Unconventional reservoir fracturing drainage and mining collaboration method and unconventional reservoir fracturing drainage and mining collaboration device

Abstract

The invention discloses an unconventional reservoir fracturing drainage and mining cooperation method and device. The method comprises the steps that physical property data, fracturing and crack parameters and soaking drainage and mining parameters of an unconventional reservoir are determined according to fracturing data, well bottom pressure data during soaking, fracturing fluid flow-back data and well bottom pressure data during drainage and mining; according to the unconventional reservoir physical properties, the fracturing fracture parameters and the soaking drainage parameters, a horizontal well fracturing drainage database is established; and according to the data in the horizontal well fracturing drainage and mining database, fracturing drainage and mining design parameters of the unconventional reservoir are determined by using an XGBoost intelligent optimization algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unconventional reservoir oil and gas development, and particularly relates to a method and device for unconventional reservoir fracturing and production. BACKGROUND

[0002] Unconventional reservoirs refer to oil and gas reservoirs with characteristics such as low permeability, low porosity, high pressure, high temperature, etc., such as tight sandstone, shale gas, coalbed methane, etc. The development of unconventional oil and gas reservoirs is difficult, and large-scale horizontal well staged multi-cluster fracturing under the well factory mode is a key technical means to realize commercial development. It has been proved that horizontal well staged multi-cluster fracturing can easily form multiple artificial fractures in the horizontal well, increase the stimulated volume and oil and gas seepage channels, maintain high productivity and long high-yield stable production period after the treatment, and significantly improve the fracturing stimulation effect of unconventional oil and gas reservoirs.

[0003] The factors affecting the fracturing and production effect of unconventional oil and gas reservoirs can be divided into: (1) fluid properties and formation parameters: characteristic parameters related to the actual reservoir properties; (2) field fracturing-fracture and soak-well production parameters: sensitive to different unconventional reservoirs. The geology, rock mechanics and ground stress of unconventional oil and gas reservoirs have complexity and heterogeneity, and the parameters such as formation permeability, original formation pressure, fracture half-length, secondary fracture density, formation pressure in the fracturing region, initial fracture pore volume, fracture permeability, well storage, horizontal permeability, initial pressure, multi-fracture stress interference, fracture width, etc. cannot be accurately determined, which has great difficulty and many challenges in efficient simulation, resulting in that the traditional unconventional reservoir fracturing and production method cannot effectively match the reservoir characteristics and development targets, the fracturing effect is not ideal, the production cost is high, and the resource utilization rate is low. SUMMARY

[0004] The present application provides a method and device for unconventional reservoir fracturing and production, which is based on the construction of a horizontal well fracturing and production database covering the physical properties of unconventional reservoirs, fracturing and fracture parameters, soak and production parameters, and uses XGBoost intelligent algorithm as the optimization means for the collaborative design of fracturing and production parameters, to maximize the development benefit of unconventional oil and gas reservoirs.

[0005] In a first aspect, the present application provides a method for unconventional reservoir fracturing and production, which comprises:

[0006] determining the physical property data of the unconventional reservoir, the fracturing and fracture parameters, and the soak and production parameters according to the fracturing data, the wellbore pressure data during soak, the fracturing fluid flowback data, and the wellbore pressure data during production, respectively;

[0007] establishing a horizontal well fracturing and production database according to the physical property data of the unconventional reservoir, the fracturing and fracture parameters, and the soak and production parameters;

[0008] According to the data in the horizontal well fracturing drainage database, the fracturing drainage design parameters of the unconventional reservoir are determined by using an XGBoost intelligent optimization algorithm.

[0009] In a second aspect, the embodiments of the present application also provide an unconventional reservoir fracturing drainage coordination device, which comprises a memory and a processor; the memory is used to save a program for performing unconventional reservoir fracturing drainage coordination, and the processor is used to read and execute the program for performing unconventional reservoir fracturing drainage coordination, and execute the method in any one of the above embodiments.

[0010] In a third aspect, the embodiments of the present application also provide a computer readable storage medium, which stores a data processing program, and the data processing program is executed by a processor to perform the unconventional reservoir fracturing drainage coordination method in any one of the above embodiments.

[0011] Compared with the related art, the present application provides an unconventional reservoir fracturing drainage coordination method, which comprises the following steps: determining the physical property data of the unconventional reservoir, the fracturing and fracture parameters, and the soak and drainage parameters according to fracturing data, well bottom pressure data during soak, fracturing fluid flowback data, and well bottom pressure data during drainage; establishing a horizontal well fracturing drainage database according to the physical property data of the unconventional reservoir, the fracturing and fracture parameters, and the soak and drainage parameters; and determining the fracturing drainage design parameters of the unconventional reservoir by using an XGBoost intelligent optimization algorithm according to the data in the horizontal well fracturing drainage database. The present application is based on the horizontal well fracturing drainage database covering the physical property of the unconventional reservoir, the fracturing and fracture parameters, and the soak and drainage parameters, and uses the XGBoost intelligent algorithm as the optimization means for the fracturing drainage parameter coordination design, so as to maximize the development benefit of the unconventional oil and gas reservoir.

[0012] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. Other advantages of the present application can be realized and obtained by means of the solutions described in the specification and the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0013] The accompanying drawings are included to provide an understanding of the present application, and constitute a part of the specification, and together with the embodiments of the present application serve to explain the technical solutions of the present application, and do not constitute a limitation on the technical solutions of the present application.

[0014] Figure 1 The unconventional reservoir fracturing drainage coordination optimization method flowchart of the embodiments of the present application;

[0015] Figure 2Schematic diagram of an unconventional reservoir fracturing and drainage collaborative optimization device according to an embodiment of the present application;

[0016] Figure 3 A schematic diagram illustrating the effect of simulating the stress shadow of a previous crack on the expansion of a next crack in some exemplary embodiments;

[0017] Figure 4 A schematic diagram illustrating the effect of simulated fracturing fluid viscosity on fracturing crack propagation in some exemplary embodiments;

[0018] Figure 5 Schematic diagram of simulating the effect of inter-fracture interference on the propagation of hydraulic fractures in some exemplary embodiments. DETAILED DESCRIPTION

[0019] This application describes multiple embodiments, but this description is exemplary rather than restrictive, and it will be apparent to those skilled in the art that there may be more embodiments and implementations within the scope of the embodiments described herein. Although many possible feature combinations are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment may be used in combination with any other feature or element in any other embodiment, or may replace any other feature or element in any other embodiment.

[0020] This application includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features, and elements disclosed in this application may also be combined with any conventional features or elements to form a unique inventive solution defined by the claims. Any features or elements of any embodiment may also be combined with features or elements from other inventive solutions to form another unique inventive solution defined by the claims. Therefore, it should be understood that any feature shown and / or discussed in this application may be implemented individually or in any appropriate combination. Therefore, except for the limitations made according to the appended claims and their equivalents, the embodiments are not subject to other limitations. In addition, various modifications and changes may be made within the scope of protection of the appended claims.

[0021] In addition, when describing representative embodiments, the specification may have presented the method and / or process as a specific sequence of steps. However, to the extent that the method or process does not rely on the specific order of the steps described herein, the method or process should not be limited to the steps in the specific order described. As will be understood by those skilled in the art, other orders of steps are also possible. Therefore, the specific order of the steps set forth in the specification should not be interpreted as a limitation to the claims. In addition, the claims for the method and / or process should not be limited to performing their steps in the order written, and those skilled in the art can readily understand that these orders can be changed and still remain within the spirit and scope of the embodiments of the present application.

[0022] Traditional unconventional reservoir fracturing and production optimization methods cannot effectively match reservoir characteristics and development goals, resulting in unsatisfactory fracturing results, high production costs, and low resource utilization. To address this issue, many scholars have mainly used the following four processes to conduct research on unconventional reservoir fracturing and production optimization:

[0023] (1) Establish a numerical simulation model for fracturing to simulate the geological characteristics, fluid migration, stress changes and other processes of unconventional oil and gas reservoirs, and analyze the changes in production capacity after fracturing;

[0024] (2) Setting the objective function and constraints, selecting maximizing cumulative production as the optimization objective, and setting fracturing parameters, drainage parameters, etc. as constraints;

[0025] (3) Selecting intelligent algorithms to optimize and search for fracturing parameters and drainage parameters to find the optimal solution or suboptimal solution;

[0026] (4) Develop a fracturing and production plan and implement monitoring and adjustments.

[0027] However, horizontal well fracturing in unconventional reservoirs involves a plethora of uncertain parameters, and conventional fracturing models rarely consider the effects of three-dimensional geological and geomechanical heterogeneity. This makes it difficult to accurately optimize fracturing and production parameters in today's unconventional oil and gas reservoir fracturing environments. Most importantly, existing optimization designs for unconventional reservoir fracturing and production are often based on statistical analysis and dynamic reservoir optimization. These methods primarily utilize mathematical algorithms and simulation software to optimize the effectiveness of unconventional reservoir fracturing development. This approach suffers from limited computational accuracy and applicability, and fails to consider the coordinated integration of fracturing, well shut-in, and production.

[0028] To achieve the above object, the inventors consider the fracturing, soaking and drainage processes in combination, and to solve the problem of low optimization precision of fracturing and drainage parameters caused by too many uncertain parameters in the fracturing process of the horizontal well in the unconventional reservoir, the present application provides a method for the combined optimization of fracturing and drainage in the unconventional reservoir, which aims to match and invert the uncertain parameters based on the field monitoring data, to consider the fracturing, soaking and drainage processes in combination to establish a horizontal well fracturing and drainage database covering the unconventional reservoir properties, fracturing and fracture parameters, and soaking and drainage parameters, to solve the problems of low fitting precision and insufficient covered parameters of the traditional fracturing prediction model by using an improved XGBoost machine learning intelligent algorithm, and finally to realize the rapid and combined optimization of the fracturing and drainage in the unconventional reservoir in the geological engineering integration.

[0029] The present application provides a method for the combined optimization of fracturing and drainage in the unconventional reservoir, as shown in the formula (I): Figure 1 The method comprises steps S100-S120:

[0030] S100: Determine the unconventional reservoir properties, fracturing and fracture parameters, and soaking and drainage parameters according to the fracturing data, wellbore pressure data during soaking, fracturing fluid flowback data and wellbore pressure data during drainage;

[0031] S110: Establish a horizontal well fracturing and drainage database according to the unconventional reservoir properties, fracturing and fracture parameters, and soaking and drainage parameters;

[0032] S120: Determine the fracturing and drainage design parameters of the unconventional reservoir by using the XGBoost intelligent optimization algorithm according to the data in the horizontal well fracturing and drainage database.

[0033] In an exemplary embodiment, the unconventional reservoir properties include formation pressure, reservoir elastic modulus, reservoir Poisson's ratio, formation permeability, oil saturation, formation thickness, formation pressure, etc.

[0034] The fracturing and fracture parameters include fracture conductivity, fracture length, fracture width, fracture inclination, fracture density, fracturing segment number, fracturing fluid injection flow rate, fracturing fluid viscosity, fracturing pump speed, proppant mass fraction, fracturing fluid flow speed, etc.

[0035] The soaking and drainage parameters mainly include soaking time and liquid discharge rate, etc.

[0036] In an exemplary embodiment, the unconventional reservoir properties, fracturing and fracture parameters, and soaking and drainage parameters are determined based on the fracturing process data, wellbore pressure data during soaking, fracturing fluid flowback data and wellbore pressure data during drainage, which comprises:

[0037] Step 1: Invert the formation permeability, original formation pressure and fracture half-length based on the fracturing pump stop data.

[0038] Fracturing pump data is mainly the wellhead pressure, variable density and flow data on the ground;

[0039] Step 2, based on the bottom hole pressure data during the soak well, the average fracture length, secondary fracture density, and the formation pressure of the fracturing region are inverted;

[0040] Step 3, based on the fracturing fluid flowback data, the initial fracture pore volume and the fracture permeability are inverted;

[0041] Step 4, based on the bottom hole pressure data during the production, the well storage, horizontal permeability and initial pressure are inverted;

[0042] Step 5, using three-dimensional displacement discontinuity method (DDM) to calculate multi-fracture stress interference and fracture width.

[0043] In an example embodiment, in step 1, based on the fracturing pump data, the formation permeability, original formation pressure and fracture half-length are inverted, including:

[0044] Step 11, after the fracturing pump is stopped, the current variable density data and flow data are determined;

[0045] Step 12, the variable density and flow surface pressure is converted to bottom hole pressure;

[0046] Step 13, the pressure drop data after the fracturing pump is stopped is filtered to eliminate noise interference;

[0047] Step 14, using well test analysis method to interpret the filtered pressure drop data to obtain the formation permeability, original formation pressure, fracture half-length and the area of the transformed region.

[0048] In an example embodiment, in step 2, based on the bottom hole pressure data during the soak well, the average fracture length, secondary fracture density, and the formation pressure of the fracturing region are inverted, including:

[0049] Step 21, according to the flow characteristics of the three-phase medium of the main fracture-secondary fracture-matrix during the soak well process, and dividing the soak well process into wellbore communication control stage, main fracture storage stage, main fracture and secondary fracture interflow control stage, secondary fracture storage stage and matrix flow control stage;

[0050] Step 22, establishing a linear flow mathematical model after the fracture is closed for the wellbore communication control stage, the main fracture and secondary fracture interflow control stage and the matrix flow control stage;

[0051] Step 23, establishing a fracture storage control mathematical model for the main fracture storage stage and the secondary fracture storage stage;

[0052] Step 24, based on the pressure drop model during the soak period, use the bottom hole pressure data during the soak period to invert the average primary fracture length, secondary fracture density, and formation pressure in the reservoir stimulation area.

[0053] In an example embodiment, in step 3, based on the fracturing fluid flowback data, invert the initial fracture pore volume and fracture permeability, including:

[0054] Step 31, divide the fracturing fluid flowback process into two stages: pseudo-fracture linear flow and pseudo-fracture boundary control flow;

[0055] Step 32, based on the oil-water or gas-water two-phase composite flow between the fracture and the matrix, and the complex occurrence and migration mechanism in unconventional reservoirs, establish a mathematical model of fracturing fluid flowback seepage in unconventional oil and gas reservoirs;

[0056] Step 33, based on the mathematical model of fracturing fluid flowback seepage, solve the characteristic curves of the pseudo-fracture linear flow stage and the pseudo-fracture boundary control flow stage respectively;

[0057] Step 34, based on the slope and intercept of the straight line segment in the characteristic curve, invert the initial fracture pore volume and initial fracture permeability.

[0058] In an example embodiment, in step 4, the inversion of the well storage, horizontal permeability, and initial pressure based on the bottom hole pressure data during the production period includes:

[0059] Step 41, preset the inversion parameters such as well storage, horizontal permeability, initial pressure, and fracture half-length, and generate multiple initial sets composed of inversion parameters by means of a random algorithm;

[0060] Step 42, based on the reservoir model, calculate the state value under the input of each set;

[0061] Step 43, substitute the state value and the observed value into the set Kalman filter method to update the state value;

[0062] Step 44, output the final state value after multiple iterations, and perform post-processing to obtain the well storage, horizontal permeability, and initial pressure.

[0063] In an example embodiment, in step 5, the calculation of multi-fracture stress interference and fracture width using a three-dimensional displacement discontinuity method (DDM) includes:

[0064] Step 51, based on the inter-fracture stress interference and dynamic allocation of inter-cluster flow during the multi-cluster fracturing process, establish a multi-cluster fracturing fracture propagation model;

[0065] Step 52, use the three-dimensional displacement discontinuity method (DDM) to calculate the stress, strain, and displacement of any point in the fracture;

[0066] Step 53, determine the inter-slit interference stress and the crack width according to the stress, strain and displacement;

[0067] Step 54, determine the crack propagation step according to the crack tip energy release rate, the inter-slit interference stress and the crack width.

[0068] In an example embodiment, according to the data in the horizontal well fracturing and production database, the fracturing and production design parameters of the unconventional reservoir are determined by using the XGBoost intelligent optimization algorithm, including:

[0069] First step, define a maximized cumulative production objective function and add a regularization term to control the base learner structure;

[0070] Second step, simplify the maximized cumulative production objective function, and introduce a tree structure to parameterize the simplified objective function to obtain an optimal objective function;

[0071] Third step, construct an optimal tree according to the optimal objective function;

[0072] Fourth step, use the optimal tree to perform an optimization search from the fracturing and production parameter data in the horizontal well fracturing and production database to obtain the optimal fracturing and production design parameters.

[0073] Based on the unconventional reservoir fracturing and production collaborative optimization method, the fracturing effect diagrams of multiple wells under the fracturing-crack, soak-production parameter are simulated for 1 typical block, 4 well groups / platforms and 24 wells, and the single-well precision index coincidence rate of the actual block reaches more than 85% (compared with the simulation precision of the finite element fracturing simulation software Stimplan of the American NSI company). For example, the influences of the previous segment crack stress, fracturing fluid viscosity and inter-slit interference on the fracturing crack propagation are respectively shown in Figures 3-5 Figures 3 to 5 It is shown that by using the unconventional reservoir fracturing and production collaborative optimization method in the embodiment, the influences of the previous segment crack stress, fracturing fluid viscosity and inter-slit interference on the fracturing crack propagation can be clearly realized and intuitively represented by diagrams.

[0074] Compared with the prior art, the beneficial effects of the present application are as follows:

[0075] (1) When the three-dimensional displacement discontinuity algorithm (DDM) is used to calculate the multi-slit stress interference and crack width, the calculation amount is small, the speed is fast, the precision is high, and there is no boundary effect; the unconventional reservoir fracturing and production sample library established by the present application covers the reservoir properties, fracturing-crack parameters and soak-production parameters, and comprehensively considers the complexity and uncertainty of the unconventional reservoir, so that the applicability is stronger;

[0076] ​(2) In the process of establishing the sample library, the present application considers fracturing, soaking and drainage in coordination, realizes geological engineering integration, truly reflects the implementation process in the mine field, and improves the precision and efficiency of the fracturing and drainage coordination optimization;

[0077] (3) The improved XGBoost algorithm provided by the present application integrates multiple basic decision tree models, can effectively extract the relationship between the input (factors or variables) and the output (response or index) of a complex system, and is easy to construct a prediction model of the original complex actual situation with high fitting precision.

[0078] (4) According to different objective functions and constraint conditions, the optimization strategy is flexibly adjusted to meet different development needs; dynamic optimization and real-time feedback are realized, the fracturing and drainage coordination scheme is adjusted in time, and the development effect is improved.

[0079] In a second aspect, the embodiments of the present application also provide an unconventional reservoir fracturing and drainage coordination device, which comprises a memory 200 and a processor 210; the memory is used to save a program for performing unconventional reservoir fracturing and drainage coordination, and the processor is used to read and execute the program for performing unconventional reservoir fracturing and drainage coordination, and execute the method of any one of the above embodiments.

[0080] In a third aspect, the embodiments of the present application also provide a computer readable storage medium, which stores a data processing program, and the data processing program is executed by a processor to perform the unconventional reservoir fracturing and drainage coordination method of any one of the above embodiments.

[0081] Example one

[0082] In an unconventional reservoir fracturing and drainage coordination optimization method, the specific implementation process of obtaining the inter-fracture interference stress and the fracture width based on the three-dimensional displacement discontinuity algorithm (DDM) includes the following steps:

[0083] Step 001, establishing a fracturing fluid dynamic allocation model

[0084] The total fracturing fluid discharge is equal to the sum of the flow rate into each fracture, and the pressure of the fracturing fluid at the wellbore root is equal to the sum of the fracture mouth pressure, the perforation pressure drop and the wellbore friction, so:

[0085]

[0086] p w =p fw,i +p pf,i +p f,i

[0087] Wherein, the perforation hole friction p pf,iThe wellbore friction from the wellbore bottom to the ith cluster of fractures, p f,i are respectively calculated by the following formulas:

[0088]

[0089]

[0090] wherein Q T (t) is the total flow rate of the fracturing fluid at time t (m 3 / s); Q i (t) is the flow rate into the ith half-wing fracture at time t (m 3 / s); N is the number of fracture clusters; p w is the fluid pressure at the wellbore bottom (MPa), p fw,i is the fracture mouth pressure of the ith cluster of fractures (MPa), p pf,i is the perforation hole friction at the ith cluster of fractures (MPa), p f,i is the wellbore friction from the wellbore bottom to the ith cluster of fractures (MPa), n p is the number of perforation holes, d is the diameter of the perforation hole (m), C is the flow coefficient of the perforation hole, and p is the mixed density of the fracturing fluid (kg / m 3 ), D is the diameter of the wellbore (m), Q w,j is the flow rate in the wellbore between the j-1th cluster and the jth cluster (m 3 / s), L w,j is the cluster spacing between the j-1th cluster and the jth cluster (m), and K is the flow regime index of the fracturing fluid, which is dimensionless.

[0091] The fracture mouth pressure p fw,i of the ith cluster of fractures is obtained by solving the fluid flow equation in the fracture:

[0092]

[0093] wherein p(s, t) is the fluid pressure in the fracture (MPa); q(s, t) is the flow rate of the fracturing fluid in the fracture unit (m 3 / s); K is the consistency coefficient of the fracturing fluid (Pa·sn); n is the flow regime index of the fracturing fluid, which is dimensionless; H is the height of the fracture (m); W(s, t) is the width of the fracture (m); t is the construction time (s); c t is the comprehensive filtration coefficient of the fracturing fluid (m / s 0.5 ); and t is the time required for the fracturing fluid to reach s at time t (s).

[0094] Step 002, rock deformation model establishment

[0095] The fracture is discretized into m fracture elements with length of 2a, and the three-dimensional correction coefficient is used to correct the model, so as to obtain the relationship between the fracture surface displacement and stress during the fracturing process:

[0096]

[0097]

[0098]

[0099] wherein,

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107]

[0108] It is assumed that the pressure distribution in each fracture element is uniform, and the stress boundary condition is:

[0109]

[0110] In the above formula, and are the stress fields of the area around the fracture, specifically the normal stress and shear stress of the fracturing fracture in the x-x, y-y and x-y planes; D s,j is the tangential displacement discontinuity (m) of the jth fracture element, D n,j is the normal displacement discontinuity (m) of the jth fracture element; m is the number of fracture elements with length of 2a; d ij is the distance (m) between any point i of the formation and the fracture element j, v is the Poisson's ratio of the rock; σ n,j and σ s,j are the stresses in the tangential and normal directions of the fracturing fracture surface; p j is the fluid pressure in the fracture; is the normal bottom stress received by the fracture surface.

[0111] Substitute the stress boundary condition into the relationship between crack surface displacement and stress, and the normal displacement discontinuity D of each crack tip element can be obtained n and the tangential displacement discontinuity D s Note that the normal displacement discontinuity is the crack width we need.

[0112] Step 003, determine the multi-crack propagation direction and extension step

[0113] Based on the normal displacement discontinuity D n and the tangential displacement discontinuity D s Calculate the stress intensity factor K I and K II :

[0114]

[0115] In the formula, E is the elastic modulus of rock (GPa); v is the Poisson's ratio of rock; a is the half length of displacement discontinuity element (m).

[0116] At this time, the crack extension direction can be represented as:

[0117]

[0118] The energy release rate can be represented as:

[0119]

[0120] Based on the relationship between the energy release rate G of a specific crack tip and the maximum energy release rate G max of all crack tips, the extension step da i of the i-th crack tip can be represented as:

[0121]

[0122] In the formula, da max is the maximum crack extension step (m); G i is the energy release rate of the i-th crack (MPa·m); is an empirical constant, usually 1; G c is the critical energy release rate (MPa·m).

[0123] The three-dimensional displacement discontinuity algorithm (DDM) used in this example can quickly calculate the stress, strain and displacement of any point in the crack under the condition of large-scale development of unconventional reservoirs and well factory mode. It is easy to obtain the interference stress between cracks and the crack width, and overcomes the difficulty that the influence of crack distribution and mutual interference between cracks on the fracturing effect cannot be accurately characterized during the fracturing and production process of unconventional reservoirs.

[0124] Example Two

[0125] In a method for unconventional reservoir fracturing and production optimization, an XGBoost machine learning intelligent algorithm includes the following steps:

[0126] Step 01, constructing a maximized cumulative production objective function

[0127] Defining the maximized cumulative production objective function L(y i, y), adding a regularization term to control the complexity of the base learner tree and prevent overfitting, the maximized cumulative production objective function is defined as follows:

[0128]

[0129] L(yi,y) is the maximized cumulative production objective function; Ω(f k ) is the regularization term; yi is the model prediction value; y is the true value of the sample; k is the number of trees; f k is the model of the kth tree.

[0130] Substitute the true value y of the sample to obtain The complexity of the subsequent base learner tree is divided into the complexity of the first k-1 trees and the complexity of the current model tree, and at this time the objective function can be rewritten as:

[0131]

[0132] Step 02, simplifying the maximized cumulative production objective function

[0133] Simplify the maximized cumulative production objective function by means of Taylor's second-order expansion, and at this time the objective function can be simplified as:

[0134]

[0135] Among them,

[0136] Step 03, parameterizing the objective function and introducing tree structure

[0137] Parameterize f k (x i ) and Ω(f k ), and substitute the complexity of the model into the objective function:

[0138]

[0139] In the formula, w q (x i ) is the prediction value of the kth tree; w j is the value of each leaf node; J is the total number of trees; n is the total number of samples; α and λ are both parameters for controlling the degree of punishment; T is the number of leaf nodes.

[0140] Step 04, constructing a maximum cumulative yield optimal tree

[0141] Splitting to construct a tree model according to the principle of maximum information gain (max (obj new -obj old )) to construct a tree model that minimizes the objective function, construct K trees according to the same logic, and obtain a maximum cumulative yield optimal tree by weighting all the trees:

[0142]

[0143] In this example, the XGBoost intelligent optimization algorithm is used to carry out the collaborative optimization of unconventional reservoir fracturing and production, to construct a horizontal well fracturing and production database covering unconventional reservoir properties, fracturing and fracture parameters, soak and production parameters, and to use the XGBoost intelligent algorithm as the optimization means for collaborative design of fracturing and production parameters, to maximize the development benefit of unconventional oil and gas reservoirs.

[0144] Those of ordinary skill in the art can understand that all or some of the steps in the method disclosed above, the functions of the modules / units in the system, and the device can be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component can have multiple functions, or one function or step can be performed by several physical components in cooperation. Certain components or all components can be implemented as software executed by a processor, such as a digital signal processor or a microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transitory media). As is well known to those of ordinary skill in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically includes computer readable instructions, data structures, program modules or other data in modulated data signals such as carrier waves or other transport mechanisms, and can include any information delivery medium.

Claims

1. A method of unconventional reservoir fracturing and flowback coordination, characterized in that, The method comprises: According to the fracturing data, the bottom hole pressure data during the soak, the fracturing fluid flowback data and the bottom hole pressure data during the production, the physical property data of the unconventional reservoir, the parameters of the fracturing and the fractures, the soak and production parameters are determined respectively; According to the physical property data of the unconventional reservoir, the fracturing and fracture parameters, the soak and production parameters, a horizontal well fracturing and production database is established; According to the data in the horizontal well fracturing and production database, the fracturing and production design parameters of the unconventional reservoir are determined by using the XGBoost intelligent optimization algorithm.

2. The unconventional reservoir fracturing and production cooperative method according to claim 1, wherein The physical property data of the unconventional reservoir, the parameters of the fracturing and the fractures, the soak and production parameters are determined according to the fracturing data, the bottom hole pressure data during the soak, the fracturing fluid flowback data and the bottom hole pressure data during the production, respectively, comprising: The formation permeability, the original formation pressure and the fracture half-length are inverted based on the fracturing pump stop data; The average fracture half-length, the secondary fracture density and the formation pressure of the fracturing transformation area are inverted based on the bottom hole pressure data during the soak; The initial fracture pore volume and the fracture permeability are inverted based on the fracturing fluid flowback data; The well storage, the horizontal permeability and the initial pressure are inverted based on the bottom hole pressure data during the production; The multi-fracture stress interference and the fracture width are calculated by using the three-dimensional displacement discontinuity algorithm.

3. The unconventional reservoir fracturing and production cooperative method according to claim 2, wherein The formation permeability, the original formation pressure and the fracture half-length are inverted based on the fracturing pump stop data, comprising: After the fracturing pump is stopped, the current variable density data and the ground pressure of the flow rate are determined; The variable density and the ground pressure of the flow rate are converted into the bottom hole pressure; The pressure drop data after the fracturing pump is stopped are filtered; The filtered pressure drop data are interpreted to determine the formation permeability, the original formation pressure, the fracture half-length and the transformation area.

4. The unconventional reservoir fracturing and production cooperative method according to claim 2, wherein The average fracture half-length, the secondary fracture density and the formation pressure of the fracturing transformation area are inverted based on the bottom hole pressure data during the soak, comprising: The soak process is divided into the wellbore connection control stage, the main fracture reservoir stage, the main fracture and secondary fracture interflow control stage, the secondary fracture reservoir stage and the matrix flow control stage according to the flow characteristics during the soak; The linear flow mathematical model after the fracture is closed is established for the wellbore connection control stage, the main fracture and secondary fracture interflow control stage and the matrix flow control stage; The fracture reservoir control mathematical model is established for the main fracture reservoir stage and the secondary fracture reservoir stage; Based on the pressure drop model during the soak, the linear flow mathematical model after the fracture is closed and the fracture reservoir control mathematical model, the main fracture average half-length, the secondary fracture density and the formation pressure of the reservoir transformation area are inverted by using the bottom hole pressure data during the soak.

5. The unconventional reservoir fracturing and production cooperative method according to claim 2, wherein The initial fracture pore volume and the fracture permeability are inverted based on the fracturing fluid flowback data, comprising: The fracturing fluid flowback process is divided into the pseudo-fracture linear flow and the pseudo-fracture boundary control flow two stages; According to the oil-water or gas-water two-phase composite flow, occurrence and migration mechanism between the fracture and the matrix, a fracturing fluid flowback seepage mathematical model of unconventional oil and gas reservoirs is established; According to the fracturing fluid flowback seepage mathematical model, the characteristic curves of the pseudo-fracture linear flow stage and the pseudo-fracture boundary control flow stage are solved respectively; The initial pore volume of the fracture and the initial permeability of the fracture are inversely calculated based on the slope and intercept of the straight line segment in the characteristic curve.

6. The unconventional reservoir fracturing and production cooperative method of claim 2, wherein the inversion of the well storage, the horizontal permeability and the initial pressure based on the bottom hole pressure data during production includes: presetting inversion parameters of the well storage, the horizontal permeability, the initial pressure and the fracture half length, generating a plurality of initial sets composed of the inversion parameters by using a random algorithm; calculating state values under the input of each set based on a reservoir model; updating the state values by substituting the state values and the field observation values into a set Kalman filter method; outputting the final state values after a plurality of iterations; processing the final state values to obtain the well storage, the horizontal permeability and the initial pressure.

7. The unconventional reservoir fracturing and production cooperative method of claim 2, wherein the calculation of the multi-fracture stress interference and the fracture width by using a three-dimensional displacement discontinuity algorithm includes: establishing a multi-cluster fracturing fracture propagation model according to the inter-fracture stress interference and the dynamic allocation of inter-cluster flow during the multi-cluster fracturing process; calculating the stress, strain and displacement of any point in the fracture by using a three-dimensional displacement discontinuity algorithm; determining the inter-fracture interference stress and the fracture width according to the stress, strain and displacement; determining the fracture propagation step according to the energy release rate at the fracture tip, the inter-fracture interference stress and the fracture width.

8. The unconventional reservoir fracturing and production cooperative method of claim 7, wherein the design parameters of the unconventional reservoir fracturing and production are determined by using an XGBoost intelligent optimization algorithm according to the data in the horizontal well fracturing and production database, including: defining a maximized cumulative production objective function, and simplifying the objective function to obtain a simplified objective function; introducing a tree structure to parameterize the simplified objective function to obtain an optimal objective function; constructing an optimal tree according to the optimal objective function; performing optimization search from the fracturing and production parameter data in the horizontal well fracturing and production database by using the optimal tree to obtain the optimal fracturing and production design parameters. The device includes a memory and a processor; the memory is used to save a program for performing unconventional reservoir fracturing and production cooperation, and the processor is used to read and execute the program for performing unconventional reservoir fracturing and production cooperation, and execute the method of any one of claims 1-8.

10. A computer readable storage medium, a data processing program is stored on the computer readable storage medium, the data processing program is executed by a processor to perform the method of unconventional reservoir fracturing and production cooperation of any one of claims 1-8. ​ 9. A non-conventional reservoir fracturing and flowback cooperative device, characterized in that, ​ ​

Images