the initial production prediction model for oil wells provided in this embodiment may include the following steps:

[0162] Step S501: Based on well logging data and production data, determine the initial production prediction characteristic data of the exploited oil wells.

[0163] Specifically, as described in step S401 above, the average production rate of the exploited oil wells at the initial set time can be determined based on production data. As described in step S402 above, the average logging data of the exploited oil well reservoir section can be determined based on the logging data type. As described in step S403 above, based on the average production rate of the exploited oil wells at the initial set time, a regression model can be used to analyze the average logging data of the exploited oil well reservoir section to obtain the initial production prediction characteristic data of the exploited oil wells.

[0164] The initial production prediction feature data of the exploited oil wells obtained through step S403 consists of t columns. The dataset of the initial production prediction feature data of the exploited oil wells is shown below:

[0165]

[0166] in, This represents the average value of the t-th type logging curve selected from the k-th well.

[0167] Step S502: Using the initial production prediction feature data, perforation thickness data, and fracturing sand addition data of the already produced oil wells as features, and the average production value of the already produced oil wells at the initial set time as labels, a random forest model is used for model training to obtain the initial production prediction model of the oil wells.

[0168] Specifically, an oil well initial production prediction model can be established based on initial production prediction feature data, perforation thickness data, fracturing proppant dosage data, and production data. More specifically, features for model training can be generated from the initial production prediction feature data, perforation thickness data, and fracturing proppant dosage data, and labels for model training can be generated from the production data. These features and labels are then input into a training model, such as a random forest model, to train the oil well initial production prediction model.

[0169] First, the initial production prediction feature data set of the produced oil wells can be combined with the perforation thickness data of the reservoir section of the produced oil wells, the fracturing proppant addition data of the reservoir section of the produced oil wells, and the initial production value data of the produced oil wells to obtain the initial production prediction feature data of the produced oil wells corresponding to the perforation thickness, fracturing proppant addition, and initial production value data of the produced oil wells. That is, the initial production prediction feature dataset of the produced oil wells, datasetprod, is shown below:

[0170]

[0171] Among them, hp k Refers to the perforation thickness data of the k-th well; Vs k This refers to the data on the amount of proppant added during fracturing in the kth well; prod k This refers to the output of the k-th well at the initial set time of its extraction.

[0172] The aforementioned preliminary production prediction feature dataset of the exploited oil wells was standardized to obtain a standardized preliminary production prediction feature dataset of the exploited oil wells.

[0173] Specifically, the initial production prediction feature dataset `datasetprod` from already exploited oil wells is standardized using the following formula: Among them, u ij σ is a data point in the i-th row and j-th column of datasetprod; j Let u be the variance of the data in column j. ij ' represents the standardized data. The standardized dataset for the initial production prediction features of exploited oil wells is named datasetprod, as shown below:

[0174]

[0175] in, The data represents the initial production prediction characteristics of standardized produced oil wells. hp′ represents the standardized perforation thickness data of standardized produced oil wells, Vs′ represents the standardized fracturing sand addition rate of produced oil, and prod′ represents the standardized average production rate of produced oil wells at the initial set time of production.

[0176] Secondly, the standardized initial production prediction dataset is split into an initial production prediction training set, an initial production prediction validation set, and an initial production prediction test set. Using the initial production prediction training set, multiple initial production prediction models are trained by setting different combinations of training parameters.

[0177] Specifically, the standardized initial production prediction feature dataset of exploited oil wells can be split into an initial production prediction training set, an initial production prediction validation set, and an initial production prediction test set using a random splitting method, for example. The initial production prediction training set accounts for 80%, the initial production prediction test set accounts for 10%, and the initial production prediction validation set accounts for 10%.

[0178] When using the oil well initial production prediction training set to develop an oil well initial production prediction model, standardized initial production prediction feature data of already exploited oil wells can be used. The standardized perforation thickness hp′ of the reservoir section of the produced oil well and the standardized fracturing sand addition Vs′ of the produced oil well are used as features, and the standardized average production value prod′ of the initial production time of the produced oil well is used as a label for training.

[0179] Specifically, a random forest model, for example, can be used to train the initial production prediction model for oil wells. Before training, the random forest model can have its training parameters set, including tree depth and the number of trees (n_trees). Using treedepth = 10, 20, 30, 40, 50, and n_trees = 100, 200, 500, 800, 1000, a total of 25 experimental combinations (5×5), a random forest model is set up for each parameter combination, and the initial production prediction model is trained using the initial production prediction training set, generating 25 initial production prediction models for oil wells.

[0180] Finally, the prediction performance of the oil well initial production prediction model was evaluated using the initial production prediction validation set and the initial production prediction test set, and the oil well initial production prediction model with the best prediction performance was selected.

[0181] Specifically, the prediction effect of the oil well initial production prediction model can be evaluated by, for example, the model error evaluation method. The model error evaluation method can be evaluated by calculating the mean absolute error (MAE). The smaller the MAE value, the better the model prediction effect.

[0182] The initial production prediction validation set data is input into the aforementioned oil well initial production prediction model to obtain the predicted initial production result label. The MAE value of different models can be calculated based on the predicted initial production result label and the actual initial production label. The oil well initial production prediction model with the smallest MAE value is selected as the preferred oil well initial production prediction model. The test set data is input into the preferred oil well initial production prediction model to obtain the predicted initial production result label. The MAE value of the preferred oil well initial production prediction model is calculated to verify the optimization effect. The oil well initial production prediction model provided in this embodiment can be used to predict the initial production value of oil wells to be exploited. Furthermore, suitable exploitation equipment, such as pumping units and tubing, can be configured according to the initial production situation of the oil wells to be exploited. In the process of establishing this oil well initial production prediction model, not only are the preferred initial production prediction feature data selected, but different oil well initial production prediction models are generated by adjusting the parameter combination of model training. The optimal oil well initial production prediction model with the best prediction effect is selected through an optimization method, thereby improving the accuracy of oil well initial production prediction.

[0183] Figure 6 This is a schematic diagram of the structure of the oil well initial production value estimation device provided in the sixth embodiment of this application.

[0184] like Figure 6 As shown, the oil well initial production value estimation device 60 provided in this embodiment may include: an acquisition module 61, a calculation module 62, and a determination module 63.

[0185] The acquisition module 61 is used to acquire logging data from the oil well to be exploited;

[0186] The calculation module 62 is used to determine the oil content prediction data of the oil well to be developed based on the oil content prediction characteristic data of the oil wells already developed in the block where the oil well to be developed is located and the logging data of the oil well to be developed.

[0187] The calculation module 62 is also used to input the oil content prediction data of the oil well to be exploited into the oil content prediction model of the oil well, and obtain the oil content distribution data of the oil well in the depth direction of the logging data.

[0188] The calculation module 62 is also used to determine the perforation thickness data of the oil well to be developed based on the oil-bearing distribution data in the depth direction of the logging data; and to determine the fracturing sand addition data of the reservoir section of the oil well to be developed based on the fracturing sand addition data of the already developed oil well, the perforation thickness data of the already developed oil well, and the perforation thickness data of the oil well to be developed.

[0189] The calculation module 62 is also used to determine the initial production estimate data of the oil well to be developed based on the initial production prediction characteristic data of the oil wells already developed in the block where the oil well to be developed is located, the logging data of the developed oil wells, the perforation thickness data of the oil well to be developed, and the reservoir section fracturing sand addition data of the oil well to be developed.

[0190] The determination module 63 is used to input the initial production prediction data of the oil well to be exploited into the oil well initial production prediction model to obtain the initial production prediction value of the oil well to be exploited.

[0191] The apparatus provided in this embodiment can be used to execute the above-described method embodiments. Figures 1 to 5 The technical solution is similar in principle and effect, and will not be described again in this embodiment.

[0192] Figure 7 This is a schematic diagram of the structure of the oil well initial production value estimation device provided in the seventh embodiment of this application.

[0193] like Figure 7 As shown, the oil well initial production value estimation device 70 provided in this embodiment may include a processor 71, a memory 72, and a communication interface 73;

[0194] Memory 72 is used to store executable instructions of processor 71;

[0195] The processor 71 is configured to execute the above method embodiments by executing executable instructions. Figures 1 to 5 Methods for estimating the initial production value of any oil well.

[0196] This application also provides a readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the above-described method embodiments. Figures 1 to 5 Methods for estimating the initial production value of any oil well.

[0197] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0198] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for estimating the initial production value of an oil well, characterized in that, include: Collect logging data from oil wells to be exploited; Based on the oil-bearing potential characteristic data of the already produced oil wells in the block where the oil well to be produced is located and the well logging data of the oil well to be produced, the oil-bearing potential data of the oil well to be produced is determined. The oil-bearing potential prediction data of the oil well to be exploited is input into the oil well oil-bearing potential prediction model to obtain the oil-bearing potential distribution data of the oil well in the depth direction of the logging data; Based on the oil-bearing distribution data of the oil well to be exploited in the depth direction of the logging data, the perforation thickness data of the oil well to be exploited is determined; Based on the fracturing proppant addition data of the reservoir section of the already produced oil well, the perforation thickness data of the already produced oil well, and the perforation thickness data of the oil well to be produced, the fracturing proppant addition data of the reservoir section of the oil well to be produced is determined. Based on the initial production prediction characteristic data of the wells already produced in the block where the well to be developed is located, the logging data of the well to be developed, the perforation thickness data of the well to be developed, and the reservoir section fracturing and proppant addition data of the well to be developed, the initial production prediction data of the well to be developed is determined. The initial production estimate data of the oil well to be exploited is input into the oil well initial production estimate model to obtain the initial production estimate value of the oil well to be exploited. Before inputting the oil-bearing potential prediction data of the oil well to be exploited into the oil well oil-bearing potential prediction model, the method further includes: Collect reservoir data of already produced oil wells within the block where the oil well to be developed is located. The reservoir data includes: well logging data, perforation thickness data, oil testing data, fracturing sand injection data, and production data. The logging data of the extracted oil wells are merged according to data type; Principal component analysis (PCA) was used to analyze the logging data of the merged exploited oil wells to obtain the principal component weight values. Based on the principal component weight values ​​and the merged logging data, the oil-bearing potential characteristic data of the exploited oil wells are obtained. Based on the oil-bearing characteristics data of the exploited oil wells and the oil testing data of the exploited oil wells, an oil-bearing prediction model for oil wells is established.

2. The method according to claim 1, characterized in that, Before inputting the initial production estimate data of the oil well to be developed into the oil well initial production estimate model to obtain the initial production estimate value of the oil well, the process also includes: Based on the well logging data and the production data, the initial production prediction characteristics of the exploited oil wells are determined; Based on the initial production prediction characteristic data, the perforation thickness data, the fracturing sand addition data, and the production data, an oil well initial production prediction model is established.

3. The method according to claim 1, characterized in that, The step of establishing an oil well oil-bearing prediction model based on the oil-bearing prediction characteristic data and the oil testing data of the exploited oil wells includes: Extract the depth data of the logging data points corresponding to the oil-bearing prediction feature data of the already exploited oil wells; Using the oil-bearing potential feature data of the already exploited oil wells as features, and the oil-bearing potential label corresponding to the oil-bearing potential feature data and the depth data as labels, the model is trained using the support vector machine algorithm to obtain the oil well oil-bearing potential prediction model.

4. The method according to claim 2, characterized in that, The step of determining the initial production prediction characteristics of the exploited oil wells based on the logging data and the production data includes: Based on the production data, determine the average production rate of the extracted oil wells at the initial set time of extraction; The average value of the logging data for the exploited oil well reservoir section is determined based on the type of logging data. Based on the average production rate of the wells at the initial production set time, a regression model is used to analyze the average logging data of the reservoir section of the wells to obtain the initial production prediction characteristic data of the wells.

5. The method according to claim 4, characterized in that, The step of establishing an oil well initial production prediction model based on the initial production prediction characteristic data, the perforation thickness data, the fracturing sand injection data, and the production data includes: The initial production prediction feature data of the already produced oil well, the perforation thickness data, and the fracturing sand addition data are used as features, and the average production value of the already produced oil well at the initial set time of production is used as the label. A random forest model is used to train the model to obtain the initial production prediction model of the oil well.

6. A device for estimating the initial production value of an oil well, characterized in that, The method for estimating the initial production value of oil wells as described in claims 1-5 is used to implement the above-described method. The device for estimating the initial production value of the oil well includes: The acquisition module is used to acquire logging data from the oil well to be exploited. The calculation module is used to determine the estimated oil content of the oil well to be developed based on the oil content prediction characteristic data of the oil wells already developed in the block where the oil well to be developed is located and the logging data of the oil well to be developed. The calculation module is also used to input the oil-bearing prediction data of the oil well to be exploited into the oil well oil-bearing prediction model to obtain the oil-bearing distribution data of the oil well to be exploited in the depth direction of the logging data. The calculation module is also used to determine the perforation thickness data of the oil well to be developed based on the oil-bearing distribution data in the depth direction of the logging data; and to determine the reservoir fracturing sand addition data of the oil well to be developed based on the reservoir section fracturing sand addition data of the oil well already developed, the perforation thickness data of the oil well already developed, and the perforation thickness data of the oil well to be developed. The calculation module is also used to determine the initial production estimate data of the oil well to be developed based on the initial production prediction characteristic data of the oil wells already developed in the block where the oil well to be developed is located, the logging data of the oil well to be developed, the perforation thickness data of the oil well to be developed, and the reservoir section fracturing sand addition data of the oil well to be developed. The determination module is used to input the initial production prediction data of the oil well to be exploited into the oil well initial production prediction model to obtain the initial production prediction value of the oil well to be exploited. The device further includes a model training module, the model training module being used for: Collect reservoir data of already produced oil wells within the block where the oil well to be developed is located. The reservoir data includes: well logging data, perforation thickness data, oil testing data, fracturing sand injection data, and production data. The logging data of the extracted oil wells are merged according to data type; Principal component analysis (PCA) was used to analyze the logging data of the merged exploited oil wells to obtain the principal component weight values. Based on the principal component weight values ​​and the merged logging data, the oil-bearing potential characteristic data of the exploited oil wells are obtained. Based on the oil-bearing characteristics data of the exploited oil wells and the oil testing data of the exploited oil wells, an oil-bearing prediction model for oil wells is established.

7. A device for estimating the initial production value of an oil well, characterized in that, include: Processor, memory, communication interface; The memory is used to store the executable instructions of the processor; The processor is configured to execute the method for estimating the initial production value of an oil well according to any one of claims 1 to 5 by executing the executable instructions.

8. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for estimating the initial production value of an oil well as described in any one of claims 1 to 5.

9. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method for estimating the initial production value of an oil well as described in any one of claims 1 to 5.

Images