Shale gas recovery ratio determination method and device and electronic equipment

Abstract

The invention provides a shale gas recovery ratio determination method and device and electronic equipment, and belongs to the technical field of data processing. The method comprises the following steps: acquiring a block volume utilization degree calculation parameter set, a fracture network control degree calculation parameter set and a matrix recovery efficiency calculation parameter set; determining a first target parameter of which the influence degree on the recovery ratio is greater than a set threshold value in the block volume utilization degree calculation parameter set through correlation analysis; determining a second target parameter of which the influence degree of the fracture network control degree calculation parameter set on the recovery ratio is greater than a set threshold value, and determining a third target parameter of which the influence degree of the matrix recovery efficiency calculation parameter set on the recovery ratio is greater than the set threshold value; and inputting evaluation data constructed by the first target parameter, the second target parameter and the third target parameter into the shale gas recovery rate prediction model to obtain an output shale gas recovery rate prediction result. The method solves the problem that an existing shale gas recovery rate calculation method is low in accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for determining shale gas recovery rate, a device for predicting shale gas recovery rate, and an electronic device. Background Art

[0002] Shale gas is natural gas stored mainly in the form of adsorption and free state in shale reservoirs. It is rich in nature and is considered an unconventional energy source to alleviate energy shortage and reduce greenhouse gas emissions. For shale reservoirs, generally, fracturing and transformation means are used to transform the shale reservoir body into a volume fracture network. Only the shale gas in the artificial fractures and the effectively affected area of pressure can be effectively produced, and the reserve utilization rate is not high, showing the characteristic of "one well, one reservoir". The recovery rate is a comprehensive index to measure a gas reservoir, and the level of the recovery rate of a shale gas field is a key concern in the dynamic analysis of oil and gas fields.

[0003] The recovery rate is defined as the ratio of the produced amount to the original geological reserve, usually expressed as a percentage. Currently, the calculation methods for shale gas recovery rate mainly include the analogy method, desorption method, material balance method, isothermal adsorption curve method, production decline method, numerical simulation method, etc. However, the shale gas recovery rates obtained by various current calculation methods for shale gas recovery rate all have the problem of low calculation accuracy, which brings obstacles to the efficient development and large-scale production increase of shale gas. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a method, a device, and an electronic device for determining shale gas recovery rate, so as to solve the problem that the shale gas recovery rates obtained by various current calculation methods for shale gas recovery rate all have low calculation accuracy.

[0005] To achieve the above purpose, the embodiments of the present invention provide a method for determining shale gas recovery rate, including:

[0006] Obtaining a calculation parameter set for the degree of block volume utilization, a calculation parameter set for the degree of fracture network control, and a calculation parameter set for the matrix production efficiency; wherein the calculation parameter set for the degree of block volume utilization includes multiple parameters for calculating the ratio of the controlled reserves of all wells in the block to the total block reserves; the calculation parameter set for the degree of fracture network control includes multiple parameters for calculating the ratio of the fracture network controlled volume of a single well to the well-controlled volume; the calculation parameter set for the matrix production efficiency includes multiple parameters for calculating the ratio of the gas production amount in the fracture network controlled area to the original reserve;

[0007] Through correlation analysis, determine the first target parameter in the block volume utilization degree calculation parameter set whose influence degree on the recovery factor is greater than the set threshold, determine the second target parameter in the fracture network control degree calculation parameter set whose influence degree on the recovery factor is greater than the set threshold, and determine the third target parameter in the matrix production efficiency calculation parameter set whose influence degree on the recovery factor is greater than the set threshold;

[0008] Input the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter into the shale gas recovery factor prediction model to obtain the shale gas recovery factor prediction result output by the shale gas recovery factor prediction model; the shale gas recovery factor prediction model is trained through sample evaluation data and the shale gas recovery factor result label corresponding to the sample evaluation data.

[0009] Optionally, the step of through correlation analysis, determining the first target parameter in the block volume utilization degree calculation parameter set whose influence degree on the recovery factor is greater than the set threshold, determining the second target parameter in the fracture network control degree calculation parameter set whose influence degree on the recovery factor is greater than the set threshold, and determining the third target parameter in the matrix production efficiency calculation parameter set whose influence degree on the recovery factor is greater than the set threshold includes:

[0010] Calculate the Pearson correlation coefficient between each parameter in the block volume utilization degree calculation parameter set and the recovery factor, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the block volume utilization degree calculation parameter set as the first target parameter;

[0011] Calculate the Pearson correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery factor, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the fracture network control degree calculation parameter set as the second target parameter;

[0012] Calculate the Pearson correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery factor, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

[0013] Optionally, the step of through correlation analysis, determining the first target parameter in the block volume utilization degree calculation parameter set whose influence degree on the recovery factor is greater than the set threshold, determining the second target parameter in the fracture network control degree calculation parameter set whose influence degree on the recovery factor is greater than the set threshold, and determining the third target parameter in the matrix production efficiency calculation parameter set whose influence degree on the recovery factor is greater than the set threshold includes:

[0014] Calculate the Spearman correlation coefficient between each parameter in the block volume utilization degree calculation parameter set and the recovery rate, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the block volume utilization degree calculation parameter set as the first target parameter;

[0015] Calculate the Spearman correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery rate, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the fracture network control degree calculation parameter set as the second target parameter;

[0016] Calculate the Spearman correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery rate, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

[0017] Optionally, the shale gas recovery rate prediction model is trained through the following steps:

[0018] Repeat the following steps until the model prediction deviation value meets the preset conditions:

[0019] Obtain a sample data set; the sample data set includes a plurality of sample evaluation data, and each sample evaluation data is constructed based on the first sample target parameter, the second sample target parameter, and the third sample target parameter;

[0020] Determine a plurality of subsets based on the sample data set, and each subset is obtained by random sampling with replacement from the sample data set; wherein, the number of sample evaluation data in each subset is less than the number of sample evaluation data in the sample data set;

[0021] Train a decision tree based on each subset to obtain a plurality of decision trees with the same number as the number of subsets; wherein, when each node of the decision tree splits, a feature is randomly selected from a part of the features of the sample evaluation data as the splitting attribute;

[0022] Input the sample evaluation data into each decision tree, and calculate the average value based on the sample shale gas recovery rate prediction results output by all the decision trees to obtain the sample average prediction result;

[0023] Determine the model prediction deviation value based on the sample average prediction result and the shale gas recovery rate result label corresponding to the sample evaluation data.

[0024] On the other hand, an embodiment of the present invention further provides a shale gas recovery rate determination device, including:

[0025] A data acquisition module, configured to acquire a calculation parameter set for the degree of block volume utilization, a calculation parameter set for the degree of fracture network control, and a calculation parameter set for the matrix production efficiency; wherein the calculation parameter set for the degree of block volume utilization includes a plurality of parameters for calculating the ratio of the controlled reserves of all wells in the block to the total block reserves; the calculation parameter set for the degree of fracture network control includes a plurality of parameters for calculating the ratio of the fracture network controlled volume of a single well to the well controlled volume; and the calculation parameter set for the matrix production efficiency includes a plurality of parameters for calculating the ratio of the gas production volume in the fracture network controlled area to the original reserves.

[0026] A correlation analysis module, configured to determine, through correlation analysis, a first target parameter in the calculation parameter set for the degree of block volume utilization, whose influence degree on the recovery factor is greater than a set threshold, a second target parameter in the calculation parameter set for the degree of fracture network control, whose influence degree on the recovery factor is greater than the set threshold, and a third target parameter in the calculation parameter set for the matrix production efficiency, whose influence degree on the recovery factor is greater than the set threshold.

[0027] A recovery factor prediction module, configured to input the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter into a shale gas recovery factor prediction model, and obtain a shale gas recovery factor prediction result output by the shale gas recovery factor prediction model; the shale gas recovery factor prediction model is trained by sample evaluation data and the shale gas recovery factor result label corresponding to the sample evaluation data.

[0028] Optionally, the step of determining, through correlation analysis, a first target parameter in the calculation parameter set for the degree of block volume utilization, whose influence degree on the recovery factor is greater than a set threshold, a second target parameter in the calculation parameter set for the degree of fracture network control, whose influence degree on the recovery factor is greater than the set threshold, and a third target parameter in the calculation parameter set for the matrix production efficiency, whose influence degree on the recovery factor is greater than the set threshold, includes:

[0029] Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set for the degree of block volume utilization and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set for the degree of block volume utilization as the first target parameter;

[0030] Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set for the degree of fracture network control and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set for the degree of fracture network control as the second target parameter;

[0031] Calculate the Pearson correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery rate, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

[0032] Optionally, the step of determining the first target parameter in the block volume utilization degree calculation parameter set with an influence degree on the recovery rate greater than the set threshold, the second target parameter in the fracture network control degree calculation parameter set with an influence degree on the recovery rate greater than the set threshold, and the third target parameter in the matrix production efficiency calculation parameter set with an influence degree on the recovery rate greater than the set threshold through correlation analysis includes:

[0033] Calculate the Spearman correlation coefficient between each parameter in the block volume utilization degree calculation parameter set and the recovery rate, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the block volume utilization degree calculation parameter set as the first target parameter;

[0034] Calculate the Spearman correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery rate, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the fracture network control degree calculation parameter set as the second target parameter;

[0035] Calculate the Spearman correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery rate, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

[0036] Optionally, the shale gas recovery rate prediction model is trained through the following steps:

[0037] Repeat the following steps until the model prediction deviation value meets the preset conditions:

[0038] Obtain a sample data set; the sample data set includes multiple sample evaluation data, and each sample evaluation data is constructed based on the first sample target parameter, the second sample target parameter, and the third sample target parameter;

[0039] Determine multiple subsets based on the sample data set, and each subset is obtained by randomly sampling the sample data set with replacement; wherein, the number of sample evaluation data in each subset is less than the number of sample evaluation data in the sample data set;

[0040] Train a decision tree based on each of the subsets, obtaining a plurality of decision trees equal in number to the number of subsets; wherein, when each node of the decision tree is split, a feature is randomly selected from a part of the features of the sample evaluation data as the splitting attribute;

[0041] Input the sample evaluation data into each decision tree, and calculate the average value based on the sample shale gas recovery rate prediction results output by all the decision trees to obtain the sample average prediction result;

[0042] Determine the model prediction deviation value based on the sample average prediction result and the shale gas recovery rate result label corresponding to the sample evaluation data.

[0043] On the other hand, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the above-mentioned shale gas recovery rate determination method is implemented.

[0044] On the other hand, the present invention also provides a machine-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned shale gas recovery rate determination method is implemented.

[0045] Through the above technical solution, the present invention determines, through correlation analysis, the first target parameter in the block volume utilization degree calculation parameter set whose influence degree on the recovery rate is greater than the set threshold, the second target parameter in the fracture network control degree calculation parameter set whose influence degree on the recovery rate is greater than the set threshold, and the third target parameter in the matrix production efficiency calculation parameter set whose influence degree on the recovery rate is greater than the set threshold, and predicts the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter through a shale gas recovery rate prediction model to obtain a shale gas recovery rate prediction result. The present invention determines the shale gas recovery rate by combining on-site data of shale gas with a machine learning model, realizes unified description and improves the accuracy of shale gas recovery rate by comprehensively considering the integrated process of geology-engineering-development and by coupling geomechanics, engineering mechanics, and fluid mechanics.

[0046] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification. Together with the following specific implementation, they are used to explain the embodiments of the present invention, but do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0048] Figure 1 is one of the flow schematic diagrams of the shale gas recovery rate determination method provided by the present invention;

[0049] Figure 2 It is a schematic diagram of the influencing factors of shale gas recovery provided by the present invention;

[0050] Figure 3 It is a schematic diagram of the degree of block volume utilization provided by the present invention;

[0051] Figure 4 It is a schematic diagram of the degree of fracture network control provided by the present invention;

[0052] Figure 5 It is a schematic diagram of the matrix production efficiency provided by the present invention;

[0053] Figure 6 It is a schematic diagram of the random forest model adopted by the shale gas recovery prediction model provided by the present invention;

[0054] Figure 7 It is the second schematic diagram of the flow of the method for determining shale gas recovery provided by the present invention;

[0055] Figure 8 It is the third schematic diagram of the flow of the method for determining shale gas recovery provided by the present invention;

[0056] Figure 9 It is a schematic diagram of the structure of the device for determining shale gas recovery provided by the present invention;

[0057] Figure 10 It is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Embodiments

[0058] The following will describe in detail the specific embodiments of the embodiments of the present invention with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the embodiments of the present invention, and are not used to limit the embodiments of the present invention.

[0059] Method Embodiment

[0060] Please refer to Figure 1 , the embodiments of the present invention provide a method for determining shale gas recovery, including:

[0061] Step 100, obtain a set of calculation parameters for the degree of block volume utilization, a set of calculation parameters for the degree of fracture network control, and a set of calculation parameters for the matrix production efficiency.

[0062] In the embodiments of the present invention, considering the differences between shale gas recovery rate and conventional recovery rate, combined with the characteristics of one well-one reservoir in shale gas development, starting from the integration of geology-engineering-development, considering the differences and strong heterogeneity of reservoir physical properties (adsorbed gas desorption seepage and the resulting deformation will increase the shale gas recovery rate. When adsorption is not considered, its recovery rate is the lowest, and desorption is beneficial to the development of shale gas), as well as the impact of fracturing parameters (fracturing parameters affect the actual reservoir transformation situation) on well control ability, a method for determining shale gas recovery rate is formed from levels such as the degree of block volume utilization, the degree of fracture network control, and the matrix production efficiency.

[0063] Specifically, the electronic device acquires a set of calculation parameters for the degree of block volume utilization, a set of calculation parameters for the degree of fracture network control, and a set of calculation parameters for the matrix production efficiency collected at the shale gas field site. The set of calculation parameters for the degree of block volume utilization includes multiple parameters for calculating the ratio of the controlled reserves of all wells in the block to the total block reserves. The set of calculation parameters for the degree of fracture network control includes multiple parameters for calculating the ratio of the fracture network control volume of a single well to the well-controlled volume. The set of calculation parameters for the matrix production efficiency includes multiple parameters for calculating the ratio of the gas production volume in the fracture network control area to the original reserves.

[0064] In one embodiment, please refer to Figure 2 and Figure 3 , the degree of block volume utilization E ACV is related to factors such as the well pattern coefficient γ, the single-well equivalent thickness control coefficient α, and the single-well planar utilization coefficient β. Among them, the well pattern coefficient γ is the ratio of the well pattern considering actual environment, terrain and other factors to the well pattern in the ideal situation. The single-well equivalent thickness control coefficient α is the ratio of the well-controlled equivalent height to the thickness of I+II type reservoirs. The single-well planar utilization coefficient β is the ratio of the horizontal section length to (the distance before the target + the horizontal section length). The smaller the distance before the target, the larger the single-well planar utilization coefficient β. The set of calculation parameters for the degree of block volume utilization includes at least two of the well pattern considering actual environment, terrain and other factors, the well pattern in the ideal situation, the well-controlled equivalent height, the thickness of I+II type reservoirs, the horizontal section length, and the distance before the target. In one embodiment, in order to comprehensively consider as many influencing factors as possible, the set of calculation parameters for the degree of block volume utilization includes the well pattern considering actual environment, terrain and other factors, the well pattern in the ideal situation, the well-controlled equivalent height, the thickness of I+II type reservoirs, the horizontal section length, and the distance before the target.

[0065] Please refer to Figure 2 and Figure 4 , the degree of fracture network control E SRVIt is related to the fracture plane utilization coefficient ε, the fracture vertical utilization coefficient ω, the equivalent sweep coefficient η, the main reservoir encounter rate δ, and the fracture network perfection coefficient ξ. Among them, the fracture plane utilization coefficient ε is the ratio of the equivalent fracture length to the well spacing. The fracture vertical utilization coefficient ω is the ratio of the equivalent fracture height to the equivalent thickness of a single well. The equivalent sweep coefficient η is the ratio of the drainage radius to the semi-cluster length. The fracture network perfection coefficient ξ includes the degree of natural fracture development, the degree of artificial fracture network perfection, etc. The calculation parameter set of the fracture network control degree includes at least two parameters among the equivalent fracture length, well spacing, equivalent fracture height, equivalent thickness of a single well, drainage radius, semi-cluster length, degree of natural fracture development, and degree of artificial fracture network perfection. In order to comprehensively consider as many influencing factors as possible, in one embodiment, the calculation parameter set of the fracture network control degree includes the equivalent fracture length, well spacing, equivalent fracture height, equivalent thickness of a single well, drainage radius, semi-cluster length, degree of natural fracture development, and degree of artificial fracture network perfection.

[0066] Please refer to Figure 2 and Figure 5 , the matrix production efficiency E P It is related to the bottom-hole pressure coefficient λ, the adsorbed-free gas distribution coefficient υ, the matrix pressure depletion efficiency χ, and the matrix modification coefficient ι. Among them, the bottom-hole pressure coefficient λ represents (initial pressure - bottom-hole pressure) / initial pressure. The adsorbed-free gas distribution coefficient υ is the degree of influence contribution of the free / adsorbed gas ratio to the gas production. The matrix pressure depletion efficiency χ is (initial pressure - reservoir average pressure) / initial pressure. The matrix modification coefficient ι is the degree of development effect affected by matrix changes caused by imbibition, water-rock interaction, stress change, etc. The calculation parameter set of the matrix production efficiency includes at least two parameters among the initial pressure, bottom-hole pressure, degree of influence contribution of the free / adsorbed gas ratio to the gas production, reservoir average pressure, and degree of development effect affected by matrix changes caused by imbibition, water-rock interaction, stress change, etc. In order to comprehensively consider as many influencing factors as possible, in one embodiment, the calculation parameter set of the matrix production efficiency includes the initial pressure, bottom-hole pressure, degree of influence contribution of the free / adsorbed gas ratio to the gas production, reservoir average pressure, and degree of development effect affected by matrix changes caused by imbibition, water-rock interaction, stress change, etc.

[0067] Step 200: Through correlation analysis, determine the first target parameter in the calculation parameter set of the block volume utilization degree that has an influence degree on the recovery factor greater than the set threshold, determine the second target parameter in the calculation parameter set of the fracture network control degree that has an influence degree on the recovery factor greater than the set threshold, and determine the third target parameter in the calculation parameter set of the matrix production efficiency that has an influence degree on the recovery factor greater than the set threshold.

[0068] The present invention can use the analytic hierarchy process to obtain the weights of various influencing factors for shale gas recovery rate, and establish an intuitive functional relationship between each main control factor and the shale gas recovery rate. Conduct research on the physical quantity mathematical parameterization conversion of influencing factors such as the degree of block volume utilization, the degree of fracture network control, and the matrix production efficiency in the gas production model, and use statistical methods to couple multiple parameters in the mathematical model. The present invention can determine the influence degree of each parameter in the calculation parameter set of the block volume utilization degree on the recovery rate through the correlation analysis method in statistics (such as chi-square test, etc.), and screen out the parameters with an influence degree greater than the set threshold (such as 90%) as the first target parameters; determine the influence degree of each parameter in the calculation parameter set of the fracture network control degree on the recovery rate through the correlation analysis method in statistics (such as chi-square test, etc.), and screen out the parameters with an influence degree greater than the set threshold (such as 90%) as the second target parameters; determine the influence degree of each parameter in the calculation parameter set of the matrix production efficiency on the recovery rate through the correlation analysis method in statistics (such as chi-square test, etc.), and screen out the parameters with an influence degree greater than the set threshold (such as 90%) as the third target parameters.

[0069] Therefore, through the correlation analysis in the embodiments of the present invention, the influence degrees of the parameters in the calculation parameter set of the block volume utilization degree, the parameters in the calculation parameter set of the fracture network control degree, and the parameters in the calculation parameter set of the matrix production efficiency on the recovery rate can be obtained, and the parameters with relatively high correlation (i.e., the first target parameters, the second target parameters, and the third target parameters) are screened out. Screening out the parameters with relatively high correlation is beneficial to improving the accuracy of shale gas recovery rate calculation.

[0070] Step 300: Input the evaluation data constructed by the first target parameters, the second target parameters, and the third target parameters into the shale gas recovery rate prediction model, and obtain the shale gas recovery rate prediction result output by the shale gas recovery rate prediction model.

[0071] The electronic device inputs the evaluation data constructed by the first target parameters, the second target parameters, and the third target parameters into the shale gas recovery rate prediction model, and obtains the shale gas recovery rate prediction result output by the shale gas recovery rate prediction model. The shale gas recovery rate prediction model can be various supervised regression-type machine learning models, such as support vector machines, decision trees, etc.

[0072] Please refer to Figure 6, in the embodiments of the present invention, a random forest model can be selected to construct a shale gas recovery prediction model. Random forest regression is based on decision trees and is established by randomly arranging. There is no association between each decision tree in the forest. A decision tree is a tree structure (which can be a binary tree or a non-binary tree). Each non-leaf node represents a test on a feature attribute, each branch represents the output of this feature attribute in a certain value range, and each leaf node stores a category. The process of making a decision using a decision tree starts from the root node, tests the corresponding feature attributes in the item to be classified, and selects the output branch according to its value until reaching the leaf node, and takes the category stored in the leaf node as the decision result.

[0073] The shale gas recovery prediction model (the random forest model is adopted in the embodiments of the present invention) can be that the shale gas recovery prediction model is trained by sample evaluation data and the shale gas recovery result label corresponding to the sample evaluation data. The sample evaluation data is constructed based on the first sample target parameter, the second sample target parameter, and the third sample target parameter. In one embodiment, the first sample target parameter can include the well pattern in the sample ideal situation, the equivalent height of well control in the sample, the thickness of the I+II type reservoirs in the sample, the length of the horizontal section in the sample, and the distance in front of the target in the sample. The second sample target parameter can include the equivalent fracture length in the sample, the well spacing in the sample, the equivalent fracture height in the sample, the equivalent thickness of a single well in the sample, the drainage radius in the sample, the semi-cluster length in the sample, and the degree of natural fracture development in the sample. The third sample target parameter can include the initial pressure in the sample, the bottom hole pressure in the sample, the influence contribution degree of the drawdown and injection ratio on the gas production in the sample, the average reservoir pressure in the sample, the imbibition in the sample, and the water-rock interaction in the sample. The random forest model can use MAE (Mean Absolute Error) or MAPE (Mean Absolute Percentage Error) to evaluate the performance of the trained random forest model.

[0074] In this shale gas recovery prediction model, the objective function can be preset as:

[0075] U * =max(J(E R ))。

[0076] That is, in the embodiments of the present invention, the shale gas recovery prediction model is trained with the maximum shale gas recovery as the goal. Where E R is the shale gas recovery, which is defined as the ratio of the cumulative gas production of the reservoir to the total gas mass under the initial conditions:

[0077]

[0078] f q is the cumulative gas production of the reservoir, and Q Ris the initial total gas mass of the shale gas reservoir, including the gas adsorbed in the matrix, the free gas in the matrix and natural fractures, V R is the reservoir volume, ρ g is the density of free gas.

[0079]

[0080] Among them, Φ m is the matrix porosity, φ f is the fracture porosity, V L is the Langmuir volume, P i is the pressure, P L is the Langmuir pressure, M is the molar mass, Z is the gas compressibility factor, R is the molar gas constant, T is the temperature, ρ a is the density of adsorbed gas, ρ g is the density of free gas.

[0081] The present invention determines the shale gas recovery rate by combining the on-site data of shale gas with a machine learning model, realizes the unified description and improves the accuracy of the shale gas recovery rate by comprehensively considering the integrated process of geology-engineering-development, and by coupling geomechanics, engineering mechanics and fluid mechanics.

[0082] In other embodiments of the present invention, please refer to Figure 7 , step 200, through correlation analysis, determine the first target parameters in the calculation parameter set of the block volume utilization degree whose influence degree on the recovery rate is greater than the set threshold, determine the second target parameters in the calculation parameter set of the fracture network control degree whose influence degree on the recovery rate is greater than the set threshold, and determine the third target parameters in the calculation parameter set of the matrix production efficiency whose influence degree on the recovery rate is greater than the set threshold, including:

[0083] Step 210, calculate the Pearson correlation coefficient between each parameter in the calculation parameter set of the block volume utilization degree and the recovery rate, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of all parameters in the calculation parameter set of the block volume utilization degree as the first target parameter.

[0084] Taking the equivalent well control height in the calculation parameter set of the block volume utilization degree as an example, collect 5 groups of data ((x i , y i), i = 1, 2, 3, 4, 5), that is, (x1, y1), (x2, y2), (x3, y3), (x4, y4) and (x5, y5). Each set of data includes the actual well control equivalent height x and the shale gas recovery rate y under the condition of this actual well control equivalent height. Calculate the covariance and standard deviation of the 5 sets of data based on the 5 sets of data, and determine the Pearson correlation coefficient between the well control equivalent height and the shale gas recovery rate based on the quotient of the covariance of the 5 sets of data and the standard deviation of the 5 sets of data. The calculation formula of the Pearson correlation coefficient is as follows:

[0085]

[0086] Among them, ρ(x,y) represents the Pearson correlation coefficient, represents the covariance of 5 sets of data, n = 5. represents the standard deviation of 5 sets of data, n = 5.

[0087] When the Pearson correlation coefficient between the well control equivalent height and the shale gas recovery rate is higher than the set threshold (for example, 90%), determine the well control equivalent height as the first target parameter. It can be understood that other parameters of the block volume utilization degree calculation parameter set can determine the first target parameter with reference to the calculation method of the well control equivalent height. It should be noted that the finally determined first target parameter may include multiple parameters. For example, the first target parameter may include well placement considering actual environment, terrain and other factors, well placement in ideal conditions, well control equivalent height, thickness of reservoir of type I + II, horizontal section length, and distance before target.

[0088] Step 220: Calculate the Pearson correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery rate, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of all the parameters in the fracture network control degree calculation parameter set as the second target parameter.

[0089] For the specific steps of determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of all the parameters in the fracture network control degree calculation parameter set as the second target parameter, reference can be made to the calculation process of the first target parameter in step 210, which will not be elaborated here. It should be noted that the finally determined second target parameter may include multiple parameters. For example, the second target parameter may include equivalent fracture length, well spacing, equivalent fracture height, single well equivalent thickness, drainage radius, half cluster length, natural fracture development degree, and artificial fracture network perfection degree.

[0090] Step 230: Calculate the Pearson correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery rate, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of all the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

[0091] The specific steps for determining the parameters corresponding to the Pearson correlation coefficients greater than the set threshold among all the parameters in the matrix production efficiency calculation parameter set as the third target parameters can refer to the calculation process of the first target parameters in step 210, which will not be elaborated here. It should be noted that the finally determined third target parameters may include multiple parameters. For example, the third target parameters may include the initial pressure, bottom-hole pressure, influence contribution degree of the stroke-suction ratio on gas production, average reservoir pressure, and the development effect degree affected by matrix changes caused by imbibition, water-rock interaction, stress change, etc.

[0092] By calculating the Pearson correlation coefficients, relatively correlated parameters (i.e., the first target parameters, the second target parameters, and the third target parameters) are screened out from the block volume utilization degree calculation parameter set, the fracture network control degree calculation parameter set, and the matrix production efficiency calculation parameter set. It can simply and quickly screen out relatively correlated parameters, which is beneficial to improving the accuracy and efficiency of shale gas recovery calculation.

[0093] In other embodiments of the present invention, please refer to Figure 8 , step 200, through correlation analysis, determine the first target parameters in the block volume utilization degree calculation parameter set whose influence degree on the recovery rate is greater than the set threshold, determine the second target parameters in the fracture network control degree calculation parameter set whose influence degree on the recovery rate is greater than the set threshold, and determine the third target parameters in the matrix production efficiency calculation parameter set whose influence degree on the recovery rate is greater than the set threshold, including:

[0094] Step 210, calculate the Spearman correlation coefficient between each parameter in the block volume utilization degree calculation parameter set and the recovery rate, and determine the parameters corresponding to the Spearman correlation coefficients greater than the set threshold among all the parameters in the block volume utilization degree calculation parameter set as the first target parameters.

[0095] Since the distribution of the original variables is unknown and may not satisfy the linear relationship, as long as the observed values of the two variables are paired rank rating data, or the rank data transformed from the observed data of continuous variables, regardless of the overall distribution form of the two variables and the size of the sample size, the Spearman correlation coefficient can be used for correlation analysis. The Spearman correlation coefficient is also called the rank correlation coefficient. It uses the rank sizes of the two variables for linear correlation analysis, has no requirement for the variable distribution, and belongs to the non-parametric statistical method with a wide range of applications. It is calculated based on the difference between the ranks of each pair of paired ranks, so it is also called the "method of rank differences". Taking the well control equivalent height in the block volume utilization degree calculation parameter set as an example, the present invention collects 5 groups of data ((x i , y i), i = 1, 2, 3, 4, 5), that is, (x1, y1), (x2, y2), (x3, y3), (x4, y4) and (x5, y5). Each set of data includes the actual well control equivalent height x and the shale gas recovery rate y under the condition of this actual well control equivalent height. Each parameter of each set of data (x i , y i ) has a corresponding rank, and each set of data has a rank difference. For example, x1 is ranked 3rd in ascending order among x1 to x5. For example, y1 is ranked 2nd in ascending order among y1 to y5. Then the rank difference d1 between x1 and y1 is 3 - 2 = 1. Therefore, the rank differences corresponding to the 5 sets of data (x1, y1), (x2, y2), (x3, y3), (x4, y4) and (x5, y5) can be obtained. In the embodiment of the present invention, the Spearman correlation coefficient between the well control equivalent height and the shale gas recovery rate is calculated through the rank differences corresponding to the 5 sets of data. The calculation formula of the Spearman correlation coefficient is as follows:

[0096]

[0097] Among them, r s represents the Spearman correlation coefficient, d i represents the rank difference of the i-th set of data. In the embodiment of the present invention, there are rank differences of n = 5 sets of data. The Spearman correlation coefficient between the well control equivalent height and the shale gas recovery rate is calculated by substituting the rank differences corresponding to the 5 sets of data into the above formula.

[0098] When the Spearman correlation coefficient between the well control equivalent height and the shale gas recovery rate is higher than the set threshold (for example, 80%), the well control equivalent height is determined as the first target parameter. It can be understood that other parameters of the block volume utilization degree calculation parameter set can determine the first target parameter with reference to the calculation method of the well control equivalent height. It should be noted that the finally determined first target parameter may include multiple parameters. For example, the first target parameter may include well placement considering actual environment, terrain and other factors, well placement in ideal conditions, well control equivalent height, thickness of type I + II reservoirs, horizontal section length, and distance before the target.

[0099] Step 220: Calculate the Spearman correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery rate, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of all the parameters in the fracture network control degree calculation parameter set as the second target parameter.

[0100] The specific steps for determining the parameters corresponding to the Spearman correlation coefficients greater than the set threshold among all the parameters in the fracture network control degree calculation parameter set as the second target parameters can refer to the calculation process of the first target parameters in step 210, which will not be elaborated here. It should be noted that the finally determined second target parameters may include multiple parameters. For example, the second target parameters may include equivalent fracture length, well spacing, equivalent fracture height, single-well equivalent thickness, drainage radius, half cluster length, natural fracture development degree, and artificial fracture network perfection degree.

[0101] Step 230: Calculate the Spearman correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery rate, and determine the parameters corresponding to the Spearman correlation coefficients greater than the set threshold among all the parameters in the matrix production efficiency calculation parameter set as the third target parameters.

[0102] The specific steps for determining the parameters corresponding to the Spearman correlation coefficients greater than the set threshold among all the parameters in the matrix production efficiency calculation parameter set as the third target parameters can refer to the calculation process of the first target parameters in step 210, which will not be elaborated here. It should be noted that the finally determined third target parameters may include multiple parameters. For example, the third target parameters may include initial pressure, bottom-hole pressure, influence contribution degree of the drawdown and injection ratio on gas production, average reservoir pressure, and development effect degree affected by matrix changes caused by imbibition, water-rock interaction, stress change, etc.

[0103] By calculating the Spearman correlation coefficient, relatively correlated parameters (i.e., the first target parameters, the second target parameters, and the third target parameters) are screened out from the block volume utilization degree calculation parameter set, the fracture network control degree calculation parameter set, and the matrix production efficiency calculation parameter set. Relatively correlated parameters can be screened out simply, quickly, and reliably, which is beneficial to improving the accuracy, efficiency, and reliability of shale gas recovery rate calculation.

[0104] In other embodiments of the present invention, the shale gas recovery rate prediction model is trained through the following steps:

[0105] Repeat the following steps until the model prediction deviation value meets the preset conditions:

[0106] Step 10: Obtain a sample data set.

[0107] The electronic device obtains a sample data set. The sample data set includes multiple sample evaluation data, and each sample evaluation data is constructed based on a first sample target parameter, a second sample target parameter, and a third sample target parameter. In one embodiment, the first sample target parameter may include sample ideal well pattern, sample well control equivalent height, sample I+II class reservoir thickness, sample horizontal section length, and sample distance in front of the target. The second sample target parameter may include sample equivalent fracture length, sample well spacing, sample equivalent fracture height, sample single well equivalent thickness, sample drainage radius, sample half cluster length, and sample natural fracture development degree. The third sample target parameter may include sample initial pressure, sample bottom hole pressure, the influence contribution degree of sample swabbing ratio on gas production, sample reservoir average pressure, sample imbibition, and sample water-rock interaction.

[0108] Step 20: Determine multiple subsets based on the sample data set, and each subset is obtained by performing random sampling with replacement on the sample data set.

[0109] The electronic device performs random sampling with replacement on the sample data set to obtain p sample evaluation data to form a subset as a new training set. By repeatedly performing random sampling with replacement multiple times, multiple subsets can be obtained. Among them, the number of sample evaluation data in each subset is less than the number of sample evaluation data in the sample data set.

[0110] Step 30: Train a decision tree based on each subset to obtain multiple decision trees with the same number as the number of subsets. Among them, when each node of the decision tree splits, a feature is randomly selected from some of the features of the sample evaluation data as the splitting attribute.

[0111] Specifically, when each sample evaluation data has K attributes, when each node of the decision tree needs to split, k attributes are randomly selected from these K attributes, satisfying the condition k < K. Then, variance is used to select an attribute from these k attributes as the splitting attribute of this node. It should be noted that during the formation process of the decision tree, each node should split according to the rule of randomly selecting a feature from some of the features of the sample evaluation data as the splitting attribute until it can no longer split.

[0112] The objective function of each decision tree is calculated by the following formula:

[0113]

[0114] Each node goes through such a splitting process. When the size of the terminal node leaf reaches the minimum value, the learning of the first tree is completed. Among them, J k is the objective function of the first tree, n is the number of trees, left and right respectively represent the left and right child nodes. y(i) is the true value of the first Sample output variable included in the child node, refers to the average value of the data included in the child node.

[0115] The present invention trains a decision tree by using a subset and does not prune this decision tree, obtaining a plurality of decision trees having the same number as the subset.

[0116] Step 40: Input the sample evaluation data into each decision tree, calculate the average value based on the sample shale gas recovery prediction results output by all the decision trees, and obtain the sample average prediction result.

[0117] The electronic device inputs the sample evaluation data into each decision tree, calculates the average value based on the sample shale gas recovery prediction results output by all the decision trees, and obtains the sample average prediction result.

[0118] Step 50: Determine the model prediction deviation value based on the sample average prediction result and the shale gas recovery result label corresponding to the sample evaluation data.

[0119] In the embodiments of the present invention, the model prediction deviation value of the random forest model can be measured by the Mean Absolute Error (MAE) or the Mean Absolute Percentage Error (MAPE). When the mean absolute error or the mean absolute percentage error exceeds a set threshold (for example, 90%), the training ends; otherwise, steps 10 to 50 are repeatedly executed.

[0120] In view of the characteristics of the complex multi-parameter comprehensive influence on shale gas recovery in the embodiments of the present invention, the integrated process of geology-engineering-development is considered, geomechanics, engineering mechanics, and fluid mechanics are coupled, and mechanisms such as fracture network influence, stress sensitivity, viscous flow, adsorption / desorption, and Knudsen diffusion are comprehensively considered, establishing a method for unified description and improvement of shale gas recovery, filling the gap in the calculation method of shale gas recovery that does not comprehensively consider multi-parameter coupling.

[0121] Device embodiments

[0122] Please refer to Figure 9 , on the other hand, the embodiments of the present invention also provide a device for determining shale gas recovery, including:

[0123] A data acquisition module 901 is configured to acquire a set of calculation parameters for the degree of block volume utilization, a set of calculation parameters for the degree of fracture network control, and a set of calculation parameters for the matrix production efficiency. The set of calculation parameters for the degree of block volume utilization includes multiple parameters for calculating the ratio of the controlled reserves of all wells in the block to the total block reserves. The set of calculation parameters for the degree of fracture network control includes multiple parameters for calculating the ratio of the fracture network controlled volume of a single well to the well controlled volume. The set of calculation parameters for the matrix production efficiency includes multiple parameters for calculating the ratio of the gas production volume in the fracture network controlled area to the original reserves.

[0124] A correlation analysis module 902 is configured to, through correlation analysis, determine a first target parameter in the set of calculation parameters for the degree of block volume utilization whose influence degree on the recovery factor is greater than a set threshold, determine a second target parameter in the set of calculation parameters for the degree of fracture network control whose influence degree on the recovery factor is greater than the set threshold, and determine a third target parameter in the set of calculation parameters for the matrix production efficiency whose influence degree on the recovery factor is greater than the set threshold.

[0125] A recovery factor prediction module 903 is configured to input the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter into a shale gas recovery factor prediction model, and obtain a shale gas recovery factor prediction result output by the shale gas recovery factor prediction model. The shale gas recovery factor prediction model is trained by sample evaluation data and the shale gas recovery factor result label corresponding to the sample evaluation data.

[0126] Optionally, the step of, through correlation analysis, determining a first target parameter in the set of calculation parameters for the degree of block volume utilization whose influence degree on the recovery factor is greater than a set threshold, determining a second target parameter in the set of calculation parameters for the degree of fracture network control whose influence degree on the recovery factor is greater than the set threshold, and determining a third target parameter in the set of calculation parameters for the matrix production efficiency whose influence degree on the recovery factor is greater than the set threshold includes:

[0127] Calculating the Pearson correlation coefficient between each parameter in the set of calculation parameters for the degree of block volume utilization and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the set of calculation parameters for the degree of block volume utilization as the first target parameter;

[0128] Calculating the Pearson correlation coefficient between each parameter in the set of calculation parameters for the degree of fracture network control and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the set of calculation parameters for the degree of fracture network control as the second target parameter;

[0129] Calculate the Pearson correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery factor, and determine the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

[0130] Optionally, the step of determining, through correlation analysis, the first target parameter in the block volume utilization degree calculation parameter set with an influence degree on the recovery factor greater than the set threshold, the second target parameter in the fracture network control degree calculation parameter set with an influence degree on the recovery factor greater than the set threshold, and the third target parameter in the matrix production efficiency calculation parameter set with an influence degree on the recovery factor greater than the set threshold includes:

[0131] Calculate the Spearman correlation coefficient between each parameter in the block volume utilization degree calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the block volume utilization degree calculation parameter set as the first target parameter;

[0132] Calculate the Spearman correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the fracture network control degree calculation parameter set as the second target parameter;

[0133] Calculate the Spearman correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

[0134] Optionally, the shale gas recovery factor prediction model is trained through the following steps:

[0135] Repeat the following steps until the model prediction deviation value meets the preset conditions:

[0136] Obtain a sample data set; the sample data set includes multiple sample evaluation data, and each sample evaluation data is constructed based on the first sample target parameter, the second sample target parameter, and the third sample target parameter;

[0137] Determine multiple subsets based on the sample data set, and each subset is obtained by random sampling with replacement from the sample data set; wherein, the number of sample evaluation data in each subset is less than the number of sample evaluation data in the sample data set;

[0138] Train a decision tree for each of the subsets to obtain a plurality of decision trees equal in number to the number of the subsets; wherein, when each node of the decision tree is split, a feature is randomly selected from a part of the features of the sample evaluation data as the splitting attribute;

[0139] Input the sample evaluation data into each decision tree, and calculate the average value based on the sample shale gas recovery prediction results output by all the decision trees to obtain the sample average prediction result;

[0140] Determine the model prediction deviation value based on the sample average prediction result and the shale gas recovery result label corresponding to the sample evaluation data.

[0141] The shale gas recovery determination device includes a processor and a memory. The above data acquisition module 901, correlation analysis module 902, recovery prediction module 903, etc. are all stored in the memory as program units, and the corresponding functions are implemented by the processor executing the above program units stored in the memory.

[0142] The processor contains a kernel, and the corresponding program unit is retrieved from the memory by the kernel. One or more kernels can be set.

[0143] The memory may include non-permanent memory in a computer-readable medium, forms such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM), and the memory includes at least one storage chip.

[0144] Figure 10 Illustrated is a schematic diagram of the physical structure of an electronic device, such as Figure 10As shown in the figure, the electronic device may include: a processor 1010, a communications interface 1020, a memory 1030, and a communication bus 1040. Among them, the processor 1010, the communications interface 1020, and the memory 1030 communicate with each other through the communication bus 1040. The processor 1010 can call the logical instructions in the memory 1030 to execute the shale gas recovery rate determination method, which includes: obtaining a set of calculation parameters for the degree of block volume utilization, a set of calculation parameters for the degree of fracture network control, and a set of calculation parameters for the matrix production efficiency; wherein the set of calculation parameters for the degree of block volume utilization includes multiple parameters for calculating the ratio of the controlled reserves of all wells in the block to the total block reserves; the set of calculation parameters for the degree of fracture network control includes multiple parameters for calculating the ratio of the fracture network control volume of a single well to the well-controlled volume; the set of calculation parameters for the matrix production efficiency includes multiple parameters for calculating the ratio of the gas production volume in the fracture network control area to the original reserves; through correlation analysis, determining a first target parameter in the set of calculation parameters for the degree of block volume utilization whose influence on the recovery rate is greater than a set threshold, determining a second target parameter in the set of calculation parameters for the degree of fracture network control whose influence on the recovery rate is greater than the set threshold, and determining a third target parameter in the set of calculation parameters for the matrix production efficiency whose influence on the recovery rate is greater than the set threshold; inputting the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter into the shale gas recovery rate prediction model to obtain the shale gas recovery rate prediction result output by the shale gas recovery rate prediction model; the shale gas recovery rate prediction model is trained by sample evaluation data and the shale gas recovery rate result label corresponding to the sample evaluation data.

[0145] In addition, when the logical instructions in the above-mentioned memory 1030 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0146] In another aspect, the present invention also provides a machine-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is used to implement a method for determining the shale gas recovery rate. The method includes: obtaining a calculation parameter set for the degree of block volume utilization, a calculation parameter set for the degree of fracture network control, and a calculation parameter set for the matrix production efficiency; wherein the calculation parameter set for the degree of block volume utilization includes multiple parameters for calculating the ratio of the controlled reserves of all wells in the block to the total block reserves; the calculation parameter set for the degree of fracture network control includes multiple parameters for calculating the ratio of the fracture network control volume of a single well to the well-controlled volume; the calculation parameter set for the matrix production efficiency includes multiple parameters for calculating the ratio of the gas production volume in the fracture network control area to the original reserves; through correlation analysis, determining a first target parameter in the calculation parameter set for the degree of block volume utilization whose influence on the recovery rate is greater than a set threshold, determining a second target parameter in the calculation parameter set for the degree of fracture network control whose influence on the recovery rate is greater than the set threshold, and determining a third target parameter in the calculation parameter set for the matrix production efficiency whose influence on the recovery rate is greater than the set threshold; inputting the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter into a shale gas recovery rate prediction model to obtain a shale gas recovery rate prediction result output by the shale gas recovery rate prediction model; the shale gas recovery rate prediction model is trained by sample evaluation data and the shale gas recovery rate result label corresponding to the sample evaluation data.

[0147] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0148] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining shale gas recovery rate, characterized in that, Including: Obtaining a calculation parameter set for the degree of block volume utilization, a calculation parameter set for the degree of fracture network control, and a calculation parameter set for the matrix production efficiency; Wherein the calculation parameter set for the degree of block volume utilization includes multiple parameters for calculating the ratio of the controlled reserves of all wells in the block to the total block reserves; the calculation parameter set for the degree of fracture network control includes multiple parameters for calculating the ratio of the fracture network controlled volume of a single well to the well controlled volume; the calculation parameter set for the matrix production efficiency includes multiple parameters for calculating the ratio of the gas production volume in the fracture network controlled area to the original reserves; Through correlation analysis, determining a first target parameter in the calculation parameter set for the degree of block volume utilization whose influence degree on the recovery factor is greater than a set threshold, determining a second target parameter in the calculation parameter set for the degree of fracture network control whose influence degree on the recovery factor is greater than a set threshold, and determining a third target parameter in the calculation parameter set for the matrix production efficiency whose influence degree on the recovery factor is greater than a set threshold; Inputting the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter into a shale gas recovery factor prediction model to obtain a shale gas recovery factor prediction result output by the shale gas recovery factor prediction model; the shale gas recovery factor prediction model is trained by sample evaluation data and the shale gas recovery factor result label corresponding to the sample evaluation data.

2. The method for determining shale gas recovery rate according to claim 1, wherein The step of, through correlation analysis, determining a first target parameter in the calculation parameter set for the degree of block volume utilization whose influence degree on the recovery factor is greater than a set threshold, determining a second target parameter in the calculation parameter set for the degree of fracture network control whose influence degree on the recovery factor is greater than a set threshold, and determining a third target parameter in the calculation parameter set for the matrix production efficiency whose influence degree on the recovery factor is greater than a set threshold, includes: Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set for the degree of block volume utilization and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set for the degree of block volume utilization as the first target parameter; Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set for the degree of fracture network control and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set for the degree of fracture network control as the second target parameter; Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set for the matrix production efficiency and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set for the matrix production efficiency as the third target parameter.

3. The method for determining shale gas recovery rate according to claim 1, characterized in that The step of, through correlation analysis, determining a first target parameter in the calculation parameter set for the degree of block volume utilization whose influence degree on the recovery factor is greater than a set threshold, determining a second target parameter in the calculation parameter set for the degree of fracture network control whose influence degree on the recovery factor is greater than a set threshold, and determining a third target parameter in the calculation parameter set for the matrix production efficiency whose influence degree on the recovery factor is greater than a set threshold, includes: Calculate the Spearman correlation coefficient between each parameter in the block volume utilization degree calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the block volume utilization degree calculation parameter set as the first target parameter; Calculate the Spearman correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the fracture network control degree calculation parameter set as the second target parameter; Calculate the Spearman correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

4. The shale gas recovery rate determination method according to claim 1, wherein, The shale gas recovery factor prediction model is trained through the following steps: Repeat the following steps until the model prediction deviation value meets the preset conditions: Obtain a sample data set; the sample data set includes multiple sample evaluation data, and each sample evaluation data is constructed based on the first sample target parameter, the second sample target parameter, and the third sample target parameter; Determine multiple subsets based on the sample data set, and each subset is obtained by randomly sampling the sample data set with replacement; wherein, the number of sample evaluation data in each subset is less than the number of sample evaluation data in the sample data set; Train a decision tree based on each subset to obtain a plurality of decision trees with the same number as the number of subsets; wherein, when each node of the decision tree splits, a feature is randomly selected from some features of the sample evaluation data as the splitting attribute; Input the sample evaluation data into each decision tree, and calculate the average value based on the sample shale gas recovery factor prediction results output by all the decision trees to obtain the sample average prediction result; Determine the model prediction deviation value based on the sample average prediction result and the shale gas recovery factor result label corresponding to the sample evaluation data.

5. A device for determining the shale gas recovery rate, characterized in that Including: A data acquisition module for acquiring a block volume utilization degree calculation parameter set, a fracture network control degree calculation parameter set, and a matrix production efficiency calculation parameter set; Wherein the block volume utilization degree calculation parameter set includes multiple parameters for calculating the ratio of the controlled reserves of all wells in the block to the total reserves of the block; the fracture network control degree calculation parameter set includes multiple parameters for calculating the ratio of the single-well fracture network controlled volume to the well-controlled volume; the matrix production efficiency calculation parameter set includes multiple parameters for calculating the ratio of the gas production volume in the fracture network controlled area to the original reserves. A correlation analysis module, configured to determine, through correlation analysis, a first target parameter in the calculation parameter set of the block volume utilization degree, whose influence degree on the recovery factor is greater than a set threshold, a second target parameter in the calculation parameter set of the fracture network control degree, whose influence degree on the recovery factor is greater than the set threshold, and a third target parameter in the calculation parameter set of the matrix production efficiency, whose influence degree on the recovery factor is greater than the set threshold; A recovery factor prediction module, configured to input the evaluation data constructed by the first target parameter, the second target parameter, and the third target parameter into a shale gas recovery factor prediction model, and obtain a shale gas recovery factor prediction result output by the shale gas recovery factor prediction model; the shale gas recovery factor prediction model is trained by sample evaluation data and the shale gas recovery factor result label corresponding to the sample evaluation data.

6. The shale gas recovery rate determination device according to claim 5, characterized in that The determining, through correlation analysis, a first target parameter in the calculation parameter set of the block volume utilization degree, whose influence degree on the recovery factor is greater than a set threshold, a second target parameter in the calculation parameter set of the fracture network control degree, whose influence degree on the recovery factor is greater than the set threshold, and a third target parameter in the calculation parameter set of the matrix production efficiency, whose influence degree on the recovery factor is greater than the set threshold, includes: Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set of the block volume utilization degree and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set of the block volume utilization degree as the first target parameter; Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set of the fracture network control degree and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set of the fracture network control degree as the second target parameter; Calculating the Pearson correlation coefficient between each parameter in the calculation parameter set of the matrix production efficiency and the recovery factor, and determining the parameter corresponding to the Pearson correlation coefficient greater than the set threshold among all the Pearson correlation coefficients of the parameters in the calculation parameter set of the matrix production efficiency as the third target parameter.

7. The shale gas recovery rate determination device according to claim 5, wherein, The determining, through correlation analysis, a first target parameter in the calculation parameter set of the block volume utilization degree, whose influence degree on the recovery factor is greater than a set threshold, a second target parameter in the calculation parameter set of the fracture network control degree, whose influence degree on the recovery factor is greater than the set threshold, and a third target parameter in the calculation parameter set of the matrix production efficiency, whose influence degree on the recovery factor is greater than the set threshold, includes: Calculating the Spearman correlation coefficient between each parameter in the calculation parameter set of the block volume utilization degree and the recovery factor, and determining the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the calculation parameter set of the block volume utilization degree as the first target parameter; Calculate the Spearman correlation coefficient between each parameter in the fracture network control degree calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the fracture network control degree calculation parameter set as the second target parameter; Calculate the Spearman correlation coefficient between each parameter in the matrix production efficiency calculation parameter set and the recovery factor, and determine the parameter corresponding to the Spearman correlation coefficient greater than the set threshold among all the Spearman correlation coefficients of the parameters in the matrix production efficiency calculation parameter set as the third target parameter.

8. The shale gas recovery rate determination device according to claim 5, characterized in that, The shale gas recovery factor prediction model is trained through the following steps: Repeat the following steps until the model prediction deviation value meets the preset conditions: Obtain a sample data set; the sample data set includes a plurality of sample evaluation data, and each sample evaluation data is constructed based on the first sample target parameter, the second sample target parameter, and the third sample target parameter; Determine a plurality of subsets based on the sample data set, and each subset is obtained by random sampling with replacement from the sample data set; wherein, the number of sample evaluation data in each subset is less than the number of sample evaluation data in the sample data set; Train a decision tree based on each subset to obtain a plurality of decision trees with the same number as the number of subsets; wherein, when each node of the decision tree splits, a feature is randomly selected from a part of the features of the sample evaluation data as the splitting attribute; Input the sample evaluation data into each decision tree, and calculate the average value based on the sample shale gas recovery factor prediction results output by all the decision trees to obtain the sample average prediction result; Determine the model prediction deviation value based on the sample average prediction result and the shale gas recovery factor result label corresponding to the sample evaluation data.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the shale gas recovery factor determination method according to any one of claims 1 to 4.

10. A machine-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the shale gas recovery factor determination method according to any one of claims 1 to 4.

Images