Nonlinear constraint coupling driven implicit modeling optimization method and system and memory

Abstract

The invention discloses a non-linear constraint coupling driven implicit modeling optimization method and system and a memory. The method comprises the steps of firstly obtaining nonlinear constraint data of a target geological interface, then generating initial implicit function representation through a Hermite radial basis function based on an initial control point of the target geological interface, and then obtaining a nonlinear coupling driving optimization objective function about implicit function representation of the target geological interface according to the data type of nonlinear constraint, and optimizing through an implicit gradient flow algorithm to obtain a geological interface reconstruction model and a global optimization target geological interface model. The invention further provides a computer readable memory, the memory contains a computer program for realizing the implicit modeling optimization method for nonlinear constraint coupling driving, the computer program can be read and executed, and the computer program, a programmable logic controller and a processor form an optimization system. According to the system, nonlinear constraint data of a large number of geological interfaces are fully coupled, an implicit modeling framework is constrained, and therefore high-precision modeling of the three-dimensional geological interfaces is achieved.

Description

Technical Field

[0001] The present invention relates to an implicit modeling optimization method, and particularly to an implicit modeling optimization method, system and memory driven by non-linear constraint coupling, belonging to the technical field of 3D geological modeling. Background Art

[0002] The 3D geological interface model can intuitively represent the geometric shape and spatial distribution characteristics of the 3D interface, and has important significance for applications such as predicting geological disasters, optimizing engineering designs, simulating geological processes, predicting mineral resources, and oil and gas exploration. However, the current modeling framework can only perform iterative reconstruction by combining linear constraint information, resulting in redundancy of a large amount of non-linear constraint information in geology, geophysics, and prior knowledge, making it difficult to constrain the 3D geological modeling process. For 3D geological interfaces with complex morphological characteristics, the current modeling methods will make the modeling results have greater uncertainty and are difficult to simulate the actual geological interface well, especially reflected in unclear detailed features, resulting in low resolution of the model structure features, bringing a greater risk of error to subsequent actual exploration applications. Summary of the Invention

[0003] Aiming at the problems existing in the prior art, the first object of the present invention is to provide an implicit modeling optimization method driven by non-linear constraint coupling. The method first obtains the non-linear constraint data of the target geological interface, then generates an initial implicit function representation through the Hermite radial basis function, and then obtains a non-linear coupling-driven optimization objective function for the implicit function expression of the target geological interface according to the data type of the non-linear constraint, and optimizes it through the implicit gradient flow algorithm to obtain a geological interface reconstruction model and a globally optimized target geological interface model.

[0004] The second object of the present invention is to provide a computer-readable memory, which contains a computer program for implementing the above-mentioned implicit modeling optimization method driven by non-linear constraint coupling and can be read and executed.

[0005] The third object of the present invention is to provide an implicit modeling optimization system driven by non-linear constraint coupling. The system realizes high-precision modeling of the 3D geological interface by fully coupling a large amount of non-linear constraint data of the geological interface, effectively solving the defect that the geological interface model in the prior art can only combine linear constraint information. Further, the method also converts the non-linear problem into a linear least-squares problem through the implicit gradient flow algorithm, thereby realizing the compatibility between non-linear constraints and linear constraints, and greatly improving the accuracy and reliability of geological modeling.

[0006] To achieve the above technical objectives, the present invention provides an implicit modeling optimization method driven by non-linear constraint coupling, including:

[0007] Step S1: Obtain the non - linear constraint data of the target geological interface to obtain a modeling constraint database;

[0008] Step S2: Extract the exploration control points of the target geological interface according to the exploration data, and generate an initial implicit function representation of the target geological interface through the Hermite radial basis function;

[0009] Step S3: According to the data types in the modeling constraint database, obtain a non - linear coupling - driven optimization objective function for the implicit function expression of the target geological interface, and give the relationship expression between each non - linear constraint in the objective function and the morphological characteristics of the geological interface;

[0010] Step S4: According to the principle of Newton's method, obtain a model implicit function with an iterative step size through the implicit gradient flow algorithm, and then obtain the globally optimized target geological interface model through the soft linear search algorithm;

[0011] Step S5: Substitute the model implicit function of each iteration into the objective function, draw the curve of the change of the objective function value. When the oscillation amplitude of the curve is within the preset minimum oscillation amplitude range and the error is less than the error threshold, output the current geological interface model as the reconstructed model.

[0012] The method provided by the present invention fully couples a large amount of non - linear constraint information of the geological interface. By constraining the modeling process, it minimizes the uncertainty of the obtained model to fundamentally solve the technical problem that the current geological modeling process can only combine linear constraints, thereby greatly improving the accuracy of the reconstructed model and increasing the reliability and accuracy of subsequent exploration applications.

[0013] As a preferred solution, the non - linear constraint data includes: geological interface continuity constraint data, lithology spatial distribution constraint data, seismic data reflecting the physical properties of the hanging wall and footwall of the geological interface, rock physics data, tectonic evolution history constraint data, and sedimentation rate data. In order to improve the accuracy of the subsequent obtained model, the present invention collects a large amount of non - linear constraint data of the geological interface. In 3D geological modeling, any data that can characterize the morphological and spatial distribution characteristics of the 3D geological interface can be used in the modeling workflow.

[0014] As a preferred solution, the extraction process of the exploration control points is as follows: Extract the 3D geological interface boundary from the plan view, section view, and borehole record map including the target geological interface, and represent the boundary in the form of spatial coordinate points, that is, obtain it.

[0015] As a preferred solution, the acquisition process of the initial implicit function representation of the target geological interface is as follows:

[0016] Step S2 - 1: Use the HRBF modeling method to represent the initial implicit function of the geological interface, that is, in the 3D geological space In it, the initial implicit function of the three-dimensional interface is defined , representing any point in the three-dimensional space at the scalar function . Combining the coordinate constraints of exploration control points, the initial implicit function of any geological interface is expressed as: Equation 1: ; Equation 2: ;

[0017] In Equation 1 and Equation 2, is the number of known points of the geological interface in the three-dimensional geological space, is the linear coefficient expressed by the HRBF function, is the radial basis function, is the exploration control point;

[0018] Step S2-2: Substitute the exploration control points into the initial implicit function of the geological interface to make it meet the classification conditions, and solve the linear coefficient , that is, the analytical representation of the initial implicit function of the geological interface is obtained, where the classification condition is: Equation 3: .

[0019] As a preferred solution, the non-linear coupling drive optimization objective function is constructed based on the data in the modeling constraint database and the initial implicit function representation of the target geological interface.

[0020] As a preferred solution, the data in the modeling constraint database are continuity constraint data, seismic data and sedimentation rate data.

[0021] As a preferred solution, the expression of the non-linear coupling drive optimization objective function is: Equation 4: ;

[0022] In Equation 4, is the displacement amount feature on both sides of the three-dimensional geological interface position, represents the number of volume elements on both sides of the three-dimensional interface, represents the coordinates of the spatial volume element with displacement observed in the current geological exploration, represents the spatial coordinates of the above volume element changing during the iteration, is the seismic wave at the corresponding position of the three-dimensional geological interface, represents the seismic wave velocity value observed at the surface observation point during geophysical exploration, represents the number of discrete points in the geological space, expresses the sedimentation rate of the corresponding coordinate point in the three-dimensional geological space, represents the sedimentation rate value of the researcher's prior knowledge or after geological verification.

[0023] As shown in Equation 4, it is mainly obtained by adding three parts. Among them, represents that the formation displacements on both sides of the three-dimensional interface geometry (such as faults and folds) have the characteristic of non-linear superposition and obey the exponential decay law in three-dimensional space. This decay form has a great influence on the three-dimensional geological interface morphology, so it is used as one of the non-linear constraint terms in three-dimensional modeling; the second part characterizes the physical property distribution on both sides of the geological interface. There is a power-law non-linear characteristic between the physical property distribution and the seismic wave velocity. Therefore, the interface morphology can be adjusted in the subsequent iterative process to make the physical property distribution on both sides more fit the geophysical seismic wave velocity observation, so as to explore the optimal spatial distribution of the three-dimensional geological interface; the third part is that during the geological structure evolution process, the sedimentation rate is closely related to the slope of the three-dimensional geological interface and can show a non-linear relationship. Therefore, this prior knowledge can be combined to constrain and optimize the morphology of the three-dimensional geological interface.

[0024] As a preferred solution, the model implicit function with an iteration step size is the solution expression form of the geological interface implicit function in the objective function, and its acquisition process is as follows: Step S4-1: Set the iteration step size for the implicit function . Therefore, the minimization solution of the above can be converted into the minimization solution of the iteration step size , that is: Equation 5: ; Step S4-2: To avoid the underdetermination of the objective function during the solution process, a smoothing term is added here, and its expression is: Equation 6: ; Step S4-3: Adopt the implicit gradient flow algorithm, perform Taylor expansion on the function expression obtained in Step S4-2, and construct the gradient representation of the geological interface implicit function in the objective function, and its expression is: Equation 7: ; Step S4-4: Perform implicit gradient calculation on the gradient representation obtained in Step S4-3, and finally obtain the model implicit function with an iteration step size, that is: Equation 8: ; In Equations 5 to 8, represents the initial implicit function value of the corresponding point , characterizes the displacement amount characteristics of the initial observation point and can be represented as a constant. Subsequently, will be used for representation, represents the weight of the smoothing term, , , Jacobian matrices respectively representing geological interface continuity constraints, seismic constraints, and sedimentation rate constraints representing the model implicit function of the th iteration, representing the step size of the

[0025] As a preferred solution, the process of the global optimization target geological interface model is as follows: Step S5-1: Adopt the weight value of the soft linear search step size, and set the iteration step size weight to , during the iterative optimization process of the objective function, this weight can be regarded as a constant value, expressed as: Equation 9: ; Step S5-2: For the initial objective function , its form with the iteration step size is: Equation 10: ; Equation 11: ; Equation 12: ; Equation 13: ; Step S5-3: According to Step S5-1 and Step S5-2, through the objective function value and the weight value of the iteration step size in each iteration, that is, in the solution process, the model implicit function with the iteration step size is expressed as: Equation 14: ; In Equations 9 to 14, is the updated value corresponding to in each iteration process, is a hyperparameter that needs to satisfy the condition of Equation 12, is a hyperparameter that needs to satisfy the condition of Equation 13.

[0026] The present invention also provides a computer-readable memory, including a computer program, and the computer program can implement the implicit modeling optimization method described in any one of the above.

[0027] The present invention also provides an implicit modeling optimization system driven by non-linear constraint coupling, including: a programmable logic controller, a processor, and the above-readable memory; the programmable logic controller includes a database construction module, an initial model construction module, a model optimization and reconstruction module, and an optimal model judgment module; the processor reads and executes the computer program on the readable memory according to the programmable logic controller.

[0028] Among them, the database construction module is used to collect non-linear constraint data related to the three-dimensional geological interface and integrate them to form a standardized database; The initial model construction module is used to extract exploration control points based on the obtained geological plane maps, sectional views, borehole records and other data, and realize the implicit function expression of the three-dimensional geological interface based on the HBRF radial basis function.

[0029] The model optimization and reconstruction module forms the objective function for model optimization and reconstruction according to the obtained non-linear constraint data types, and uses the proposed implicit gradient flow algorithm to transform the non-linear optimization problem in the objective function into a linear least squares problem through Taylor expansion and implicit gradient solution to solve the step size of each iterative optimization, and realize the implicit function expression of the model with step size during the iterative process. In addition, this module adds a soft linear search process to solve the weight of the iterative step size to avoid the optimization model falling into the local minimum of the objective function; The optimal model judgment module is used to draw the objective function values of each iteration, form the objective function change curve, judge the oscillation amplitude of the curve, and when the minimum oscillation amplitude and the minimum objective function value are reached, stop the model iteration process in time and output the implicit function expression of the optimized model.

[0030] Compared with the prior art, the beneficial effects brought by the technical solution provided by the present invention are as follows: 1) The implicit modeling optimization method provided by the present invention first obtains the non-linear constraint data of the target geological interface, then generates the initial implicit function representation through the Hermite radial basis function, and then obtains the non-linear coupled driving optimization objective function for the implicit function expression of the target geological interface according to the data type of the non-linear constraint, and optimizes it through the implicit gradient flow algorithm to obtain the geological interface reconstruction model and the globally optimized target geological interface model; 2) In the technical solution provided by the present invention, by fully coupling a large amount of non-linear constraint data of the geological interface, the implicit modeling framework is constrained, thereby realizing the high-precision modeling of the three-dimensional geological interface, effectively solving the defect that the geological interface model in the prior art can only combine linear constraint information. Further, this method also transforms the non-linear problem into a linear least squares problem through the implicit gradient flow algorithm, thereby realizing the compatibility between non-linear constraints and linear constraints, and greatly improving the accuracy and reliability of geological modeling. Description of the Drawings

[0031] Figure 1 It is a schematic diagram of the standard model and the initial model of the simple structure in Embodiment 1 of the present invention; Among them, Figure 1 (a) is the standard geological interface model of the simple structure, Figure 1 (b) is the initial geological model of the simple structure; Figure 2 It is a schematic diagram of the fine reconstruction result of the simple structure model in Embodiment 1 of the present invention; Figure 3 It is an example diagram of the standard model and the initial model of the complex salt dome in Embodiment 1 of the present invention; Among them, Figure 3 (a) is the standard geological model of the complex salt dome, Figure 3 (b) is an example of the initial model of the complex salt dome; Figure 4 It is a result diagram of the fine reconstruction of the model of the complex salt dome in Embodiment 1 of the present invention. Specific implementation manners

[0032] The technical solution of the present invention will be further described in detail below in conjunction with specific embodiments and the accompanying drawings. To facilitate the understanding of the present invention, the present invention will be described more comprehensively and meticulously below in conjunction with the specification drawings and preferred embodiments. It should be noted that the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0033] Embodiment 1

[0034] In this embodiment, aiming at the problem that the traditional three-dimensional geological modeling framework is difficult to combine non-linear constraints, resulting in redundancy of a large amount of geological and geophysical exploration information and prior knowledge of predecessors, an implicit modeling optimization method driven by non-linear constraint coupling is provided. The specific process is as follows: Step 1: In three-dimensional geological modeling, any data that can characterize the morphology and spatial distribution characteristics of the three-dimensional geological interface can be used in the modeling workflow. Here, common data related to the expression of the three-dimensional geological interface characteristics can be collected, such as geological interface continuity constraint data and lithology spatial distribution constraint data in geological exploration, seismic data and rock physics data reflecting the physical properties of the upper and lower plates of the geological interface in geophysical exploration, and tectonic evolution history constraint data and sedimentation rate data related to the morphological evolution characteristics of the geological interface in the geological process simulation, etc., to form a three-dimensional modeling constraint database; Step 2: For the collected intuitive geological plan and profile data, as well as borehole record diagrams, extract the boundaries of the three-dimensional geological interface, and represent the boundaries in the form of spatial coordinate points, that is, form exploration control points. The HRBF modeling method can be used to represent the initial implicit function of the three-dimensional interface, that is, in the three-dimensional geological space Define the initial implicit function of the three-dimensional interface , representing the scalar function at any point x in the three-dimensional space . Combining the coordinate constraints of the exploration control points, the implicit function of any geological interface can be expressed as: Equation 1: ; Equation 2: ; In Equation 1 and Equation 2, is the number of known points of the geological interface in the three-dimensional geological space, is the linear coefficient expressed by the HRBF function, is the radial basis function, is the exploration control point; Step S2-2: Substitute the exploration control points into the initial implicit function of the geological interface to make it satisfy the classification conditions, and solve the linear coefficient , that is, the analytical characterization of the initial implicit function of the geological interface is obtained. Among them, the classification condition is: Equation 3: ; For the above implicit modeling optimization method driven by non-linear constraint coupling, for the linear coefficients in the implicit function of the three-dimensional geological interface model, the coordinates of the known points of the geological interface in the three-dimensional geological space can be substituted to make the implicit function 0, so as to solve the linear coefficient , and obtain the expression of the initial implicit function; Step 3: Based on the implicit function characterization of the above three-dimensional geological interface, the objective function for reconstructing the three-dimensional geological interface model can be constructed according to the known modeling constraint data. Taking the geological interface continuity constraint data (geological data), seismic data (geophysical data), and sedimentation rate data (prior knowledge) as examples, since the non-linear constraints are independent of each other, their coupling driving effect can be regarded as the sum of the error mismatch functions between each constraint data and the forward modeling value of the model. That is, the following objective function is constructed to realize the optimized reconstruction of the three-dimensional geological interface model: Equation 4: ; For the first term in the above objective function, expresses the displacement characteristics on both sides of the corresponding position of the three-dimensional geological interface, characterizes the number of volume elements on both sides of the three-dimensional interface, represents the coordinates of the spatial volume element with displacement observed in the current geological exploration, characterizes the spatial coordinates of the above volume element changing during the iteration process. This term indicates that the stratum displacements on both sides of the three-dimensional interface geometry (such as faults and folds) have non-linear superposition characteristics and obey the exponential decay law in the three-dimensional space. This decay form has a great influence on the shape of the three-dimensional geological interface, so it is one of the non-linear constraint terms for three-dimensional modeling; For the second term in the above objective function: expresses the seismic wave velocity at the corresponding position of the three-dimensional geological interface, characterizes the seismic wave velocity value observed at the surface observation point during geophysical exploration, Characterize the number of discrete points in the geological space. This objective function characterizes the physical property distribution on both sides of the geological interface. There is a power-law non-linear relationship between the physical property distribution and the seismic wave velocity. Therefore, the interface shape can be adjusted in the subsequent iteration process to make the physical property distribution on both sides fit the geophysical seismic wave velocity observation better, and explore the optimal spatial distribution of the three-dimensional geological interface.

[0035] For the third term in the objective function: Express the sedimentation rate of the corresponding coordinate points in the three-dimensional geological space, Characterize the sedimentation rate value of the researcher's prior knowledge or after geological verification. During the geological structure evolution process, the sedimentation rate is closely related to the slope of the three-dimensional geological interface and can show a non-linear relationship. Therefore, this prior knowledge can be combined to constrain and optimize the shape of the three-dimensional geological interface.

[0036] Based on the above objective function, the error between the three-dimensional geological interface and various types of exploration data can be expressed in the form of an implicit function. By solving the minimization of the above objective function, the reconstructed three-dimensional geological interface can be made close to the true observation value, and thus a high-precision three-dimensional geological interface model can be obtained.

[0037] For the displacement characteristics on both sides of the geological interface position in the first term of the objective function in the implicit modeling optimization method driven by non-linear constraint coupling of the above embodiments of the present invention, the displacement of the geological interface in a homogeneous elastic medium is considered here. The displacement can be calculated by the dislocation theory, and its analytical solution can be determined by the dislocation and the geometric parameters of the geological interface, that is: Equation 5: ; In Equation 5, and are the medium elastic constant tensors, is the distance from the observation point to the geological interface, is the integration variable; it can be seen from this equation that the displacement on both sides of the geological interface has a non-linear relationship with the geometric characteristics of the geological interface. Therefore, the geometric shape of the geological interface can be constrained by this data to make the modeling result more accurate; For the seismic wave velocity data constraint in the second term of the objective function, through the empirical correlation formula, it can be known that the seismic wave velocity and the physical properties on both sides of the geological interface have a non-linear relationship, which can be expressed as: Equation 6: ; In Equation 6, and is an empirical constant. According to the geophysical interface inversion hypothesis, during the iterative reconstruction of the model, constant physical property parameters are set on both sides of the three-dimensional geological interface. By adjusting the position of the geological interface, different physical property parameters are given to the voxels on both sides of the geological interface, so as to generate different forward responses of seismic wave velocity, and observe the forward model that best fits the observed seismic wave velocity value as the optimal model for iterative reconstruction, thereby constraining the three-dimensional geological modeling framework; Regarding the prior knowledge of the third term of the objective function about the sedimentation rate, previous studies have shown that there is a close relationship between the sedimentation rate and the undulation of the geological interface. The sedimentation rate can be obtained from the diffusion equation and the interface slope The relationship between them is as follows: Equation 7: ; In Equation 7, is the diffusion coefficient, is the non-linear exponent. The above formula can show the non-linear relationship between the sedimentation rate and the slope of the three-dimensional geological interface. Therefore, the sedimentation rate obtained from the geological observations and simulation information in the target study area can be added to the three-dimensional geological modeling framework and used to constrain the iterative optimization direction of the model; The non-linear constraint information is closely related to the shape and physical property characteristics of the three-dimensional geological interface. These spatial shape and physical property distribution characteristics can all be characterized by the implicit function of the three-dimensional geological interface. Finally, an association expression between the three-dimensional geological interface and the non-linear constraint information can be established from the perspective of the implicit function, thereby realizing the expression of the objective function for the reconstruction and optimization model; Step 4: To obtain the implicit function expression of the optimized surface, an implicit gradient flow algorithm is proposed to solve the optimal value of in the above objective function. Since the constraint data in the modeling process are all non-linear constraints, the basic principle of Newton's method is borrowed to set the iteration step for the implicit function . Therefore, the minimization solution of the above can be converted into the minimization solution of the iteration step , that is: Equation 8: ; In Equation 8, represents the initial implicit function value of the corresponding point , characterizes the displacement characteristic of the initial observation point and can be represented as a constant, which will be represented by subsequently; To avoid the underdetermination of the objective function during the solution process, a smoothing term is added here to ensure the smoothness of the reconstructed three-dimensional geological interface and ensure the overdetermination of the objective function solution. After adding the smoothing term, the objective function can be expressed as: Equation 9: ; Subsequently, the implicit gradient flow algorithm is used to perform a Taylor expansion of the above objective function to construct a gradient representation of the implicit function of the three-dimensional geological interface: Equation 10: ; In Equations 9 and 10, represents the weight of the smoothing term, , , respectively represent the Jacobian matrices of the geological interface continuity constraint, seismic constraint, and sedimentation rate constraint, which are mathematically expressed as the first derivative form of the corresponding non-linear function expression, and their specific forms can be derived from the non-linear relationship between the above non-linear constraint data and the three-dimensional geological interface. Subsequently, an implicit gradient calculation is performed on the Taylor expansion of the above objective function to solve the non-linear least squares problem of the objective function for each iteration, and finally the expression form of the iteration step size is solved. Since the above non-linear relationships can all be characterized by non-linear analytical expressions, the solved expression form of the iteration step size can also be expressed as an analytical function form. Finally, the reconstructed three-dimensional geological interface can be expressed as an implicit function representation with an iteration step size, that is: Equation 11: ; In Equation 11, represents the number of iterations, represents the model implicit function of the th iteration, represents the step size of the th iteration; For the implicit modeling optimization method driven by non-linear constraint coupling in this embodiment, for the Jacobian matrices corresponding to the geological interface continuity constraint, seismic constraint, and sedimentation rate constraint, an analytical expression of the non-linear relationship between the implicit function and the constraint information can be constructed through the non-linear relationship between the constraint information and the geological interface distribution represented in the specific implementation of Step 4. The analytical expression form of the Jacobian matrix can be obtained through the first derivative of this analytical expression. Subsequently, the Taylor expansion of the objective function ε(f(x)) can be differentiated, and the derivative function is set equal to 0 to obtain the minimum value of the objective function under different non-linear constraints; For the non-linear constraints in the objective function, by taking the derivative, we can obtain: Equation 12: ; In order to realize the analytical function representation of the implicit gradient flow, the DiracDelta function is used here to express the discrete volume elements in its implicit function, that is: Equation 13: ; To find the minimum value of the objective function, by solving the analytical expression of the iteration step size when , a solution form in the form of a linear combination of spatial influence kernel functions can be obtained, that is: Equation 14: ; Step 5. To avoid the iteration process falling into the local minimum of the objective function during the solution process of the objective function, a weighted step size is used for iterative optimization of the objective function, and the weight value of the soft linear search step size is used, that is, during the solution process, the function is expressed as: Equation 15: ; In Equation 15, is the iteration step weight. During the iterative optimization process of the objective function, this weight can be regarded as a constant value, that is: Equation 16: ; This constant will not affect the implicit gradient flow solution algorithm under non - linear constraints. Therefore, the value of this step weight after each iteration will be mainly discussed; For the initial objective function , its form with the iteration step can be written as: Equation 17: ; This function needs to satisfy: ; Among them: ; ; Through the above - mentioned conditions to be satisfied, it is ensured that The value of the objective function needs to decrease effectively each time of iteration, and needs to be large enough to ensure that the step size is large enough to jump out of the local minimum range. Thus, through the objective function value of each iteration, the weight value of the iteration step size is obtained through soft linear search to represent the implicit function expression of the final optimized model; Step 6. For the implicit function expression of the three - dimensional geological interface in each above - mentioned iteration, it can be substituted into the objective function to obtain the corresponding objective function value. Draw the curve of the objective function value during the iteration process. When the oscillation amplitude of the curve is within the preset minimum oscillation amplitude range and the error is less than the error threshold, output the current geological interface model as the reconstructed model.

[0038] Based on the above - mentioned optimization method, this embodiment also provides a computer - readable memory, which contains a computer program for implementing the above - mentioned implicit modeling optimization method for non - linear constraint coupling drive and can be read and executed; This embodiment also provides an implicit modeling optimization system driven by non - linear constraint coupling, including: a programmable logic controller, a processor, and the above - mentioned readable memory; the programmable logic controller includes a database construction module, an initial model construction module, a model optimization and reconstruction module, and an optimal model judgment module; the processor reads and executes the computer program on the readable memory according to the programmable logic controller.

[0039] To more fully illustrate the excellent technical effects of the technical solution of the present invention, the present invention also adopts the non - linear constraint coupling - driven implicit modeling optimization method described in the above - mentioned embodiment to perform three - dimensional modeling on a complex salt dome. During the construction process, for the non - linear constraints of the custom model, a non - linear coupling - driven optimization objective function is constructed, and the implicit gradient flow algorithm is used to solve the objective function. In this process, the solution process is avoided from falling into local minima, and finally a fine three - dimensional geological interface reconstruction model is obtained. As shown in the example, Figure 1 in (a) and (b) are respectively a relatively simple standard geological interface model and the initial input model of the algorithm. Based on the formation displacement constraints, physical property distribution constraints, and slope constraints on both sides of the geological body interface, a non - linear constraint coupling - driven implicit model reconstruction objective function is constructed, and through iterative optimization, the fine reconstruction of the closed geological interface model is realized. The reconstruction result is as Figure 2 shown; in addition, the present invention also performs non - linear constraint coupling - driven reconstruction on the salt dome model to illustrate the applicability of the method of the present invention to the reconstruction of geological interfaces with complex shapes, where Figure 3 in (a) and (b) are respectively the standard model of the salt dome model and the initial input model of the algorithm. From the reconstruction results of the above two example models, it can be seen that the selected initial models are all ideal and uniform simple closed - surface models, such as spheres and ellipsoids. Under the non - linear constraint coupling - driven implicit modeling framework, non - linear data constraints closely related to the shape can be combined to realize the fine reconstruction of any complex and smooth model. For simple geological interface models, Figure 2 the reconstruction results can intuitively show that under the non - linear constraint coupling drive, the reconstructed model returns to the standard model form. In addition, for the salt dome model with complex shape and many local details, in the case where there is no surface shape fluctuation in the initial input model, the reconstructed model can still restore the surface shape details and at the same time consider the smoothness of the salt dome model surface, with high accuracy and reliability.

[0040] Through Figures 1 to 4It can be seen that the implicit modeling optimization method driven by non-linear constraint coupling provided by the present invention realizes the coupling of different types of non-linear constraint information, can be applied to the 3D modeling framework of any type of information, breaks through the limitation that it is difficult for the traditional modeling framework to combine non-linear constraint information, can directly extract the features related to the 3D geological interface form and spatial distribution from multivariate non-linear information, and uses this to constrain the modeling process, enhancing the accuracy and effectiveness of the 3D geological interface reconstruction model.

[0041] The above is the preferred implementation manner of the present invention. It should be pointed out that for those of ordinary skill in the art of the present technology, several improvements and refinements can be made without departing from the principle described in the present invention, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. An implicit modeling optimization method driven by non-linear constraint coupling, characterized in that, Including: Step S1: Obtain the non-linear constraint data of the target geological interface to obtain a modeling constraint database; Step S2: Extract the exploration control points of the target geological interface from the exploration data, and generate an initial implicit function representation of the target geological interface through the Hermite radial basis function; Step S3: According to the data types in the modeling constraint database, obtain a non-linear coupled driving optimization objective function for the implicit function expression of the target geological interface, and give the relationship expression between each non-linear constraint in the objective function and the morphological characteristics of the geological interface; Step S4: According to the principle of Newton's method, obtain a model implicit function with an iteration step size through the implicit gradient flow algorithm, and then obtain the globally optimized target geological interface model through the soft linear search algorithm; Step S5: Substitute the model implicit function of each iteration into the objective function, draw a curve of the change of the objective function value. When the oscillation amplitude of the curve is within the preset minimum oscillation amplitude range and the error is less than the error threshold, output the current geological interface model as the reconstructed model.

2. The implicit modeling optimization method driven by non-linear constraint coupling according to claim 1, characterized in that: The non-linear constraint data includes: geological interface continuity constraint data, lithology spatial distribution constraint data, seismic data reflecting the physical properties of the upper and lower plates of the geological interface, rock physics data, tectonic evolution history constraint data, and sedimentation rate data.

3. The implicit modeling optimization method driven by non - linear constraint coupling according to claim 1, characterized in that: The process of extracting the exploration control points is: extract the three-dimensional geological interface boundary from the plan view, sectional view and borehole record map including the target geological interface, and represent the boundary in the form of spatial coordinate points, that is, obtain.

4. The implicit modeling optimization method with non-linear constraint coupling drive according to claim 3, wherein: The process of obtaining the initial implicit function representation of the target geological interface is: Step S2-1: Characterize the initial implicit function of the geological interface using the HRBF modeling method, that is, in the three-dimensional geological space define the initial implicit function of the three-dimensional interface to characterize the scalar function at any point in the three-dimensional space. Combining the coordinate constraints of exploration control points, express the initial implicit function of any geological interface as: Formula 1: ; Formula 2: ; In Formula 1 and Formula 2, is the number of known points of the geological interface in the three-dimensional geological space, is the linear coefficient expressed by the HRBF function, is the radial basis function, is the exploration control point; Step S2-2: Substitute the exploration control points into the initial implicit function of the geological interface to make it satisfy the classification conditions, and solve the linear coefficients therein , thus obtaining the analytical representation of the initial implicit function of the geological interface, where the classification conditions are as follows: Formula 3: .

5. A non-linear constraint coupling-driven implicit modeling optimization method according to claim 1, characterized in that: The non-linear coupled driving optimization objective function is constructed according to the data in the modeling constraint database and the initial implicit function representation of the target geological interface; The data in the modeling constraint database are continuity constraint data, seismic data and sedimentation rate data.

6. The implicit modeling optimization method with non - linear constraint coupling drive according to claim 5, characterized in that: The expression of the non-linear coupled driving optimization objective function is: Formula 4: ; In Equation 4, is the displacement quantity feature on both sides of the three-dimensional geological interface position, characterizes the number of volume elements on both sides of the three-dimensional interface, represents the spatial volume element coordinates with displacement observed in the current geological exploration, characterizes the spatial coordinates of the above volume elements changing during the iteration process, is the seismic wave at the corresponding position of the three-dimensional geological interface, characterizes the seismic wave velocity value observed at the surface observation point during geophysical exploration, characterizes the number of discrete points in the geological space, expresses the sedimentation rate of the corresponding coordinate points in the three-dimensional geological space, characterizes the sedimentation rate value of the researcher's prior knowledge or after geological verification.

7. An implicit modeling optimization method driven by non-linear constraint coupling according to claim 1, characterized in that: The model implicit function with an iteration step size is the solution expression form of the geological interface implicit function in the objective function, and its obtaining process is: Step S4-1: Set the iteration step for the implicit function , so the above minimization solution for can be converted into the minimization solution for the iteration step , that is: Formula 5: ; Step S4-2: To avoid the underdetermination of the objective function in the solution process, a smoothing term is added here, and its expression is: Formula 6: ; Step S4-3: Adopt the implicit gradient flow algorithm, perform Taylor expansion on the function expression obtained in Step S4-2, and construct a gradient representation of the geological interface implicit function in the objective function, and its expression is: Formula 7: ; Step S4-4: Perform implicit gradient calculation on the gradient representation obtained in Step S4-3, and finally obtain a model implicit function with an iteration step size, that is: Formula 8: ; In Formulas 5 to 8, represents the initial implicit function value of the corresponding point , characterizes the displacement feature of the initial observation point and can be characterized as a constant, which will be represented by subsequently, represents the weight of the smoothing term, , , are the Jacobian matrices characterizing the geological interface continuity constraint, the seismic constraint, and the sedimentation rate constraint, respectively, characterizes the model implicit function of the th iteration, characterizes the step size of the th iteration.

8. A method for implicit modeling optimization with non - linear constraint coupling drive according to claim 1, characterized in that: The process of the globally optimized target geological interface model is: Step S5-1: Adopt a weight value with a soft linear search step size. Let the iteration step size weight be , which can be regarded as a constant value during the iterative optimization of the objective function and is expressed as: Formula 9: ; Step S5-2: For the initial objective function , its form with an iteration step size is as follows: Formula 10: ; Formula 11: ; Formula 12: ; Formula 13: ; Step S5-3: According to Step S5-1 and Step S5-2, through the objective function value of each iteration and the weight value of the iteration step size, that is, represent the model implicit function with an iteration step size in the solution process as: Formula 14: ; In Formulas 9 to 14, is the corresponding update value in each iteration process , is a hyperparameter that needs to satisfy the condition of Formula 12, is a hyperparameter that needs to satisfy the condition of Formula 13.

9. A computer-readable memory comprising a computer program, characterized in that: The computer program can implement the implicit modeling optimization method described in any one of claims 1 to 8.

10. An implicit modeling optimization system driven by non - linear constraint coupling, characterized in that, Including: A programmable logic controller, a processor, and a readable memory as claimed in claim 9; the programmable logic controller includes a database construction module, an initial model construction module, a model optimization and reconstruction module, and an optimal model judgment module; the processor reads and executes the computer program on the readable memory according to the programmable logic controller.

Images