Foundation surface defect identification and evaluation method, system and equipment based on monitoring data and medium

Abstract

The invention discloses a foundation surface defect identification and evaluation method, system, equipment and medium based on monitoring data, and belongs to the technical field of hydraulic structure nondestructive testing and operation period health diagnose.The method comprises the steps that an interface model of a foundation surface is constructed, the foundation surface is modeled into a two-dimensional discontinuous interface, and an interface constitutive equation and a seepage model are explicitly introduced; collecting operation monitoring data, performing feature extraction on the operation monitoring data, and establishing bidirectional mapping from defect parameters to monitoring responses; and identifying foundation surface defect parameters through reverse learning, substituting an identification result back to the forward model, and outputting engineering evaluation indexes of foundation surface levels. According to the method, the identifiability of the defect position, the scale and the interface force-water property is remarkably improved, distortion caused by indirect inference only based on overall measurement is avoided, the robustness is kept under noise disturbance and working condition fluctuation, and the result has uncertainty definition and engineering interpretability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of non-destructive testing and health diagnosis of hydraulic structures during operation, and particularly relates to a base surface defect identification and evaluation method, system, device and medium based on monitoring data. BACKGROUND

[0002] Existing methods mainly use drilling, geophysical exploration or baseline statistical threshold discrimination to identify base surface abnormalities. The first two methods have limitations in spatial coverage and consistency of operating conditions, and it is difficult to convert the contact state, joint porosity and permeation channel geometry of the base surface into a function relationship of the operating monitoring quantity. The latter relies on a single index threshold, making it difficult to complete the reversible mapping of "monitoring quantity to base surface defect parameter". Some research uses three-dimensional finite elements to analyze the "dam-base whole" scenario, but if the base surface is not considered as an independent force-water coupled discontinuous interface, it is difficult to simultaneously depict the opening and closing of the contact, the in-plane water conduction, the out-of-plane seepage, and the resulting uplift pressure redistribution, thereby being limited in terms of identifiability and engineering usability. The above shortcomings directly lead to large errors in the reconstruction of uplift pressure distribution and large uncertainties in defect positioning, making it difficult to form a reviewable safety criterion at the base surface level. The present application models and parameterizes the base surface as the first object, and cooperates with the interface selection and learning-inversion-back substitution closed loop of the operating monitoring quantity to provide a solution to the above shortcomings. SUMMARY

[0003] In view of the above problems, the present application is proposed.

[0004] Therefore, the present application aims to model and parameterize the base surface as the first object, cooperate with the interface selection and learning-inversion-back substitution closed loop of the operating monitoring quantity, and provide a solution to the above shortcomings.

[0005] To solve the above technical problems, the present application provides the following technical scheme: a base surface defect identification and evaluation method based on monitoring data, which comprises, An interface model of the base surface is constructed, the base surface is modeled as a two-dimensional discontinuous interface, and an interface constitutive equation and a seepage model are explicitly introduced; operating monitoring data are collected and feature extraction is performed on the operating monitoring data to establish a bidirectional mapping of defect parameters to monitoring responses; the base surface defect parameters are identified through reverse learning, and the identification results are substituted back into the forward model to output engineering evaluation indexes at the base surface level.

[0006] As a preferred solution of the base surface defect identification and evaluation method based on monitoring data, the interface model of the base surface comprises: the base surface is modeled as a two-dimensional discontinuous interface, and the control equations in the dam body and the bedrock body use classical balance equations and Biot consolidation theory; The interface constitutive equation is explicitly introduced at the foundation surface, the in-plane seepage at the foundation surface is described by the seepage model, and the foundation surface level is given priority constraints.

[0007] As a preferred embodiment of the foundation surface defect identification and evaluation method based on monitoring data described in this invention, the interface model of the foundation surface further includes uniform parameterization of the hidden defects of the foundation surface. Defect patches are defined on the base surface using level sets and smoothly embedded on the interface parameters using regularization functions.

[0008] As a preferred embodiment of the foundation surface defect identification and assessment method based on monitoring data described in this invention, the collected operational monitoring data includes foundation surface uplift pressure, corridor segment flow rate, drainage hole outflow, and near-foundation displacement. The feature extraction includes preserving the principal components of the uplift pressure distribution on the foundation surface, the phase characteristics of the uplift pressure gradient and the flow rate-head along the surface.

[0009] The preferred technical solution in the embodiments of the present invention has the following advantages: by modeling the foundation surface as a two-dimensional discontinuous interface and explicitly introducing the interface constitutive equation and seepage model, accurate simulation of the mechanics and seepage behavior of the foundation surface is achieved, thereby improving the accuracy of defect identification and computational efficiency.

[0010] As a preferred embodiment of the foundation surface defect identification and evaluation method based on monitoring data described in this invention, the step of identifying foundation surface defect parameters through reverse learning includes generating a sample library consisting of foundation surface parameter-monitoring response pairs; Sampling in the parameter space using Latin hypercubes includes defect centers on the basal plane. Surface coordinates, patch size and direction, interface opening Method, tangential interface stiffness External leakage coefficient The parameters were obtained and subjected to batch forward modeling under representative water level historical conditions to form a large-scale coverage. gather; Reverse identification uses a probabilistic regression model for output. The training objective is to obtain the Gaussian negative log-likelihood plus a regularization term, while also introducing a forward consistency penalty term: Defect parameter field Substituting back into the forward model constrains the reproduction error of the basal surface observation features, performs closed-loop correction at the basal surface level, and simultaneously provides a physical inversion alternative. Using the weighted residuals of the basal surface observations as the objective function, and combining this with the adjoint sensitivity of the discrete equations, the model is directly solved. Two paths are used to perform bidirectional mapping at the foundation plane level, which are mutually verified.

[0011] The beneficial effects of the preferred technical solutions in the embodiments of the present application are that through the probability regression model and the forward consistency check, bidirectional mapping and closed-loop correction of defect parameters are realized, thereby improving the reliability of defect parameter identification and the physical rationality of inversion results.

[0012] As a preferred scheme of the foundation surface defect identification and evaluation method based on monitoring data, the output engineering evaluation index of the foundation surface level comprises identification result back substitution review and engineering evaluation strict limitation in the foundation surface level output. Calculate the foundation surface anti-sliding safety factor : Wherein, is the foundation surface area, is the microelement area on the foundation surface, is the equivalent cohesion, is the internal friction angle, , is the interface normal and tangential traction under the review working condition, is the tangential traction module; The uplift pressure amplification coefficient is defined as: Wherein, is the pore pressure at the foundation surface position under the defect parameter field identified, is the pore pressure at the foundation surface position under the defect parameter field identified.

[0013] As a preferred scheme of the foundation surface defect identification and evaluation method based on monitoring data, the output engineering evaluation index of the foundation surface level further comprises the average water pressure uplift amplitude of the foundation surface caused by defects measured by the contact loss area ratio: Wherein, is the area measure collected on the foundation surface, is the normal displacement jump, is the opening criterion threshold value; The spatial proportion of the foundation surface cracking or void is described, and for the seepage gushing risk, the in-plane dimensionless slope index : Wherein, is the along-surface pressure gradient on the foundation surface, is the critical hydraulic slope, ​​To convert the pressure gradient into a dimensionless slope indicator, To take the maximum value on the base surface; Local high gradient, potential gully channel and corridor anomaly outflow point are identified from the base surface level, all indicators and defect-free baseline difference are combined to form defect influence level, and Spread to all indicators to give interval estimation.

[0014] The beneficial effects of the preferred technical solutions in the embodiments of the present application are: by calculating the anti-skid safety factor, uplift pressure amplification coefficient and other indicators, the specific quantitative evaluation of the base surface level is provided, thereby supporting the comprehensive analysis and decision making of the engineering safety state.

[0015] Another object of the present application is to provide a base surface defect identification and evaluation system based on monitoring data.

[0016] To solve the above technical problems, the present application provides the following technical solutions: a base surface defect identification and evaluation system based on monitoring data, comprising: a modeling module, a data acquisition module, an identification and evaluation module; The modeling module constructs an interface model of the base surface, the base surface modeling is a two-dimensional discontinuous interface, and the interface constitutive equation and seepage model are explicitly introduced; The data acquisition module acquires operation monitoring data and extracts features from the operation monitoring data, and establishes a bidirectional mapping of defect parameters to monitoring responses; The identification and evaluation module identifies the base surface defect parameters through reverse learning, and outputs the engineering evaluation indicators of the base surface level by back substitution of the identification results to the forward modeling.

[0017] The present application provides a computer device comprising a memory and a processor, the memory stores a computer program, characterized in that the processor implements the steps of the base surface defect identification and evaluation method based on monitoring data when executing the computer program.

[0018] The present application provides a computer readable storage medium having a computer program stored thereon, characterized in that the computer program is executed by a processor to implement the steps of the base surface defect identification and evaluation method based on monitoring data.

[0019] The beneficial effects of the present application: the interface modeling of the present application with the base surface as the center establishes a direct and derivable function relationship between the monitoring quantity and the defect parameter, significantly improving the recognizability of the defect position, size and interface force property; through the closed loop of "recognition-back substitution-evaluation", the engineering quantification indexes such as anti-skid safety, pressure amplification and contact loss are directly output at the base surface level, avoiding the distortion caused by indirect inference relying on the overall measurement; the probability regression and the physical inversion are mutually checked, and remain stable under noise disturbance and working condition fluctuation, and the results have uncertainty definition and engineering interpretability.

[0020] The force-water coupling characteristics of the base surface are expressed by a two-dimensional discontinuous interface, and the defect parameter is parameterized by an opening-stiffness-leakage coefficient ternary group; the probability inverse model is trained by the base surface measurement priority feature system, and the forward consistency penalty term is closed at the base surface level; the four types of indexes of base surface anti-skid safety, uplift pressure amplification, contact loss and in-plane slope are completed for quantitative evaluation and are matched with uncertainty output. The above points constitute the overall scheme which is implementable, reviewable and engineering usable. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0022] Figure 1 The overall flowchart of the base surface defect identification and evaluation method based on monitoring data provided by an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present application.

[0024] Embodiment 1, refer to Figure 1 For an embodiment of the present application, the embodiment provides a base surface defect identification and evaluation method based on monitoring data, comprising: S100, constructing an interface model of the base surface, modeling the base surface as a two-dimensional discontinuous interface, and explicitly introducing an interface constitutive equation and a seepage model; S200, collecting operation monitoring data and extracting features from the operation monitoring data, and establishing a bidirectional mapping from the defect parameter to the monitoring response; S300, identifying the foundation surface defect parameter through reverse learning, and outputting the engineering evaluation index of the foundation surface level by back substitution of the identification result to the forward model; It should be noted that the existing method mainly uses drilling, geophysical exploration or baseline statistical threshold to identify the abnormality of the foundation surface. The first two methods have limitations in spatial coverage and operating condition consistency, and it is difficult to convert the contact state, joint porosity and permeation channel geometry at the foundation surface into a functional relationship of the operating monitoring quantity. The latter relies on a single index threshold, and it is difficult to complete the reversible mapping of "monitoring quantity → foundation surface defect parameter". Some researches use three-dimensional finite elements to analyze the "dam-foundation whole" in scenarios, but if the foundation surface is not regarded as an independent force-water coupled discontinuous interface, it is difficult to simultaneously depict the contact opening and closing, in-plane water conduction, out-of-plane seepage and the resulting uplift pressure redistribution, thereby being limited in recognizability and engineering usability. The above shortcomings directly lead to large errors in uplift pressure distribution reconstruction and large uncertainties in defect positioning, making it difficult to form a reviewable safety criterion at the foundation surface level.

[0025] Therefore, in view of the above problems, through the steps of S100-S300, the present application establishes a seepage-mechanical coupling model of the foundation surface interface and realizes parameterized identification and quantitative evaluation combined with operating monitoring data. The technical key lies in constructing a seepage-mechanical coupling numerical model of the discontinuous interface dominated by the foundation surface, and establishing a bidirectional mapping of the defect parameter to the monitoring response combined with the operating monitoring data, so as to realize the parameterized identification and quantitative evaluation of the defects such as the foundation surface cavity, soft interlayer, filling fracture zone and contact deterioration.

[0026] Embodiment 2, with reference to Figure 1 For an embodiment of the present application, the embodiment provides a foundation surface defect identification and evaluation method based on monitoring data, comprising: In the embodiment of the present application, the interface model of the foundation surface is constructed in S100, the foundation surface is modeled as a two-dimensional discontinuous interface, and the interface constitutive equation and the seepage model are explicitly introduced, including the following steps S101-S103: S101, the present application focuses on the foundation surface .

[0027] The dam body domain and the bedrock domain are described as linear elastic body and saturated porous medium respectively, and the foundation surface is modeled as a two-dimensional discontinuous interface, having contact-friction and in-plane water conduction characteristics.

[0028] The control equations in the dam body and the bedrock body use the classical balance equation and the Biot consolidation theory, denoting the dam body domain , the foundation domain , the displacement , the pore pressure and the stress satisfy: in, The divergence operator takes the divergence of a tensor field, and the result is a vector. Let be the stress tensor of the dam body domain. The density of the dam material. The vector of gravitational acceleration. For the elastic (stiffness) tensor of the dam material, For double dot product / tensor contraction operators, For strain operators acting on the displacement field The obtained strain tensor Let be the displacement vector of the dam body domain. The total stress tensor of the fundamental domain The total density of the base material (saturated porous media), The elastic tensor of the basic domain (solid skeleton), For skeleton displacement The strain tensor The displacement vector of the basic domain (solid skeleton) This is the Biot coefficient (the isotropic influence coefficient of pore pressure on total stress). This refers to the pore water pressure (pore pressure). For a unit tensor, make Indicates isotropic action; The storage coefficient / specific storage coefficient characterizes the ability to change water volume. For time The partial derivative, For volumetric strain, For pressure gradient, For penetration rate, The pore fluid dynamic viscosity coefficient; The seepage divergence term is obtained by applying Darcy's law and the law of conservation of mass. For source terms, the injection / extraction rate per unit volume; For volumetric strain, For trace operators, the sum of the diagonal elements of the tensor.

[0029] Meanwhile, the present invention in Explicitly introduce interface constitutives: Interface normal and tangential displacement jump , With corresponding traction satisfy in, , To verify the interface normal and tangential traction under the operating conditions, and For the normal and tangential interface stiffness, is the Coulomb friction coefficient, corresponding to the "softening" and "separation" effects when the base surface contacts deteriorate; S102. In-plane seepage at the foundation surface is described using a unified approach of the cubic law of gaps and two-dimensional Darcy, with the interface aperture recorded. With equivalent in-plane conductivity ,exist: in, For the surface gradient, The out-of-plane leakage coefficient reflects the strength of the connection between the foundation surface and the bedrock leakage and drainage system. For fluid dynamic viscosity, For the in-plane Darcy flux (tangential), the volumetric flux density along the tangential direction of the interface is given by the two-dimensional Darcy law. ; The out-of-plane leakage flux (normal) is the normal leakage from the foundation surface to the bedrock / drainage system, using a linear leakage boundary. , For the surface gradient operator, at the interface tangential upward relative to scalar Find the gradient; This refers to the interfacial pore water pressure, i.e., the water pressure within the foundation surface. For bedrock / drainage system pressure; Therefore, the hidden defects of the foundation surface are uniformly parameterized into an interface parameter field. Local patch The deviation above, among which, The coordinates are on the base surface.

[0030] Cavities or filled fracture zones correspond to Enlarge reduce, The value should be high or low depending on the permeability of the filling material; The weak interlayer is manifested as Lower and A combination that can be high or low.

[0031] Numerical robustness and consistency of differentiation are achieved through level sets in this invention. exist Upper-bound defect patch Then, a regularized Heaviside function is used to smoothly embed the interface parameters, achieving differentiable coupling between geometry and parameters.

[0032] S103. Boundary and load case settings are prioritized at the foundation surface level: Uplift pressure and pressure measuring holes are located on the foundation surface or in the adjacent area, and the drainage gallery and curtain system are expressed by a mixture of constant head or constant flow conditions. Changes in upstream and downstream water levels serve as the head boundary input; Temperature contraction and self-weight are expressed through body load; In terms of time progression, the seepage equation is discretized using backward Euler discretization and iterated with the mechanical field to ensure numerical robustness even under drastic changes in the parameters of the foundation surface.

[0033] In this embodiment of the invention, step S200 involves collecting operational monitoring data and extracting features from the operational monitoring data to establish a bidirectional mapping from defect parameters to monitoring responses, including the following step S201: S201. Monitoring data and feature characterization revolve around the measurement of the foundation surface.

[0034] Observation vector It consists of the head sequence of the uplift line at the foundation surface, the segmented flow of the gallery, the outflow from the drainage holes, and the minute displacement of the dam body near the foundation.

[0035] Using observation operators State Mapping to these graphical observations, we obtain And estimated based on historical stable periods ,in, Represents the displacement field. Typically, s represents the pore pressure field, and s represents the vector of the system's complete state, i.e., s = , The observation noise represents the unavoidable errors and interferences during the actual measurement process.

[0036] To enhance identifiability, this invention prioritizes preserving the principal components of the uplift pressure distribution on the foundation surface, the uplift pressure gradient, and the phase characteristics of the surface flow-head ratio in the feature space, thus constructing a processed feature vector. Used for reverse learning.

[0037] In an embodiment of the present invention, in step S300, defect parameters of the foundation surface are identified through reverse learning, and the identification results are substituted back into the forward model to output engineering evaluation indicators at the foundation surface level, including the following steps S301-S302: S301. Generate a sample library, consisting of base surface parameters and monitoring response pairs; Sampling in the parameter space using Latin hypercubes includes defect centers on the basal plane. Surface coordinates, patch size and direction, interface opening Normal and tangential interface stiffness and External leakage coefficient The parameters were obtained and subjected to batch forward modeling under representative water level historical conditions to form a large-scale coverage. gather; Reverse identification uses a probabilistic regression model for output. The training objective is to obtain the Gaussian negative log-likelihood plus a regularization term, while also introducing a forward consistency penalty term: Defect parameter field Substituting back into the forward model constrains the reproduction error of the basal surface observation features, performs closed-loop correction at the basal surface level, and simultaneously provides a physical inversion alternative. Using the weighted residuals of the basal surface observations as the objective function, and combining this with the adjoint sensitivity of the discrete equations, the model is directly solved. Two paths are used to perform bidirectional mapping at the foundation plane level, which are mutually verified.

[0038] S302. The identification results back-substitution verification and engineering evaluation are strictly limited to the output at the construction foundation level. Calculate the anti-slip safety factor of the foundation surface : in, For the construction of the base area, for The area of ​​the infinitesimal element on the surface, For equivalent cohesion, It is the internal friction angle. , To verify the interface normal and tangential traction under the operating conditions, The mold is for tangential traction; Define the uplift pressure amplification factor: in, To identify the defect parameter field Below, the location of the foundation surface Pore ​​pressure; The face pressure of the base plate under a defect-free baseline parameter field.

[0039] The increase in average water pressure on the foundation surface caused by defects is measured by the contact loss area ratio: in, For the set in Area measurement on For normal displacement jump, The threshold for opening criterion For interface opening; To characterize the spatial proportion of cracks or voids in the foundation surface, and to assess the risk of sudden seepage, an in-plane dimensionless slope index is used. : in, For the surface pressure gradient on the foundation surface, The critical hydraulic gradient. This is used to convert pressure gradients into dimensionless gradient indices. To obtain the maximum value on the foundation surface; Identify local high gradients, potential convection channels, and abnormal outflow points in corridors at the foundation surface level. Combine all indicators with the difference from the defect-free baseline to construct the defect impact level. Propagation to all indicators provides an interval estimate.

[0040] Finally, a hybrid finite element discretization method combining three-dimensional volume elements and two-dimensional interface elements was adopted. The volume-interface force-water coupling was solved by splitting iterative methods, and the interface parameters were embedded by level sets and regularized interpolation.

[0041] Sample management, feature compression, probabilistic regression, and adjoint inversion are all organized with a traceable data structure to ensure that parameters and results can be replayed and reviewed.

[0042] Example 3 is an embodiment of the present invention, which provides a method for identifying and evaluating foundation surface defects based on monitoring data. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0043] The implementation path is illustrated using gravity dams as an example: First, the geometry of the foundation surface is reconstructed based on the construction completion data, using surface parametric coordinates. Describe the location of the foundation surface and the extent of the defect patch; in Arrange the two-dimensional interface units and initialize them. The baseline distribution was determined; corresponding mixed boundary conditions were applied at the locations of existing piezometers and drainage holes in the area from the dam heel to the dam toe, and the segmented outflow of the gallery was represented by source and sink terms. Subsequently, uplift pressure, outflow, and near-base displacement data were collected during representative periods of water level fluctuation and stabilization, characterized according to the observation noise model, and presented in parameter space. Tens of thousands of forward modeling samples are generated through low-discrepancy sampling. A probabilistic regression model is trained using this sample library, and a forward modeling consistency penalty is introduced. The model is frozen after convergence on the validation set and in-plane residuals. The latest monitoring data is input during the deployment phase to obtain... Its uncertainty, and immediately back-substitute for verification, output. Indicators and confidence intervals; when the following occurs , , Predetermined threshold or In cases of exceeding limits, locate the most unfavorable area on the foundation surface and provide targeted suggestions for grouting improvement, drainage optimization, or increased monitoring.

[0044] Example 4 is an embodiment of the present invention. The above is an illustrative scheme of the foundation surface defect identification and evaluation method based on monitoring data. It should be noted that the technical solution of the foundation surface defect identification and evaluation system based on monitoring data belongs to the same concept as the technical solution of the foundation surface defect identification and evaluation method based on monitoring data described above. For details not described in detail in the technical solution of the foundation surface defect identification and evaluation system based on monitoring data in this embodiment, please refer to the description of the technical solution of the foundation surface defect identification and evaluation method based on monitoring data described above.

[0045] This embodiment provides a foundation surface defect identification and evaluation system based on monitoring data, including: a modeling module, a data acquisition module, and an identification and evaluation module; The modeling module constructs an interface model of the foundation surface, which is modeled as a two-dimensional discontinuous interface, and explicitly introduces the interface constitutive equation and seepage model. The data acquisition module collects operational monitoring data and extracts features from the operational monitoring data to establish a two-way mapping from defect parameters to monitoring responses; The identification and evaluation module identifies foundation surface defect parameters through reverse learning, substitutes the identification results back into the forward model, and outputs engineering evaluation indicators at the foundation surface level.

[0046] This embodiment also provides an electronic device applicable to the method for identifying and evaluating foundation surface defects based on monitoring data, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the method for identifying and evaluating foundation surface defects based on monitoring data as proposed in the above embodiment.

[0047] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the foundation surface defect identification and evaluation method based on monitoring data as proposed in the above embodiments.

[0048] The storage medium proposed in this embodiment and the method for identifying and evaluating foundation surface defects based on monitoring data proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0049] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0050] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for foundation defect identification and assessment based on monitoring data, characterized in that: The application relates to a method for identifying and evaluating defects of a foundation surface based on monitoring data. The interface model of the foundation surface comprises the following steps: the foundation surface is modeled as a two-dimensional discontinuous interface, and the interface constitutive equation and the seepage model are explicitly introduced. The operation monitoring data are collected and characteristic extraction is performed on the operation monitoring data to establish a bidirectional mapping of defect parameters to monitoring responses. The defect parameters of the foundation surface are identified through reverse learning, and the identification results are back-substituted into the forward model to output engineering evaluation indexes of the foundation surface level.

2. The method for foundation defect identification and evaluation based on monitoring data according to claim 1, wherein: The interface model of the foundation surface comprises the following steps: the foundation surface is modeled as a two-dimensional discontinuous interface, and the interface constitutive equation and the seepage model are explicitly introduced. The interface model of the foundation surface further comprises the following steps: hidden defects of the foundation surface are uniformly parameterized.

3. The method of claim 2, wherein: The defects are defined on the foundation surface by using a level set, and a regularization function is used to smoothly embed the interface parameters. The operation monitoring data collected by the foundation surface comprise uplift pressure, gallery sectional flow, drainage hole outflow and near-base displacement.

4. The method of foundation defect identification and assessment based on monitoring data according to claim 3, wherein: The characteristic extraction comprises the following steps: principal components of the uplift pressure distribution of the foundation surface, uplift pressure gradients and phase characteristics of the face flow-water head are reserved. The defect parameters of the foundation surface are identified through reverse learning, and the identification results are back-substituted into the forward model to output engineering evaluation indexes of the foundation surface level.

5. The method of foundation defect identification and assessment based on monitoring data as claimed in claim 4, wherein: The output engineering evaluation indexes of the foundation surface level further comprise the following steps: the identification results are back-substituted for rechecking, and engineering evaluation is strictly limited to the output of the foundation surface level. Sampling in parameter space by Latin hypercubes including parameters of defect center on base surface , patch size and orientation, interface opening , normal and tangent interface stiffness , out-of-plane leakage coefficient under representative water table histories, forming large-scale cover ed ensembles; The inverse recognition adopts a probability regression model to output The training target is a Gaussian negative log likelihood plus a regularization term, and a forward consistency penalty term is introduced. Put the defect parameter field Back to the forward model, constraint the reproduction error of the base surface observation characteristics, closed-loop correction in the base surface level, at the same time, provide physical inversion alternatives, with the weighted residual error of the base surface observation as the objective function, combined with the adjoint sensitivity of the discrete equation, directly solve , through two paths, bidirectional mapping in the base surface level, mutual check.

6. The method of foundation defect identification and assessment based on monitoring data according to claim 5, wherein: An uplift pressure amplification coefficient is defined. Computing a base surface safety factor against sliding : wherein, is the base area, is is the infinitesimal area on the base, is the equivalent cohesion, is the internal friction angle, , is the normal and tangential traction on the interface under the checking condition, is the tangential traction modulus; The output engineering evaluation indexes of the foundation surface level further comprise the following steps: the uplift pressure of the foundation surface is lifted by a magnitude caused by defects, and the magnitude is measured by using a contact loss area ratio: wherein, is the defect-free baseline parameter field, is the defect-free baseline parameter field, is the defect-free baseline parameter field, is the defect-free baseline parameter field.

7. The method of foundation defect identification and assessment based on monitoring data as claimed in claim 6, wherein: The application relates to a method for identifying and evaluating defects of a foundation surface based on monitoring data. wherein, is the area measure on , is the normal displacement jump, is the opening criterion threshold value; The space proportion of the cracking or voiding of the base surface is depicted, and the in-plane dimensionless slope index is used for the seepage burst risk : wherein is the along-face pressure gradient on the building surface, is the critical hydraulic slope, is the dimensionless slope indicator for converting the pressure gradient, is taken as the maximum value on the building surface; From the base level of construction to identify local high gradient, potential gully channel and corridor anomaly outflow point, all indicators and defect-free baseline difference value together constitute the defect influence level, and The interval estimation is given to all indicators.

8. A system for identification and evaluation of defects of a foundation based on monitoring data, applying the method for identification and evaluation of defects of a foundation based on monitoring data according to any one of claims 1 to 7, characterized in that The modeling module is used for constructing an interface model of the foundation surface, modeling the foundation surface as a two-dimensional discontinuous interface, and explicitly introducing an interface constitutive equation and a seepage model. The data acquisition module is used for collecting operation monitoring data, performing characteristic extraction on the operation monitoring data, and establishing a bidirectional mapping of defect parameters to monitoring responses. The identification and evaluation module is used for identifying defect parameters of the foundation surface through reverse learning, back-substituting identification results into a forward model, and outputting engineering evaluation indexes of the foundation surface level. The processor executes the computer program to realize the steps of the method for identifying and evaluating defects of a foundation surface based on monitoring data according to any one of claims 1 to 7. The computer program is executed by the processor to realize the steps of the method for identifying and evaluating defects of a foundation surface based on monitoring data according to any one of claims 1 to 7. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. ​ 10. A computer-readable storage medium having stored thereon a computer program, characterized in that, ​

Images