Dynamic quantification method and device for seepage risk

Abstract

The invention provides a seepage risk dynamic quantification method and device, and relates to the technical field of engineering prediction.The method comprises the steps that seepage parameter observation data of different observation points in a target area and geological environment factor data related to seepage parameters are obtained; constructing a multi-scale geographically weighted regression model based on the permeability parameter observation data and the geological environment factor data to obtain a deterministic space trend field of permeability parameters; calculating a spatial residual error of the penetration parameter observation value relative to the deterministic spatial trend field, and performing Kriging interpolation on the spatial residual error to obtain a random residual error field and uncertainty information; fusing the deterministic space trend field and the random residual field to obtain a permeability parameter space distribution field; and based on the seepage parameter space distribution field and the uncertainty information, utilizing a Bayesian assimilation algorithm to obtain a seepage risk dynamic quantification result. According to the method, the core defects of insufficient description of seepage parameter spatial heterogeneity and single risk prediction static state in a traditional method can be solved.

Description

Technical Field

[0001] This invention relates to the field of engineering prediction technology, and in particular to a method and apparatus for dynamic quantification of seepage risk. Background Technology

[0002] Seepage risk assessment is widely used in engineering scenarios such as underground cavern groups, tunnels, and surrounding rock of hydraulic structures. It is used to predict indicators such as water inflow, water head, pore water pressure, and hydraulic gradient caused by changes in the permeability of the surrounding rock, thereby providing a basis for construction organization, drainage design, and safety early warning. The reliability of seepage risk assessment largely depends on the accurate construction of the spatial field of seepage parameters and the way the risk results are expressed.

[0003] On the one hand, seepage parameters such as permeability coefficient exhibit significant spatial heterogeneity, influenced by factors such as lithological differences, fracture zones, fissure development, burial depth, and external water pressure. Furthermore, this influence varies across different regions. Traditional methods often employ constant assignment, empirical zoning assignment, or simple interpolation to construct the seepage parameter field, which struggles to accurately characterize these spatial heterogeneities. In particular, they easily overlook local high-permeability channels and abrupt change zones, leading to a mismatch between the parameter inputs of the seepage numerical model and actual geological conditions. This results in distorted input parameters for numerical seepage calculations, consequently causing significant deviations in predicted results such as inflow, hydraulic head, or pore water pressure.

[0004] On the other hand, when the parameter field is uncertain and evolves with construction disturbances, traditional risk prediction usually uses one-time calculation results as the conclusion. The output is mostly deterministic single value or fixed zoning level. It lacks quantitative expression of parameter estimation error and its impact on risk indicators. It is also difficult to form an effective linkage and update with the new monitoring information added during construction / operation. As a result, risk prediction has static and single characteristics, which makes it difficult to reflect the changes and uncertainties of risk level at different stages and has limited decision support capabilities. Summary of the Invention

[0005] To address the core shortcomings of traditional methods in failing to adequately characterize the spatial heterogeneity of seepage parameters and in providing static and singular risk predictions, this invention offers a method and apparatus for dynamic quantification of seepage risk.

[0006] According to a first aspect of the present invention, a dynamic quantification method for seepage risk is provided, applied to seepage risk assessment of a target area, the dynamic quantification method for seepage risk comprising: Acquire observation data of permeability parameters and geological environmental factors related to the permeability parameters at different observation points in the target area; Based on the observed data of permeability parameters and geological environmental factor data, a multi-scale geographically weighted regression model is constructed to obtain the deterministic spatial trend field of permeability parameters; Calculate the spatial residual of the permeation parameter observations relative to the deterministic spatial trend field, and perform Kriging interpolation on the spatial residual to obtain the stochastic residual field and uncertainty information; By fusing the deterministic spatial trend field and the stochastic residual field, the spatial distribution field of the penetration parameters is obtained; Based on the spatial distribution field of the permeability parameters and the uncertainty information, the dynamic quantification result of seepage risk is obtained using the Bayesian assimilation algorithm.

[0007] In some exemplary embodiments of the present invention, based on the foregoing scheme, the permeability parameter observation data includes spatial coordinates; Based on the aforementioned permeability parameter observation data and geological environmental factor data, a multi-scale geographically weighted regression model is constructed, including: Based on the spatial coordinates of each observation point Establish spatial weights to weight neighborhood samples; Weighted regression is performed under the aforementioned spatial weights to obtain the local intercept that varies with spatial location. and the local regression coefficients corresponding to various geological environmental factors ; For any observation point Calculate the measured values ​​of geological environmental factors and corresponding local regression coefficients The product of the product and the sum of the local intercepts are used to obtain the trend value of the permeability coefficient. ; The permeability coefficient deterministic spatial trend field is obtained based on the permeability coefficient trend values ​​of all observation points.

[0008] In some exemplary embodiments of the present invention, based on the foregoing scheme, calculating the spatial residual of the permeability coefficient observation relative to the deterministic spatial trend field includes: Permeability coefficient measurements were obtained at each observation point. And determine the permeability coefficient trend value corresponding to the spatial coordinates of the observation point in the deterministic spatial trend field. ; The difference between the observed permeability coefficient value and the trend value of the permeability coefficient is taken as the spatial residual of the observation point. .

[0009] In some exemplary embodiments of the present invention, based on the foregoing scheme, kriging interpolation is performed on the spatial residual to obtain a stochastic residual field and uncertainty information, including: based on Spatial residuals of known sample points For unknown points Perform linear unbiased optimal estimation to obtain the unknown points. Kriging residuals ,in in, To be assigned to the The weighting coefficients for each sample point are determined by solving the fundamental Kriging equations under unbiased constraints. , Based on the obtained weight coefficients, calculate the unknown points. Kriging variance As the uncertainty information, wherein in, Unknown point With the The theoretical semivariance value between sample points, wherein the theoretical semivariance value is calculated by the theoretical semivariance function based on the distance between the two points; Lagrange multipliers introduced to satisfy the unbiased constraint; the Kriging residuals at each unknown point. The random residual field is formed by the Kriging variance of each unknown point. This constitutes the variance field.

[0010] In some exemplary embodiments of the present invention, based on the foregoing scheme, the uncertainty information includes at least the Kriging variance field; Based on the spatial distribution field of the permeability parameters and the uncertainty information, the dynamic quantification results of seepage risk are obtained using the Bayesian assimilation algorithm, including: Based on the spatial distribution field of the permeability parameters, a physical seepage numerical model of the target area is established; Using the spatial distribution field of the permeation parameters as the mean and the Kriging variance field as the variance, as the prior information for Bayesian inference, a Bayesian data assimilation framework for the physical seepage numerical model is constructed. By inputting the design conditions into the Bayesian data assimilation framework of the physical seepage numerical model, the prior prediction probability distribution of seepage risk indicators at different stages is obtained. Real-time seepage observation data is acquired, and the real-time seepage observation data is used as the observation to perform Bayesian update on the prior prediction probability distribution to obtain the posterior probability distribution of seepage risk indicators at different stages. The posterior probability distribution is output as the dynamic quantification result of the seepage risk.

[0011] In some exemplary embodiments of the present invention, based on the foregoing scheme and the spatial distribution field of the permeability parameters, establishing a physical seepage numerical model of the target area includes: The computational domain of the target region is meshed to form multiple discrete units; The spatial distribution field of the permeability parameters is mapped to the discrete unit, so that the permeability coefficient corresponding to each discrete unit is the unit permeability coefficient that varies with spatial location; Based on the unit permeability coefficient and Darcy's law, the physical seepage numerical model is obtained on the discrete unit.

[0012] In some exemplary embodiments of the present invention, based on the foregoing scheme, the physical seepage numerical model obtained on the discrete unit, based on the unit permeability coefficient and Darcy's law, includes: Within each discrete unit, the unit permeability coefficient corresponding to that discrete unit is substituted into Darcy's law to establish the permeability expression of that discrete unit. The seepage control equations for the discrete unit are established by combining the mass conservation relationship, and the seepage control equations are discretized. The discrete equations of each discrete unit are assembled, and boundary conditions are applied at the boundary of the computational domain to obtain the physical seepage numerical model.

[0013] According to a second aspect of the present invention, a dynamic quantification device for seepage risk is provided, comprising: The data acquisition module is used to acquire observation data of permeability parameters at different observation points in the target area, as well as geological environmental factor data related to the permeability parameters. The model building module is used to construct a multi-scale geographically weighted regression model based on the observed data of the permeability parameters and the geological environmental factor data, so as to obtain the deterministic spatial trend field of the permeability parameters; The interpolation module is used to calculate the spatial residual of the penetration parameter observations relative to the deterministic spatial trend field, and to perform Kriging interpolation on the spatial residual to obtain the stochastic residual field and uncertainty information. The data fusion module is used to fuse the deterministic spatial trend field and the stochastic residual field to obtain the spatial distribution field of the penetration parameters; The result generation module is used to obtain the dynamic quantification result of seepage risk based on the spatial distribution field of the permeability parameters and the uncertainty information using a Bayesian assimilation algorithm.

[0014] According to a third aspect of the present invention, an electronic device is provided, comprising: a processor; and a memory storing computer-readable instructions that, when executed by the processor, implement the method of the first aspect.

[0015] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method of the first aspect.

[0016] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: On the one hand, by acquiring permeability coefficient observation data from multiple observation points and combining it with geological environmental factor data, a deterministic spatial trend field of permeability coefficients is obtained using a multi-scale geographically weighted regression model. Then, Kriging interpolation is performed on the spatial residuals of the observed values ​​relative to the trend field to obtain a stochastic residual field and output the corresponding uncertainty information. Finally, the deterministic spatial trend field and the stochastic residual field are fused to form a spatial distribution field of permeability parameters. This approach can simultaneously take into account the spatial non-stationary trend driven by geological environmental factors and the spatial correlation structure of the residuals. Compared with traditional construction methods such as constant assignment, empirical partitioning, or simple interpolation, it can more precisely characterize the spatial heterogeneity of permeability parameters. In particular, it can improve the ability to identify and represent key parts such as local high-permeability channels and abrupt change zones, reduce parameter input distortion in seepage numerical calculations from the source, and thus improve the accuracy and stability of predicting seepage risk indicators such as inflow, head, pore water pressure, and hydraulic gradient.

[0017] On the other hand, based on the spatial distribution field of the aforementioned seepage parameters and the uncertainty information obtained by Kriging interpolation, a Bayesian assimilation algorithm is used to output dynamic quantitative results of seepage risk. This allows risk prediction to move beyond one-time, deterministic single-value or fixed-level classifications, enabling the expression of risk levels in a probabilistic manner. Furthermore, it allows for continuous updates and phased assessments as new monitoring / operating condition information arrives. This overcomes the shortcomings of traditional risk prediction—its static, singular nature, difficulty in quantifying errors and uncertainties—and improves the reliability and interpretability of risk warnings and construction decisions.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the specification, serve to explain the principles of the invention.

[0020] Figure 1 A schematic diagram of the system architecture of an exemplary application environment in which a dynamic quantification method and apparatus for seepage risk can be applied according to embodiments of the present invention is shown; Figure 2 The schematic diagram illustrates a flow chart of a dynamic quantification method for seepage risk according to some embodiments of the present invention; Figure 3 A schematic diagram of a seepage risk dynamic quantification device according to some embodiments of the present invention is shown; Figure 4The schematic diagram illustrates the structure of a computer system of an electronic device according to some embodiments of the present invention; Figure 5 A schematic diagram of a computer-readable storage medium according to some embodiments of the present invention is shown. Detailed Implementation

[0021] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0022] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” used in this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0023] It should be understood that although the terms first, second, third, etc., may be used in this invention to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of this invention, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0024] Figure 1 A schematic diagram of the system architecture of an exemplary application environment in which a dynamic quantification method and apparatus for seepage risk can be applied according to embodiments of the present invention is shown.

[0025] like Figure 1As shown, system architecture 100 may include one or more terminal devices such as desktop computer 101, portable computer 102, and smartphone 103, network 104, and server 105. Network 104 is used as a medium to provide a communication link between the terminal devices and server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables. Terminal devices may be various electronic devices with data processing capabilities, and these devices have a display screen for displaying users the dynamic quantitative results of seepage risk, such as a dynamic probability distribution map of seepage risk along the tunnel, including but not limited to the aforementioned desktop computer, portable computer, and smartphone. It should be understood that... Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, there can be any number of terminal devices, networks, and servers. For example, server 105 could be a sub-server cluster composed of multiple sub-servers.

[0026] The dynamic quantification method for seepage risk provided in this embodiment of the invention can generally be executed by a terminal device, and correspondingly, the dynamic quantification device for seepage risk is generally installed in the terminal device. However, it is readily understood by those skilled in the art that the dynamic quantification method for seepage risk provided in this embodiment of the invention can also be executed by a server 105, and correspondingly, the dynamic quantification device for seepage risk can also be installed in the server 105. This exemplary embodiment does not impose any special limitations on this.

[0027] Furthermore, it should be understood that the dynamic quantification method for seepage risk according to embodiments of the present invention can be configured as a software module. In some implementation scenarios, the dynamic quantification scheme for seepage risk of the present invention can be deployed independently to dynamically quantify the seepage risk of different target areas. In other implementation scenarios, the dynamic quantification scheme for seepage risk of the present invention can be deployed within other software as a functional module of that software, such as in seepage parameter analysis software. The present invention does not impose any particular restrictions on the application of the dynamic quantification method for seepage risk.

[0028] The embodiments of the present invention will now be described in detail.

[0029] like Figure 2 As shown, Figure 2 This invention discloses a dynamic quantification method for seepage risk according to an exemplary embodiment, applied to seepage risk assessment of a target area. The target area is used to define the spatial scope of the seepage risk assessment, which can be a three-dimensional spatial domain corresponding to the surrounding rock of underground cavern groups, tunnels, or hydraulic structures, or an influence range domain surrounding the tunnel axis / dam foundation.

[0030] In some implementations, the dynamic quantification method for seepage risk includes: S210: Obtain observation data of permeability parameters and geological environmental factors related to the permeability parameters at different observation points in the target area; S220: Based on the observed data of the permeability parameters and the geological environmental factor data, a multi-scale geographically weighted regression model is constructed to obtain the deterministic spatial trend field of the permeability parameters; S230: Calculate the spatial residual of the permeation parameter observations relative to the deterministic spatial trend field, and perform Kriging interpolation on the spatial residual to obtain the stochastic residual field and uncertainty information; S240: By fusing the deterministic spatial trend field and the stochastic residual field, the spatial distribution field of the penetration parameters is obtained; S250: Based on the spatial distribution field of the permeability parameters and the uncertainty information, the dynamic quantification result of seepage risk is obtained using the Bayesian assimilation algorithm.

[0031] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: On the one hand, by acquiring permeability coefficient observation data from multiple observation points and combining it with geological environmental factor data, a deterministic spatial trend field of permeability coefficients is obtained using a multi-scale geographically weighted regression model. Then, Kriging interpolation is performed on the spatial residuals of the observed values ​​relative to the trend field to obtain a stochastic residual field and output the corresponding uncertainty information. Finally, the deterministic spatial trend field and the stochastic residual field are fused to form a spatial distribution field of permeability parameters. This approach can simultaneously take into account the spatial non-stationary trend driven by geological environmental factors and the spatial correlation structure of the residuals. Compared with traditional construction methods such as constant assignment, empirical partitioning, or simple interpolation, it can more precisely characterize the spatial heterogeneity of permeability parameters. In particular, it can improve the ability to identify and represent key parts such as local high-permeability channels and abrupt change zones, reduce parameter input distortion in seepage numerical calculations from the source, and thus improve the accuracy and stability of predicting seepage risk indicators such as inflow, head, pore water pressure, and hydraulic gradient.

[0032] On the other hand, based on the spatial distribution field of the aforementioned seepage parameters and the uncertainty information obtained by Kriging interpolation, a Bayesian assimilation algorithm is used to output dynamic quantitative results of seepage risk. This allows risk prediction to move beyond one-time, deterministic single-value or fixed-level classifications, enabling the expression of risk levels in a probabilistic manner. Furthermore, it allows for continuous updates and phased assessments as new monitoring / operating condition information arrives. This overcomes the shortcomings of traditional risk prediction—its static, singular nature, difficulty in quantifying errors and uncertainties—and improves the reliability and interpretability of risk warnings and construction decisions.

[0033] In S210, permeability parameter observation data and geological environmental factor data related to the permeability parameters are acquired at different observation points in the target area.

[0034] Different observation points refer to multiple spatial sampling locations distributed within the target area, used to provide samples of permeability parameters. Observation points can be determined by boreholes / test holes, corresponding to the location of the borehole opening or a certain depth section within the borehole (e.g., divided by layers or by borehole segments); observation points can also be determined by monitoring points, including the locations of piezometers, manometers, water level observation wells, etc.; observation points can also be determined by monitoring locations on the engineering surface, such as monitoring sections of tunnel wall water inflow, locations of water collection wells, or locations of drainage ditch sections.

[0035] Permeability parameter observation data are used to characterize the permeability of the medium at each observation point. The permeability parameter can be the permeability coefficient. Penetration rate Hydraulic conductivity Or anisotropic permeability parameters (e.g.) At least one of the following.

[0036] The sources of permeability parameter observation data can be field test data and their calculation results, such as raw records of flow rate, pressure difference, and duration obtained from water pressure tests, pumping tests, and injection tests, and the permeability parameter observation values ​​calculated according to preset conversion rules; permeability parameter observation data can also be equivalent permeability parameter observation values ​​output by monitoring platforms or inversion systems; permeability parameter observation data can also be existing exploration reports, test records, or permeability parameter results stored in databases. To enhance data usability, in addition to permeability parameter observation values, permeability parameter observation data can further include observation timestamps, test condition identifiers, data quality identifiers, and / or error range information for subsequent assimilation updates or uncertainty control.

[0037] Geological environmental factor data are used to characterize explanatory variables affecting the spatial variation of permeability parameters. These variables are correlated with permeability parameters, and the correlation can be mechanistic (e.g., fractures / faults control permeability channels) or statistical (e.g., factors significantly correlated with permeability parameters in historical data). Geological environmental factor data may include at least one of the following: lithology or lithological indices, distance from faults / breccia zones, distribution of tectonic breccia zones, fracture density or rock quality designation (RQD), degree of weathering, burial depth / overburden thickness, topographic elevation, external water pressure / boundary head, recharge and discharge conditions, etc.; it may also include geophysical property factors, such as resistivity, wave velocity, or other spatial properties that can serve as proxies for permeability. Geological environmental factor data can be stored in the form of raster layers, vector layers, zoning maps, or attributes of 3D geological models.

[0038] To ensure a one-to-one correspondence between geological environmental factor data and observation points, spatial matching and extraction are performed between the data and the points: using the spatial coordinates of the observation points as an index, the factor values ​​at the corresponding locations are extracted from the geological environmental factor data and used as the measured factor values ​​for those observation points. When the factor data is in raster or grid format, the grid index can be used to directly retrieve the values; when the coordinates of the observation point do not coincide with the grid nodes, nearest neighbor retrieval, linear interpolation, or neighborhood-based statistical retrieval (e.g., taking the mean / maximum / quantile within a preset radius around the observation point) can be used to determine the values. When the target area is a three-dimensional domain and the factors vary with depth, the factor values ​​of the corresponding strata can be extracted according to the relationship between the depth segment of the observation point and the stratigraphic interface to ensure that the factor values ​​are consistent with the medium in which the observation point is located. For discrete category factors (such as lithology categories), they can be converted into numerical factors that can participate in modeling through preset coding rules.

[0039] In S220, a multi-scale geographically weighted regression model is constructed based on the observed data of the permeability parameters and the geological environmental factor data to obtain the deterministic spatial trend field of the permeability parameters.

[0040] Multiscale geographically weighted regression models are used to characterize the spatial nonstationary trend of permeability parameters driven by geological environmental factors.

[0041] In some embodiments, the permeability coefficient observation data includes at least the permeability coefficient observation values ​​from multiple observation points and their corresponding spatial coordinates. These spatial coordinates can be two-dimensional or three-dimensional. In this embodiment of the invention, the spatial coordinates are used as three-dimensional coordinates. , Taking the numbering of observation points as an example, the following explanation is provided. For ease of regression modeling, the number of observation points can be... Geological environmental factors at the observation point The measured value is expressed as The total number of environmental factors is .

[0042] To reflect the spatial non-stationarity of the relationship between permeability coefficient and geological environmental factors, spatial weights can be established based on the spatial coordinates of observation points to weight neighboring samples. Specifically, for the target observation point... This allows for the calculation of spatial distances between the target observation point and other observation points, and the construction of weight values ​​based on these distances. This results in higher weights for sample points closer to the target observation point and lower weights for sample points farther away, forming a set of spatial weights for weighted regression. The introduction of these spatial weights allows the regression parameters to vary with spatial location, thus characterizing the differences in local patterns across different regions.

[0043] Spatial weights can be determined by distance attenuation methods, such as using Gaussian kernels, biquadratic kernels, or exponential kernels, so that sample points closer to the target observation point are given greater weights, and sample points farther away are given smaller weights. To reflect multi-scale characteristics, in some implementations, different bandwidth parameters can be set for different geological environmental factors, so that the local regression coefficients corresponding to different factors have different spatial scales of influence.

[0044] Under the aforementioned spatial weighting, a weighted regression analysis is performed on the observed permeability coefficient and the measured geological environmental factors to obtain the local intercept that varies with spatial location. and the local regression coefficients corresponding to various geological environmental factors Among them, the local intercept Used to characterize observation points when geological environmental factors are not considered. Baseline level of permeability coefficient; local regression coefficient Used to characterize the Geological environmental factors at the observation point The influence intensity on the permeability coefficient is determined, and this influence intensity varies with spatial location to reflect the non-stationary effects caused by differences in geological conditions in different regions. In specific implementation, the multi-scale nature of multi-scale geographic weighted regression can be reflected by using different spatial scales for different geological environmental factors. Thus, during the solution process, corresponding bandwidths or weighted neighborhood ranges are applied to different factors to obtain a local regression coefficient distribution that better reflects actual geological differences.

[0045] After obtaining the local intercept and local regression coefficients, for any observation point Calculate its permeability coefficient trend value Specifically, the measured values ​​of various geological environmental factors will be... With corresponding local regression coefficients Multiplying and summing these factors yields the combined contribution of each factor to the permeability coefficient. This contribution is then superimposed with the local intercept to obtain the observation point. Permeability coefficient trend value: The above trend values ​​characterize the spatial variation trend of the permeability coefficient driven by geological environmental factors, and can reflect the non-stationary trend characteristics caused by the different intensity of factors in different regions.

[0046] The permeability coefficient trend value was calculated for each observation point. Furthermore, the trend value can be extended or mapped within the target area according to a preset spatial sampling method (e.g., grid points or cell center points) to form a deterministic spatial trend field for the permeability coefficient. This deterministic spatial trend field characterizes the deterministic trend distribution of the permeability coefficient within the target area and provides a benchmark for subsequent residual calculations and spatial correlation compensation.

[0047] In S230, the spatial residual of the permeation parameter observations relative to the deterministic spatial trend field is calculated, and the spatial residual is kriging interpolated to obtain the stochastic residual field and uncertainty information. For each observation point Read the permeability coefficient observation value at this observation point. Simultaneously, based on the spatial coordinates of the observation point, the corresponding location is located in the deterministic spatial trend field, and the trend value of the permeability coefficient at that location is extracted. When the deterministic spatial trend field is stored in the form of discrete grids or cells, the extraction process can be directly obtained through the grid index; when the coordinates of the observation point do not completely coincide with the grid nodes, the corresponding trend value can be determined by nearest neighbor taking or linear interpolation.

[0048] In obtaining and Then, the spatial residual at that observation point is obtained by subtracting the observed permeability coefficient from the permeability coefficient trend value. ,For example: .

[0049] Next, kriging interpolation is performed on the spatial residuals to obtain the stochastic residual field and uncertainty information within the target region. Specifically, let there be a total Given _n_ sample points, their spatial residuals are respectively ,in For any unknown point within the target area. The linear unbiased optimal estimation is used to predict its residuals, and the unknown points are... The Kriging residuals are expressed as a weighted sum of the residuals of known sample points: in, To be assigned to the The weighting coefficients for each sample point. To ensure the unbiasedness of the estimate and minimize the variance, the weighting coefficients are obtained by solving the fundamental Kriging equations; unbiased constraints are introduced during the solution process, and the constraints are solved using Lagrange multipliers.

[0050] In the specific implementation, a theoretical semivariance function is first constructed based on the spatial distance between sample points, and then the unknown points are calculated using the theoretical semivariance function. The theoretical semivariance values ​​between each sample point and between sample points are used to form the semivariance term in the fundamental Kriging equations. Among these, the unknown points... With the The theoretical semivariance between 1 sample points is denoted as . The weights are calculated from the theoretical semivariance function and the distance between the two points. Then, the fundamental Kriging equations are solved to obtain the optimal weights for each sample point. and Lagrange multipliers .

[0051] After obtaining the weighting coefficients, in addition to calculating the unknown points Kriging residuals In addition, it also calculates unknown points. Kriging variance As information of uncertainty, Kriging variance can be expressed as: in, Lagrange multipliers are introduced to satisfy the unbiasedness constraint. By repeating the above Kriging interpolation and variance calculation process for each unknown point within the target region, the distribution of Kriging residuals and the distribution of Kriging variance covering the target region can be obtained: from the Kriging residuals of each unknown point... The residual field is composed of the kriging variance of each unknown point. This constitutes the variance field, which is used to characterize the uncertainty of stochastic residual estimation.

[0052] In S240, the deterministic spatial trend field and the stochastic residual field are fused to obtain the spatial distribution field of the penetration parameters.

[0053] In other words, for any spatial location point within the target area (such as a grid node or the center point of a discrete unit) Determine the trend value of the permeation parameter corresponding to this location in the deterministic spatial trend field. And determine the random residual value corresponding to that position in the random residual field. Subsequently, the trend values ​​of the permeability parameters are superimposed with the random residual values ​​to obtain the spatial distribution values ​​of the permeability parameters at that location. ,For example: The above fusion process is repeated for each spatial location point within the target area to form a spatial distribution field of permeability parameters covering the target area. Through this fusion method, the spatial distribution field of permeability parameters includes both spatially non-stationary trend components driven by geological environmental factors and random fluctuation components obtained by spatial residual correlation structure compensation, thereby improving the ability to express the details of spatial heterogeneity of permeability parameters.

[0054] In S250, based on the spatial distribution field of the permeability parameters and the uncertainty information, a Bayesian assimilation algorithm is used to obtain the dynamic quantification result of seepage risk.

[0055] After obtaining the spatial distribution field of the permeability parameters and the uncertainty information by completing the aforementioned steps, this embodiment uses a Bayesian assimilation algorithm to output the dynamic quantification result of seepage risk, as follows: First, a physical seepage numerical model of the target area is established based on the spatial distribution field of the seepage parameters.

[0056] In some implementations, the physical seepage numerical model can be achieved by inputting the spatial distribution field of seepage parameters into a commercial or open-source seepage numerical simulation software, such as GeoStudio (seepage analysis software) or COMSOL (multiphysics simulation software). The numerical simulation software meshes the computational domain of the target region to form millions of discrete cells, each with corresponding element geometry and spatial location. The spatial distribution field of seepage parameters is then mapped onto these discrete cells, making the permeability coefficient of each discrete cell a unit permeability coefficient that varies with spatial location. Specifically, using the center point, integration point, or node coordinates of the discrete cell as the mapping location, the permeability coefficient value corresponding to the mapping location is determined in the spatial distribution field of seepage parameters, and this value is assigned to the corresponding discrete cell, resulting in the unit permeability coefficient for each discrete cell. Thus, discrete cells at different spatial locations have different permeability coefficient values, avoiding the simplification of the permeability coefficient to a constant or an artificially partitioned constant. Next, the physical seepage numerical model is obtained on these discrete cells based on the unit permeability coefficient and Darcy's law. That is, within each discrete unit, the unit permeability coefficient is used as the permeability characteristic parameter of the medium. This coefficient is substituted into Darcy's law to establish a permeability expression at the unit scale, and combined with the mass conservation relationship to form the seepage control equations. These seepage control equations are discretized and assembled on the discrete units to form a set of model equations for numerical solution, thus constituting the physical seepage numerical model. By solving the physical seepage numerical model, the head field and / or pore water pressure field and flow direction information within the computational domain can be obtained, and further used to calculate seepage risk indicators such as inflow rate.

[0057] Secondly, using the spatial distribution field of the permeability parameters as the mean and the kriging variance field as the variance, as prior information for Bayesian inference, a Bayesian data assimilation framework for the physical seepage numerical model is constructed. Specifically, the permeability parameter field or its parameterized representation is taken as the state variable / parameter vector to be assimilated, the spatial distribution field of the permeability parameters is taken as the prior mean field of this state variable, and the kriging variance field is taken as the prior variance field of this state variable, thereby forming a prior probability model of the permeability parameters. The prior probability model and the physical seepage numerical model together constitute the Bayesian data assimilation framework, enabling the uncertainty of the permeability parameters to propagate to the uncertainty of the risk indicators through forward seepage calculations. To ensure the physical feasibility of the permeability parameters, a prior distribution of the permeability parameters can be constructed in the original domain or the logarithmic domain, and the prior mean and prior variance are defined in the same domain.

[0058] Then, the design conditions are input into the Bayesian data assimilation framework of the physical seepage numerical model to obtain the prior prediction probability distribution of seepage risk indicators at different stages.

[0059] In some implementations, design conditions such as boundary head, external water pressure, and recharge / discharge conditions can be used as model inputs, and the physical seepage numerical model can be driven to make predictions sequentially according to the stage division order. Simultaneously, multiple sets of seepage parameter field samples (or state variable samples) are generated based on the prior probability model, and each sample is input into the physical seepage numerical model for forward solving to obtain the seepage risk index sample set for the corresponding stage. The sample set is then statistically analyzed to obtain the prior prediction probability distribution of the seepage risk index for different stages. The seepage risk index includes at least the inflow rate, and may further include at least one of the following: head, pore water pressure, and hydraulic gradient.

[0060] Next, real-time seepage observation data is acquired, and the real-time seepage observation data is used as the observation to perform Bayesian update on the prior prediction probability distribution to obtain the posterior probability distribution of seepage risk indicators at different stages.

[0061] The real-time monitored seepage observation data corresponds to the seepage risk indicators. The observation data can be at least one of the following: wall inflow monitoring value, hydraulic head monitoring value, pore water pressure monitoring value, hydraulic gradient monitoring value, or a combination of the above observations. An observation error model is established by combining the observation data with the output of the prior prediction model, and a likelihood function is constructed to update the prior prediction results under observation constraints. Bayesian updates can be implemented through analytical updates or numerical updates. When the physical seepage numerical model is nonlinear or has a high dimension, Bayesian updates can be completed using a set update method (e.g., obtaining the posterior set through set gain correction), or using a sampling and reweighting method (e.g., obtaining the posterior particle set through particle weighting and resampling), or using a posterior sampling method (e.g., obtaining posterior samples through Markov chain sampling), thereby obtaining the posterior probability distribution of the seepage risk indicators at different stages.

[0062] Different stages can be divided by time windows, with real-time monitoring data triggering Bayesian updates according to a preset update cycle. Alternatively, stages can be divided by changes in construction conditions or monitoring events, triggering Bayesian updates when a sudden increase in water inflow, abnormal water head, or sudden change in pore pressure is detected, thus achieving event-driven dynamic quantification.

[0063] Finally, the posterior probability distribution is output as the dynamic quantification result of the seepage risk. Specifically, the output includes the posterior mean, posterior variance, and confidence interval of the seepage risk index, and can further output the exceedance probability of the risk index exceeding a preset threshold for use in risk warning classification and decision support.

[0064] Here, in some implementations, the Bayesian assimilation algorithm can be implemented using ensemble Kalman filtering: multiple random field samples of permeability parameters satisfying the prior mean field and prior variance field are generated as set members, each set member is input into the physical seepage numerical model to obtain a risk index set, and the set members are updated according to the observation residuals to obtain the posterior set, and the posterior probability distribution is obtained by statistical analysis of the posterior set.

[0065] In other implementations, the Bayesian assimilation algorithm is implemented using particle filtering: the random field samples of the permeation parameters are used as particles, and the particles are weighted based on the observation likelihood and resampling is performed so that the weights are concentrated on a set of particles that better match the real-time observation data, thereby obtaining the posterior probability distribution of the seepage risk index.

[0066] Bayesian assimilation algorithms can also be implemented using Markov chain Monte Carlo: the permeability parameter field or its low-dimensional representation (e.g., partition parameters, principal component coefficients, or basis function coefficients) is used as the parameters to be estimated, the posterior samples of the parameters are obtained by sampling the posterior distribution, and the posterior probability distribution of the seepage risk index is calculated accordingly.

[0067] After obtaining the dynamic quantitative results of seepage risk According to a second aspect of the present invention, a dynamic quantification device for seepage risk 300 is also provided, with reference to Figure 3 As shown, the seepage risk dynamic quantification device 300 includes: Data acquisition module 310 is used to acquire observation data of permeability parameters at different observation points in the target area and geological environmental factor data related to the permeability parameters; The model building module 320 is used to build a multi-scale geographically weighted regression model based on the permeability parameter observation data and geological environmental factor data to obtain the deterministic spatial trend field of the permeability parameter. Interpolation module 330 is used to calculate the spatial residual of the penetration parameter observations relative to the deterministic spatial trend field, and to perform Kriging interpolation on the spatial residual to obtain the stochastic residual field and uncertainty information; The data fusion module 340 is used to fuse the deterministic spatial trend field and the stochastic residual field to obtain the spatial distribution field of the penetration parameters; The result generation module 350 is used to obtain the dynamic quantification result of seepage risk based on the spatial distribution field of the permeability parameters and the uncertainty information using a Bayesian assimilation algorithm.

[0068] It should be noted that although several modules of the dynamic quantification device for seepage risk have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more modules described above can be embodied in a single module or unit. Conversely, the features and functions of a single module described above can be further divided into multiple modules or sub-modules.

[0069] Furthermore, in an exemplary embodiment of the present invention, an electronic device capable of implementing the above-described dynamic quantification method for seepage risk is also provided.

[0070] Those skilled in the art will understand that various aspects of the present invention can be implemented as systems, methods, or program products. Therefore, various aspects of the present invention can be specifically implemented as entirely hardware embodiments, entirely software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, collectively referred to herein as “circuit,” “module,” or “system.”

[0071] The following reference Figure 4 To describe an electronic device 400 according to such an embodiment of the present invention. Figure 4 The electronic device 400 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0072] like Figure 4As shown, the electronic device 400 is presented in the form of a general-purpose computing device. The components of the electronic device 400 may include, but are not limited to: at least one processing unit 410, at least one storage unit 420, a bus 430 connecting different system components (including storage unit 420 and processing unit 410), and a display unit 440.

[0073] The storage unit stores program code that can be executed by the processing unit 410, causing the processing unit 410 to perform the steps described in the "Exemplary Method" section above, based on various exemplary embodiments of the present invention. For example, the processing unit 410 can perform actions such as... Figure 2 As shown in S210, in step S220, in step S220, in step S220, in step S230, in step S240, in step S250, in step S260, in step S270, in step S280, in step S29 ...60, in step S270, in step S280, in step S290, in step S220, in step S260, in step S270, in step S280, in step S290, in step S220, in step S260, in step S270, in step S280, in step S290, in step S220, in step S260, in step S270, in step S280, in step S290, in step S220, in step S260, in step S270, in step S280, in step S290

[0074] Storage unit 420 may include readable media in the form of volatile storage units, such as random access memory (RAM) 421 and / or cache memory 422, and may further include read-only memory (ROM) 423.

[0075] Storage unit 420 may also include a program / utility 424 having a set (at least one) of program modules 425, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0076] Bus 430 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0077] Electronic device 400 can also communicate with one or more external devices 470 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 400, and / or with any device that enables electronic device 400 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 450. Furthermore, electronic device 400 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 460. As shown, network adapter 460 communicates with other modules of electronic device 400 via bus 430. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0078] Through the description of the above embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions of the embodiments of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, portable hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of the present invention.

[0079] In exemplary embodiments of the present invention, a computer-readable storage medium is also provided, on which a program product capable of implementing the methods described above is stored. In some possible embodiments, various aspects of the present invention may also be implemented as a program product comprising program code, which, when the program product is run on a terminal device, causes the terminal device to perform the steps of the various exemplary embodiments of the present invention described in the "Exemplary Methods" section above.

[0080] refer to Figure 5 As shown, a program product 500 for implementing the above-described dynamic quantification method for seepage risk according to an embodiment of the present invention is described. It may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In the present invention, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0081] The program product may employ any combination of one or more readable storage media. Readable storage media may be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0082] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0083] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0084] Through the description of the above embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions of the embodiments of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, portable hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the methods according to the embodiments of the present invention.

[0085] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the claims.

[0086] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for dynamic quantification of seepage risk, characterized in that, The seepage risk dynamic quantification method is applied to seepage risk assessment of a target area, and comprises the following steps: Obtaining observation data of a permeability parameter and geological environmental factor data related to the permeability parameter at different observation points in the target area; Constructing a multi-scale geographically weighted regression model based on the observation data of the permeability parameter and the geological environmental factor data to obtain a deterministic spatial trend field of the permeability parameter; Calculating spatial residuals of the observation value of the permeability parameter relative to the deterministic spatial trend field, and performing Kriging interpolation on the spatial residuals to obtain a random residual field and uncertainty information; Fusing the deterministic spatial trend field and the random residual field to obtain a spatial distribution field of the permeability parameter; Based on the spatial distribution field of the permeability parameter and the uncertainty information, a seepage risk dynamic quantification result is obtained by using a Bayesian assimilation algorithm.

2. The method for dynamic quantification of the filtration risk according to claim 1, characterized in that, The observation data of the permeability parameter includes spatial coordinates; The multi-scale geographically weighted regression model is constructed based on the observation data of the permeability parameter and the geological environmental factor data, which comprises the following steps: based on spatial coordinates of the observation points spatial weights are established to weight the neighborhood samples; The weighted regression is solved under the spatial weight to obtain a local intercept varying with a spatial position and a local regression coefficient corresponding to each geological environment factor ; For any observation point , the product of the measured value of the geologic environmental factor and the corresponding local regression coefficient is added to the local intercept to obtain the trend value of the permeability coefficient ; The deterministic spatial trend field of the permeability coefficient is obtained according to the permeability coefficient trend values of all observation points.

3. The method of dynamic quantification of the filtration risk according to claim 1, characterized in that, The spatial residuals of the observation value of the permeability coefficient relative to the deterministic spatial trend field are calculated, which comprises the following steps: obtaining permeability coefficient observation values at observation points and determining a permeability coefficient trend value corresponding to the spatial coordinates of the observation points in the deterministic spatial trend field ; differencing the permeability observation values from the permeability trend values and using the difference as a spatial residual for the observation point .

4. The method for dynamic quantification of the flow risk according to claim 1, characterized in that, The Kriging interpolation is performed on the spatial residuals to obtain a random residual field and uncertainty information, which comprises the following steps: Based on the spatial residual of the known sample points , the Kriging residual value of the unknown point is obtained by performing linear unbiased optimal estimation on the unknown point , wherein ​ wherein, is a weight coefficient assigned to the i-th sample point, the weight coefficient being determined by solving the Kriging base equation set and under the unbiasedness constraint condition; , On the basis of the weight coefficients, the Kriging variance of the unknown point is calculated as the uncertainty information, wherein wherein is the theoretical semivariogram value between the unknown point and the first sample point, calculated by the theoretical semivariogram function from the distance between the two points; is the Lagrange multiplier introduced to satisfy the unbiasedness constraint; is the kriging residual value at each unknown point constitutes the stochastic residual field, is the kriging variance at each unknown point constitutes the variance field.

5. The method for dynamic quantification of flow risk according to claim 1, characterized in that, The uncertainty information at least includes a Kriging variance field; Based on the spatial distribution field of the permeability parameter and the uncertainty information, a seepage risk dynamic quantification result is obtained by using a Bayesian assimilation algorithm, which comprises the following steps: Based on the spatial distribution field of the permeability parameter, a physical seepage numerical model of the target area is established; The Bayesian data assimilation framework of the physical seepage numerical model is constructed by taking the spatial distribution field of the permeability parameter as a mean value and the Kriging variance field as a variance, as prior information of Bayesian inference; The prior predictive probability distribution of the seepage risk index at different stages is obtained by inputting the design working condition into the Bayesian data assimilation framework of the physical seepage numerical model; Real-time monitored seepage observation data is obtained, and the real-time monitored seepage observation data is taken as an observation to perform Bayesian updating on the prior predictive probability distribution to obtain the posterior probability distribution of the seepage risk index at different stages; The posterior probability distribution is taken as the seepage risk dynamic quantification result and output.

6. The method for dynamic quantification of the filtration risk according to claim 5, characterized in that, Based on the spatial distribution field of the permeability parameter, a physical seepage numerical model of the target area is established, which comprises the following steps: The calculation domain of the target area is meshed to form a plurality of discrete units; The spatial distribution field of the permeability parameter is mapped to the discrete units, so that the permeability coefficient corresponding to each discrete unit is a unit permeability coefficient varying with spatial position; Based on the unit permeability coefficient and Darcy's law, the physical seepage numerical model is obtained on the discrete units.

7. The method for dynamic quantification of the filtration risk according to claim 6, characterized in that, Based on the unit permeability coefficient and Darcy's law, the physical seepage numerical model is obtained on the discrete units, which comprises the following steps: In each discrete unit, the unit permeability coefficient corresponding to the discrete unit is substituted into Darcy's law to establish a seepage flux expression of the discrete unit; The seepage control equations of the discrete units are established in combination with the mass conservation relationship, and the seepage control equations are discretized; The discrete equations of each discrete unit are assembled, and boundary conditions at boundaries of a calculation domain are applied, to obtain the physical seepage numerical model.

8. A seepage flow risk dynamic quantification device, characterized in that, The method comprises the following steps: The data acquisition module is configured to acquire seepage parameter observation data of different observation points in a target region and geological environment factor data related to the seepage parameter; The model construction module is configured to construct a multi-scale geographic weighted regression model based on the seepage parameter observation data and the geological environment factor data, to obtain a deterministic spatial trend field of the seepage parameter; The interpolation module is configured to calculate spatial residual errors of the seepage parameter observation values relative to the deterministic spatial trend field, and to perform Kriging interpolation on the spatial residual errors, to obtain a random residual field and uncertainty information; The data fusion module is configured to fuse the deterministic spatial trend field and the random residual field, to obtain a seepage parameter spatial distribution field; The result generation module is configured to obtain a seepage risk dynamic quantification result by using a Bayesian assimilation algorithm based on the seepage parameter spatial distribution field and the uncertainty information.

9. An electronic device, comprising: The method comprises the following steps: A processor; and A memory having computer readable instructions stored thereon, the computer readable instructions being executed by the processor to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored thereon, the computer program being executed by a processor to implement the method of any one of claims 1 to 7.

Images