A multi-field coupling complex fractured formation fracture width distribution inversion method

Abstract

The present application relates to the technical field of oil and gas drilling engineering, and specifically discloses a multi-field coupling complex fractured formation fracture width distribution inversion method, based on the theory of porous elasticity, a drilling fluid loss model considering fluid-solid coupling effect is constructed, a roughness correction factor is introduced to correct the traditional cubic law, and the application of Darcy's law in the matrix is improved, a drilling fluid loss mathematical model coupled with the matrix and the fracture is established, through numerical simulation, the influence law of key parameters such as fluid viscosity, pressure difference and fracture width on drilling fluid loss is obtained, and on this basis, a fracture width multivariate linear regression inversion equation with fluid viscosity, pressure difference and cumulative loss as independent variables is established, the inversion equation has high prediction accuracy and practicability, can be used for quickly and accurately inverting the fracture width on site, and provides theoretical and technical support for the understanding of the loss mechanism of multi-scale fractured formation and the optimization of leakage prevention and treatment.

Description

Technical Field

[0001] This invention relates to the field of oil and gas drilling engineering technology, and specifically to a method for inverting the fracture width distribution in complex fractured formations using multi-field coupling. Background Technology

[0002] Drilling fluid loss in complex fractured formations is a common and difficult-to-solve global problem in oil and gas well drilling, severely impacting drilling efficiency, safety, and cost. Accurately predicting the pattern of loss and the width of the loss fractures is a crucial foundation for scientifically designing plugging schemes and improving the success rate of plugging.

[0003] Currently, scholars both domestically and internationally have conducted extensive research on predicting fracture width due to leakage. Common methods for calculating fracture width mainly include: interpreting fracture width based on well logging data, calculating fracture width forward from rock mechanical properties, and inverting fracture width based on leakage parameters (such as leakage rate and leakage volume). Among these, the inversion method based on leakage parameters is more widely used because it is directly related to in-situ leakage behavior.

[0004] Existing fluid models used to predict fracture width mainly include Newtonian fluid models, Bingham fluid models, and Herba fluid models. For example, Sanfillippo et al. calculated fracture width based on the Newtonian fluid model, but the Newtonian fluid model has limitations in describing the complex rheological properties of actual drilling fluids, thus limiting its application. Lietard and Verga et al., building upon the Bingham fluid model, established the relationship between theoretical leakage characteristic curves or leakage volume and fracture width and engineering parameters through dimensionless variation methods or analytical formulas, and confirmed that non-Newtonian fluid models are more accurate than Newtonian fluid models in describing drilling fluid leakage. Majidi and ShahriMP et al. further studied the flow behavior of Herba fluids within fractures and established a predictive model.

[0005] However, existing models still have shortcomings. Some models do not comprehensively describe the rheological properties of drilling fluids, failing to fully consider the impact of their non-Newtonian characteristics on leakage behavior. Most models do not adequately consider the reduction effect of fracture surface roughness on fluid conductivity, while actual formation fractures are often rough and uneven. They also do not adequately consider the fluid-structure interaction effect between complex fracture networks and the matrix, failing to accurately reflect the dynamic influence of formation deformation on fracture aperture and the filtering effect of the matrix on fractures. Although three-dimensional fracture network drilling fluid leakage models have been proposed, achieving accurate prediction of the size distribution characteristics of multi-scale leakage channels within fracture systems, especially in terms of rapid and accurate fracture width inversion in engineering applications, without relying on geological parameters (such as fracture density) that are difficult to obtain directly, still presents certain challenges. Summary of the Invention

[0006] The purpose of this invention is to provide a multi-field coupled method for inverting the fracture width distribution in complex fractured formations, so as to overcome the limitations of existing technologies in fracture width prediction.

[0007] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution:

[0008] A method for inverting fracture width distribution in complex fractured formations using multi-field coupling includes the following steps:

[0009] Step 100: Obtain field data related to drilling fluid loss in fractured formations, including rock physical parameters characterizing matrix seepage characteristics and drilling fluid rheological parameters;

[0010] Step 200: Based on the theory of porous elasticity, the bottom layer is divided into a porous matrix and a discrete fracture network, and a set of fluid-solid coupling equations for the porous matrix is ​​constructed, taking into account the interaction between drilling fluid and formation.

[0011] After processing the discrete fracture network using the extended finite element method, the displacement field approximation expression and the fluid pressure field expression obtained by approximating with the Heaviside jump function are substituted into the weak integral form of the fluid-structure interaction control equations to obtain the drilling fluid leakage model in discrete calculation format.

[0012] The drilling fluid leakage model achieves coupled calculation of displacement field and fluid pressure field through iterative solution, solving the strong nonlinear problem of crack deformation and fluid flow in drilling fluid leakage, and providing a theoretical basis for establishing the relationship in the future.

[0013] Step 300: Construct the drilling fluid leakage model into a fracture network model, and combine the parameters contained in the field data, input the parameter combination into the constructed fracture network model for numerical simulation, obtain the influence of the parameters contained in the field data on the trend of drilling fluid leakage over time and the cumulative leakage over a fixed time, obtain the cumulative leakage data under different parameter combinations, and form the training dataset of the model.

[0014] Step 400: Based on the training dataset, use the multiple linear regression method to establish a model with fluid viscosity as the modulus. Pressure difference and cumulative leakage Input parameters, invert crack width For the prediction model of output parameters;

[0015] The expression for the prediction model is:

[0016] ;

[0017] in, Represents the intercept term. These are partial regression coefficients.

[0018] As a preferred embodiment of the present invention, the matrix fluid-structure interaction equation set includes the stress balance equation of the matrix, the mass conservation equation of the fluid in the porous matrix, and the fluid flow control equation in the crack.

[0019] In constructing the governing equations for fluid flow within the fracture, the fracture is assumed to be a parallel plate model with a constant aperture. The flow of drilling fluid within the fracture is set to satisfy the cubic law, and the cubic law is modified by introducing a geometric correction factor. The modified cubic law is as follows:

[0020] ;

[0021] In the formula, This represents the cross-sectional flow rate per unit crack length within the crack. This represents the current crack width; This is expressed as the fluid pressure within the crack; This is the geometric correction factor for rough cracks, with a value ranging from 0 to 1.

[0022] As a preferred embodiment of the present invention, taking into account the interaction between drilling fluid and formation, the specific implementation is as follows:

[0023] Assuming the drilling fluid conforms to the Bingham rheological model, the relationship between its shear stress and shear rate is as follows:

[0024] ;

[0025] In the formula, Expressed as shear stress; Represented as dynamic shear force; Expressed as plastic viscosity; Expressed as shear rate;

[0026] Assuming the crack aperture conforms to a linear deformation law, its calculation formula is as follows:

[0027] ;

[0028] In the formula, This represents the current crack width; This represents the original crack width; This is expressed as the change in pressure within the crack; Expressed as the normal phase stiffness coefficient;

[0029] The setting is for planar fractures, where drilling fluid flows along the fracture surface.

[0030] As a preferred embodiment of the present invention, the fluid pressure field within the fracture is described by a partial differential equation based on the modified cubic law and the setting that the drilling fluid conforms to the Bingham rheological model:

[0031] ;

[0032] In the formula, Represented as the gradient operator in the fluid domain, Represented as geometric / unit conversion factors, Expressed as the equivalent viscosity of Bingham fluid; Expressed as pressure in the fluid phase; This is expressed as effective porosity; This is expressed as the filtration loss from the crack to the matrix; This represents the current crack width.

[0033] As a preferred embodiment of the present invention, when constructing the fracture network model, the three-dimensional discrete fracture network plugin in COMSOL Multiphysics software is used to generate the coordinates of the fracture center point, and a generation domain larger than the analysis domain is set to eliminate boundary effects.

[0034] As a preferred embodiment of the present invention, the inverted crack width output by the prediction model This includes the minimum crack width, the maximum crack width, the average crack width, and the probability density of the average crack width.

[0035] Based on the data measured by the inversion equation of the minimum crack width, the maximum crack width, the average crack width, and the probability density of the average crack width, a crack width prediction distribution map is drawn.

[0036] In a preferred embodiment of the present invention, when constructing the fluid-structure interaction equations for a porous matrix, the stress balance equation of the matrix, the mass conservation equation of the fluid in the porous matrix, and the fluid flow control equation within the crack are coupled through set initial and boundary conditions:

[0037] The initial conditions are that the formation pressure is equal to the initial pore pressure;

[0038] The boundary conditions are set as follows: the fracture boundary in contact with the wellbore is set to wellbore pressure, and the outer boundary is set to a no-flow boundary condition.

[0039] As a preferred embodiment of the present invention, the parameters included in the acquired field data influence the trend of drilling fluid loss over time and the cumulative loss over a fixed time, specifically including:

[0040] Fluid viscosity is negatively correlated with leakage rate;

[0041] Pressure differential, crack width, and crack density are positively correlated with leakage.

[0042] Crack width has a significant nonlinear growth effect on leakage.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] This invention, based on porous elasticity theory, constructs a three-dimensional discrete fracture network model considering fluid-structure interaction effects. A roughness correction factor is introduced to correct the traditional cubic law, and Darcy's law is improved for application in the matrix, thus establishing a mathematical model for drilling fluid loss in matrix-fracture coupling. Through numerical simulation, the influence of key parameters such as fluid viscosity, differential pressure, and fracture width on drilling fluid loss is systematically studied. Based on this, a novel multiple linear regression inversion equation for fracture width is established, with fluid viscosity, differential pressure, and cumulative loss as independent variables. This inversion equation exhibits high predictive accuracy and practicality, and can be used for rapid and accurate inversion of fracture width in the field, providing theoretical and technical support for understanding the leakage mechanism of multi-scale fractured formations and optimizing leakage prevention and control. Attached Figure Description

[0045] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0046] Figure 1 A flowchart illustrating the drilling fluid loss pattern and fracture width prediction in complex fractures according to an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram of a fractured formation physical model according to an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of crack distribution prediction according to an embodiment of the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] like Figure 1 , Figure 2 and Figure 3As shown, this invention provides a method for inverting the fracture width distribution in complex fractured formations using multi-field coupling, including the following specific steps:

[0051] Step 100: Obtain field data related to drilling fluid loss in fractured formations, including rock physical parameters characterizing matrix seepage characteristics and drilling fluid rheological parameters;

[0052] Step 200: Based on the theory of porous elasticity, the bottom layer is divided into a porous matrix and a discrete fracture network, and a set of fluid-solid coupling equations for the porous matrix is ​​constructed, taking into account the interaction between drilling fluid and formation.

[0053] After processing the discrete fracture network using the extended finite element method, the displacement field approximation expression and the fluid pressure field expression obtained by approximating with the Heaviside jump function are substituted into the weak integral form of the fluid-structure interaction control equations to obtain the drilling fluid leakage model in discrete calculation format.

[0054] The drilling fluid leakage model achieves coupled calculation of displacement field and fluid pressure field through iterative solution;

[0055] Step 300: Construct the drilling fluid leakage model into a fracture network model, and combine the parameters contained in the field data, input the parameter combination into the constructed fracture network model for numerical simulation, obtain the influence of the parameters contained in the field data on the trend of drilling fluid leakage over time and the cumulative leakage over a fixed time, obtain the cumulative leakage data under different parameter combinations, and form the training dataset of the model.

[0056] Step 400: Based on the training dataset, use the multiple linear regression method to establish a model with fluid viscosity as the modulus. Pressure difference and cumulative leakage For input parameters, crack width The prediction model for output parameters is used for rapid inversion of crack width;

[0057] The expression for the prediction model is:

[0058] ;

[0059] in, Represents the intercept term. These are partial regression coefficients.

[0060] In this embodiment, the displacement field specifically refers to the spatial deformation distribution of fractures under drilling fluid pressure, and the compression of formation pores is quantified through the displacement field. A displacement-enhancing shape function (Heaviside function) is used to simulate the displacement discontinuity of the fracture surface. The fluid pressure field refers to the pressure distribution of drilling fluid in the fractures and formation, and is the direct driving force for dynamic behaviors such as leakage rate and filtration loss. Similarly, a translation-enhancing shape function method is used to weaken the pressure field mixing elements.

[0061] The displacement approximation and fluid pressure approximation after translation correction are substituted into the fluid-structure interaction control equations, including the stress balance equation of the matrix, the mass conservation equation, and the fluid flow equation within the fracture. The aim is to construct a computable numerical model through mathematical discretization and physical field coupling to solve the strongly nonlinear problem of fracture deformation and fluid flow in drilling fluid loss.

[0062] The cumulative leakage rate in this implementation directly determines the feature dimension, temporal resolution, and physical constraints of the training dataset.

[0063] The dataset must systematically cover the key variables of the leakage dynamic process, and through rigorous data cleaning, leakage prevention, and hybrid modeling techniques, ensure that the model can accurately capture the physical essence of the leakage evolution. Otherwise, the model may only fit surface correlations and fail to guide actual leakage plugging decisions.

[0064] For example, rock physical parameters such as formation porosity and matrix permeability of the lost circulation zone can be obtained from well logging reports; rheological parameters of the drilling fluid used, such as dynamic viscosity, plastic viscosity, and dynamic shear force, can be obtained from well logging reports; and real-time or cumulative drilling fluid loss, wellbore pressure, and original formation pressure determined through pressure testing or experience can be obtained from well control or drilling parameter monitoring systems. These data ensure the accuracy of parameter settings in numerical simulations and inversion models.

[0065] Well logging data mainly includes porosity and permeability;

[0066] The logging data mainly includes drilling fluid viscosity, cumulative loss, formation pressure, and wellbore pressure.

[0067] In step 200, the three-dimensional discrete fracture network model uses the extended finite element method to handle the discontinuous interface of the fracture, and combines the translational strengthening shape function to construct approximate expressions for the displacement field and fluid pressure field, thus obtaining the discrete calculation format of the drilling fluid leakage model.

[0068] The extended finite element method (EFM) is employed to address fracture discontinuities, and approximate expressions for the displacement and fluid pressure fields are constructed by translating enhanced shape functions. These approximate expressions are then substituted into the weak integral form of the fluid-structure interaction (FSI) governing equations to obtain the final discrete calculation scheme for the drilling fluid leakage model, forming a coupled set of algebraic equations. This set of equations is solved iteratively to achieve simultaneous calculation of the displacement and pressure fields.

[0069] Initial conditions: The formation pressure is the initial pore pressure.

[0070] Boundary conditions: The fracture boundary in contact with the wellbore is set to wellbore pressure, and the outer boundary is set to a no-flow boundary condition.

[0071] In step 300, the three-dimensional discrete fracture network model treats the formation as a composite matrix consisting of a porous matrix and a discrete fracture network embedded therein;

[0072] When constructing the three-dimensional discrete crack network model, the three-dimensional discrete crack network plugin in COMSOL Multiphysics software is used to generate the coordinates of the crack center point, and a generation domain larger than the analysis domain is set to eliminate boundary effects.

[0073] In step 300, a multi-parameter combined simulation is performed on fluid viscosity, pressure difference, crack width, and crack density. According to the simulation scheme, the multi-parameter combined simulation is performed on fluid viscosity, pressure difference, crack width, and crack density. For example, fluid viscosity is taken as 10, 30, 60, and 90 mPa·s; pressure difference is taken as 1, 6, 10, and 15 MPa; crack width is taken as 0.1, 0.5, 1, 2, 3, 4, and 5 mm; and crack density is taken as 0.01, 0.1, and 0.5 cracks / m². 2 .

[0074] Subsequently, the solver built into COMSOL Multiphysics solves the coupled displacement and pressure fields simultaneously using an iterative method.

[0075] During the simulation, the leakage rate of drilling fluid from the wellbore into the fracture network was continuously monitored and integrated over time to obtain the cumulative leakage at different times.

[0076] Analysis of the simulation results reveals the influence of various parameters on leakage. For example, the study shows that fluid viscosity is negatively correlated with leakage, while pressure difference, crack width, and crack density are positively correlated with leakage. In particular, crack width has a significant nonlinear growth effect on leakage and is the primary factor controlling the leakage level.

[0077] All input parameters of the simulated operating conditions (fluid viscosity, pressure difference, crack width, crack density) and the corresponding cumulative leakage over a fixed time are compiled into a large training dataset for subsequent regression analysis.

[0078] Crack width output by the prediction model This includes the minimum crack width, the maximum crack width, the average crack width, and the probability density of the average crack width.

[0079] Based on the data measured by the inversion equations of the minimum crack width, maximum crack width, average crack width, and probability density of the average crack width, a crack width prediction distribution map was drawn.

[0080] Establishment of the inversion equation:

[0081] (1) Minimum crack width Establishment of the inversion equation:

[0082] ;

[0083] In the formula, The minimum crack width is expressed in mm. The viscosity is expressed in mPa·s. The pressure difference is expressed in MPa. For the cumulative leakage, m 3 .

[0084] (2) Maximum crack width Establishment of the inversion equation:

[0085] ;

[0086] In the formula, The maximum crack width is expressed in mm. The viscosity is expressed in mPa·s. Pressure difference, MPa; For the cumulative leakage, m 3 .

[0087] (3) Average crack width Establishment of the inversion equation:

[0088] ;

[0089] In the formula, The average crack width is in mm; The viscosity is expressed in mPa·s. Pressure difference, MPa; For the cumulative leakage, m 3 .

[0090] (4) Probability density of average crack width Establishment of the inversion equation:

[0091] ;

[0092] In the formula, The average crack width is in mm; The viscosity is expressed in mPa·s. Pressure difference, MPa; For the cumulative leakage, m 3 .

[0093] Based on the data predicted by the above inversion equations, a normal distribution map of the predicted crack width can be drawn.

[0094] This invention achieves the following objectives:

[0095] 1. Construct a three-dimensional discrete fracture network leakage model that considers fluid-structure interaction, rough fracture flow, and matrix-fracture interaction to more realistically simulate drilling fluid leakage behavior in complex fractured formations.

[0096] 2. Through systematic numerical simulation, the influence of key parameters such as fluid viscosity, differential pressure, fracture width and fracture density on drilling fluid loss is revealed, providing a theoretical basis for understanding the loss mechanism.

[0097] 3. Develop a practical, efficient and high-precision crack width inversion equation. This equation can quickly predict crack width using parameters that are readily available on-site (such as fluid viscosity, pressure difference, and cumulative leakage), thereby assisting in on-site leak sealing decisions.

[0098] Inversion equation for minimum crack width: coefficient of determination The coefficient of determination of the inversion equation for the maximum crack width reaches 0.887. The coefficient of determination of the inversion equation for the average crack width reaches 0.872. The coefficient of determination of the inversion equation for the average probability density of crack width reaches 0.809. The result of 0.842 indicates that the model has high prediction accuracy and practicality.

[0099] In actual drilling operations, when lost circulation occurs, drilling engineers can obtain the current viscosity of the drilling fluid based on logging data. The pressure difference is calculated based on the wellbore pressure and formation pressure. And monitor the cumulative leakage in real time. Then, by directly substituting these parameters into the fracture width inversion equation provided by this invention, the width of the leakage fracture path can be quickly calculated. Based on the inverted fracture width, such as micro-fractures of 0.1-0.5 mm, medium-sized fractures of 0.5-2 mm, or extra-large fractures greater than 2 mm, the drilling team can selectively choose appropriate plugging materials (such as plugging agents with different particle size distributions) and construction schemes, thereby improving the success rate of plugging and effectively controlling leakage.

[0100] Using 48 sets of leakage data from an oil spill site as an example, and taking known parameters as model input data, the proposed inversion model was validated and compared with the Sanfillippo, Verga, and Griffiths models. The results show that the average relative error of the proposed coupled-method road width prediction model is 3.76% and 3.984% compared to the Sanfillippo and Verga models, respectively, significantly lower than the average relative error of 7.39% for the GriffithsCubic model. This demonstrates the high prediction accuracy and generalization ability of the proposed model in field applications.

[0101] The construction of the equation system in this embodiment specifically includes:

[0102] The stress balance equation of the matrix, considering the Biot effect:

[0103] ;

[0104] In the formula, the total stress tensor , For the effective stress tensor, For Biot coefficient, Pore ​​fluid pressure, For unit tensors, For matrix density, Let be the acceleration due to gravity. This equation reveals the effect of the pore pressure gradient on matrix deformation.

[0105] Constitutive relation: The constitutive relation between matrix deformation and effective stress follows the generalized Hooke's law.

[0106] Mass conservation equation for fluids in porous matrices:

[0107] ;

[0108] In the formula, Porosity For fluid density, This is the Darcy flow velocity.

[0109] Darcy's equation for fluid flow in the matrix:

[0110] ;

[0111] In the formula, For Darcy velocity, For matrix permeability, This represents the fluid viscosity.

[0112] Pore ​​pressure diffusion equation:

[0113] ;

[0114] In the formula, The overall compression ratio is 1. For matrix volume strain, For source and sink items.

[0115] Relationship between fluid pressure and fluid volume increment: Considering Biot modulus, matrix volumetric strain will change its porosity and permeability, thus affecting fluid flow.

[0116] (2) Control equations for fluid flow within the crack:

[0117] Modified Cubic Law: For rough cracks, a geometric correction factor is introduced. Revised traditional cube law:

[0118] ;

[0119] In the formula, For crack opening, This is the equivalent viscosity of Bingham fluid. The pressure is the fluid pressure inside the crack. Determined by fractal parameters, it is usually less than 1, reflecting the reduction of permeability due to roughness.

[0120] The geometric correction factor here is an ideal value obtained from the model, and the specific value can be selected based on field data. For the cubic law, the geometric correction factor can correct the seepage error of rough cracks, take into account the influence of tortuosity on the seepage path, and improve the seepage prediction ability of crack networks. For the coupled control equations, the geometric correction factor can enhance physical consistency, improve the dynamic interaction of fluid-structure interaction, improve nonlinear processing ability, and enhance engineering applicability.

[0121] These effects make the cubic law and the coupled control equations more applicable to engineering problems in complex fractured formations (such as drilling fluid loss and oil and gas field fracturing).

[0122] Mass conservation equation for fluid within the crack:

[0123] ;

[0124] In the formula, The equivalent porosity of the crack. The flow velocity within the crack, This represents the amount of fluid lost from the cracks into the matrix.

[0125] Bingham fluid rheological model: The relationship between drilling fluid shear stress and shear rate follows the Bingham rheological model.

[0126] Linear deformation law of crack aperture: Considering the stress sensitivity of cracks, the crack aperture changes linearly with the pressure inside the crack:

[0127] ;

[0128] In the formula, The original crack width. This represents the pressure change within the crack. This is the normal stiffness coefficient.

[0129] The governing equations for the fluid pressure field within the fracture: Considering the above factors, the partial differential equations describing the fluid pressure field within the fracture are obtained:

[0130] ;

[0131] In the formula, Here is the stiffness matrix of the solid. and This is a fluid-structure interaction term; The fluid mass matrix; Here is the fluid stiffness matrix; The displacement degree of freedom vector; The fluid pressure degree of freedom vector; and These are the solid and fluid load vectors, respectively. This equation is solved iteratively to achieve coupled calculation of the displacement field and the pressure field.

[0132] Alternatively, it can be expressed using the following partial differential equation:

[0133] ;

[0134] In the formula, Represented as the gradient operator in the fluid domain, Represented as geometric / unit conversion factors, Expressed as the equivalent viscosity of Bingham fluid; Expressed as pressure in the fluid phase; This is expressed as effective porosity; This is expressed as the filtration loss from the crack to the matrix; This represents the current crack width.

[0135] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for inverting fracture width distribution in complex fractured formations using multi-field coupling, characterized in that, Including specific steps: Step 100: Obtain field data related to drilling fluid loss in fractured formations, including rock physical parameters characterizing matrix seepage characteristics and drilling fluid rheological parameters; Step 200: Based on the theory of porous elasticity, the bottom layer is divided into a porous matrix and a discrete fracture network, and a set of fluid-solid coupling equations for the porous matrix is ​​constructed, taking into account the interaction between drilling fluid and formation. After processing the discrete fracture network using the extended finite element method, the displacement field approximation expression and the fluid pressure field expression obtained by approximating with the Heaviside jump function are substituted into the weak integral form of the fluid-structure interaction control equations to obtain the drilling fluid leakage model in discrete calculation format. The drilling fluid leakage model achieves coupled calculation of displacement field and fluid pressure field through iterative solution; Step 300: Construct the drilling fluid leakage model into a drilling fluid leakage fracture network model, and combine the parameters contained in the field data, input the parameter combination into the constructed drilling fluid leakage fracture network model for numerical simulation, obtain the influence of the parameters contained in the field data on the trend of drilling fluid leakage over time and the cumulative leakage over a fixed time, obtain the cumulative leakage data under different parameter combinations, and form the training dataset of the model; Step 400: Based on the training dataset, use the multiple linear regression method to establish a model with fluid viscosity as the modulus. Pressure difference and cumulative leakage Input parameters, invert crack width For the prediction model of output parameters; The expression for the prediction model is: ； in, Represents the intercept term. These are partial regression coefficients.

2. The method for inverting the fracture width distribution in complex fractured formations using multi-field coupling as described in claim 1, characterized in that, The matrix fluid-structure interaction equation set includes the stress balance equation of the matrix, the mass conservation equation of the fluid in the porous matrix, and the fluid flow control equation in the crack. In constructing the governing equations for fluid flow within the fracture, the fracture is assumed to be a parallel plate model with a constant aperture. The flow of drilling fluid within the fracture is set to satisfy the cubic law, and the cubic law is modified by introducing a geometric correction factor. The modified cubic law is as follows: ； In the formula, This represents the cross-sectional flow rate per unit crack length within the crack. This represents the current crack width; This is expressed as the fluid pressure within the crack; This is the geometric correction factor for rough cracks, with a value ranging from 0 to 1.

3. The method for inverting the fracture width distribution in complex fractured formations using multi-field coupling as described in claim 2, characterized in that, Taking into account the interaction between drilling fluid and formation, specifically: Assuming the drilling fluid conforms to the Bingham rheological model, the relationship between its shear stress and shear rate is as follows: ； In the formula, Expressed as shear stress; Represented as dynamic shear force; Expressed as plastic viscosity; Expressed as shear rate; Assuming the crack aperture conforms to a linear deformation law, its calculation formula is as follows: ； In the formula, This represents the current crack width; This represents the original crack width; This is expressed as the change in pressure within the crack; Expressed as the normal phase stiffness coefficient; The setting is for planar fractures, where drilling fluid flows along the fracture surface.

4. The method for inverting the fracture width distribution in complex fractured formations using multi-field coupling as described in claim 3, characterized in that, Using the modified cubic law and the assumption that the drilling fluid conforms to the Bingham rheological model, the pressure field of the fluid within the fracture is described by partial differential equations: ； In the formula, Represented as the gradient operator in the fluid domain, Represented as geometric / unit conversion factors, Expressed as the equivalent viscosity of Bingham fluid; Expressed as pressure in the fluid phase; This is expressed as effective porosity; This is expressed as the filtration loss from the crack to the matrix; This represents the current crack width.

5. The method for inverting the fracture width distribution in complex fractured formations using multi-field coupling as described in claim 1, characterized in that, When constructing the drilling fluid leakage fracture network model, the three-dimensional discrete fracture network plugin in COMSOL Multiphysics software was used to generate the coordinates of the fracture center point, and a generation domain larger than the analysis domain was set to eliminate boundary effects.

6. The method for inverting the fracture width distribution in complex fractured formations using multi-field coupling as described in claim 1, characterized in that, The inversion crack width output by the prediction model This includes the minimum crack width, the maximum crack width, the average crack width, and the probability density of the average crack width. Based on the data measured by the inversion equation of the minimum crack width, the maximum crack width, the average crack width, and the probability density of the average crack width, a crack width prediction distribution map is drawn.

7. The method for inverting the fracture width distribution in complex fractured formations using multi-field coupling as described in claim 1, characterized in that, When constructing the fluid-structure interaction equations for porous matrices, the stress balance equations of the matrix, the mass conservation equations of the fluid in the porous matrix, and the fluid flow control equations within the cracks are coupled through set initial and boundary conditions: Initial conditions: Formation pressure is the initial pore pressure; Boundary conditions: The fracture boundary in contact with the wellbore is set to wellbore pressure, and the outer boundary is set to a no-flow boundary condition.

8. The method for inverting the fracture width distribution in complex fractured formations using multi-field coupling as described in claim 1, characterized in that, The parameters included in the acquired field data influence the trend of drilling fluid loss over time and the cumulative loss over a fixed period, specifically including: Fluid viscosity is negatively correlated with leakage rate; Pressure differential, crack width, and crack density are positively correlated with leakage. Crack width has a significant nonlinear growth effect on leakage.

Images