Panel leakage amount prediction method and system suitable for concrete face rockfill dam

Abstract

The present application provides a panel leakage amount prediction method suitable for a concrete face rockfill dam, comprising: S1, counting the first crack parameter information of the panel of the concrete face rockfill dam, and then obtaining the probability distribution information corresponding to the first crack parameter information through a kernel density estimation method; S2, establishing a plurality of optimization water conservancy professional proxy models respectively corresponding to each small block panel of the concrete face rockfill dam; S3, obtaining the leakage amount of each small block panel through the one-to-one corresponding prediction of each optimization water conservancy professional proxy model; S4, summing up the leakage amount of each small block panel obtained in S3, thereby obtaining the total leakage amount of the panel of the concrete face rockfill dam under the condition of existing cracks; the present application also provides a panel leakage amount prediction system suitable for a concrete face rockfill dam. The present application can accurately and efficiently predict the panel leakage amount of the concrete face rockfill dam with panel cracks.

Description

Technical Field

[0001] This invention relates to the field of intelligent analysis technology for concrete panel leakage of rockfill dams, and specifically to a method and system for predicting the leakage of concrete panel rockfill dams. Background Technology

[0002] Concrete-faced rockfill dams have become one of the most widely used dam types in water conservancy and hydropower projects due to their advantages such as simple construction, short construction period, strong seismic resistance, and low cost. However, under the influence of water level, ambient temperature, complex geological conditions, and dam filling quality, the concrete face may experience complex stress-strain states, ultimately leading to face cracks. Face cracks have a significant impact on the integrity and performance of the concrete face structure, greatly reducing its seepage prevention performance. Meanwhile, with the rapid development of digital twin technology in recent years, higher requirements have been placed on finite element calculations of dam seepage under different operating conditions. Rapid calculation and analysis of reservoir dam seepage based on actual engineering data is needed to assess the dam's safety status in real time. Therefore, establishing a proxy model that can accurately and quickly calculate the seepage volume in the face crack area under conditions of face cracks, and predicting the overall face seepage volume of the dam, is of great significance for analyzing the operating status of concrete-faced rockfill dams and establishing a digital twin system.

[0003] For calculating the seepage of concrete-faced rockfill dams with panel cracks, commonly used numerical methods include the finite element method (FEM), finite difference method (FDM), and finite volume method (FVM). These methods, based on Darcy's law and the continuity equation, can simulate complex seepage field distributions by discretizing the computational domain and solving the governing equations. For crack treatment, the equivalent permeability coefficient method and discrete crack models are mainly used. The equivalent permeability coefficient method treats the cracked area of ​​the concrete panel as a high-permeability material, assigning it an equivalent permeability coefficient; the discrete crack model treats the panel cracks as independent entities, modeling them separately and considering their geometric features and permeability characteristics. Although existing methods have made significant progress in seepage analysis, some limitations remain. For example, the complexity of crack geometry, the uncertainty of material properties, and multi-physics coupling limit computational accuracy and efficiency. Furthermore, existing methods are computationally expensive when dealing with large-scale engineering problems, making them difficult to meet practical engineering needs. Therefore, developing an efficient and accurate calculation method for predicting the seepage of concrete-faced rockfill dams with panel cracks has significant engineering implications and application value. Summary of the Invention

[0004] In view of this, it is necessary to propose a method and system for predicting the leakage of concrete-faced rockfill dams, which is applicable to the above-mentioned problems, in order to overcome some shortcomings of the background technology and thus solve the following technical problems:

[0005] How to accurately and efficiently predict the leakage of concrete-faced rockfill dams with panel cracks, so as to provide a precise and reliable data foundation for the digital twin model or operational status analysis of the concrete-faced rockfill dam.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] This invention proposes a method for predicting the leakage of concrete-faced rockfill dam panels. The method involves statistical analysis of each small block of the concrete-faced rockfill dam panel. The method is characterized by the following features:

[0008] Step S1: Statistically analyze the first crack parameter information of the concrete-faced rockfill dam panel, and then obtain the probability distribution information of the corresponding first crack parameter information through a kernel density estimation method; the first crack parameter information includes the total number of cracks in the concrete-faced rockfill dam panel and the crack shape parameters of the concrete-faced rockfill dam panel; the crack shape parameters of the concrete-faced rockfill dam panel include the panel crack length, the equivalent width of the panel crack, and the absolute roughness of the panel crack surface.

[0009] Step S2: Establish multiple optimized hydraulic professional agent models that correspond one-to-one with each small block panel of the concrete-faced rockfill dam; the optimized hydraulic professional agent models are optimized and adjusted by a co-kriging model and the Genghis Khan shark optimizer algorithm to predict the leakage of the small block panels of the concrete-faced rockfill dam.

[0010] Step S3: The leakage amount of each small block panel is predicted one-to-one through the optimized water conservancy professional agent model.

[0011] Step S4 involves summing the leakage amounts of each small panel obtained in step S3 to obtain the total leakage amount of the concrete panel rockfill dam under cracked conditions.

[0012] Furthermore, in step S2, the process of establishing the optimized water conservancy professional agency model includes the following steps:

[0013] Step S21: Based on the probability distribution information obtained in step S1, a Latin hypercube sampling method is used to sample the number of cracks in the small block panels of the concrete-faced rockfill dam, thereby obtaining a first data set including the number of cracks in the small block panels of the concrete-faced rockfill dam.

[0014] Step S22: Based on the total number of cracks in the concrete-faced rockfill dam obtained in step S1, the Latin hypercube sampling method is used to sample the crack shape parameters of the small block panels of the concrete-faced rockfill dam, thereby obtaining a second data set including the crack shape parameters of the small block panels of the concrete-faced rockfill dam; the crack shape parameters of the small block panels include the crack length, the equivalent width of the crack, and the absolute roughness of the crack surface.

[0015] Step S23: Combine the first data set and the second data set to establish a proxy model input parameter dataset;

[0016] Step S24: Based on the proxy model input parameter dataset obtained in step S23, obtain the equivalent permeability coefficient of the small block panel area of ​​the corresponding concrete panel rockfill dam.

[0017] Step S25: Based on the equivalent permeability coefficient obtained in step S23, the leakage of a small block panel of a concrete-faced rockfill dam is calculated using a finite element method.

[0018] Step S26: The number of cracks in the small block panel, the crack shape parameters of the small block panel, and the finite element calculation of the leakage of the small block panel are integrated to obtain integrated data including training and test sets. The integrated data is input into and the co-kriging model is adjusted. Then, the hyperparameters of the adjusted co-kriging model are optimized by a Genghis Khan Shark optimizer algorithm to establish multiple optimized hydraulic professional agent models for predicting the leakage of the small block panel of the concrete-faced rockfill dam.

[0019] Furthermore, step S25 includes the following sub-steps:

[0020] Step S251: Obtain the COMSOL calculation file from the nonlinear finite element analysis software;

[0021] Step S252: Call the COMSOL calculation file to record the finite element calculation leakage of the small block panel of the concrete panel rockfill dam, and merge it with the proxy model input parameter dataset generated in step S23.

[0022] Step S253: Determine whether the number of times COMSOL is called to calculate the file has reached a preset threshold. If the determination is yes, proceed to step S26; otherwise, return to step S252.

[0023] Furthermore, between steps S25 and S26, the method for predicting the leakage of concrete-faced rockfill dam panels also includes:

[0024] Step S2526: The number of cracks in the small block panel obtained in step S21, the crack shape parameters of the small block panel obtained in step S22, and the finite element calculation leakage amount of the small block panel obtained in step S25 are combined into a set of panel rockfill dam small block panel crack parameters - panel rockfill dam small block panel leakage amount data set.

[0025] Step S26 includes the following sub-steps:

[0026] Step S261 involves coordinating the panel crack parameters and leakage data of the panel rockfill dam block obtained in step S2526 into a training set and a test set, thereby obtaining training set and test set data. The training set and test set data include the coordinated second crack parameter information and the finite element calculation leakage of the block panel corresponding to the coordinated second crack parameter information. The second crack parameter information includes the coordinated number of cracks in the block panel and the coordinated crack shape parameters of the block panel.

[0027] Step S262: The second crack parameter information obtained in step S261 is used as the independent variable, and the finite element calculation of the leakage amount of the small block panel corresponding to the second crack parameter information after overall planning is used as the dependent variable. The independent variable and the dependent variable are input into a first co-kriging model to establish an adjusted second co-kriging model for expressing the relationship between the leakage amount of the small block panel with panel cracks and the crack parameters. The root mean square error of the leakage amount between the second co-kriging model and the test set is calculated.

[0028] Step S263: Using the root mean square error obtained in step S262 as the fitness function value, the hyperparameters of the second co-kriging model are optimized several times using the Genghis Khan Shark Optimizer algorithm, thereby obtaining the third co-kriging model with the minimum root mean square error of the test set under the condition of the optimal number of optimizations.

[0029] Further, in step S263, the upper and lower limits of the hyperparameter values ​​of the second co-kriging model are determined.

[0030] Furthermore, step S263 includes the following sub-steps:

[0031] Step S2631: Determine the maximum number of iterations k in the Genghis Khan Shark Optimizer algorithm. max Set the shark population size to S. n ;

[0032] Step S2632: The Genghis Khan Shark Optimizer algorithm is used to iteratively call the relationship between input and output sample data. By solving the optimal solution of an objective function, the optimal hyperparameters of the co-kriging model are determined, thereby obtaining the optimal optimized water conservancy professional agency model.

[0033] Furthermore, the objective function in step S2632 is:

[0034]

[0035] In formula (1), n1 is the number of test set data; h i h represents the value calculated using the finite element method. i 'Calculate values ​​for the proxy model.'

[0036] Furthermore, in step S24, when water exchange is considered to occur no longer along the horizontal direction of the concrete panel, the water flow through a crack of equal width (a) exhibits laminar flow characteristics, as shown in the following formula:

[0037]

[0038] In formulas (2) and (3), v x (y) represents the water flow velocity along the crack length x; p represents the water flow pressure; μ represents the dynamic viscosity coefficient of the water flow; q represents the water flow rate per unit width of the crack; J represents the hydraulic gradient; γ represents the specific weight of water; y represents the longitudinal coordinate value in the coordinate system of water flow in a crack of equal width.

[0039] Formulas (2) and (3) are derived based on the assumption of smooth surface cracks, with the roughness correction coefficient C and the equivalent crack width b. e The calculation method is as follows:

[0040]

[0041] In formulas (4) and (5), b0 and b t Δ represents the inlet and outlet widths of the crack; Δ represents the absolute roughness of the crack surface. The relative roughness of the crack;

[0042] The equivalent permeability coefficient K of a single fracture is calculated using formulas (2) to (5). fe The calculation formula is as follows:

[0043]

[0044] In formula (6), if there are a large number of cracks on the concrete panel, by selecting a small area on the panel, assuming that the length of the area is l, the width is d, the number of cracks in the area is n, the length of the cracks is m, and γ is the unit weight of water; according to the principle that the seepage flow through the cracks in the small area of ​​the panel is equal to the seepage flow through the equivalent intact panel, the permeability coefficient K of the equivalent small area panel can be obtained. y The calculation formula is as follows:

[0045]

[0046] In formula (7), Ci is the roughness correction coefficient for the i-th crack; b ei Let be the equivalent crack width of the i-th crack. Since the length m of the crack on the panel may not be equal to the length l of the selected small-sized region, the formula for calculating the equivalent permeability coefficient for different crack lengths within the small-sized region of the panel is as follows:

[0047]

[0048] In formula (8), m i γ is the length of the i-th crack; γ is the specific gravity of water.

[0049] This invention further proposes a panel leakage prediction system suitable for concrete-faced rockfill dams, comprising:

[0050] The probability distribution information analysis module statistically analyzes the first crack parameter information of the concrete-faced rockfill dam panel, and then obtains the probability distribution information of the corresponding first crack parameter information through a kernel density estimation method. The first crack parameter information includes the total number of cracks in the concrete-faced rockfill dam panel and the crack shape parameters of the concrete-faced rockfill dam panel. The crack shape parameters of the concrete-faced rockfill dam panel include the panel crack length, the equivalent width of the panel crack, and the absolute roughness of the panel crack surface.

[0051] The sampling analysis module is used to obtain a first data set and a second data set, and combine the first data set and the second data set to establish a proxy model input parameter dataset; based on the probability distribution information obtained by the probability distribution information analysis module, a Latin hypercube sampling method is used to sample the number of cracks in the small block panels of the concrete-faced rockfill dam, thereby obtaining a first data set including the number of cracks in the small block panels of the concrete-faced rockfill dam; based on the total number of cracks in the panels of the concrete-faced rockfill dam obtained in step S1, the Latin hypercube sampling method is used to sample the crack shape parameters in the small block panels of the concrete-faced rockfill dam, thereby obtaining a second data set including the crack shape parameters of the small block panels of the concrete-faced rockfill dam; the crack shape parameters of the small block panels include the crack length, the equivalent width of the crack, and the absolute roughness of the crack surface.

[0052] The equivalent permeability coefficient acquisition module is used to obtain the equivalent permeability coefficient of the small block panel area of ​​the corresponding concrete panel rockfill dam based on the proxy model input parameter dataset obtained by the sampling analysis module.

[0053] The finite element analysis module is used to obtain the finite element calculation leakage of a small block panel of a concrete-faced rockfill dam based on the equivalent permeability coefficient obtained by the equivalent permeability coefficient acquisition module and through a finite element method.

[0054] An optimized hydraulic engineering agent model acquisition module is used to integrate the number of cracks in the concrete block panel, the crack shape parameters of the concrete block panel, and the finite element calculation of the leakage of the concrete block panel to obtain integrated data including training and testing sets. This integrated data is input into and the co-kriging model is adjusted. Then, a Genghis Khan shark optimizer algorithm is used to optimize the hyperparameters of the adjusted co-kriging model, thereby establishing multiple optimized hydraulic engineering agent models for predicting the leakage of concrete block panels in rockfill dams. The co-kriging model is used to establish a predictive model for the leakage of concrete block panels in rockfill dams; the Genghis Khan shark optimizer algorithm is used to optimize the hyperparameters of the co-kriging model, thereby improving the predictive accuracy of the co-kriging model.

[0055] The leakage prediction module for small block panels is used to predict the leakage of each small block panel by means of the optimized hydraulic professional agent model obtained by the optimized hydraulic professional agent model acquisition module.

[0056] The total leakage acquisition module is used to sum the leakage amounts of each small block panel obtained by the leakage prediction module of the small block panel, so as to obtain the total leakage amount of the concrete panel rockfill dam panel under the condition of cracks.

[0057] The present invention further proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the panel leakage prediction method for concrete-faced rockfill dams as described in any of the preceding claims.

[0058] The beneficial effects of this invention are as follows:

[0059] This invention can accurately and efficiently predict the leakage of concrete-faced rockfill dams with panel cracks, and can also quickly perform calculation and analysis of reservoir dam seepage based on engineering measurement data to make real-time assessments of the dam's safety status. This provides a precise and reliable data foundation for the digital twin model or operational status analysis of the concrete-faced rockfill dam. This invention greatly improves the accuracy and efficiency of calculating the leakage of concrete-faced rockfill dams with panel cracks, and reduces the calculation cost. Attached Figure Description

[0060] The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the invention and, together with the description, serve to explain the principles of the invention. These drawings are for illustrative purposes only and are not intended to limit the scope of the invention.

[0061] Figure 1 This is a flowchart of the method for predicting the leakage of concrete-faced rockfill dams according to the present invention.

[0062] Figure 2 This invention relates to a structural cross-sectional diagram of a concrete-faced rockfill dam for calculating the equivalent permeability coefficient of the crack region in the concrete panel.

[0063] Figure 3 This invention relates to a comparative structural diagram of two models applicable to concrete-faced rockfill dams;

[0064] Figure 4 This invention relates to a schematic diagram of a three-dimensional simulation model of a concrete-faced rockfill dam under different water head pressures.

[0065] Figure 5 It corresponds Figure 4 The display shows a cross-sectional view of the largest section of the three-dimensional simulation model, used to reflect the seepage calculation results when cracks exist in the panel;

[0066] Figure 6 This invention relates to a fitness function process graph of the Genghis Khan Shark Optimizer algorithm for optimizing a co-Kriging model;

[0067] Figure 7 This invention relates to a comparative analysis diagram of the predicted leakage value and the finite element calculation value based on the hydraulic professional agent model.

[0068] Figure 8 This is a structural principle diagram of a panel leakage prediction system for concrete-faced rockfill dams according to the present invention.

[0069] Explanation of reference numerals in the attached figures:

[0070] 10. Concrete-faced rockfill dam; 101. Face panel; 1011. Crack; F1. Water load; 102. Equivalent concrete face panel; 104. Cracked area; 103. Subbase; 105. Transition layer; 100. Dam body. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be further described clearly and completely below in conjunction with the embodiments of this invention. It should be noted that the described embodiments are merely some embodiments of this invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0072] The terms “first,” “second,” “third,” and “fourth” are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, the use of “first,” “second,” “third,” and “fourth” to designate a feature may explicitly or implicitly include one or more of that feature.

[0073] The following is a detailed description of embodiments of the invention depicted in the accompanying drawings. The embodiments are detailed in order to clearly convey the invention. However, the amount of detail provided is not intended to limit the contemplative variations of the embodiments; rather, it is intended to cover all modifications, equivalents, and substitutions falling within the spirit and scope of the invention as defined by the appended claims.

[0074] In the following description, numerous specific details are set forth in order to provide a thorough understanding of embodiments of the invention. It will be apparent to those skilled in the art that embodiments of the invention may be practiced without some of these specific details.

[0075] Embodiments of the present invention include various steps, which will be described below. These steps may be performed by hardware components or may be contained in machine-executable instructions that can be used by a general-purpose or special-purpose processor programmed with those instructions to perform these steps. Alternatively, the steps may be performed by a combination of hardware, software, and firmware and / or by a human operator.

[0076] The various methods described herein can be practiced by combining one or more machine-readable storage media containing code according to the invention with suitable standard computer hardware to execute the code contained therein. Apparatus for implementing the various embodiments of the invention may include one or more computers (or one or more processors within a single computer) and a storage system containing or having network access to computer programs encoded according to the various methods described herein, and the method steps of the invention may be performed by modules, routines, subroutines, or sub-parts of a computer program product.

[0077] If the specification states that a component or feature "may", "can", "may" include or have the feature, then it is not necessary to include that particular component or feature or have that feature.

[0078] As used in this specification and the following claims, the words “a,” “an,” and “the” have the meaning of plural reference unless the context clearly indicates otherwise. Furthermore, as used in the description herein, unless the context clearly indicates otherwise, “in” has the meaning of both “in…” and “on…”.

[0079] Exemplary embodiments will now be described more fully below with reference to the accompanying drawings, which illustrate exemplary embodiments. These exemplary embodiments are provided for illustrative purposes only and to make the invention thorough and complete, and to fully convey the scope of the invention to those skilled in the art. However, the disclosed invention can be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Various modifications will be apparent to those skilled in the art. The general principles defined herein can be applied to other embodiments and applications without departing from the spirit and scope of the invention. Furthermore, all statements regarding embodiments of the invention and specific examples thereof described herein are intended to cover their structural and functional equivalents. Additionally, these equivalents are intended to include currently known equivalents as well as those developed in the future (i.e., any element developed that performs the same function, regardless of its structure). Moreover, the terminology and wording used are for the purpose of describing exemplary embodiments and should not be considered limiting. Therefore, the invention is to be endowed with the broadest scope, including various substitutions, modifications, and equivalents consistent with the disclosed principles and features. For clarity, details of technical materials known in the art related to this invention have not been described in detail so as not to unnecessarily obscure the invention.

[0080] Therefore, for example, those skilled in the art will understand that schematic diagrams, schematics, illustrations, etc., represent conceptual views or processes embodying the systems and methods of the present invention. The functionality of the various elements shown in the figures can be provided using dedicated hardware and hardware capable of executing the relevant software. Similarly, any switches shown in the figures are merely conceptual. Their functionality can be performed through the operation of program logic, through dedicated logic, through interaction between program control and dedicated logic, or even manually; specific techniques may be chosen by the entity implementing the invention. Those skilled in the art should further understand that the exemplary hardware, software, processes, methods, and / or operating systems described herein are for illustrative purposes and are therefore not intended to be limited to any particular named element.

[0081] Embodiments of the present invention may provide a computer program product that may include a machine-readable storage medium on which instructions are tangibly implemented, which may be used to program a computer (or other electronic device) to perform processing. The terms "machine-readable storage medium" or "computer-readable storage medium" include, but are not limited to, fixed (hardware) drives, magnetic tape, floppy disks, optical discs, optical disc read-only memory (CD-ROM) and magneto-optical discs, semiconductor memories such as ROMs, PROMs, random access memories (RAM), programmable read-only memories (PROMs), erasable PROMs (EPROMs), electrically erasable PROMs (EEPROMs), flash memory, magnetic or optical cards, or other types of media / machine-readable media suitable for storing electronic instructions (e.g., computer programming code, such as software or firmware). Machine-readable media may include non-transitory media in which data can be stored and does not include carrier waves and / or transient electronic signals propagated via wireless or wired connections. Examples of non-transitory media may include, but are not limited to, magnetic disks or magnetic tapes, optical storage media such as compact discs (CDs) or digital universal discs (DVDs), flash memory, memory, or memory devices. Computer program products may include code and / or machine-executable instructions, which may represent any combination of procedures, functions, subroutines, programs, routines, subroutines, modules, software packages, classes, or instructions, data structures, or program statements. Code segments may be coupled to other code segments or hardware circuitry by passing and / or receiving information, data, variables, parameters, or memory contents. Information, variables, parameters, data, etc., may be passed, forwarded, or transmitted by any suitable means, including memory sharing, message passing, token passing, network transmission, etc.

[0082] Furthermore, embodiments can be implemented using hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware, or microcode, program code or code segments (e.g., a computer program product) that perform the necessary tasks can be stored on a machine-readable medium. The processor can then perform the necessary tasks.

[0083] The systems depicted in the figures can be provided in various configurations. In some embodiments, the system can be configured as a distributed system, wherein one or more components of the system are distributed across one or more networks of a cloud computing system.

[0084] Each of the appended claims defines a separate invention, which, for infringement purposes, is considered to include the various elements or limited equivalents specified in the claims. Depending on the context, all references to "invention" below may refer only to certain specific embodiments in some cases. In other cases, it should be recognized that references to "invention" will refer to one or more, but not necessarily all, the subject matter described in the claims.

[0085] Unless otherwise stated herein or the context clearly contradicts it, all methods described herein may be performed in any suitable order. The use of any and all examples or exemplary language (e.g., “such as”) provided with respect to certain embodiments herein is intended only to better illustrate the invention and not to limit the scope of the claimed invention. No language in the specification should be construed as indicating any unclaimed element essential to the implementation of the invention.

[0086] The various terms used herein are as follows. Where no term is defined below as used in the claims, the broadest definition should be given, and those skilled in the art have provided the term as reflected in printed publications and granted patents at the time of filing.

[0087] Example 1

[0088] like Figures 1-7 As shown:

[0089] in Figure 2 The (g) area graphic is used to display cracks in the panels of a concrete-faced rockfill dam. Figure 2 The (a) area graphic is used for local magnification display. Figure 2 A region of the (g) area graphic, Figure 2 The graph in region (b) is used to reflect the equivalent permeability coefficient of the concrete-faced rockfill dam panel; where Figure 3 The (c) section of the graphic is used to display a model of a concrete-faced rockfill dam. Figure 3 The (d) section diagram is used to display the cross-sectional structure of the model of the concrete-faced rockfill dam. Figure 3 The (e) section of the graphic is used to display the overall finite element model of the concrete-faced rockfill dam. Figure 3 The (f) region diagram is used to display the crack model of a local area in the overall finite element model of the concrete-faced rockfill dam; where Figure 4 and Figure 5 The "Water Head" refers to the water head; among which... Figure 6 The horizontal axis represents the number of iterations. Figure 6 The ordinate of is the fitness function value; where Figure 7 The horizontal axis coordinates are the values ​​calculated using the finite element method. Figure 7 The vertical axis coordinate is the value calculated by the surrogate model. Figure 7 The “Scope” refers to the range;

[0090] This embodiment proposes a method for predicting the leakage of concrete-faced rockfill dam panels. Statistical analysis is performed on each small block of the concrete-faced rockfill dam 10. The method for predicting leakage includes:

[0091] Step S1: Statistically analyze the parameter information of a first crack in the panel 101 of the concrete-faced rockfill dam 10, and then obtain the probability distribution information corresponding to the first crack parameter information using a kernel density estimation method. The first crack parameter information includes the total number n of cracks in the panel 101 of the concrete-faced rockfill dam 10 and the crack shape parameters of the panel 101 of the concrete-faced rockfill dam 10. The crack shape parameters of the panel 101 of the concrete-faced rockfill dam 10 include the crack length m and the equivalent width b of the crack. e The absolute roughness Δ of the panel crack surface c ;

[0092] Step S2: Establish multiple optimized hydraulic professional agent models that correspond one-to-one with each small block panel of the concrete-faced rockfill dam 10; the optimized hydraulic professional agent models are optimized and adjusted by a co-Kriging model and the Genghis Khan Shark Optimizer algorithm to predict the leakage of the small block panels of the concrete-faced rockfill dam.

[0093] Step S3: The leakage amount of each small block panel is predicted one-to-one through the optimized water conservancy professional agent model.

[0094] Step S4 involves summing the leakage amounts of each small panel obtained in step S3 to obtain the total leakage amount of the panel 101 of the concrete panel rockfill dam 10 under the condition of cracks.

[0095] After obtaining accurate information on panel cracks, this invention can quickly and accurately calculate the leakage of a concrete panel rockfill dam in a small block panel area, thereby obtaining the total leakage of the concrete panel rockfill dam panel under the condition of cracks, providing strong support for the prediction of leakage of concrete panel rockfill dams and the establishment of a digital twin system.

[0096] Ideally, in step S2, the process of establishing the optimized water conservancy professional agency model includes the following steps:

[0097] Step S21: Based on the probability distribution information obtained in step S1, a Latin hypercube sampling method is used to sample the number of cracks in the small block panels of the concrete-faced rockfill dam 10, thereby obtaining a first data set including the number of cracks in the small block panels of the concrete-faced rockfill dam 10.

[0098] Step S22: Based on the total number of cracks in the panel 101 of the concrete-faced rockfill dam 10 obtained in step S1, the Latin hypercube sampling method is used to sample the crack shape parameters of the small block panels of the concrete-faced rockfill dam 10, thereby obtaining a second data set including the crack shape parameters of the small block panels of the concrete-faced rockfill dam 10; the crack shape parameters of the small block panels include the crack length of the small block panels, the equivalent width of the cracks in the small block panels, and the absolute roughness of the crack surface in the small block panels.

[0099] Step S23: Combine the first data set and the second data set to establish a proxy model input parameter dataset;

[0100] Step S24: Based on the proxy model input parameter dataset obtained in step S23, obtain the equivalent permeability coefficient of the small block panel area of ​​the corresponding concrete panel rockfill dam 10.

[0101] Step S25: Based on the equivalent permeability coefficient obtained in step S23, the leakage of the small block panel of the concrete-faced rockfill dam 10 is calculated by finite element method.

[0102] Step S26: The number of cracks in the small block panel, the crack shape parameters of the small block panel, and the finite element calculation of the leakage of the small block panel are integrated to obtain integrated data including training and test sets. The integrated data is input into and the co-kriging model is adjusted. Then, the hyperparameters of the adjusted co-kriging model are optimized by a Genghis Khan Shark optimizer algorithm to establish multiple optimized hydraulic professional agent models for predicting the leakage of the small block panel of the concrete-faced rockfill dam.

[0103] Optimally, step S25 includes the following sub-steps:

[0104] Step S251: Obtain the COMSOL calculation file from the nonlinear finite element analysis software;

[0105] Step S252: Call the COMSOL calculation file to record the finite element calculation leakage of the small block panel of the concrete panel rockfill dam 10, and merge it with the proxy model input parameter dataset generated in step S23.

[0106] Step S253: Determine whether the number of times COMSOL is called to calculate the file has reached a preset threshold. If the determination is yes, proceed to step S26; otherwise, return to step S252.

[0107] Specifically, the COMSOL calculation file covers the entire process of model building and calculation. The COMSOL calculation file includes preprocessing information such as dam model, dam body partitions, material parameter information, and boundary condition information.

[0108] Optimally, between steps S25 and S26, the method for predicting the leakage of concrete-faced rockfill dam panels further includes:

[0109] Step S2526: The number of cracks in the small block panel obtained in step S21, the crack shape parameters of the small block panel obtained in step S22, and the finite element calculation leakage amount of the small block panel obtained in step S25 are combined into a set of panel rockfill dam small block panel crack parameters - panel rockfill dam small block panel leakage amount data set.

[0110] Step S26 includes the following sub-steps:

[0111] Step S261 involves coordinating the panel crack parameters and leakage data of the panel rockfill dam block obtained in step S2526 into a training set and a test set, thereby obtaining training set and test set data. The training set and test set data include the coordinated second crack parameter information and the finite element calculation leakage of the block panel corresponding to the coordinated second crack parameter information. The second crack parameter information includes the coordinated number of cracks in the block panel and the coordinated crack shape parameters of the block panel.

[0112] Step S262: The second crack parameter information obtained in step S261 is used as the independent variable, and the finite element calculation of the leakage amount of the small block panel corresponding to the second crack parameter information after overall planning is used as the dependent variable. The independent variable and the dependent variable are input into a first co-kriging model to establish an adjusted second co-kriging model for expressing the relationship between the leakage amount of the small block panel with panel cracks and the crack parameters. The root mean square error of the leakage amount between the second co-kriging model and the test set is calculated.

[0113] Step S263: Using the root mean square error obtained in step S262 as the fitness function value, the hyperparameters of the second co-kriging model are optimized several times using the Genghis Khan Shark Optimizer algorithm, thereby obtaining the third co-kriging model with the minimum root mean square error of the test set under the condition of the optimal number of optimizations.

[0114] Specifically, in step S2, the leakage of the small block panel is recorded by calling the COMSOL calculation file. Combined with the generated panel crack parameters, the panel crack parameter-leakage dataset is obtained. Then, the co-kriging method is used to construct a surrogate model for calculating the panel crack parameter-leakage flow. The hyperparameters of the surrogate model are tuned by the Genghis Khan Sand Shark optimizer to obtain the optimal fitting model. This not only allows for the rapid and reliable acquisition of the optimal surrogate model, but also ensures the accuracy of subsequent calculations of the leakage of concrete panel rockfill dams with panel cracks.

[0115] Ideally, in step S263, the upper and lower limits of the hyperparameter θ of the second co-kriging model are determined.

[0116] Optimally, step S263 includes the following sub-steps:

[0117] Step S2631: Determine the maximum number of iterations k in the Genghis Khan Shark Optimizer algorithm. max Set the shark population size to S. n ;

[0118] Step S2632: The Genghis Khan Shark Optimizer algorithm is used to iteratively call the relationship between input and output sample data. By solving the optimal solution of an objective function, the optimal hyperparameters of the co-kriging model are determined, thereby obtaining the optimal optimized water conservancy professional agency model.

[0119] Ideally, the objective function in step S2632 is:

[0120]

[0121] In formula (1), n1 is the number of test set data; h i h represents the value calculated using the finite element method. i 'Calculate values ​​for the proxy model.'

[0122] Optimally, in step S24, when water exchange is considered to occur no longer along the horizontal direction of the concrete panel 101, the water flow through a crack of equal width with length a and width b exhibits laminar flow characteristics, and the water flow characteristics within the crack are as follows:

[0123]

[0124]

[0125] In formulas (2) and (3), v x (y) represents the water flow velocity along the crack length x; p represents the water flow pressure; μ represents the dynamic viscosity coefficient of the water flow; q represents the water flow rate per unit width of the crack; J represents the hydraulic gradient; γ represents the specific weight of water; y represents the longitudinal coordinate value in the coordinate system of water flow in a crack of equal width.

[0126] Formulas (2) and (3) are derived based on the assumption of smooth surface cracks, with the roughness correction coefficient C and the equivalent crack width b. e The calculation method is as follows:

[0127]

[0128] In formulas (4) and (5), b0 and b tΔ represents the inlet and outlet widths of the crack; Δ represents the absolute roughness of the crack surface. The relative roughness of the crack is used; however, in actual engineering, the cracks penetrate the concrete panel horizontally, and the crack width may change along the direction of water flow. The crack surface has rough undulations. Therefore, the equivalent crack width and roughness correction method is used to correct the water flow characteristics in the actual crack.

[0129] The equivalent permeability coefficient K of a single fracture is calculated using formulas (2) to (5). fe The calculation formula is as follows:

[0130]

[0131] In formula (6), if there are a large number of cracks on the concrete panel 101, a small area on panel 101 is selected. Assuming the length of this area is l, the width is d, the number of cracks in the area is n, the length of the cracks is m, and γ is the unit weight of water; based on the principle that the seepage flow through the cracks in the small area panel is equal to the seepage flow through the equivalent intact panel, the permeability coefficient K of the equivalent small area panel can be obtained. y The calculation formula is as follows:

[0132]

[0133] In formula (7), Ci is the roughness correction coefficient for the i-th crack; b ei Let m be the equivalent crack width of the i-th crack. Since the length m of the crack on panel 101 may not be equal to the length l of the selected small-sized region, the formula for calculating the equivalent permeability coefficient for different crack lengths within the small-sized region of panel 101 is as follows:

[0134]

[0135] In formula (8), m i γ is the length of the i-th crack; γ is the specific gravity of water.

[0136] Example 2

[0137] like Figures 2-8 As shown:

[0138] This embodiment proposes a panel leakage prediction system suitable for concrete-faced rockfill dams, including:

[0139] The probability distribution information analysis module statistically analyzes the parameter information of a first crack in the panel 101 of the concrete-faced rockfill dam 10, and then obtains the probability distribution information of the corresponding first crack parameter information through a kernel density estimation method. The first crack parameter information includes the total number of cracks in the panel 101 of the concrete-faced rockfill dam 10 and the crack shape parameters of the panel 101 of the concrete-faced rockfill dam 10. The crack shape parameters of the panel 101 of the concrete-faced rockfill dam 10 include the panel crack length, the equivalent width of the panel crack, and the absolute roughness of the panel crack surface.

[0140] The sampling analysis module is used to obtain a first data set and a second data set, and combine the first data set and the second data set to establish a proxy model input parameter dataset; based on the probability distribution information obtained by the probability distribution information analysis module, a Latin hypercube sampling method is used to sample the number of cracks in the small block panels of the concrete-faced rockfill dam 10, thereby obtaining a first data set including the number of cracks in the small block panels of the concrete-faced rockfill dam 10; based on the total number of cracks in the panel 101 of the concrete-faced rockfill dam 10 obtained in step S1, the Latin hypercube sampling method is used to sample the crack shape parameters in the small block panels of the concrete-faced rockfill dam 10, thereby obtaining a second data set including the crack shape parameters of the small block panels of the concrete-faced rockfill dam 10; the crack shape parameters of the small block panels include the crack length, the equivalent width of the crack, and the absolute roughness of the crack surface.

[0141] The equivalent permeability coefficient acquisition module is used to obtain the equivalent permeability coefficient of the small block panel area of ​​the concrete panel rockfill dam 10 based on the proxy model input parameter dataset obtained by the sampling analysis module.

[0142] The finite element analysis module is used to obtain the finite element calculation leakage of the small block panel of the concrete-faced rockfill dam 10 based on the equivalent permeability coefficient obtained by the equivalent permeability coefficient acquisition module and through a finite element method.

[0143] An optimized hydraulic engineering agent model acquisition module is used to integrate the number of cracks in the concrete block panel, the crack shape parameters of the concrete block panel, and the finite element calculation of the leakage of the concrete block panel to obtain integrated data including training and testing sets. This integrated data is input into and the co-kriging model is adjusted. Then, a Genghis Khan shark optimizer algorithm is used to optimize the hyperparameters of the adjusted co-kriging model, thereby establishing multiple optimized hydraulic engineering agent models for predicting the leakage of concrete block panels in rockfill dams. The co-kriging model is used to establish a predictive model for the leakage of concrete block panels in rockfill dams; the Genghis Khan shark optimizer algorithm is used to optimize the hyperparameters of the co-kriging model, thereby improving the predictive accuracy of the co-kriging model.

[0144] The leakage prediction module for small block panels is used to predict the leakage of each small block panel by means of the optimized hydraulic professional agent model obtained by the optimized hydraulic professional agent model acquisition module.

[0145] The total leakage acquisition module is used to sum the leakage amounts of each small block panel obtained by the leakage prediction module of the small block panel, so as to obtain the total leakage amount of the concrete panel rockfill dam panel under the condition of cracks.

[0146] In a further optimized manner, each software module of the panel leakage prediction system for a concrete-faced rockfill dam in this embodiment performs the steps of the panel leakage prediction method for a concrete-faced rockfill dam 10 as described in any one of Embodiment 1.

[0147] Example 3

[0148] This embodiment also proposes a computer-readable storage medium storing a computer program, characterized in that: when the computer program is executed by a processor, it implements the steps of the panel leakage prediction method applicable to concrete panel rockfill dam 10 as described in any one of Embodiments 1.

[0149] Example 4

[0150] Example 4 is a further optimized design of Example 1;

[0151] In step S1, basic information on the number of panel cracks and the range of crack morphology parameters is determined by summarizing a large number of literature and engineering examples.

[0152] In step S21, the Latin hypercube sampling method is used to generate a combination of crack quantity parameters that conforms to the number of panels in the small block; according to the sum of crack quantity parameters, the Latin hypercube sampling method is used to generate a specified number of parameter datasets.

[0153] In step S252, the COMSOL software is called to calculate the file and record the calculation results of the leakage amount of the small block panel and the leakage amount of the overall panel area.

[0154] In step S261, 80% of the data set of panel crack parameters and finite element calculation of seepage flow of panel rockfill dam block is set as the training set of the co-kriging model, and 20% of the data set of panel crack parameters and finite element calculation of seepage flow of panel rockfill dam block is set as the test set of the model, so as to complete the division of the data set of panel crack parameters and finite element calculation of seepage flow of panel rockfill dam block.

[0155] In step S262, the relationship between input and output sample data is established through the co-kriging algorithm to construct a proxy model for crack parameters and leakage of small block panels, and the root mean square error between the predicted value of the proxy model and the calculated value by the finite element method is calculated.

[0156] In step S263, using the root mean square error calculated in step S262 as the fitness function value, the Genghis Khan Shark Optimizer algorithm is used to optimize the selection of hyperparameters for the co-kriging model. Figure 6 Specifically, this involves the fitness function value change curve during the optimization process; after reaching the maximum number of optimization iterations, a proxy model for calculating the leakage of concrete-faced rockfill dam panels in small blocks based on the co-kriging model is constructed. The prediction results of the proxy model are shown below. Figure 7 Finally, the predicted leakage of concrete-faced rockfill dams under the condition of panel cracks was calculated.

[0157] Example 5

[0158] Example 5 is a further optimized design of Example 1;

[0159] In step S2, the number of calculations in step S2 is set to W. When constructing the water conservancy professional agent model, it is necessary to establish a dataset for training the model. Here, the number of calculations W is set to the number of data groups in the training dataset; preferably, W is a value of 1000.

[0160] In step S21, the Latin hypercube sampling method is used to sample the number of cracks in the panel. The number of samples is the number of concrete panel rockfill dam blocks, and the total number of cracks is N.

[0161] In step S22, the Latin hypercube sampling method is used to sample the crack shape parameters, generating N combinations of crack shape parameters, and generating a dataset S = {S1, S2, S3, ..., S...}. N}, where the i-th group of data in dataset S is S_i. ii = 1, 2, 3, ..., N Where, m i Let be the length of the i-th crack. Let Δ be the equivalent width of the i-th crack. c Let be the relative roughness of the i-th crack.

[0162] Specifically, dataset S is selected. * 80% of the data sets were used as the training set, and dataset S was selected. * 20% of the data sets were used as the test set.

[0163] In step S26, the upper limit range of the hyperparameter θ is set to [1e-7, 1] and the lower limit range is set to [1e2, 1e4]. The maximum number of iterations k is set to [1e-7, 1]. max The number of trials is set to 200, and the shark population size is Sn = 50.

[0164] Example 6

[0165] Example 6 is a further optimized design of Example 5;

[0166] In step S24, according to data group S i The equivalent permeability coefficient set D of the panel crack region was calculated. i ;

[0167] In step S25, the equivalent permeability coefficient set D of the calculated panel crack region is... i The data was input into the nonlinear finite element analysis software COMSOL for numerical calculation. The COMSOL calculation file was obtained, and the leakage of the small block panel was extracted from the calculation file. The total leakage of the dam panel was then calculated.

[0168] In step S263, the root mean square error obtained in step S262 is used as the fitness function value. The hyperparameters of the co-kriging model are optimized using the Genghis Khan Shark Optimizer algorithm. After reaching the maximum number of iterations, a hydraulic professional agent model for calculating the leakage in the cracked area of ​​the concrete-faced rockfill dam is established. Finally, the predicted leakage values ​​of each small block are summed to realize the prediction of leakage in the concrete-faced rockfill dam under the condition of cracked panels.

[0169] Specifically, determine the maximum number of iterations k in the Genghis Khan shark optimizer algorithm. max Set the shark population size to S. n The Genghis Khan Shark Optimizer algorithm is used to iteratively call the relationship between input and output sample data. By solving the optimal solution of the objective function, the optimal hyperparameters of the surrogate model are determined, thus obtaining the optimal surrogate model.

[0170] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for predicting the amount of leakage of a concrete face slab of a concrete face rockfill dam, characterized by statistically analyzing each small block face slab of the concrete face rockfill dam, respectively. The panel leakage amount prediction method comprises: Step S1, a first crack parameter information of a concrete face rockfill dam panel is counted, and then a probability distribution information corresponding to the first crack parameter information is obtained by a kernel density estimation method; the first crack parameter information comprises a total number of cracks of the concrete face rockfill dam panel and a crack shape parameter of the concrete face rockfill dam panel; the crack shape parameter of the concrete face rockfill dam panel comprises a panel crack length, an equivalent width of the panel crack and an absolute roughness of the panel crack surface; Step S2, a plurality of optimized water conservancy professional proxy models respectively corresponding to each small block panel of the concrete face rockfill dam are established; the optimized water conservancy professional proxy model is optimized and adjusted by a collaborative Kriging model and a Genghis Khan shark optimizer algorithm, and is used for predicting a leakage amount of the small block panel of the concrete face rockfill dam; Step S3, the leakage amount of each small block panel is obtained by one-to-one corresponding prediction through each optimized water conservancy professional proxy model; Step S4, the leakage amount of each small block panel obtained in step S3 is summed, so as to obtain a total leakage amount of the concrete face rockfill dam panel under the condition of existing cracks; In step S2, the establishment process of the optimized water conservancy professional proxy model comprises the following steps: Step S21, according to the probability distribution information obtained in step S1 and by using a Latin hypercube sampling method, the small block panel of the concrete face rockfill dam is sampled with respect to the crack number, so as to obtain a first data set comprising the crack number of the small block panel of the concrete face rockfill dam; Step S22, according to the total number of cracks of the concrete face rockfill dam panel counted in step S1 and by using the Latin hypercube sampling method, the small block panel of the concrete face rockfill dam is sampled with respect to the crack shape parameter, so as to obtain a second data set comprising the crack shape parameter of the small block panel of the concrete face rockfill dam; Step S23, the first data set and the second data set are combined, so as to establish a proxy model input parameter data set; Step S24, according to the proxy model input parameter data set obtained in step S23, an equivalent permeability coefficient corresponding to the small block panel area of the concrete face rockfill dam is obtained; Step S25, according to the equivalent permeability coefficient obtained in step S23, a finite element calculation leakage amount of the small block panel of the concrete face rockfill dam is obtained by a finite element method; Step S26, the crack number of the small block panel, the crack shape parameter of the small block panel and the finite element calculation leakage amount of the small block panel are integrated to obtain integrated data comprising a training set and a test set, the integrated data is input and adjusted into a collaborative Kriging model, and then a Genghis Khan shark optimizer algorithm is used to optimize the hyperparameters of the adjusted collaborative Kriging model, so as to establish a plurality of optimized water conservancy professional proxy models for predicting the leakage amount of the small block panel of the concrete face rockfill dam.

2. The method for predicting the amount of seepage of a face slab of a concrete face rockfill dam according to claim 1, characterized by, Step S25 comprises the following substeps: Step S251, a COMSOL calculation file of a nonlinear finite element analysis software is obtained; Step S252, calling the COMSOL calculation file records the finite element calculation leakage of the small block panel of the concrete face rockfill dam, and merges with the proxy model input parameter data set generated in step S23 for processing; Step S253, judging whether the number of times of calling the COMSOL calculation file reaches the preset threshold, if yes, executing step S26, otherwise returning to execute step S252.

3. The method for predicting the amount of seepage of a face slab of a concrete face rockfill dam according to claim 1 or 2, characterized in that, Between steps S25 and S26, the panel leakage prediction method suitable for the concrete face rockfill dam further comprises: Step S2526, merging the small block panel crack number obtained in step S21, the small block panel crack shape parameter obtained in step S22 and the finite element calculation leakage of the small block panel obtained in step S25 into a set of small block panel crack parameter-small block panel leakage data set of the panel rockfill dam; Step S26 comprises the following sub-steps: Step S261, integrating the small block panel crack parameter-small block panel leakage data set of the panel rockfill dam obtained in step S2526 into a training set and a test set, thereby obtaining training set and test set data; the training set and test set data include a second crack parameter information after integration and the finite element calculation leakage of the small block panel corresponding to the second crack parameter information after integration; the second crack parameter information includes the small block panel crack number after integration and the small block panel crack shape parameter after integration; Step S262, taking the second crack parameter information obtained in step S261 as the independent variable, taking the finite element calculation leakage of the small block panel corresponding to the second crack parameter information after integration as the dependent variable, inputting the independent variable and the dependent variable into a first co-Kriging model, thereby establishing a second co-Kriging model for expressing the relationship between the small block panel leakage and the crack parameter of the panel rockfill dam with the crack, and calculating the root mean square error of the second co-Kriging model with the test set about the leakage; Step S263, taking the root mean square error obtained in step S262 as the fitness function value, and using the Genghis Khan shark optimizer algorithm to optimize the hyperparameters of the second co-Kriging model for several times, thereby obtaining a third co-Kriging model with the minimum test set root mean square error under the optimal optimization times.

4. The method for predicting the amount of seepage of a face slab of a concrete face rockfill dam according to claim 3, characterized by, In step S263, the upper limit and lower limit of the hyperparameter value of the second co-Kriging model are determined.

5. The method for predicting the amount of seepage of a face slab of a concrete face rockfill dam according to claim 3, characterized by, Step S263 comprises the following sub-steps: Step S2631, determine the maximum number of iterations in the Genghis Shark optimizer algorithm , set the number of shark population as S n ; Step S2632, using the Genghis Khan shark optimizer algorithm to continuously iterate the relationship between the input and output sample data, determining the optimal hyperparameters of the co-Kriging model by solving the optimal solution of an objective function, thereby obtaining the optimal optimization water conservancy professional proxy model.

6. The method for predicting the amount of seepage of a face slab of a concrete face rockfill dam according to claim 5, characterized in that, The objective function in step S2632 is: （1） In equation (1), is the number of data in the test set; is the finite element calculation value; is the agent model calculation value.

7. The method for predicting the amount of seepage of a face slab of a concrete face rockfill dam according to claim 3, characterized by, In step S24, when it is considered that no water exchange occurs along the horizontal direction of the concrete panel, for an equal-width crack with a length of a and a width of b, the water flow through the crack presents a laminar flow characteristic, and the water flow characteristics in the crack are as follows: (2) (3) In formula (2) and formula (3), is the water flow velocity in the length direction of the crack; p is the fluid pressure of the water flow; is the dynamic viscosity coefficient of the water flow; q is the water flow rate per unit width of the crack; is the hydraulic gradient; is the unit weight of water; y is the longitudinal coordinate value in the coordinate system of the equal-width crack water flow; Equations (2) and (3) are derived based on the assumption of surface-smooth cracks, and the calculation method of the roughness correction coefficient C and the equivalent crack width b e is as follows: (4) (5) In Equations (4) and (5), b0and b t is the crack entrance width and exit width; is the absolute roughness of the crack surface; is the relative roughness of the crack; The equivalent permeability coefficient of single fracture is calculated by formula (2)-(5) The calculation formula is as follows: (6) In formula (6), if there are a large number of cracks on the concrete panel, a small size area on the panel is selected, it is assumed that the length of the area is l, the width is d, the number of cracks existing in the area is n, and the length of the crack is m; according to the principle that the seepage flow through the small size area panel crack is equal to the seepage flow through the equivalent complete panel, the seepage coefficient of the small area panel after the equivalent can be obtained The calculation formula is as follows: (7) In formula (7), Ci is the roughness correction factor of the ith crack; is the equivalent crack width of the ith crack. Since the length m of the crack on the panel can not be equal to the length l of the selected small-size region, the equivalent permeability coefficient calculation formula for different panel crack lengths in the small-size region of the panel is as follows: (8) In equation (8), L is the length of the ith crack.

8. A system for predicting the amount of seepage from a concrete face panel suitable for use in a concrete face rockfill dam, characterized by, The panel leakage amount prediction system performs the panel leakage amount prediction method for a concrete face rockfill dam according to any one of claims 1 to 7, and the panel leakage amount prediction system comprises: a probability distribution information analysis module, which is configured to statistically acquire a first crack parameter information of a panel of the concrete face rockfill dam, and then acquire probability distribution information corresponding to the first crack parameter information by using a kernel density estimation method; a sampling analysis module, which is configured to acquire the first data set and the second data set, combine the first data set and the second data set to establish an agent model input parameter data set, sample the small block panels of the concrete face rockfill dam with respect to the crack number according to the probability distribution information acquired by the probability distribution information analysis module and by using a Latin hypercube sampling method, so as to acquire the first data set including the crack number of the small block panels of the concrete face rockfill dam, and sample the small block panels of the concrete face rockfill dam with respect to the crack shape parameter according to the total number of cracks of the panel of the concrete face rockfill dam statistically acquired in step S1 and by using the Latin hypercube sampling method, so as to acquire the second data set including the crack shape parameter of the small block panels of the concrete face rockfill dam; an equivalent permeability coefficient acquisition module, which is configured to acquire the equivalent permeability coefficient corresponding to the small block panel area of the concrete face rockfill dam according to the agent model input parameter data set acquired by the sampling analysis module; a finite element analysis module, which is configured to acquire the finite element calculation leakage amount of the small block panels of the concrete face rockfill dam according to the equivalent permeability coefficient acquired by the equivalent permeability coefficient acquisition module and by using a finite element method; an optimized water conservancy professional agent model acquisition module, which is configured to integrate the crack number of the small block panels, the crack shape parameter of the small block panels and the finite element calculation leakage amount of the small block panels to acquire integrated data including a training set and a test set, input the integrated data into a co-Kriging model and adjust the co-Kriging model, and then optimize the hyperparameters of the adjusted co-Kriging model by using a Gisotto Shark optimizer algorithm, so as to establish a plurality of optimized water conservancy professional agent models for predicting the leakage amount of the small block panels of the concrete face rockfill dam; the co-Kriging model is used to establish a prediction model of the leakage amount of the small block panels of the concrete face rockfill dam; and the Gisotto Shark optimizer algorithm is used to optimize the hyperparameters of the co-Kriging model, so as to improve the prediction accuracy of the co-Kriging model; a small block panel leakage amount prediction module, which is configured to correspondingly predict the leakage amount of each small block panel by using each of the optimized water conservancy professional agent models acquired by the optimized water conservancy professional agent model acquisition module; a total leakage amount acquisition module, which is configured to sum the leakage amounts of the small block panels acquired by the small block panel leakage amount prediction module, so as to acquire the total leakage amount of the panel of the concrete face rockfill dam under the condition of existing cracks.

9. A computer readable storage medium, the computer readable storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the steps of the panel leakage amount prediction method for a concrete face rockfill dam according to any one of claims 1 to 7.

Images