Method and device for optimizing coal bed gas fracturing construction parameters

Abstract

The invention relates to the field of coal bed gas development, and discloses a coal bed gas fracturing construction parameter optimization method and device.Geological data, engineering data and productivity data of multiple single wells in a coal bed gas development area are collected, and the multiple single wells are classified into multiple well groups based on the geological data; determining a plurality of key first-level engineering parameters according to the parameter value of the first-level engineering parameter and the productivity data of each single well in the well group, if at least one key first-level engineering parameter has a corresponding second-level engineering parameter, continuing to determine at least one key second-level engineering parameter, and until each key multi-level engineering parameter does not have a corresponding sub-level engineering parameter. And training a coalbed methane productivity prediction model based on the key geological parameters of each single well in the well group, the parameter values of the key multi-stage engineering parameters and the productivity data so as to predict the coalbed methane productivity of the single wells in the well group, optimize single well fracturing construction parameter values, improve the optimization effect of the single well fracturing construction parameter values and improve the coalbed methane development efficiency.

Description

Technical Field

[0001] The present invention relates to the field of coalbed methane development, and in particular to a method and device for optimizing coalbed methane fracturing construction parameters. Background Art

[0002] With the continuous growth of global energy demand and the increasing depletion of traditional oil and gas resources, the development of unconventional natural gas, especially coalbed methane, has become a research hotspot in the energy field.

[0003] Among them, in the process of coalbed methane fracturing development, fracturing transformation is the core technical means to increase the production of coalbed methane wells, and is crucial to the improvement of coalbed methane production capacity.

[0004] The optimization of fracturing construction parameters in related technologies generally relies on empirical judgment or simple statistical analysis. Insufficient optimization may lead to unreasonable settings of fracturing construction parameters and reduce the efficiency of coalbed methane development. Summary of the Invention

[0005] The present invention provides a method and device for optimizing coalbed methane fracturing construction parameters, which are used to solve the problem that the optimization of fracturing construction parameters in related technologies generally relies on empirical judgment or simple statistical analysis, and insufficient optimization may lead to unreasonable setting of fracturing construction parameters and reduce the efficiency of coalbed methane development. The method optimizes the fracturing construction parameter values ​​of a single well based on a trained coalbed methane production capacity prediction model, which can effectively improve the optimization effect of the fracturing construction parameter values ​​of a single well and improve the efficiency of coalbed methane development.

[0006] In a first aspect, the present invention provides a method for optimizing coalbed methane fracturing operation parameters, comprising: Collecting geological data, engineering data, and production capacity data for multiple single wells in a coalbed methane development area; wherein the geological data for each single well includes parameter values ​​of multiple geological parameters, and the engineering data for each single well includes parameter values ​​of multiple primary engineering parameters; According to the parameter value of each geological parameter of each single well and the productivity data of each single well, screening out a plurality of key geological parameters with the greatest correlation with the productivity data from the plurality of geological parameters; Clustering the plurality of single wells based on the plurality of key geological parameters to obtain a plurality of well groups; wherein each of the well groups includes at least one of the single wells; For any of the well groups, a plurality of key primary engineering parameters are determined based on the parameter values ​​of the primary engineering parameters and the production capacity data of each of the individual wells in the well group; if at least one of the key primary engineering parameters has a corresponding secondary engineering parameter, at least one key secondary engineering parameter is continuously determined until no corresponding sub-level engineering parameter exists for each key multi-level engineering parameter; and a trained coalbed methane production capacity prediction model is generated based on the key geological parameters, the parameter values ​​of the key multi-level engineering parameters, and the production capacity data of each of the individual wells in the well group; Based on the trained coalbed methane production capacity prediction model, the fracturing construction parameter values ​​of the single well are optimized.

[0007] Optionally, the step of selecting, based on the parameter value of each geological parameter of each single well and the productivity data of each single well, a plurality of key geological parameters having the greatest correlation with the productivity data from the plurality of geological parameters includes: For any of the geological parameters, determining the correlation between the geological parameter and the productivity data according to the parameter value of the geological parameter of each of the single wells and the productivity data of each of the single wells; Among the correlations between each of the geological parameters and the production capacity data, the largest N correlations are determined, and the geological parameters corresponding to the N correlations are all determined as the key geological parameters; wherein N is an integer greater than 1.

[0008] Optionally, the first-level engineering parameter is an independent engineering parameter whose parameter value is obtained by directly collecting the parameter value independently of other parameters, or the first-level engineering parameter is the sum of similar engineering parameters; Multi-level engineering parameters are calculated based on the parameters of the previous level engineering.

[0009] Optionally, the determining of a plurality of key primary engineering parameters based on the parameter value and production capacity data of the primary engineering parameter of each single well in the well group includes: Based on the parameter value of each primary engineering parameter of each single well in the well group and the productivity data of each single well in the well group, screening out a plurality of key primary engineering parameters having the greatest correlation with the productivity data of the well group from the plurality of primary engineering parameters; Determining at least one key secondary engineering parameter includes: Among the secondary engineering parameters corresponding to all the key primary engineering parameters of the well group, at least one key secondary engineering parameter having the greatest correlation with the productivity data of the well group is screened out.

[0010] Optionally, generating a trained coalbed methane productivity prediction model based on the key geological parameters, parameter values ​​of the key multi-level engineering parameters, and productivity data of each single well in the well group includes: For any single well in the well group, the parameter value of each key geological parameter, each key primary engineering parameter, and each key multi-level engineering parameter of the single well are taken as a training sample, and the productivity data of the single well is taken as the actual productivity corresponding to the training sample; Inputting each of the training samples and the corresponding actual production capacity into the coalbed methane production capacity prediction model to be trained to perform production capacity prediction, thereby obtaining the predicted production capacity corresponding to each of the training samples; Based on the difference between the predicted production capacity and the actual production capacity corresponding to each training sample, the parameters in the coalbed methane production capacity prediction model to be trained are updated to obtain a trained coalbed methane production capacity prediction model.

[0011] Optionally, the fracturing construction parameter value of the single well includes the parameter value of each of the key geological parameters, each of the key first-level engineering parameters, and each of the key multi-level engineering parameters of the single well; Optimizing the fracturing operation parameter values ​​of the single well based on the trained coalbed methane production capacity prediction model includes: Determining the trained target coalbed methane production capacity prediction model corresponding to each of the single wells; For any of the single wells, multiple groups of fracturing construction parameter values ​​for the single well are determined based on the optimization algorithm, and the target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of fracturing construction parameter values, respectively, with the goal of maximizing the production capacity prediction value. Based on the optimization algorithm and the production capacity prediction value of each group of fracturing construction parameter values, each group of pressure construction parameters is optimized to obtain multiple groups of optimized fracturing construction parameter values. The target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of optimized fracturing construction parameter values, respectively, until the set iterative optimization end conditions are met, and the fracturing construction parameter value corresponding to the current maximum production capacity prediction value is determined as the optimal fracturing construction parameter value for the single well.

[0012] Optionally, before determining a plurality of key primary engineering parameters based on the parameter value of the primary engineering parameter and the production capacity data of each single well in the well group, the method further includes: Determining whether the number of individual wells in the well group is less than a set threshold; If the number of single wells in the well group is less than the set threshold, performing sample enhancement on the single wells in the well group to increase the number of single wells in the well group to no less than the set threshold, and then performing the step of determining a plurality of key primary engineering parameters based on the parameter value of the primary engineering parameter and the production capacity data of each of the single wells in the well group; If the number of single wells in the well group is not less than the set threshold, the step of determining multiple key primary engineering parameters based on the parameter values ​​and production capacity data of the primary engineering parameters of each single well in the well group is directly executed.

[0013] Optionally, the multiple geological parameters include at least two of coal seam depth, coal seam thickness, coal seam gas content, coal seam permeability, coal seam porosity, coal rock mechanical properties and ground stress; The production capacity data refers to the unimpeded flow rate, average daily gas production or cumulative gas production during the target production period.

[0014] Optionally, the primary engineering parameters include construction displacement, sand addition volume, net liquid volume, comprehensive sand-liquid ratio, single-stage length of fracturing, number of clusters, and number of perforations; the secondary engineering parameters include pre-fluid volume ratio, sand-carrying liquid volume ratio, displacement liquid volume ratio, first-mesh quartz sand volume, second-mesh quartz sand volume, first-viscosity slickwater volume, and second-viscosity slickwater volume; Among them, the proportion of pre-liquid volume, the proportion of sand-carrying liquid volume, the proportion of displacement liquid volume, the first viscosity slick water dosage, and the second viscosity slick water dosage are the secondary engineering parameters corresponding to the net liquid volume; the first mesh size quartz sand dosage and the second mesh size quartz sand dosage are the secondary engineering parameters corresponding to the sand addition volume.

[0015] In a second aspect, the present invention provides a device for optimizing coalbed methane fracturing construction parameters, comprising: A collection unit, configured to collect geological data, engineering data, and production capacity data of a plurality of single wells in a coalbed methane development area; wherein the geological data of each single well includes parameter values ​​of a plurality of geological parameters, and the engineering data of each single well includes parameter values ​​of a plurality of primary engineering parameters; a screening unit, configured to screen out a plurality of key geological parameters having the greatest correlation with the productivity data from the plurality of geological parameters based on the parameter value of each geological parameter of each single well and the productivity data of each single well; a clustering unit, configured to cluster the plurality of single wells based on the plurality of key geological parameters to obtain a plurality of well groups; wherein each of the well groups includes at least one of the single wells; A first determining unit is configured to determine, for any of the well groups, a plurality of key primary engineering parameters based on the parameter value of the primary engineering parameter and the production capacity data of each of the single wells in the well group; a second determining unit configured to, if at least one of the key first-level engineering parameters has a corresponding second-level engineering parameter, continue to determine at least one key second-level engineering parameter until no corresponding sub-level engineering parameter exists for each key multi-level engineering parameter; a generating unit, configured to generate a trained coalbed methane productivity prediction model based on the key geological parameters, parameter values ​​of the key multi-level engineering parameters, and productivity data of each single well in the well group; The optimization unit is used to optimize the fracturing construction parameter values ​​of the single well based on the trained coalbed methane production capacity prediction model.

[0016] The coalbed methane fracturing construction parameter optimization method and device provided by the present invention can collect geological data, engineering data and production capacity data of multiple single wells in the coalbed methane development area, and classify the multiple single wells into multiple well groups based on the geological data. For any well group, multiple key first-level engineering parameters are determined according to the parameter values ​​and production capacity data of the first-level engineering parameters of each single well in the well group. If at least one key first-level engineering parameter has a corresponding second-level engineering parameter, at least one key second-level engineering parameter is continued to be determined until each key multi-level engineering parameter has no corresponding sub-level engineering parameter. Based on the key geological parameters, parameter values ​​and production capacity data of each single well in the well group, a trained coalbed methane production capacity prediction model is generated. The trained coalbed methane production capacity prediction model of the present invention can be used to predict the coalbed methane production capacity of a single well in the corresponding well group. It can effectively adapt to the complex and changeable geological and engineering conditions of coalbed methane, reveal the relationship between geological data, engineering data and production capacity data, realize the prediction of the coalbed methane production capacity of a single well, and improve the accuracy of coalbed methane production capacity prediction. Based on the trained coalbed methane production capacity prediction model, the fracturing construction parameter value of a single well is optimized, which can effectively improve the optimization effect of the fracturing construction parameter value of a single well and improve the efficiency of coalbed methane development. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or related technologies, the following is a brief introduction to the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flow chart of a method for optimizing coalbed methane fracturing construction parameters provided by an embodiment of the present invention; Figure 2 A flow chart of another method for optimizing coalbed methane fracturing construction parameters provided by an embodiment of the present invention; Figure 3 A schematic diagram of the correlation between multiple primary engineering parameters and production capacity data of a well group provided in an embodiment of the present invention; Figure 4 A schematic diagram of the correlation between multiple primary engineering parameters and production capacity data of a Class II well group provided by an embodiment of the present invention; Figure 5 A schematic diagram of the correlation between multiple primary engineering parameters and production capacity data for three types of well groups provided in an embodiment of the present invention; Figure 6 A schematic diagram of the correlation between multiple secondary engineering parameters and production capacity data of a well group provided by an embodiment of the present invention; Figure 7 A schematic diagram of the correlation between multiple secondary engineering parameters and production capacity data of a second type of well group provided by an embodiment of the present invention; Figure 8 A schematic diagram of the correlation between multiple secondary engineering parameters and production capacity data for three types of well groups provided in an embodiment of the present invention; Figure 9 A schematic structural diagram of a device for optimizing coalbed methane fracturing construction parameters provided by an embodiment of the present invention; Figure 10 A schematic structural diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0020] The following combination Figures 1-8 The present invention describes the method for optimizing coalbed methane fracturing construction parameters.

[0021] like Figure 1 As shown, this embodiment proposes a first method for optimizing coalbed methane fracturing construction parameters, which may include the following steps: S101. Collect geological data, engineering data, and production capacity data of multiple single wells in a coalbed methane development area; wherein the geological data of each single well includes parameter values ​​of multiple geological parameters, and the engineering data of each single well includes parameter values ​​of multiple primary engineering parameters.

[0022] Specifically, the coalbed methane development area may be a region where deep coalbed methane development is being carried out, and the coalbed methane development area may include multiple single wells.

[0023] Optionally, a first-level engineering parameter is an independent engineering parameter whose value is obtained by directly collecting data without relying on other parameters, or a first-level engineering parameter is the sum of similar engineering parameters. A first-level engineering parameter may have corresponding second-level engineering parameters, third-level engineering parameters, and other multi-level engineering parameters.

[0024] Multi-level engineering parameters are calculated based on the parameters of the previous level engineering.

[0025] Optional first-level engineering parameters include construction displacement, sand addition volume, net liquid volume, comprehensive sand-liquid ratio, single-stage length of fracturing, number of clusters, and number of perforations; second-level engineering parameters include pre-fluid volume ratio, sand-carrying liquid volume ratio, displacement liquid volume ratio, first-mesh quartz sand volume, second-mesh quartz sand volume, first-viscosity slickwater volume, and second-viscosity slickwater volume. Among them, the proportion of pre-liquid volume, the proportion of sand-carrying liquid volume, the proportion of displacement liquid volume, the first viscosity slick water consumption, and the second viscosity slick water consumption are the secondary engineering parameters corresponding to the net liquid volume; the first mesh quartz sand consumption and the second mesh quartz sand consumption are the secondary engineering parameters corresponding to the sand addition volume.

[0026] Each well has corresponding geological data, engineering data, and production capacity data. The geological data of each well may include the same values ​​for multiple geological parameters and multiple primary engineering parameters. For example, the geological data of each well may include the values ​​for geological parameters A, B, and C, and the engineering data of each well may include the values ​​for primary engineering parameters D, E, and F. The values ​​for the same geological parameter in each well may be equal or unequal. The values ​​for the same primary engineering parameter in each well may be equal or unequal.

[0027] The production capacity data of a single well refers to the coalbed methane production capacity data of a single well, such as the cumulative gas production or average daily gas production during the target production period.

[0028] Optionally, the multiple geological parameters include at least two of coal seam depth, coal seam thickness, coal seam gas content, coal seam permeability, coal seam porosity, coal rock mechanical properties and ground stress; The production capacity data is the unobstructed flow, average daily gas production or cumulative gas production during the target production period.

[0029] Specifically, this embodiment can collect geological data, engineering data and production capacity data of each single well separately.

[0030] S102 , based on the parameter value of each geological parameter of each single well and the productivity data of each single well, select a plurality of key geological parameters with the greatest correlation with the productivity data from among the plurality of geological parameters.

[0031] Specifically, this embodiment can use the parameter values ​​of all geological parameters of all single wells and the productivity data of all single wells to screen out multiple geological parameters with the greatest correlation with the productivity data from all geological parameters, and each screened geological parameter is a geological parameter.

[0032] Optionally, step S102 may include: For any geological parameter, the correlation between the geological parameter and the productivity data is determined based on the parameter value of the geological parameter of each single well and the productivity data of each single well; Among the correlations between each geological parameter and the production capacity data, the largest N correlations are determined, and the geological parameters corresponding to the N correlations are all determined as key geological parameters; wherein N is an integer greater than 1.

[0033] Specifically, this embodiment can determine the correlation between each geological parameter and the production capacity data through correlation analysis or principal component analysis. This embodiment can then sort all the determined correlations from largest to smallest, determine the top N correlations, and determine the geological parameters corresponding to the top N correlations as key geological parameters.

[0034] S103. Clustering the multiple single wells based on multiple key geological parameters to obtain multiple well groups; wherein each well group includes at least one single well.

[0035] Specifically, this embodiment can use all the selected key geological parameters and a clustering algorithm to cluster all individual wells, calculate the distance between each individual well in the geological feature space, and cluster wells with similar geological conditions into one category based on the distance. This will generate well groups of different geological types, and then divide all individual wells into multiple well groups. Each well group can include one or more individual wells.

[0036] Optionally, the clustering algorithm may be a K-means clustering algorithm or a hierarchical clustering algorithm.

[0037] S104. For any well group, determine a plurality of key primary engineering parameters based on the parameter values ​​and production capacity data of the primary engineering parameters of each single well in the well group.

[0038] Specifically, in this embodiment, for any well group, multiple key primary engineering parameters can be determined based on the parameter values ​​and production capacity data of the primary engineering parameters of each single well in the well group.

[0039] Optionally, step S104 may include: Based on the parameter value of each primary engineering parameter of each single well in the well group and the productivity data of each single well in the well group, multiple key primary engineering parameters with the greatest correlation with the productivity data of the well group are screened out from multiple primary engineering parameters.

[0040] Specifically, in this embodiment, for any well group, the correlation between each primary engineering parameter and the production capacity data of each individual well in the well group can be determined based on the parameter value of each primary engineering parameter of each individual well in the well group, and the production capacity data of each individual well in the well group. The largest P correlations are then determined, and the primary engineering parameters corresponding to the largest P correlations are determined as the multiple key primary engineering parameters of the well group. P is a positive integer that can be set by a technician based on actual needs and is not limited in this embodiment.

[0041] It should be noted that this embodiment can refer to the process of determining key geological parameters to screen key first-level engineering parameters from multiple first-level engineering parameters, and the relevant execution process will not be repeated here.

[0042] S105: If at least one key first-level engineering parameter has a corresponding second-level engineering parameter, continue to determine at least one key second-level engineering parameter until each key multi-level engineering parameter has no corresponding sub-level engineering parameter.

[0043] Optionally, the above-mentioned determination of at least one key secondary engineering parameter includes: Among the secondary engineering parameters corresponding to all key primary engineering parameters of the well group, at least one key secondary engineering parameter having the greatest correlation with the productivity data of the well group is selected.

[0044] Specifically, for any well group, if one or more key primary engineering parameters of the well group have corresponding secondary engineering parameters, then this embodiment can screen out multiple key secondary engineering parameters with the greatest correlation with production capacity data from all secondary engineering parameters corresponding to these key primary engineering parameters.

[0045] Specifically, for any well group, if one or more key secondary engineering parameters of the well group have corresponding third-level engineering parameters, then this embodiment can screen out multiple key third-level engineering parameters with the greatest correlation with production capacity data from all third-level engineering parameters corresponding to these key secondary engineering parameters, until each key multi-level engineering parameter cannot be further classified.

[0046] It should be noted that this embodiment can refer to the process of determining key geological parameters to screen key secondary engineering parameters from multiple secondary engineering parameters, or screen key tertiary engineering parameters from multiple tertiary engineering parameters. The relevant execution process will not be repeated here.

[0047] S106. Generate a trained coalbed methane production capacity prediction model based on the key geological parameters, parameter values ​​of key multi-level engineering parameters, and production capacity data of each single well in the well group.

[0048] Specifically, for any well group, this embodiment can train the coalbed methane production capacity prediction model to be trained based on the parameter values ​​of all key geological parameters of all single wells in the well group, as well as the parameter values ​​and production capacity data of key multi-level engineering parameters of all single wells in the well group, to obtain a trained coalbed methane production capacity prediction model.

[0049] It is understandable that each well group has a corresponding trained coalbed methane production capacity prediction model.

[0050] Optionally, step S106 may include: For any single well in the well group, the parameter values ​​of each key geological parameter, each key first-level engineering parameter, and each key multi-level engineering parameter of the single well are taken as a training sample, and the productivity data of the single well is taken as the actual productivity corresponding to the training sample; Input each training sample and the corresponding actual production capacity into the coalbed methane production capacity prediction model to be trained to perform production capacity prediction, and obtain the predicted production capacity corresponding to each training sample; Based on the difference between the predicted capacity and the actual capacity corresponding to each training sample, the parameters in the coalbed methane capacity prediction model to be trained are updated to obtain a trained coalbed methane capacity prediction model.

[0051] Specifically, this embodiment can use each selected key geological parameter, each key primary engineering parameter, and each key multi-level engineering parameter as input variables and coalbed methane production capacity as output variables to establish and train a coalbed methane production capacity prediction model.

[0052] Among them, the coalbed methane production capacity prediction model to be trained can be a neural network model with preliminary data prediction performance, or it can be a linear regression model, a decision tree regression model, a random forest regression model, a support vector machine regression model, etc.

[0053] It should be noted that this embodiment can establish a variety of coalbed methane production capacity regression prediction models, such as linear regression models, decision tree regression models, random forest regression models, support vector machine regression models, etc. These models are trained and verified using historical data. By calculating evaluation indicators such as mean square error, mean absolute error, and coefficient of determination, the performance of each model is compared, and the model with the highest prediction accuracy and the strongest generalization ability is selected as the final production capacity prediction model.

[0054] It is understandable that some first-level engineering parameters may not have corresponding multi-level engineering parameters, while some first-level engineering parameters do have corresponding multi-level engineering parameters. For first-level engineering parameters that do not have corresponding multi-level engineering parameters, this embodiment can directly use the first-level engineering parameter as part of the input variable to construct training samples during the coalbed methane production capacity prediction model training process. For first-level engineering parameters that have multi-level engineering parameters, this embodiment can use only the last-level engineering parameter of the first-level engineering parameter as part of the input variable, or all multi-level engineering parameters in the first-level engineering parameter classification process as part of the input variable, or even the first-level engineering parameter as part of the input variable to construct training samples.

[0055] S107. Based on the trained coalbed methane production capacity prediction model, optimize the fracturing construction parameter values ​​of a single well.

[0056] Specifically, this embodiment can collaboratively use a trained coalbed methane production capacity prediction model to optimize the fracturing construction parameter values ​​of a single well.

[0057] Optionally, the fracturing construction parameter values ​​of a single well include the parameter values ​​of each key geological parameter, each key primary engineering parameter, and each key multi-level engineering parameter of the single well. In this case, step S107 may include: Determine the trained target coalbed methane production capacity prediction model corresponding to each single well; For any single well, multiple groups of fracturing construction parameter values ​​for the single well are determined based on the optimization algorithm, and the target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of fracturing construction parameter values ​​respectively. With the goal of maximizing the production capacity prediction value, each group of pressure construction parameters is optimized based on the optimization algorithm and the production capacity prediction value of each group of fracturing construction parameter values ​​to obtain multiple groups of optimized fracturing construction parameter values. The target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of optimized fracturing construction parameter values ​​respectively until the set iterative optimization end conditions are met, and the fracturing construction parameter value corresponding to the current maximum production capacity prediction value is determined as the optimal fracturing construction parameter value for the single well.

[0058] Specifically, this embodiment can utilize a coalbed methane production capacity prediction model in conjunction with a particle swarm optimization algorithm to optimize fracturing parameters for each individual well. Each particle in the particle swarm optimization algorithm represents a set of fracturing parameter values ​​(including key geological and engineering parameters). Through continuous iterative search within the parameter space, with maximizing coalbed methane production capacity as the objective function, the coalbed methane production capacity prediction model calculates the corresponding production capacity prediction value for each particle. Based on the predicted value, the particle's position and velocity are adjusted until the optimal fracturing parameter solution is found, outputting the optimal fracturing parameter values.

[0059] The coalbed methane fracturing construction parameter optimization method proposed in this embodiment can collect geological data, engineering data and production capacity data of multiple single wells in the coalbed methane development area, and classify the multiple single wells into multiple well groups based on the geological data. For any well group, multiple key first-level engineering parameters are determined according to the parameter values ​​and production capacity data of the first-level engineering parameters of each single well in the well group. If at least one key first-level engineering parameter has a corresponding second-level engineering parameter, at least one key second-level engineering parameter is continued to be determined until each key multi-level engineering parameter has no corresponding sub-level engineering parameter. Based on the key geological parameters, parameter values ​​and production capacity data of each single well in the well group, a trained coalbed methane production capacity prediction model is generated. The trained coalbed methane production capacity prediction model in this embodiment can be used to predict the coalbed methane production capacity of a single well in the corresponding well group. It can effectively adapt to the complex and changeable geological and engineering conditions of coalbed methane, reveal the relationship between geological data, engineering data and production capacity data, realize the prediction of the coalbed methane production capacity of a single well, and improve the accuracy of coalbed methane production capacity prediction. Optimizing the fracturing construction parameter values ​​of a single well based on the trained coalbed methane production capacity prediction model can effectively improve the optimization effect of the fracturing construction parameter values ​​of a single well and improve the efficiency of coalbed methane development.

[0060] based on Figure 1 This embodiment proposes a second method for optimizing coalbed methane fracturing construction parameters. Before step S104, the method may further include: Determine whether the number of single wells in the well group is less than the set threshold; If the number of single wells in the well group is less than the set threshold, sample enhancement is performed on the single wells in the well group to increase the number of single wells in the well group to no less than the set threshold, and then step S104 is executed.

[0061] If the number of single wells in the well group is not less than the set threshold, step S104 is directly executed.

[0062] It should be noted that the number of individual wells (i.e., the number of samples) within the clustered well groups may be unbalanced, with some types having a higher number of samples and others having a lower number. To mitigate the impact of sample imbalance on subsequent analysis, a synthetic minority oversampling technique can be used to synthesize minority class samples to balance the number of samples across all types. Specifically, for each minority class sample, several samples are selected from its nearest neighbors and synthesized using methods such as linear interpolation to increase the number of minority class samples and achieve a relative balance between the number of samples across all types.

[0063] Specifically, this embodiment can perform engineering feature selection on different types of samples after sample balancing. First, feature selection methods (such as mutual information method, recursive feature elimination method, model-based feature selection method, etc.) can be used on the primary engineering parameters to analyze the relationship between the primary engineering parameters and coalbed methane production capacity, and determine the most important primary engineering parameters for each type, namely the key primary engineering parameters. Based on the top-ranked parameters determined by the primary engineering parameters, feature selection methods are further used to analyze the relationship between the secondary engineering parameters and coalbed methane production capacity, and determine the most important secondary engineering parameters for each type, namely the key secondary engineering parameters. For example, for a certain well group, fracturing fluid viscosity and fracturing pump pressure and total sand content may be determined as the most important primary engineering parameters. Further analysis shows that high-viscosity fracturing fluid and 70-140 mesh quartz sand usage are the most important secondary engineering parameters, allowing for the next step of parameter optimization design.

[0064] Specifically, this embodiment can select key geological parameters and key secondary engineering parameters as input variables and coalbed methane production capacity as the output variable to establish multiple coalbed methane production capacity regression prediction models, such as linear regression models, decision tree regression models, random forest regression models, and support vector machine regression models. These models are trained and validated using historical data. The performance of each model is compared by calculating evaluation indicators such as mean square error, mean absolute error, and coefficient of determination. The model with the highest prediction accuracy and the strongest generalization ability is selected as the final production capacity prediction model.

[0065] The coalbed methane fracturing construction parameter optimization method proposed in this embodiment can classify wells into different types by preliminary feature selection and geological feature clustering of geological data and engineering data of deep coalbed methane, taking into account the main influence of geological parameters; to address the problem of imbalanced cluster samples, sample enhancement technology is used to balance samples, thereby improving the accuracy of analysis; engineering features are selected for different types of samples respectively, so that the determination of engineering parameters is more accurate; multiple production capacity regression prediction models are established and the optimal model is selected, thereby improving the accuracy of production capacity prediction; finally, the production capacity prediction model is used in conjunction with the particle swarm optimization algorithm to optimize the construction parameters of a single well, with the goal of maximizing production capacity, and the optimal construction parameter plan can be output, thereby improving the development efficiency and production capacity of deep coalbed methane, which has significant economic benefits and application value.

[0066] The optimization methods for fracturing construction parameters in related technologies often rely on empirical judgment or simple statistical analysis, making it difficult to fully utilize the complex data of deep coalbed methane geological engineering. In particular, when processing data, if the geological parameters and engineering parameters are not systematically selected and analyzed, many parameters are often treated equally, resulting in the submerging of key influencing factors and an inability to accurately grasp the actual impact of different parameters on deep coalbed methane production capacity. In terms of cluster analysis, related technologies often ignore the problem of sample imbalance, and the characteristics of minority samples are not fully explored. The clustering results lack reliability, which affects the targeted optimization of subsequent fracturing construction parameters. In the production capacity prediction stage, the models used in related technologies are too simplistic and cannot adapt to the complex and changing geological and engineering conditions of deep coalbed methane. The prediction accuracy is low, making it difficult to effectively guide the adjustment of fracturing construction parameters. Moreover, most optimization methods in related technologies do not closely integrate production capacity prediction and parameter optimization, and cannot achieve precise optimization with the goal of maximizing production capacity. This leads to unreasonable construction parameter settings, reduced deep coalbed methane development efficiency, and caused resource waste and increased costs.

[0067] The inventors of the present invention have found that the effect of fracturing transformation directly depends on the matching between geological and engineering parameters. However, the main controlling factors of deep coalbed methane post-compression production and their interaction mechanisms are not yet fully understood, which greatly limits the improvement of fracturing technology level and post-compression production. The development of data analysis, mining and machine learning technology has provided a new approach to solving this problem. Through in-depth analysis of a large amount of geological-engineering data, machine learning can reveal hidden correlations and more accurately identify key parameters that affect the effect of deep coalbed methane fracturing. In addition, the production prediction model based on machine learning can provide a scientific basis for the optimization of fracturing parameters, help to improve the level of deep coalbed methane fracturing technology and post-compression production, and reduce development costs.

[0068] In order to solve the problems of difficulty in evaluating the effect of deep coalbed methane fracturing, fuzzy main controlling factors of post-fracturing production capacity, and time-consuming calculation of fracturing process parameter optimization design, this embodiment can be based on the intelligent optimization method of fracturing construction parameters based on geological classification and classification. Based on machine learning methods, it utilizes the analysis and mining of a large amount of reservoir geological, engineering, and production data to form a hierarchical classification of reservoirs in different areas of the target block and intelligent optimization matching of their fracturing parameters, providing theoretical guidance and technical support for the high-quality design of deep coalbed methane fracturing plans, iteration of fracturing process technology, on-site construction adjustment decision-making, and improvement of post-fracturing production capacity.

[0069] This embodiment can generate a large amount of data after deep coalbed methane reservoir transformation and fracturing development, introduce machine learning to analyze the main controlling factors affecting production capacity in a layered manner and optimize fracturing construction parameters, achieve scientific and effective application, improve the development efficiency and production capacity of deep coalbed methane, have significant economic benefits and application value, and are of great significance to ensuring the safe and efficient development of oil and gas reservoirs.

[0070] like Figure 2 As shown, in the other coalbed methane fracturing operation parameter optimization method proposed in this embodiment, the following steps may be included: Collect data related to deep coalbed methane geological parameters, engineering parameters, and production parameters. Production parameter data refers to production capacity data. Pre-process the collected data, including cleaning and standardization, to remove outliers and missing values.

[0071] Classification based on geological parameters: In this embodiment, multiple key geological parameters with the greatest correlation with production capacity data can be selected from multiple geological parameters, and multiple single wells can be classified into different types of well groups based on the multiple key geological parameters.

[0072] Determine whether the samples in the well cluster are balanced, that is, whether the number of individual wells in each well cluster is less than a set threshold. For well clusters with a small number of individual well samples, the Synthetic Minority Over-sampling Technique (SMOTE) can be used to enhance the data of the samples in the well cluster.

[0073] Afterwards, the main controlling factors of different types of multi-stage engineering parameters are determined, that is, the key primary engineering parameters and key multi-stage engineering parameters of different well groups are determined respectively.

[0074] Based on the key multi-level engineering parameters and key geological parameters of different well groups, corresponding coalbed methane production capacity prediction models are established. Based on the production capacity prediction model and intelligent optimization algorithm, the fracturing operation parameters of individual wells are collaboratively optimized.

[0075] Taking a deep coalbed methane (CBM) block as an example, geological and engineering data were collected from multiple wells within the block, including coal seam depth, thickness, permeability, drilling fluid density, and fracturing fluid viscosity. Correlation analysis revealed a high correlation between the depth, thickness, and permeability of the geological data and CBM production data, identifying them as key geological features. Using the K-means clustering algorithm, all wells were classified into three geological types. By calculating the Euclidean distance between each well in these three geological feature spaces, wells with similar distances were clustered together.

[0076] Regarding sample balance and engineering feature selection. Assuming that after clustering, the number of samples in one type is small, SMOTE can be used. For each sample of this type, samples are selected from its five nearest neighbors and synthesized through linear interpolation to ensure that the number of samples of this type is similar to that of the other types. Engineering features are then selected for each type of sample. The mutual information method is used to calculate the mutual information value between engineering parameters and production capacity, identifying fracturing fluid viscosity and fracturing pump pressure as key engineering parameters.

[0077] like Figures 3 to 5 As shown in Figure 2, the key primary engineering parameters corresponding to different well groups are shown in Figure 2. Figure 3 Reflects the correlation between multiple primary engineering parameters and production capacity data of a type of well group, Figure 4 Reflects the correlation between multiple primary engineering parameters and production capacity data of the second type of well group, Figure 5 Reflects the correlation between multiple primary engineering parameters and production capacity data of three types of well groups.

[0078] like Figures 6 to 8 As shown in Figure 2, the key secondary engineering parameters corresponding to different well groups are shown in Figure 2. Figure 6 Reflects the correlation between multiple secondary engineering parameters and production capacity data of a type of well group, Figure 7 Reflects the correlation between multiple secondary engineering parameters and production capacity data of the second type of well group, Figure 8 Reflects the correlation between multiple secondary engineering parameters and production capacity data of three types of well groups.

[0079] In this example, a support vector machine regression model, a random forest regression model, and a lightweight gradient boosting regression model can be established, using 80% of the historical data for training and 20% of the data for validation. The mean squared error of each model is calculated, and the lightweight gradient boosting regression model has the smallest mean squared error, so it is preferred as the final model.

[0080] Specifically, this embodiment can also use the preferred random forest regression model as the objective function and employ a particle swarm optimization algorithm to optimize the fracturing fluid viscosity and fracturing pump pressure for a single well, setting the particle swarm size to 50 and the number of iterations to 100. Through iterative search, a parameter combination that maximizes the predicted production capacity is found.

[0081] The coalbed methane fracturing construction parameter optimization method proposed in this embodiment can first obtain deep coalbed methane geological engineering parameter data, determine the dominant position of geological parameters through preliminary feature selection, and classify wells based on key geological features using a clustering algorithm; to address the imbalance of cluster samples, use SMOTE technology to balance samples, then perform engineering feature selection on different types of samples to clarify important engineering parameters; establish multiple coalbed methane production capacity regression prediction models and select the optimal model; finally, combine the production capacity prediction model with the particle swarm optimization algorithm to optimize the single well construction parameters with the goal of maximizing production capacity, and output the optimal solution. This embodiment uses multi-link optimization to accurately analyze deep coalbed methane geological engineering parameters, effectively solve the problem of sample imbalance, significantly improve the accuracy of coalbed methane production capacity prediction and the optimization effect of construction parameters, improve the efficiency and production capacity of deep coalbed methane development, and has good economic benefits and broad application prospects.

[0082] like Figure 9 As shown, this embodiment provides a device for optimizing coalbed methane fracturing operation parameters, which may include: The collection unit 901 is used to collect geological data, engineering data, and production capacity data of multiple single wells in the coalbed methane development area; wherein the geological data of each single well includes parameter values ​​of multiple geological parameters, and the engineering data of each single well includes parameter values ​​of multiple primary engineering parameters; A screening unit 902 is configured to screen out a plurality of key geological parameters having the greatest correlation with the productivity data from a plurality of geological parameters based on the parameter value of each geological parameter of each single well and the productivity data of each single well; A clustering unit 903 is configured to cluster multiple single wells based on multiple key geological parameters to obtain multiple well groups, wherein each well group includes at least one single well; The first determining unit 904 is configured to determine, for any well group, a plurality of key primary engineering parameters based on the parameter values ​​and production capacity data of the primary engineering parameters of each single well in the well group; The second determining unit 905 is configured to, if at least one key first-level engineering parameter has a corresponding second-level engineering parameter, continue to determine at least one key second-level engineering parameter until no corresponding sub-level engineering parameter exists for each key multi-level engineering parameter; A generating unit 906 is configured to generate a trained coalbed methane productivity prediction model based on the key geological parameters, parameter values ​​of key multi-level engineering parameters, and productivity data of each single well in the well group; The optimization unit 907 is used to optimize the fracturing operation parameter values ​​of a single well based on the trained coalbed methane production capacity prediction model.

[0083] It should be noted that the processing of the collection unit 901, the screening unit 902, the clustering unit 903, the first determination unit 904, the second determination unit 905, the generation unit 906 and the optimization unit 907 and the beneficial effects thereof can be referred to in detail. Figure 1 Steps S101 to S107 in the above are not described in detail.

[0084] Optionally, the screening unit 902 is further configured to: For any geological parameter, the correlation between the geological parameter and the productivity data is determined based on the parameter value of the geological parameter of each single well and the productivity data of each single well; Among the correlations between each geological parameter and the production capacity data, the largest N correlations are determined, and the geological parameters corresponding to the N correlations are all determined as key geological parameters; wherein N is an integer greater than 1.

[0085] Optionally, the first-level engineering parameter is an independent engineering parameter that does not depend on other parameters and obtains its parameter value by directly collecting, or the first-level engineering parameter is the sum of similar engineering parameters; Multi-level engineering parameters are calculated based on the parameters of the previous level engineering.

[0086] Optionally, the first determining unit 904 is further configured to: Based on the parameter value of each primary engineering parameter of each single well in the well group and the productivity data of each single well in the well group, multiple key primary engineering parameters with the greatest correlation with the productivity data of the well group are screened out from the multiple primary engineering parameters; The second determining unit 905 is further configured to: Among the secondary engineering parameters corresponding to all key primary engineering parameters of the well group, at least one key secondary engineering parameter having the greatest correlation with the productivity data of the well group is selected.

[0087] Optionally, the generating unit 906 is further configured to: For any single well in the well group, the parameter values ​​of each key geological parameter, each key first-level engineering parameter, and each key multi-level engineering parameter of the single well are taken as a training sample, and the productivity data of the single well is taken as the actual productivity corresponding to the training sample; Input each training sample and the corresponding actual production capacity into the coalbed methane production capacity prediction model to be trained to perform production capacity prediction, and obtain the predicted production capacity corresponding to each training sample; Based on the difference between the predicted capacity and the actual capacity corresponding to each training sample, the parameters in the coalbed methane capacity prediction model to be trained are updated to obtain a trained coalbed methane capacity prediction model.

[0088] Optionally, the fracturing construction parameter values ​​of a single well include the parameter values ​​of each key geological parameter, each key first-level engineering parameter, and each key multi-level engineering parameter of the single well; The optimization unit 907 is further configured to: For any single well, multiple groups of fracturing construction parameter values ​​for the single well are determined based on the optimization algorithm, and the target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of fracturing construction parameter values ​​respectively. With the goal of maximizing the production capacity prediction value, each group of pressure construction parameters is optimized based on the optimization algorithm and the production capacity prediction value of each group of fracturing construction parameter values ​​to obtain multiple groups of optimized fracturing construction parameter values. The target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of optimized fracturing construction parameter values ​​respectively until the set iterative optimization end conditions are met, and the fracturing construction parameter value corresponding to the current maximum production capacity prediction value is determined as the optimal fracturing construction parameter value for the single well.

[0089] Optionally, the above device further comprises: a sample enhancement unit; Sample enhancement unit for: Before determining a plurality of key primary engineering parameters based on the parameter values ​​and production capacity data of the primary engineering parameters of each single well in the well group, determining whether the number of single wells in the well group is less than a set threshold; If the number of individual wells in the well group is less than a set threshold, sample enhancement is performed on the individual wells in the well group to increase the number of individual wells in the well group to no less than the set threshold, and then a step of determining a plurality of key primary engineering parameters based on the parameter value and production capacity data of the primary engineering parameter of each individual well in the well group is performed; If the number of individual wells in the well group is not less than the set threshold, the step of determining multiple key primary engineering parameters based on the parameter values ​​and production capacity data of the primary engineering parameters of each individual well in the well group is directly executed.

[0090] Optionally, the multiple geological parameters include at least two of coal seam depth, coal seam thickness, coal seam gas content, coal seam permeability, coal seam porosity, coal rock mechanical properties and ground stress; The production capacity data is the unobstructed flow, average daily gas production or cumulative gas production during the target production period.

[0091] Optional first-level engineering parameters include construction displacement, sand addition volume, net liquid volume, comprehensive sand-liquid ratio, single-stage length of fracturing, number of clusters, and number of perforations; second-level engineering parameters include pre-fluid volume ratio, sand-carrying liquid volume ratio, displacement liquid volume ratio, first-mesh quartz sand volume, second-mesh quartz sand volume, first-viscosity slickwater volume, and second-viscosity slickwater volume. Among them, the proportion of pre-liquid volume, the proportion of sand-carrying liquid volume, the proportion of displacement liquid volume, the first viscosity slick water consumption, and the second viscosity slick water consumption are the secondary engineering parameters corresponding to the net liquid volume; the first mesh quartz sand consumption and the second mesh quartz sand consumption are the secondary engineering parameters corresponding to the sand addition volume.

[0092] The coalbed methane fracturing construction parameter optimization device proposed in this embodiment and the trained coalbed methane production capacity prediction model can be used to predict the coalbed methane production capacity of a single well in the corresponding well group. It can effectively adapt to the complex and changeable geological and engineering conditions of coalbed methane, reveal the relationship between geological data, engineering data and production capacity data, realize the prediction of the coalbed methane production capacity of a single well, and improve the accuracy of coalbed methane production capacity prediction. Optimizing the fracturing construction parameter value of a single well based on the trained coalbed methane production capacity prediction model can effectively improve the optimization effect of the fracturing construction parameter value of a single well and improve the efficiency of coalbed methane development.

[0093] The coalbed methane fracturing construction parameter optimization device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0094] The embodiment of the present invention also provides a computer device having the above Figure 9The coalbed methane fracturing construction parameter optimization device shown.

[0095] See also Figure 10 , a structural diagram of a computer device provided by an optional embodiment of the present invention, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 10 A processor 10 is taken as an example.

[0096] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0097] The memory 20 stores instructions that can be executed by at least one processor 10, so as to enable at least one processor 10 to execute the method shown in the above embodiment.

[0098] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function. The data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0099] The memory 20 may include volatile memory, such as random access memory. The memory may also include non-volatile memory, such as flash memory, a hard disk, or a solid-state drive. The memory 20 may also include a combination of the above types of memory.

[0100] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0101] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for optimizing coalbed methane fracturing construction parameters, characterized in that: include: Collecting geological data, engineering data, and production capacity data for multiple single wells in a coalbed methane development area; wherein the geological data for each single well includes parameter values ​​of multiple geological parameters, and the engineering data for each single well includes parameter values ​​of multiple primary engineering parameters; According to the parameter value of each geological parameter of each single well and the productivity data of each single well, screening out a plurality of key geological parameters with the greatest correlation with the productivity data from the plurality of geological parameters; Clustering the plurality of single wells based on the plurality of key geological parameters to obtain a plurality of well groups; wherein each of the well groups includes at least one of the single wells; For any of the well groups, a plurality of key primary engineering parameters are determined based on the parameter values ​​of the primary engineering parameters and the production capacity data of each of the individual wells in the well group; if at least one of the key primary engineering parameters has a corresponding secondary engineering parameter, at least one key secondary engineering parameter is continuously determined until no corresponding sub-level engineering parameter exists for each key multi-level engineering parameter; and a trained coalbed methane production capacity prediction model is generated based on the key geological parameters, the parameter values ​​of the key multi-level engineering parameters, and the production capacity data of each of the individual wells in the well group; Based on the trained coalbed methane production capacity prediction model, the fracturing construction parameter values ​​of the single well are optimized.

2. The method according to claim 1, characterized in that The step of selecting, based on the parameter value of each geological parameter of each single well and the productivity data of each single well, a plurality of key geological parameters having the greatest correlation with the productivity data from the plurality of geological parameters includes: For any of the geological parameters, determining the correlation between the geological parameter and the productivity data according to the parameter value of the geological parameter of each of the single wells and the productivity data of each of the single wells; Among the correlations between each of the geological parameters and the production capacity data, the largest N correlations are determined, and the geological parameters corresponding to the N correlations are all determined as the key geological parameters; wherein N is an integer greater than 1.

3. The method according to claim 1, characterized in that The first-level engineering parameter is an independent engineering parameter whose parameter value is obtained by directly collecting the parameter value independently of other parameters, or the first-level engineering parameter is the sum of engineering parameters of the same type; Multi-level engineering parameters are calculated based on the parameters of the previous level engineering.

4. The method according to claim 3, characterized in that Determining a plurality of key primary engineering parameters based on the parameter values ​​and production capacity data of the primary engineering parameters of each single well in the well group includes: Based on the parameter value of each primary engineering parameter of each single well in the well group and the productivity data of each single well in the well group, screening out a plurality of key primary engineering parameters having the greatest correlation with the productivity data of the well group from the plurality of primary engineering parameters; Determining at least one key secondary engineering parameter includes: Among the secondary engineering parameters corresponding to all the key primary engineering parameters of the well group, at least one key secondary engineering parameter having the greatest correlation with the productivity data of the well group is screened out.

5. The method according to claim 1, wherein Generating a trained coalbed methane production capacity prediction model based on the key geological parameters, parameter values ​​of the key multi-level engineering parameters, and production capacity data of each single well in the well group includes: For any single well in the well group, the parameter value of each key geological parameter, each key primary engineering parameter, and each key multi-level engineering parameter of the single well are taken as a training sample, and the productivity data of the single well is taken as the actual productivity corresponding to the training sample; Inputting each of the training samples and the corresponding actual production capacity into the coalbed methane production capacity prediction model to be trained to perform production capacity prediction, thereby obtaining the predicted production capacity corresponding to each of the training samples; Based on the difference between the predicted production capacity and the actual production capacity corresponding to each training sample, the parameters in the coalbed methane production capacity prediction model to be trained are updated to obtain a trained coalbed methane production capacity prediction model.

6. The method according to claim 1, wherein The fracturing construction parameter values ​​of the single well include the parameter values ​​of each of the key geological parameters, each of the key primary engineering parameters and each of the key multi-level engineering parameters of the single well; Optimizing the fracturing operation parameter values ​​of the single well based on the trained coalbed methane production capacity prediction model includes: Determining the trained target coalbed methane production capacity prediction model corresponding to each of the single wells; For any of the single wells, multiple groups of fracturing construction parameter values ​​for the single well are determined based on the optimization algorithm, and the target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of fracturing construction parameter values, respectively, with the goal of maximizing the production capacity prediction value. Based on the optimization algorithm and the production capacity prediction value of each group of fracturing construction parameter values, each group of pressure construction parameters is optimized to obtain multiple groups of optimized fracturing construction parameter values. The target coalbed methane production capacity prediction model corresponding to the single well is used to predict the production capacity prediction value corresponding to each group of optimized fracturing construction parameter values, respectively, until the set iterative optimization end conditions are met, and the fracturing construction parameter value corresponding to the current maximum production capacity prediction value is determined as the optimal fracturing construction parameter value for the single well.

7. The method according to claim 1, characterized in that Before determining a plurality of key primary engineering parameters based on the parameter values ​​of the primary engineering parameters and the production capacity data of each of the single wells in the well group, the method further includes: Determining whether the number of individual wells in the well group is less than a set threshold; If the number of single wells in the well group is less than the set threshold, performing sample enhancement on the single wells in the well group to increase the number of single wells in the well group to no less than the set threshold, and then performing the step of determining a plurality of key primary engineering parameters based on the parameter value of the primary engineering parameter and the production capacity data of each of the single wells in the well group; If the number of single wells in the well group is not less than the set threshold, the step of determining multiple key primary engineering parameters based on the parameter values ​​and production capacity data of the primary engineering parameters of each single well in the well group is directly executed.

8. The method according to claim 1, characterized in that The plurality of geological parameters include at least two of coal seam depth, coal seam thickness, coal seam gas content, coal seam permeability, coal seam porosity, coal rock mechanical properties and ground stress; The production capacity data refers to the unimpeded flow rate, average daily gas production or cumulative gas production during the target production period.

9. The method according to claim 3, characterized in that The first-level engineering parameters are construction displacement, sand addition volume, net liquid volume, comprehensive sand-liquid ratio, single-stage length of fracturing, number of clusters, and number of perforations; the second-level engineering parameters are pre-fluid volume ratio, sand-carrying liquid volume ratio, displacement liquid volume ratio, first-mesh quartz sand volume, second-mesh quartz sand volume, first-viscosity slickwater volume, and second-viscosity slickwater volume; Among them, the proportion of pre-liquid volume, the proportion of sand-carrying liquid volume, the proportion of displacement liquid volume, the first viscosity slick water dosage, and the second viscosity slick water dosage are the secondary engineering parameters corresponding to the net liquid volume; the first mesh size quartz sand dosage and the second mesh size quartz sand dosage are the secondary engineering parameters corresponding to the sand addition volume.

10. A coalbed methane fracturing construction parameter optimization device, characterized in that: include A collection unit, configured to collect geological data, engineering data, and production capacity data of a plurality of single wells in a coalbed methane development area; wherein the geological data of each single well includes parameter values ​​of a plurality of geological parameters, and the engineering data of each single well includes parameter values ​​of a plurality of primary engineering parameters; a screening unit, configured to screen out a plurality of key geological parameters having the greatest correlation with the productivity data from the plurality of geological parameters based on the parameter value of each geological parameter of each single well and the productivity data of each single well; a clustering unit, configured to cluster the plurality of single wells based on the plurality of key geological parameters to obtain a plurality of well groups; wherein each of the well groups includes at least one of the single wells; A first determining unit is configured to determine, for any of the well groups, a plurality of key primary engineering parameters based on the parameter value of the primary engineering parameter and the production capacity data of each of the single wells in the well group; a second determining unit configured to, if at least one of the key first-level engineering parameters has a corresponding second-level engineering parameter, continue to determine at least one key second-level engineering parameter until no corresponding sub-level engineering parameter exists for each key multi-level engineering parameter; a generating unit, configured to generate a trained coalbed methane productivity prediction model based on the key geological parameters, parameter values ​​of the key multi-level engineering parameters, and productivity data of each single well in the well group; The optimization unit is used to optimize the fracturing construction parameter values ​​of the single well based on the trained coalbed methane production capacity prediction model.

Images