Fracturing parameter design method, device and equipment based on data driving

Abstract

The invention discloses a fracturing parameter design method, device and equipment based on data driving, and the method comprises the steps: building a sample database of the same oil reservoir type based on the oil and gas reservoir type of a block where a target well is located; a data analysis method is adopted to analyze the sample database, and yield influence factors of all the fracturing sections are screened out; drawing a box tassel map of yield influence factors in each fracturing section in the sample database, and drawing a box tassel map of at least one well logging parameter, a well logging parameter and a well drilling parameter included in different fracturing sections of the target well; similarity comparison is carried out based on the box whisker map, so that horizontal well fracturing sections corresponding to different fracturing sections of the target well are determined in a sample database; and on the basis of the fracturing parameters of the horizontal well fracturing sections corresponding to the different fracturing sections of the target well, the fracturing parameters of the different fracturing sections of the target well are designed. The method can realize design automation, reduces design workload, is suitable for staged fracturing parameter optimization design of an inclined shaft or a horizontal well of an oil and gas well, and improves working efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas well reservoir transformation in oil and gas development, and in particular to a data-driven fracturing parameter design method, device and equipment. Background Art

[0002] Digitalization and intelligentization are key development directions for the oil and gas industry. Data-driven fracturing parameter design is a key component of intelligent fracturing design. Currently, two main approaches are used for horizontal well fracturing parameter design. One approach combines reservoir numerical simulation with fracturing fracture simulation. This approach optimizes hydraulic fracture geometry and conductivity through reservoir numerical simulation. Reservoir numerical simulation utilizes software such as Eclipse and CMG. Different hydraulic fracture geometries and conductivity settings are used to optimize hydraulic fracture length, height, and conductivity, targeting cumulative production, recovery factor, or economic benefits. Through fracturing simulation optimization, fracturing construction parameters are obtained, including cluster spacing, sand addition intensity, fluid intensity, construction displacement, proppant particle size combination and ratio, etc. Fracturing fracture simulation uses common fracturing simulation software such as FracproPT, Gohfer, and Stimplan. Different construction parameters are set. The hydraulic fracture geometry and fracture conductivity obtained from reservoir numerical simulation are used as the target to optimize the key fracturing construction parameters such as sand addition intensity, fluid intensity, construction displacement, proppant particle size combination and ratio. The second is the geological engineering integrated software simulation method. This method is based on reservoir characteristics, constructs a geomechanical model, and uses Mangrove, CYPHER, and geological engineering integrated simulation software to predict production under different fracture numbers and fracture characteristics. According to the set fracturing design goals, the number of fractures, hydraulic fracture geometry and conductivity are optimized. The geological engineering integrated simulation software is used to optimize parameters such as sand addition intensity, fluid intensity, construction displacement, cluster spacing, proppant particle size combination and ratio. Summary of the Invention

[0003] In order to automate parameter design, reduce the design workload, and thereby improve the efficiency of the optimization design of staged fracturing parameters in oil and gas wells, while enriching the process routes and increasing the selection space, the embodiments of the present invention provide a data-driven fracturing parameter design method, device and equipment.

[0004] In a first aspect, an embodiment of the present invention provides a data-driven fracturing parameter design method, which may include:

[0005] Based on the reservoir type of the target well, the well logging parameters, mud logging parameters, drilling parameters, and fracturing parameters and production data of horizontal wells of the same reservoir type are obtained to build a sample database of the same reservoir type.

[0006] Analyzing the sample database using a data analysis method to screen out factors influencing the production of each fracturing stage from the well logging parameters, the mud logging parameters, and the drilling parameters;

[0007] Drawing a box-and-whisker plot of factors affecting production in each fracturing stage in the sample database, and drawing a box-and-whisker plot of at least one well logging parameter, mud logging parameter, and drilling parameter included in different fracturing stages of the target well;

[0008] Performing a similarity comparison based on the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database, so as to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database;

[0009] Based on the fracturing parameters of the horizontal well fracturing sections corresponding to the different fracturing sections of the target well, the fracturing parameters of the different fracturing sections of the target well are designed.

[0010] Optionally, performing similarity comparison based on the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database may include:

[0011] Calculating similarity coefficients of the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database to obtain a similarity coefficient ranking result;

[0012] The horizontal well fracturing section corresponding to the highest similarity coefficient value in the similarity coefficient sorting results is used as the horizontal well fracturing section with optimized fracturing parameters for matching the fracturing section in the target well.

[0013] Optionally, similarity coefficients of the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database are calculated based on a similarity metric analogy method to obtain a similarity coefficient ranking result.

[0014] Optionally, the box-and-whisker plot may include at least the following parameters: a lower limit value, an upper limit value, a lower quartile, an upper quartile, and a median.

[0015] Optionally, after designing the fracturing parameters of different fracturing stages of the target well, the method may further include: estimating the production ranges of different fracturing stages of the target well.

[0016] Optionally, the logging parameters may include at least one of the following parameters: natural gamma, natural potential, well diameter, deep resistivity, shallow resistivity, longitudinal wave time difference, acoustic wave time difference, neutron porosity and density; the logging parameters may include at least one of the following parameters: total hydrocarbons, components C1-C10 and element logging values; the drilling parameters may include at least one of the following parameters: drilling time, bit pressure and mud density; the fracturing parameters may include at least one of the following parameters: displacement, sand addition intensity, fluid addition intensity, sand concentration, liquid combination and proppant combination; the production data may include at least one of the test production or the cumulative production within the same time.

[0017] Optionally, the data analysis method may include at least one of the following methods: random forest method, distributed gradient boosting library, Pearson correlation coefficient analysis method, maximum information coefficient method, distance correlation coefficient method, grey relational analysis method and multiple regression method.

[0018] In a second aspect, an embodiment of the present invention provides a data-driven fracturing parameter design device, which may include:

[0019] A construction module is used to obtain the logging parameters, mud logging parameters, drilling parameters, and fracturing parameters and production data of horizontal wells of the same oil and gas reservoir type, based on the oil and gas reservoir type of the block where the target well is located, so as to build a sample database of the same oil and gas reservoir type;

[0020] a screening module, configured to analyze the sample database using a data analysis method, so as to screen out factors influencing the production of each fracturing stage from the well logging parameters, the mud logging parameters, and the drilling parameters;

[0021] a drawing module, configured to draw a box-and-whisker plot of factors influencing production in each fracturing stage in the sample database, and a box-and-whisker plot of at least one logging parameter, mud logging parameter, and drilling parameter included in different fracturing stages of the target well;

[0022] a determination module, configured to perform similarity comparison based on the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database, so as to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database;

[0023] The design module is used to design the fracturing parameters of the different fracturing sections of the target well based on the fracturing parameters of the horizontal well fracturing sections corresponding to the different fracturing sections of the target well.

[0024] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the data-driven fracturing parameter design method as described in the first aspect.

[0025] In a fourth aspect, an embodiment of the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the data-driven fracturing parameter design method as described in the first aspect is implemented.

[0026] The beneficial effects of the above technical solutions provided by the embodiments of the present invention include at least:

[0027] An embodiment of the present invention provides a data-driven fracturing parameter design method, device and equipment. The method constructs a sample database based on the logging parameters, mud logging parameters, drilling parameter data, fracturing operation parameters and production data of horizontal wells, and then automatically designs the fracturing operation parameters and predicts the production capacity of new wells based on the similarity criterion based on the data-driven design method of similarity analogy. The method can realize the automation of design, reduce the design workload, and is suitable for the optimization design of staged fracturing parameters of inclined wells or horizontal wells in oil and gas wells, thereby improving work efficiency.

[0028] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0029] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0031] Figure 1 A flow chart of a data-driven fracturing parameter design method provided in an embodiment of the present invention;

[0032] Figure 2 This is a box-and-whisker plot of the logging parameters of the 15th section of a sample well provided in an embodiment of the present invention;

[0033] Figure 3 This is a box-and-whisker plot of the 15th section of mud logging parameters of a sample well provided in an embodiment of the present invention;

[0034] Figure 4 This is a box-and-whisker plot of drilling parameters for the 15th section of a sample well provided in an embodiment of the present invention;

[0035] Figure 5 Schematic diagram of the structure of a data-driven fracturing parameter design device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0037] The inventors studied the existing fracturing parameter design methods in the actual work process. For example, the patent with publication number CN114547953A discloses a fracturing construction parameter optimization method and system based on the optimization design plate. The method includes: classifying the target reservoirs in different well areas of the target block and dividing different types of reservoirs; generating different numerical simulation schemes for different types of reservoirs and establishing different fracturing production integrated numerical simulation methods; using the fracturing production integrated numerical simulation method to simulate and calculate the numerical simulation schemes to obtain production dynamic data of different numerical simulation schemes; calculating the corresponding economic net present value based on the production dynamic data of different numerical simulation schemes; drawing a fracturing construction parameter optimization design plate according to the economic net present value corresponding to different numerical simulation schemes; optimizing the fracturing construction parameters of the target reservoir according to the fracturing construction parameter optimization design plate, and determining the optimal fracturing construction parameters. The method of the present invention can optimize the fracturing construction parameters conveniently and quickly with economic benefits as the goal. This patent uses an integrated fracturing and production numerical simulation method to determine the production dynamics of different fracturing schemes. The method then calculates the economic net present value (NPV) based on this production dynamic data, plots the NPV against fracturing operation parameters, and optimizes fracturing parameters based on economic benefits. The inventors discovered that this method has the following drawbacks: ① The optimization objective is the NPV, which is a relatively simple objective; ② Simulating different fracturing schemes requires manual configuration, which may result in unreasonable configurations.

[0038] For example, the patent with publication number CN110469303A discloses a method for optimizing the design of volumetric fracturing parameters based on four types of transformation volumes. The method includes the following steps: S1, establishing a geological grid attribute model of a reservoir with natural fractures; S2, modifying the stress field distribution model; S3, establishing a hydraulic fracturing simulation model; S4, calculating the sizes of four different types of transformation volumes of hydraulic fracturing fractures; S5, studying the stress sensitivity of hydraulic fracturing fracture conductivity; S6, coupling multiple fields such as pore pressure field, ground stress field, and fracture conductivity change to predict production capacity; S7, optimizing volumetric fracturing parameters. The beneficial effect of the present invention is that it can compare the four different types of transformation volumes of hydraulic fracturing fractures generated during the hydraulic fracturing process for fractured oil and gas reservoirs, analyze the changes in the pore pressure field, ground stress field, and fracture conductivity stress sensitivity of the fractures generated under different fracturing construction parameters during the production process, predict the production capacity changes of the construction well section, and thus guide the optimization of volumetric fracturing construction parameters. This patent compares the hydraulic fracturing volume of fractured oil and gas reservoirs, analyzes the changes in pore pressure, geostress, fracture conductivity, and stress sensitivity under different fracturing parameters, predicts production capacity changes in the well section under construction, and guides the optimization of volume fracturing construction parameters. The inventors discovered that this method has the following drawbacks: ① The method provided in this patent is only applicable to the optimization of fracturing parameters in fractured oil and gas reservoirs; ② This method requires analysis of the pore pressure field and geostress field, which places high demands on the quality of the geological model and stress model of the oil and gas reservoir.

[0039] For another example, the patent with publication number CN116562428A discloses a method for optimizing fracturing construction parameters based on machine learning. The method includes: obtaining geological factors and engineering factors that affect production; using machine learning methods to sort geological factors and engineering factors, and screening the main controlling factors based on the sorting results; performing dimensionality reduction processing on the screened geological main controlling factors and engineering main controlling factors; using the entropy weight method combined with heuristic automatic search to determine the weights of the geological main controlling factors and engineering main controlling factors after dimensionality reduction; based on the screening results of the main controlling factors, combining machine learning methods and optimization algorithms, establishing an intelligent fracturing process parameter optimization model with the goal of optimizing single well production, and optimizing the optimal construction parameters. The fracturing construction parameter optimization method provided by this invention improves computational efficiency and optimization accuracy compared to the existing technology. The method uses machine learning methods to sort geological and engineering factors that affect production and perform dimensionality reduction processing, uses the entropy weight method combined with heuristic automatic search to determine the weights of the geological and engineering main controlling factors after dimensionality reduction, and combines machine learning to establish an intelligent fracturing parameter optimization model with the goal of optimizing single well production, and optimizes fracturing parameters. The inventors found that this method has the following defects: ① The optimization goal is to achieve the optimal single well production, and the optimization target is relatively single; ② The method provided by this patent reduces the dimensionality of geological and engineering factors. There are differences in the factors affecting the production of different types of reservoirs. The main controlling factors of a reservoir in a certain area after dimensionality reduction may be less adaptable to other areas.

[0040] In view of the above problems, the present invention is proposed to provide a data-driven fracturing parameter design method, device and equipment that overcome the above problems or at least partially solve the above problems.

[0041] The present invention provides a data-driven fracturing parameter design method, referring to Figure 1 As shown, the method may include the following steps:

[0042] Step S11: Based on the oil and gas reservoir type of the block where the target well is located, the well logging parameters, mud logging parameters, drilling parameters, and fracturing parameters and production data of horizontal wells of the same oil and gas reservoir type are obtained to build a sample database of the same reservoir type.

[0043] This step involves building a sample database. In practice, different types of sample databases can be created based on reservoir type. These reservoir types include shale gas, shale oil, tight oil, and tight gas, and the wells deployed in these reservoirs are shale gas horizontal wells, shale oil horizontal wells, tight oil horizontal wells, and tight gas horizontal wells, respectively. Preferably, the sample database is built based on horizontal wells within the target well's block. This allows for more accurate design of fracturing parameters for the target well and higher production estimates due to the stronger correlation within the same block.

[0044] In an embodiment of the present invention, the well logging parameters include at least one of the following parameters: natural gamma, natural potential, well diameter, deep resistivity, shallow resistivity, compressional wave time difference, sonic wave time difference, neutron porosity and density; the logging parameters include at least one of the following parameters: total hydrocarbons, components C1-C10 and element logging values; the drilling parameters include at least one of the following parameters: drilling time, bit pressure and mud density; the fracturing parameters include at least one of the following parameters: displacement, sand addition intensity, fluid addition intensity, sand concentration, liquid combination and proppant combination; the production data include at least one of the test production or the cumulative production within the same period of time.

[0045] It should be noted that, in the embodiments of the present invention, when constructing the sample database, sample horizontal wells are preferably selected from those with the most complete data on well logging parameters, mud logging parameters, and drilling parameters. Sample horizontal well fracturing sections are preferably selected from those with the most complete data on fracturing parameters and production. This allows for more accurate weighting of factors influencing production during parameter screening in subsequent steps. If necessary, this step requires data preprocessing to fill in parameter data or to delete horizontal wells or fracturing sections with significant missing parameter data, thereby ensuring a more complete sample database.

[0046] Step S12: Analyze the sample database using a data analysis method to screen out factors affecting the production of each fracturing stage from well logging parameters, mud logging parameters, and drilling parameters.

[0047] It should be noted that the data analysis method may include at least one of the following methods: random forest method, distributed gradient boosting library (XGBoost), Pearson correlation coefficient analysis method, maximum information coefficient method, distance correlation coefficient method, grey relational analysis method and multiple regression method.

[0048] This step is to screen the sample data in the above-mentioned sample database, analyze the sample data using data analysis methods, and then select the production influencing factors from the above-mentioned parameters, that is, the factors with the largest weight influencing the production. After screening the parameter data using the data analysis method, the parameters with the largest weight are selected as the production influencing factors. For example, the seven parameters of wellbore diameter, natural gamma, natural potential, longitudinal wave time difference, shear wave time difference, deep resistivity and neutron porosity in the well logging parameters in step S11 are used as the production influencing factors in the well logging parameters; the four parameters of total hydrocarbon, C1, Si (silicon), and Ca (calcium) in the logging parameters are used as the production influencing factors in the logging parameters; and the two parameters of mud density and drilling time in the drilling parameters are used as the production influencing factors in the drilling parameters.

[0049] Step S13: draw a box-and-whisker plot of the production influencing factors in each fracturing stage in the sample database, and draw a box-and-whisker plot of at least one well logging parameter, mud logging parameter, and drilling parameter included in different fracturing stages of the target well.

[0050] In this step, when drawing the box-and-whisker plot, each production-influencing factor in each fracturing stage requires its own box-and-whisker plot. This step is automatically drawn by a computer program. Due to the large size of the sample database, a large number of box-and-whisker plots can be drawn, which is a significant difference from the existing technology. Similarly, box-and-whisker plots are drawn for different fracturing stages of the target well using the same production-influencing factor parameters. Similarity comparisons are performed on these numerous box-and-whisker plots, and the fracturing parameters for the target fracturing stage of the target well are then automatically matched based on the similarity criterion.

[0051] It should be noted that the box-and-whisker plot in the embodiment of the present invention includes at least the following parameters: a lower limit, an upper limit, a lower quartile, an upper quartile, and a median.

[0052] Step S14: performing similarity comparison based on the box-whisker plots corresponding to different fracturing stages of the target well and the box-whisker plots corresponding to each fracturing stage in the sample database, so as to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database.

[0053] Step S15: designing the fracturing parameters of the different fracturing sections of the target well based on the fracturing parameters of the horizontal well fracturing sections corresponding to the different fracturing sections of the target well.

[0054] The above-mentioned data-driven fracturing parameter design method provided in the embodiment of the present invention constructs a sample database based on the logging parameters, mud logging parameters, drilling parameter data, fracturing operation parameters and production data of the horizontal well, and then automatically designs the fracturing operation parameters and predicts the production capacity of the new well according to the similarity criterion based on the data-driven design method of similarity analogy; this method can realize the automation of design, reduce the design workload, and is suitable for the optimization design of segmented fracturing parameters of inclined wells or horizontal wells of oil and gas wells, thereby improving work efficiency.

[0055] In an optional embodiment, the above-mentioned step S14 may specifically include: first, calculating the similarity coefficients of the box-whisker plots corresponding to different fracturing sections of the target well and the box-whisker plots corresponding to each fracturing section in the sample database to obtain a similarity coefficient ranking result; then, the horizontal well fracturing section corresponding to the highest similarity coefficient value in the similarity coefficient ranking result is used as the horizontal well fracturing section with optimized fracturing parameters for matching the fracturing sections in the target well.

[0056] In a more specific embodiment, in the above step S14, similarity coefficients of the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database are calculated based on a similarity metric analogy method to obtain a similarity coefficient ranking result.

[0057] In the specific implementation of the embodiment of the present invention, the similarity coefficient adopts a high-dimensional clustering similarity measurement function, and the expression of the similarity measurement function sim(X, Y) is:

[0058]

[0059] where X=(x1,…,x d ) and Y=(y1,…,y d ) are two vectors in d-dimensional space, It represents the interval length of the i-th dimension, that is, the maximum distance of the data in the i-th dimension. The purpose is to make the similarity measure depend not only on X and Y, but also on the entire data. ε is a very small constant that can be taken as 0.001 to ensure that the algorithm is not divided by 0.

[0060] ① The similarity between two objects is obtained by calculating the similarity of each dimension of data and then taking the average value. The value is between 0 and 1. The larger the value, the more similar the two objects are.

[0061] ②The minimum value of the function is 0, which means that the difference between objects X and Y is the largest in each dimension, and the similarity between X and Y is the smallest.

[0062] ③ The maximum value of the function is 1, which means that the values ​​of objects X and Y are equal in all dimensions, that is, X and Y overlap in the d-dimensional space. At this time, the similarity between X and Y is the highest.

[0063] According to the characteristics of this function, as long as the values ​​of X and Y are close in certain dimensions, they will show a certain degree of similarity. The degree of similarity increases as the number of dimensions with similar values ​​increases. According to the above formula, the different fracturing stages of the target well (target stage) and the horizontal well fracturing stages in the sample database (sample stage) are sorted by similarity coefficient. The fracturing parameters of the sample stage with the highest similarity coefficient are selected as the optimized fracturing parameters matching the different fracturing stages of the target well.

[0064] In another optional embodiment, also referring to Figure 1 As shown, after designing the fracturing parameters of different fracturing stages of the target well in step S15, the following steps may also be performed:

[0065] Step S16: estimating the production ranges of different fracturing stages of the target well.

[0066] It should be noted that in the sample database described above, not all fracturing stages have independent production data. This would result in an incomplete box-and-whisker plot of production. When plotting the box-and-whisker plot of production, the box-and-whisker plot of the fracturing stages with production data is first drawn. The production range of the target stage (target stage) in the target well is estimated based on the fracturing stage with the highest similarity value among the fracturing stages with production data. For example, the production data of the top 10% of similarity coefficient values ​​is used as the data basis to estimate the production range of the target stage.

[0067] In a specific example, a shale gas block is used as an example for explanation, wherein the target well is the shale gas horizontal well WY-3H. After the WY-3H horizontal well is drilled and completed, the method provided by the present invention is used to design fracturing parameters and estimate production. The design steps are as follows:

[0068] Step (1) constructing a sample database using the well logging parameters, mud logging parameters, drilling parameters, fracturing parameters and production parameters of 315 fracturing stages of 15 wells in the WY well area (block) where WY-3H is located;

[0069] Step (2): Based on the sample database constructed above, the production influencing factors are optimized in 315 fracturing stages; the optimized logging parameters include wellbore diameter, natural gamma, natural potential, P-wave time difference, S-wave time difference, deep resistivity and neutron porosity; the logging parameters include total hydrocarbon, C1 and Si (silicon), Ca (calcium) elements; the drilling parameters include mud density and drilling time;

[0070] Referring to Table 1, the optimization results of logging parameters for 315 fracturing stages in 15 horizontal wells in the WY well area are shown. Various data analysis methods are used to analyze the influencing factors of logging parameters, including seven parameters: wellbore diameter, natural gamma, natural potential, compressional wave time difference, shear wave time difference, deep resistivity, and neutron porosity. Box-whisker plots of these seven logging parameters are drawn.

[0071] Table 1 Factors affecting yield and their weights

[0072]

[0073] Step (3) draws a box-and-whisker diagram of the factors affecting yield in the sample database, for example, Figure 2 As shown, the box and whisker diagram of the logging parameters of the 15th section of a sample well in the WY well area; Figure 3 As shown, the box and whisker plot of the logging parameters of the 15th section of the sample well; Figure 4 As shown in the figure, the box-and-whisker plot of the drilling parameters of the 15th section of the sample well. In this example, the target well is divided into 22 fracturing sections. Referring to the above 13 parameters, the box-and-whisker plot of the well logging parameters, mud logging parameters, and drilling parameters of the 22 sections in the WY-3H well, such as the 15th section of the sample well, is drawn;

[0074] Step (4) uses a similarity metric analogy method to calculate the similarity coefficient between the 13 parameters in the box-whisker plot of each of the 315 well sections in the sample database and the 13 parameters in the box-whisker plot of each of the 22 well sections in the WY-3H well; preferably, the similarity coefficient is greater than or equal to 0.9.

[0075] Step (5), sort by similarity coefficient, and select the fracturing parameters of the sample well section with the highest similarity coefficient value as the optimized fracturing parameters of each section of the WY-3H well, which are:

[0076] The similarity coefficient between the first section of WY-3H and the fifth section of WY20A-X1H well is the highest, 0.9213. The optimized parameters adopt the fracturing parameters of the fifth section of WY20A-X1H well, namely: displacement 13.5m3 / min, sand addition intensity 1.80t / m, fluid intensity 24.0m3 / (m·section), and sand concentration 185kg / m3.

[0077] The similarity coefficient between the second section of WY-3H and the 17th section of WY20A-X7H well is the highest, 0.9311. The optimized parameters adopt the fracturing parameters of the 17th section of WY20A-X7H well, namely: displacement 13.8m3 / min, sand addition intensity 2.32t / m, fluid intensity 25.5m3 / (m·section), and sand concentration 193kg / m3.

[0078] The similarity coefficient between the 3rd section of WY-3H and the 23rd section of WY20A-X3H well is the highest at 0.8994. The optimized parameters adopt the fracturing parameters of the 23rd section of WY20A-X3H well, namely: displacement 13.8m3 / min, sand addition intensity 2.36t / m, fluid intensity 26.2m3 / (m·section), and sand concentration 195kg / m3.

[0079] The similarity coefficient between the 4th section of WY-3H and the 22nd section of WY20A-X3H well is the highest at 0.9087. The optimized parameters adopt the fracturing parameters of the 22nd section of WY20A-X3H well, namely: displacement 13.8m3 / min, sand addition intensity 2.45t / m, fluid intensity 26.9m3 / (m·section), and sand concentration 205kg / m3.

[0080] The similarity coefficient between the 5th stage of WY-3H and the 12th stage of WY20A-X3H well is the highest at 0.9344. The optimized parameters adopt the fracturing parameters of the 12th stage of WY20A-X3H well, namely: displacement 13.5m3 / min, sand addition intensity 2.55t / m, fluid intensity 27.3m3 / (m·stage), and sand concentration 220kg / m3.

[0081] The similarity coefficient between the 6th section of WY-3H and the 4th section of WY20B-X2H well is the highest at 0.8749. The optimized parameters adopt the fracturing parameters of the 4th section of WY20B-X2H well, namely: displacement 12.5m3 / min, sand addition intensity 1.88t / m, fluid intensity 25.0m3 / (m·section), and sand concentration 187kg / m3.

[0082] The similarity coefficient between the 7th section of WY-3H and the 5th section of WY20B-X2H well is the highest at 0.9222. The optimized parameters adopt the fracturing parameters of the 5th section of WY20B-X2H well, namely: displacement 12.5m3 / min, sand addition intensity 1.80t / m, fluid intensity 26.0m3 / (m·section), and sand concentration 180kg / m3.

[0083] The similarity coefficient between the 8th section of WY-3H and the 7th section of WY20A-X7H well is the highest at 0.8745. The optimized parameters adopt the fracturing parameters of the 7th section of WY20A-X7H well, namely: displacement 14.2m3 / min, sand addition intensity 2.55t / m, fluid intensity 25.5m3 / (m·section), and sand concentration 220kg / m3.

[0084] The similarity coefficient between the 9th section of WY-3H and the 8th section of WY20A-X7H well is the highest at 0.9007. The optimized parameters adopt the fracturing parameters of the 8th section of WY20A-X7H well, namely: displacement 14.2m3 / min, sand addition intensity 2.61t / m, fluid intensity 26.4m3 / (m·section), and sand concentration 225kg / m3.

[0085] The similarity coefficient between the 10th section of WY-3H and the 9th section of WY20A-X7H well is the highest at 0.8396. The optimized parameters adopt the fracturing parameters of the 9th section of WY20A-X7H well, namely: displacement 14.3m3 / min, sand addition intensity 2.50t / m, fluid intensity 26.0m3 / (m·section), and sand concentration 218kg / m3.

[0086] The similarity coefficient between the 11th section of WY-3H and the 17th section of WY20A-X6H well is the highest at 0.8726. The optimized parameters adopt the fracturing parameters of the 17th section of WY20A-X6H well, namely: displacement 14.0m3 / min, sand addition intensity 2.73t / m, fluid intensity 26.8m3 / (m·section), and sand concentration 227kg / m3.

[0087] The similarity coefficient between the 12th section of WY-3H and the 18th section of WY20A-X6H well is the highest at 0.9401. The optimized parameters adopt the fracturing parameters of the 18th section of WY20A-X6H well, namely: displacement 14.0m3 / min, sand addition intensity 2.70 / m, fluid intensity 27.5m3 / (m·section), and sand concentration 224kg / m3.

[0088] The similarity coefficient between the 13th section of WY-3H and the 19th section of WY20A-X6H well is the highest at 0.9029. The optimized parameters adopt the fracturing parameters of the 19th section of WY20A-X6H well, namely: displacement 13.8m3 / min, sand addition intensity 2.68t / m, fluid intensity 27.4m3 / (m·section), and sand concentration 220kg / m3.

[0089] The similarity coefficient between the 14th section of WY-3H and the 20th section of WY20A-X6H well is the highest, 0.9114. The optimized parameters adopt the fracturing parameters of the 20th section of WY20A-X6H well, namely: displacement 13.8m3 / min, sand addition intensity 2.68t / m, fluid intensity 27.5m3 / (m·section), and sand concentration 219kg / m3.

[0090] The similarity coefficient between the 15th section of WY-3H and the 3rd section of WY20A-X2H well is the highest at 0.8797. The optimized parameters adopt the fracturing parameters of the 3rd section of WY20A-X2H well, namely: displacement 13.0m3 / min, sand addition intensity 2.16t / m, fluid intensity 23.2m3 / (m·section), and sand concentration 197kg / m3.

[0091] The similarity coefficient between the 16th section of WY-3H and the 5th section of WY20A-X2H well is the highest at 0.8821. The optimized parameters adopt the fracturing parameters of the 5th section of WY20A-X2H well, namely: displacement 13.2m3 / min, sand addition intensity 2.25t / m, fluid intensity 24.9m3 / (m·section), and sand concentration 202kg / m3.

[0092] The similarity coefficient between the 17th section of WY-3H and the 6th section of WY20A-X2H well is the highest, 0.8385. The optimized parameters adopt the fracturing parameters of the 6th section of WY20A-X2H well, namely: displacement 13.2m3 / min, sand addition intensity 2.25t / m, fluid intensity 25.0m3 / (m·section), and sand concentration 200kg / m3.

[0093] The similarity coefficient between the 18th section of WY-3H and the 7th section of WY20B-X4H well is the highest at 0.9229. The optimized parameters adopt the fracturing parameters of the 7th section of WY20B-X4H well, namely: displacement 12.8m3 / min, sand addition intensity 2.18t / m, fluid intensity 25.6m3 / (m·section), and sand concentration 197kg / m3.

[0094] The similarity coefficient between the 19th section of WY-3H and the 8th section of WY20B-X4H well is the highest, 0.9200. The optimized parameters adopt the fracturing parameters of the 8th section of WY20B-X4H well, namely: displacement 13.0m3 / min, sand addition intensity 2.11t / m, fluid intensity 26.0m3 / (m·section), and sand concentration 182kg / m3.

[0095] The similarity coefficient between the 20th section of WY-3H and the 9th section of WY20B-X4H well is the highest at 0.9117. The optimized parameters adopt the fracturing parameters of the 9th section of WY20B-X4H well, namely: displacement 13.1m3 / min, sand addition intensity 2.16t / m, fluid intensity 26.4m3 / (m·section), and sand concentration 183kg / m3.

[0096] The similarity coefficient between the 21st section of WY-3H and the 18th section of WY20A-X2H well is the highest at 0.8664. The optimized parameters adopt the fracturing parameters of the 18th section of WY20A-X2H well, namely: displacement 14.5m3 / min, sand addition intensity 2.55t / m, fluid intensity 26.8m3 / (m·section), and sand concentration 227kg / m3.

[0097] The similarity coefficient between the 22nd stage of WY-3H and the 20th stage of WY20A-X2H well is the highest at 0.8499. The optimized parameters adopt the fracturing parameters of the 20th stage of WY20A-X2H well, namely: displacement 14.3m3 / min, sand addition intensity 2.54t / m, fluid intensity 26.2m3 / (m·stage), and sand concentration 218kg / m3.

[0098] Step (6) estimates the production range of different fracturing stages of the target well.

[0099] The WY-3H well was optimized for fracturing parameters using the method provided by the present invention. The daily gas production after fracturing was 21% higher than that of adjacent wells on the same platform.

[0100] The above-mentioned method provided in an embodiment of the present invention constructs a sample database based on horizontal well logging data, mud logging data, drilling data, fracturing data, and production data; based on the sample database, uses a data analysis method to optimize production influencing factors; draws a box-and-whisker plot of the production influencing factor data in the sample database; draws a box-and-whisker plot of the target well's logging, mud logging, and drilling data; calculates the similarity coefficient between the box-and-whisker plot of one or more of the target well's logging, mud logging, and drilling data and the box-and-whisker plot in the sample database, and obtains a similarity coefficient ranking result; and optimizes the designed fracturing parameters based on the similarity coefficient ranking result. This method can realize design automation, reduce design workload, and is suitable for optimizing the design of staged fracturing parameters for deviated or horizontal wells in oil and gas wells, thereby improving work efficiency.

[0101] Based on the same inventive concept, the present invention also provides a data-driven fracturing parameter design device, referring to Figure 5 As shown, the device may include: a construction module 51, a screening module 52, a drawing module 53, a determination module 54 and a design module 55, and its working principle is as follows:

[0102] The construction module 51 is used to obtain the logging parameters, mud logging parameters, drilling parameters, and fracturing parameters and production data of horizontal wells of the same oil and gas reservoir type, based on the oil and gas reservoir type of the block where the target well is located, to build a sample database of the same oil reservoir type;

[0103] The screening module 52 is used to analyze the sample database using a data analysis method to screen out the production influencing factors of each fracturing stage from the well logging parameters, mud logging parameters, and drilling parameters;

[0104] The drawing module 53 is used to draw a box-and-whisker plot of the factors affecting production in each fracturing stage in the sample database, and draw a box-and-whisker plot of at least one logging parameter, mud logging parameter, and drilling parameter included in different fracturing stages of the target well;

[0105] The determination module 54 is used to perform similarity comparison based on the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database, so as to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database;

[0106] The design module 55 is used to design the fracturing parameters of different fracturing sections of the target well based on the fracturing parameters of the horizontal well fracturing sections corresponding to the different fracturing sections of the target well.

[0107] In another optional embodiment, also referring to Figure 5 As shown, the device may further include: an estimation module 56, which is used to estimate the production ranges of different fracturing stages of the target well.

[0108] In an optional embodiment, the determination module 54 is specifically used to: first, calculate the similarity coefficients of the box-and-whisker plots corresponding to different fracturing sections of the target well and the box-and-whisker plots corresponding to each fracturing section in the sample database to obtain a similarity coefficient ranking result; then, based on the similarity coefficient ranking result, the horizontal well fracturing section corresponding to the highest similarity coefficient value is used as the horizontal well fracturing section with optimized fracturing parameters for matching the fracturing sections in the target well.

[0109] In a more specific embodiment, the determination module 54 is specifically configured to calculate similarity coefficients of the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database based on a similarity metric analogy method to obtain a similarity coefficient ranking result.

[0110] In the specific implementation of the embodiment of the present invention, the similarity coefficient adopts a high-dimensional clustering similarity measurement function, and the expression of the similarity measurement function sim(X, Y) is:

[0111]

[0112] where X=(x1,…,x d ) and Y=(=1,…,y d ) are two vectors in d-dimensional space, It represents the interval length of the i-th dimension, that is, the maximum distance of the data in the i-th dimension. The purpose is to make the similarity measure depend not only on X and Y, but also on the entire data. ε is a very small constant that can be taken as 0.001 to ensure that the algorithm is not divided by 0.

[0113] ① The similarity between two objects is obtained by calculating the similarity of each dimension of data and then taking the average value. The value is between 0 and 1. The larger the value, the more similar the two objects are.

[0114] ②The minimum value of the function is 0, which means that the difference between objects X and Y is the largest in each dimension, and the similarity between X and Y is the smallest.

[0115] ③ The maximum value of the function is 1, which means that the values ​​of objects X and Y are equal in all dimensions, that is, X and Y overlap in the d-dimensional space. At this time, the similarity between X and Y is the highest.

[0116] According to the characteristics of this function, as long as the values ​​of X and Y are close in certain dimensions, they will show a certain degree of similarity. The degree of similarity increases as the number of dimensions with similar values ​​increases. According to the above formula, the different fracturing stages of the target well (target stage) and the horizontal well fracturing stages in the sample database (sample stage) are sorted by similarity coefficient. The fracturing parameters of the sample stage with the highest similarity coefficient are selected as the optimized fracturing parameters matching the different fracturing stages of the target well.

[0117] Based on the same inventive concept, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the above-mentioned data-driven fracturing parameter design method is implemented.

[0118] Based on the same inventive concept, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned data-driven fracturing parameter design method is implemented.

[0119] The principles of the problems solved by the above-mentioned devices, media, related equipment and systems in the embodiments of the present invention are similar to those of the above-mentioned methods. Therefore, their implementation can refer to the implementation of the above-mentioned methods, and the repeated parts will not be repeated.

[0120] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer-usable program code.

[0121] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0122] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0124] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A data-driven fracturing parameter design method, characterized in that: include: Based on the reservoir type of the target well, the well logging parameters, mud logging parameters, drilling parameters, and fracturing parameters and production data of horizontal wells of the same reservoir type are obtained to build a sample database of the same reservoir type. Analyzing the sample database using a data analysis method to screen out factors influencing the production of each fracturing stage from the well logging parameters, the mud logging parameters, and the drilling parameters; Drawing a box-and-whisker plot of factors affecting production in each fracturing stage in the sample database, and drawing a box-and-whisker plot of at least one well logging parameter, mud logging parameter, and drilling parameter included in different fracturing stages of the target well; Performing a similarity comparison based on the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database, so as to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database; Based on the fracturing parameters of the horizontal well fracturing sections corresponding to the different fracturing sections of the target well, the fracturing parameters of the different fracturing sections of the target well are designed.

2. The method according to claim 1, characterized in that Performing a similarity comparison based on the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database, including: Calculating similarity coefficients of the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database to obtain a similarity coefficient ranking result; The horizontal well fracturing section corresponding to the highest similarity coefficient value in the similarity coefficient sorting results is used as the horizontal well fracturing section with optimized fracturing parameters for matching the fracturing section in the target well.

3. The method according to claim 2, characterized in that Based on the similarity metric analogy method, similarity coefficients of the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database are calculated to obtain a similarity coefficient ranking result.

4. The method according to any one of claims 1 to 3, characterized in that The box-and-whisker plot includes at least the following parameters: a lower limit value, an upper limit value, a lower quartile, an upper quartile and a median.

5. The method according to any one of claims 1 to 3, characterized in that After designing the fracturing parameters of different fracturing sections of the target well, the method further includes: estimating the production ranges of different fracturing sections of the target well.

6. The method according to any one of claims 1 to 3, characterized in that The logging parameters include at least one of the following parameters: natural gamma, natural potential, well diameter, deep resistivity, shallow resistivity, compressional wave time difference, acoustic wave time difference, neutron porosity and density; the logging parameters include at least one of the following parameters: total hydrocarbon, component C1-C 10 and element logging values; the drilling parameters include at least one of the following parameters: drilling time, bit pressure and mud density; the fracturing parameters include at least one of the following parameters: displacement, sand addition intensity, fluid addition intensity, sand concentration, liquid combination and proppant combination; the production data includes at least one of the test production or the cumulative production in the same period.

7. The method according to any one of claims 1 to 3, characterized in that The data analysis method includes at least one of the following methods: random forest method, distributed gradient boosting library, Pearson correlation coefficient analysis method, maximum information coefficient method, distance correlation coefficient method, grey relational analysis method and multiple regression method.

8. A data-driven fracturing parameter design device, characterized in that: include: A construction module is used to obtain the logging parameters, mud logging parameters, drilling parameters, and fracturing parameters and production data of horizontal wells of the same oil and gas reservoir type, based on the oil and gas reservoir type of the block where the target well is located, so as to build a sample database of the same oil and gas reservoir type; a screening module, configured to analyze the sample database using a data analysis method, so as to screen out factors influencing the production of each fracturing stage from the well logging parameters, the mud logging parameters, and the drilling parameters; a drawing module, configured to draw a box-and-whisker plot of factors influencing production in each fracturing stage in the sample database, and a box-and-whisker plot of at least one logging parameter, mud logging parameter, and drilling parameter included in different fracturing stages of the target well; a determination module, configured to perform similarity comparison based on the box-and-whisker plots corresponding to different fracturing stages of the target well and the box-and-whisker plots corresponding to each fracturing stage in the sample database, so as to respectively determine the horizontal well fracturing stages corresponding to different fracturing stages of the target well in the sample database; The design module is used to design the fracturing parameters of the different fracturing sections of the target well based on the fracturing parameters of the horizontal well fracturing sections corresponding to the different fracturing sections of the target well.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the data-driven fracturing parameter design method according to any one of claims 1 to 7 is implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the data-driven fracturing parameter design method according to any one of claims 1 to 7 is implemented.

Images