Multi-source seepage driven heterogeneous reservoir slope irregular seepage path intelligent inversion method

Abstract

The invention relates to the technical field of reservoir bank slope seepage instability risk assessment, and discloses a multi-source seepage driven heterogeneous reservoir slope irregular seepage path intelligent inversion method, which comprises the following steps: acquiring a characteristic water level and rainfall intensity in a reservoir area range; acquiring a lower bound and an upper bound of a soil body saturated permeability coefficient change range, and jointly establishing a bank slope model calculation parameter database with the basic parameters of the soil body; establishing a reservoir bank slope certainty model; associating the saturated permeability coefficient of the soil body with the field variable; representing the spatial heterogeneity of the saturated permeability coefficient of the model soil body through a random field method; and displaying a seepage result obtained by calculation based on the heterogeneous reservoir bank slope model. The technical problems that in the prior art, an irregular actual seepage path in the reservoir bank slope under the multi-source complex seepage boundary condition is difficult to intelligently and finely describe and invert, the seepage rule in the near-dam reservoir bank slope cannot be effectively explored, and the seepage risk of the region cannot be efficiently evaluated are solved.

Description

Technical Field

[0001] The invention relates to the technical field of reservoir bank slope seepage instability risk assessment, and in particular to an intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage. Background Art

[0002] As the uncertainty of global climate change becomes increasingly prominent, coupled with the extremely uneven temporal and spatial distribution of water resources in my country, extreme weather and climate events such as long-term heavy rainfall have increased significantly, resulting in a significant rise in water levels in the reservoir areas of water conservancy and hydropower projects in some areas. Under the coupling effect of long-term heavy rainfall and sudden changes in reservoir water levels, combined with the complex geological structure conditions in the high mountain canyon area where the project is located, landslide disasters near the dam and reservoir banks are common, threatening the overall safety of the operation of high dams and large reservoirs to a certain extent. The water that infiltrates from the outside of the slope cannot be discharged in a short time after entering the slope. The increase in groundwater level leads to an increase in the pore water pressure of the soil, which in turn changes the original stress field in the slope, which is directly related to the infiltration path of water flow inside it. For the near-dam reservoir bank slope project under water-bearing conditions, its geometric nonlinearity and boundary condition nonlinearity are prominent, which brings great difficulties and challenges to the exploration of the seepage stability problem of this type of slope.

[0003] In addition, compared with other parameters of rock and soil, the permeability coefficient of soil shows more significant spatial variability, which directly affects the flow pattern of the irregular seepage field inside the slope, which makes the response solution related to the slope stability problem have higher uncertainty. Based on this fact, when dealing with the stability problem of reservoir bank slopes under complex water conditions, scholars tend to focus on the permeability of soil, because the seepage field will directly affect the stress field. However, most studies currently use random variable models to characterize the spatial heterogeneity of soil permeability coefficients, which is essentially based on deterministic models for analysis and cannot effectively reflect the spatial correlation characteristics of soil parameters, making it difficult to characterize the actual irregular seepage path inside the slope. The evolution law of seepage inside the heterogeneous near-dam reservoir bank slope under specific multi-source complex seepage boundary conditions needs to be further studied and clarified.

[0004] Based on this, existing methods are difficult to solve this type of problem, and there is an urgent need to develop an intelligent inversion method for irregular infiltration paths inside heterogeneous reservoir bank slopes that can effectively embed multi-source complex seepage boundaries such as reservoir water level changes and heavy rainfall. Summary of the invention

[0005] In order to solve the current technical problems, the main purpose of the present invention is to provide an intelligent inversion method for irregular seepage paths of non-homogeneous reservoir slopes driven by multi-source seepage, so as to solve the technical problem that the existing technology is difficult to intelligently and finely characterize and invert the actual irregular seepage paths inside the reservoir bank slope under multi-source complex seepage boundary conditions, and thus cannot effectively explore the seepage laws inside the reservoir bank slope near the dam and efficiently evaluate the seepage risk in the area.

[0006] The technical solution adopted by the present invention is: an intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage, characterized in that it includes the following steps: S1. Obtain characteristic water level and rainfall intensity within the reservoir area; S2, obtaining the lower and upper limits of the variation range of the soil saturated permeability coefficient, and establishing a calculation parameter database of the reservoir bank slope model together with the basic parameters of the soil; S3. Establish a deterministic model of the reservoir bank slope, set the hydraulic characteristic curves of saturated and unsaturated soil, set multi-source complex seepage boundary conditions, and divide the model unit grid; S4, correlating soil saturated permeability coefficient with field variables; S5. The spatial heterogeneity of the saturated permeability coefficient of the model soil is characterized by the random field method. The permeability coefficient of the unsaturated zone soil is converted by the soil saturation to obtain the numerical seepage model of the heterogeneous reservoir bank slope. S6. Seepage calculation, showing the result cloud map obtained based on the numerical model of seepage of heterogeneous reservoir bank slope, clarifying the evolution law of irregular seepage path inside the slope.

[0007] In S1, based on the accumulated historical hydrological and meteorological monitoring data of the reservoir area, the changing patterns of the monitoring data are mined to determine the characteristic water level and rainfall intensity within the reservoir area as the initial values ​​of the multi-source complex seepage boundary of the numerical model of the reservoir bank slope.

[0008] In S2, by conducting indoor permeability tests on soil samples, the variation range of the saturated permeability coefficient of soil is explored, the lower and upper limits of the variation range of the saturated permeability coefficient are determined, and the calculation parameter database of the reservoir bank slope model is established together with the basic soil parameters.

[0009] In S3, the deterministic model of the reservoir bank slope is established using finite element software. The hydraulic characteristic curves of saturated and unsaturated soil are set based on the hydraulic parameters of the soil. The transient pure seepage type analysis step is established. The multi-source complex seepage boundary conditions of reservoir water level and rainfall infiltration are set. The plane strain four-node quadrilateral pore pressure unit CPE4P is used to refine the unit grid of the slope model. The mathematical expression formula of the hydraulic characteristic curve of saturated and unsaturated soil in the deterministic model of the reservoir bank slope is as follows: (1); In formula (1): S is the soil saturation; i w is the volumetric water content; i s and i r are the saturated volumetric water content and the residual volumetric water content, respectively; u w is the pore water pressure; α w It is the inverse of the air intake value and is related to the air intake pressure of the soil; n w is the pore distribution parameter, which describes the uniformity of soil pore size distribution; m w is the shape parameter; The permeability coefficient of the saturated and unsaturated soil area of ​​the reservoir bank slope model is expressed mathematically as follows: (2); In formula (2): k ( S ) is the permeability coefficient of each area in the model; k s is the saturated permeability coefficient of soil; In the process of changing the external multi-source seepage boundary, the permeability coefficient of the unsaturated soil will change with the change of saturation. k s Obtained by reduction.

[0010] Setting the complex seepage boundary conditions of reservoir water level and rainfall infiltration from multiple sources includes the following steps: Reservoir water level boundary conditions: Use the spatial distribution function to set a hydrostatic pore pressure boundary that increases linearly with depth on the boundaries below the water level on the left and right sides of the model. The calculation formula is as follows: (3); In formula (3): u w is the pore water pressure at boundary nodes below the water level on both sides of the slope model; h i is the water head at the current node; z is the water head at the bottom of the model; c w is the weight of water; Rainfall infiltration boundary condition: The rainfall boundary function is expressed in terms of rainfall intensity.

[0011] In S4, the deterministic model of the reservoir bank slope is exported as a calculation source file, and the calculation source file is modified so that the saturated permeability coefficient of the soil is associated with the custom field variable of the subroutine embedded in the finite element software, so as to change the spatial properties of the saturated permeability coefficient of the soil in the model through secondary development of the subroutine.

[0012] In S5, secondary development is carried out based on the subroutine. In the subroutine custom area, the spatial heterogeneity of the saturated permeability coefficient of the model soil is characterized by adopting the random field method that obeys uniform distribution. The permeability coefficient of the soil in the unsaturated zone is converted by the soil saturation, and then the numerical model of seepage of the heterogeneous reservoir bank slope is obtained.

[0013] The random field generation method of soil saturated permeability coefficient that obeys uniform distribution includes the following steps: First, the KL series expansion method is used to generate a two-dimensional standard normal distribution random field G (u), where u is the spatial position vector, and the mathematical expression is as follows: (4); In formula (4): l i and f i (u) are the eigenvalue and eigenfunction of the autocorrelation function respectively; x i ( ψ ) is an independent standard normal random variable of the material parameter random field, where ψ Represents external space coordinates; set up X (u) is the Beta target random field defined in the probability space, then X (u) can be based on G (u) is obtained through equal probability transformation, and the mathematical expression formula of the transformation relationship is as follows: (5); In formula (5): Φ(·) and F = F X ( x ; p , q ) are the cumulative distribution functions of the standard Gaussian variable and the target Beta variable, respectively, where p and q They are the two shape parameters of Beta distribution; The shape parameter p and q If both are set to 1, the soil saturated permeability random field is uniformly distributed. k (u) can be obtained by the following formula: (6); In formula (6): k min and k max They are the lower and upper limits of the variation range of soil saturated permeability coefficient respectively.

[0014] In S6, the modified INP file and subroutine file are placed in the same file directory, the seepage calculation is submitted through the command stream, the ODB calculation result file is opened, and the custom field variable TEMP is used as the output variable to display the water flow seepage pattern and soil saturation cloud map calculated based on the numerical model of inhomogeneous reservoir bank slope seepage, clarify the evolution law of irregular seepage path inside the slope, and compare it with the corresponding results calculated based on the deterministic slope model.

[0015] The present invention has the following beneficial effects: 1. The present invention can effectively integrate multi-source complex seepage boundaries and soil heterogeneity, and provide technical support for the construction of a random seepage model of a heterogeneous reservoir bank slope under any complex terrain and geology and multi-source seepage conditions.

[0016] 2. The present invention can perform calculations based on the numerical model constructed based on the proposed technical solution on the basis of actual geological surveys and monitoring statistical data of reservoir bank slope engineering, and can intelligently invert the irregular infiltration paths inside the slope under specific multi-source complex seepage conditions, greatly reducing the uncertainty of predicting the infiltration path, and providing an effective way to reduce the risk of landslide disasters near the dam reservoir bank caused by seepage damage.

[0017] 3. The present invention helps to effectively deal with the risk challenges of landslide disasters on the slopes near the dam and reservoir banks under extreme climate backgrounds that affect the overall safety of project operations. It is also suitable for risk assessment of common seepage damage such as pipe bursts in dam projects. It has high universality and good economic, social and ecological benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a flow chart of the present invention.

[0020] Figure 2 It is a schematic diagram of the geometric dimensions and finite element mesh of the reservoir bank slope model according to an embodiment of the present invention.

[0021] Figure 3 The cloud diagrams of water flow seepage patterns obtained by calculating based on the deterministic slope model and the generated random field slope model under the conditions of a reservoir water level of 20 m and a rainfall duration of 48 h in an embodiment of the present invention, wherein Figure a represents the calculation result based on the deterministic slope model, and Figure b represents the calculation result based on the random field slope model.

[0022] Figure 4 The soil saturation cloud maps are obtained by calculating the soil saturation based on the deterministic slope model and the generated random field slope model under the conditions of a reservoir water level of 20 m and a rainfall duration of 48 h in an embodiment of the present invention, wherein Figure a represents the calculation result based on the deterministic slope model, and Figure b represents the calculation result based on the random field slope model. DETAILED DESCRIPTION

[0023] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] See also Figure 1 The embodiment of the present invention provides an intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage, comprising the following steps: S1. Obtain characteristic water level and rainfall intensity within the reservoir area.

[0025] Specifically, in S1, based on the accumulated historical hydrological and meteorological monitoring data of the reservoir area, the changing patterns of the monitoring data are mined to determine the characteristic water level within the reservoir area. H (m) and rainfall intensity q (m / h), as the initial value of the multi-source complex seepage boundary in the numerical model of the reservoir bank slope.

[0026] In the embodiment of the present invention, combined with a certain engineering case, consider H = 20 m, q = 0.012 m / h combined multi-source seepage boundary condition in the reservoir area.

[0027] S2. Obtain the lower and upper limits of the variation range of the soil saturated permeability coefficient, and establish a calculation parameter database for the reservoir bank slope model together with the basic parameters of the soil.

[0028] Specifically, in S2, by conducting indoor permeability tests on a wide range of soil samples, the variation range of the saturated permeability coefficient of the soil was explored and its lower limit was determined. k min and upper bound k max, together with basic parameters such as soil elastic modulus and Poisson's ratio, a reservoir bank slope model calculation parameter database is established.

[0029] S3. Establish a deterministic model of the reservoir bank slope, set the hydraulic characteristic curves of saturated and unsaturated soil, set multi-source complex seepage boundary conditions, and divide the model unit grid.

[0030] Specifically, in S3, see Figure 2 The deterministic model of the reservoir bank slope was established using the finite element software ABAQUS. The hydraulic characteristic curves of saturated and unsaturated soils were set based on the hydraulic parameters of the soil. The transient pure seepage Soils type analysis step was established. The multi-source complex seepage boundary conditions of reservoir water level and rainfall infiltration were set. The displacements of all nodes of the slope model were fixed, and the plane strain four-node quadrilateral pore pressure unit CPE4P was used to refine the unit grid of the slope model.

[0031] The mathematical expression formula of the hydraulic characteristic curve of saturated and unsaturated soil in the deterministic model of the reservoir bank slope is as follows: (1); In formula (1): S is the soil saturation; i w is the volumetric water content; i s and i r are the saturated volumetric water content and the residual volumetric water content, respectively; u w is the pore water pressure; α w It is the inverse of the air intake value and is related to the air intake pressure of the soil; n w is the pore distribution parameter, which describes the uniformity of soil pore size distribution; m w is the shape parameter, m w =1 1 / n w .

[0032] Thus, the permeability coefficient of the saturated and unsaturated soil area of ​​the deterministic model of the reservoir bank slope can be obtained, and the mathematical expression formula is as follows: (2); In formula (2): k ( S ) is the permeability coefficient of each area in the model; k s is the saturated permeability coefficient of soil.

[0033] It can be seen that in the process of changing the external multi-source seepage boundary, the permeability coefficient of the unsaturated soil will change with the change of saturation through k s Obtained by reduction.

[0034] The specific setting method for the complex seepage boundary conditions of reservoir water level and rainfall infiltration is as follows: 1) Reservoir water level boundary condition: Use the spatial distribution function to set a hydrostatic pore pressure boundary that increases linearly with depth on the boundaries below the water level on the left and right sides of the model. The calculation formula is as follows: (3); In formula (3): u w is the pore water pressure at boundary nodes below the water level on both sides of the slope model; h i is the water head at the current node; z is the water head at the bottom of the model; c w The weight of water.

[0035] 2) Rainfall infiltration boundary condition: The rainfall boundary function is based on rainfall intensity, i.e., unit flow q (m / h). The rainfall intensity is set at the seepage boundary at the top of the slope as q , the rainfall intensity on the slope is set to q× cos(30°).

[0036] S4. Correlate the soil saturated permeability coefficient with the field variables.

[0037] The deterministic model of the reservoir bank slope is exported as an INP calculation source file, and the INP file is modified so that the saturated permeability coefficient of the soil is associated with the custom field variable TEMP of the embedded subroutine UTEMP of the finite element software ABAQUS, so that the spatial properties of the saturated permeability coefficient of the soil in the model can be changed through secondary development of the subroutine.

[0038] S5. The spatial heterogeneity of the saturated permeability coefficient of the model soil is characterized by the random field method. The permeability coefficient of the unsaturated zone soil is converted by the soil saturation to obtain the numerical model of seepage of the heterogeneous reservoir bank slope.

[0039] Specifically, in S5, secondary development is carried out based on the subroutine UTEMP. The program code is independently compiled in the subroutine custom area. In view of the strong spatial variability and large numerical span of the soil permeability coefficient, the spatial heterogeneity of the saturated permeability coefficient of the model soil is characterized by using a random field method that obeys a uniform distribution. Based on formula (2), the permeability coefficient of the soil in the unsaturated zone is converted by the soil saturation, and then the numerical model of seepage of the heterogeneous reservoir bank slope is obtained.

[0040] The basic idea of ​​the random field generation method of soil saturated permeability coefficient that obeys uniform distribution is to obtain it through indirect conversion through Beta target random field. The specific steps are as follows: 1) First, the KL series expansion method is used to generate a two-dimensional standard normal distribution random field G (u), where u is the spatial position vector, and the mathematical expression is as follows: (4); In formula (4): l i and f i (u) are the eigenvalue and eigenfunction of the autocorrelation function respectively; x i ( ψ ) is an independent standard normal random variable of the material parameter random field, where ψ Represents external space coordinates; 2) Settings X (u) is the Beta target random field defined in the probability space, then X (u) can be based on G (u) is obtained through equal probability transformation, and the mathematical expression formula of the transformation relationship is as follows: (5); In formula (5): Φ(·) and F = F X ( x ; p , q ) are the cumulative distribution functions of the standard Gaussian variable and the target Beta variable, respectively, where p and q are the two shape parameters of the Beta distribution.

[0041] 3) Change the shape parameter p and q If both are set to 1, the soil saturated permeability random field is uniformly distributed. k (u) can be obtained by the following formula: (6); In formula (6): k min and k max are the lower and upper limits of the variation range of the soil saturated permeability coefficient, respectively. k min and k max 1 respectively × 10–6 m / s and 1 × 10 –4 m / s.

[0042] S6. Seepage calculation, showing the result cloud map obtained based on the numerical model of seepage of heterogeneous reservoir bank slope, clarifying the evolution law of irregular seepage path inside the slope.

[0043] Specifically, in S6, the modified INP file and subroutine file are placed in the same file directory, the seepage calculation is submitted through the command stream, the ODB calculation result file is opened, and the custom field variable TEMP is used as the output variable to display the water flow seepage pattern and soil saturation cloud map calculated based on the numerical model of inhomogeneous bank slope seepage, clarify the evolution law of irregular seepage path inside the slope, and compare it with the corresponding results calculated based on the deterministic slope model, see Figure 3~4 .

[0044] from Figure 3~4 It can be seen that under the same multi-source complex seepage boundary conditions, compared with the deterministic analysis results, due to the consideration of the spatial variability of the soil permeability coefficient, the water flow is more likely to flow along the area with a larger soil permeability coefficient. The water flow infiltration path in the non-homogeneous reservoir bank slope model shows obvious irregular characteristics, and this phenomenon is more significant in the saturated area below the infiltration line. Similarly, in the unsaturated area of ​​the soil, with the deepening of the wetting peak, the saturation of the unsaturated soil also changes unevenly. The above phenomenon is consistent with the seepage conditions in the actual engineering of this type of slope. This embodiment further illustrates the effectiveness of the technical method proposed in the present invention in intelligently and finely characterizing and inverting the irregular seepage paths inside the non-homogeneous reservoir bank slope under multi-source complex seepage boundary conditions. It can be combined with actual geological survey conditions to help improve the reliability of capturing irregular seepage paths, and provide strong technical support for the seepage risk assessment of near-dam reservoir bank slopes in closed geological environments.

[0045] The above embodiments are only used to illustrate the present invention, wherein the structure, connection mode and manufacturing process of each component may be changed. Any equivalent transformations and improvements based on the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims

1. Intelligent inversion method for irregular seepage paths in heterogeneous reservoir slopes driven by multi-source seepage, characterized in that: The following steps are involved: S1. Obtain characteristic water level and rainfall intensity within the reservoir area; S2, obtaining the lower and upper limits of the variation range of the soil saturated permeability coefficient, and establishing a calculation parameter database of the reservoir bank slope model together with the basic parameters of the soil; S3. Establish a deterministic model of the reservoir bank slope, set the hydraulic characteristic curves of saturated and unsaturated soil, set multi-source complex seepage boundary conditions, and divide the model unit grid; S4, correlating soil saturated permeability coefficient with field variables; S5. The spatial heterogeneity of the saturated permeability coefficient of the model soil is characterized by the random field method. The permeability coefficient of the unsaturated zone soil is converted by the soil saturation to obtain the numerical seepage model of the heterogeneous reservoir bank slope. S6. Seepage calculation, showing the result cloud map obtained based on the numerical model of seepage of heterogeneous reservoir bank slope, clarifying the evolution law of irregular seepage path inside the slope.

2. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 1 is characterized in that: In S1, based on the accumulated historical hydrological and meteorological monitoring data of the reservoir area, the changing patterns of the monitoring data are mined to determine the characteristic water level and rainfall intensity within the reservoir area as the initial values ​​of the multi-source complex seepage boundary of the numerical model of the reservoir bank slope.

3. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 1 is characterized in that: In S2, by conducting indoor permeability tests on soil samples, the variation range of the saturated permeability coefficient of soil is explored, the lower and upper limits of the variation range of the saturated permeability coefficient are determined, and the calculation parameter database of the reservoir bank slope model is established together with the basic soil parameters.

4. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 1 is characterized in that: In S3, the deterministic model of the reservoir bank slope is established using finite element software. The hydraulic characteristic curves of saturated and unsaturated soil are set based on the hydraulic parameters of the soil. The transient pure seepage type analysis step is established. The multi-source complex seepage boundary conditions of reservoir water level and rainfall infiltration are set. The plane strain four-node quadrilateral pore pressure unit CPE4P is used to refine the unit grid of the slope model.

5. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 4 is characterized in that: The mathematical expression formula of the hydraulic characteristic curve of saturated and unsaturated soil in the deterministic model of the reservoir bank slope is as follows: （1）； In formula (1): S is the soil saturation; θ w is the volumetric water content; θ s and θ r are the saturated volumetric water content and the residual volumetric water content, respectively; u w is the pore water pressure; α w It is the inverse of the air intake value and is related to the air intake pressure of the soil; n w is the pore distribution parameter, which describes the uniformity of soil pore size distribution; m w is the shape parameter; The permeability coefficient of the saturated and unsaturated soil area of ​​the reservoir bank slope model is expressed mathematically as follows: （2）； In formula (2): k ( S ) is the permeability coefficient of each area in the model; k s is the saturated permeability coefficient of soil; In the process of changing the external multi-source seepage boundary, the permeability coefficient of the unsaturated soil will change with the change of saturation. k s Get by reduction.

6. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 4 is characterized in that: Setting the complex seepage boundary conditions of reservoir water level and rainfall infiltration from multiple sources includes the following steps: 1) Reservoir water level boundary condition: Use the spatial distribution function to set a hydrostatic pore pressure boundary that increases linearly with depth on the boundaries below the water level on the left and right sides of the model. The calculation formula is as follows: （3）； In formula (3): u w is the pore water pressure at boundary nodes below the water level on both sides of the slope model; h i is the water head at the current node; z is the water head at the bottom of the model; γ w is the weight of water; 2) Rainfall infiltration boundary condition: The rainfall boundary function is expressed in terms of rainfall intensity.

7. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 1 is characterized in that: In S4, the deterministic model of the reservoir bank slope is exported as a calculation source file, and the calculation source file is modified so that the saturated permeability coefficient of the soil is associated with the custom field variable of the subroutine embedded in the finite element software, so as to change the spatial properties of the saturated permeability coefficient of the soil in the model through secondary development of the subroutine.

8. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 7 is characterized in that: In S5, secondary development is carried out based on the subroutine. In the subroutine custom area, the spatial heterogeneity of the saturated permeability coefficient of the model soil is characterized by adopting the random field method that obeys uniform distribution. The permeability coefficient of the soil in the unsaturated zone is converted by the soil saturation, and then the numerical model of seepage of the heterogeneous reservoir bank slope is obtained.

9. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 8 is characterized in that: The random field generation method of soil saturated permeability coefficient that obeys uniform distribution includes the following steps: 1) First, the KL series expansion method is used to generate a two-dimensional standard normal distribution random field G (u), where u is the spatial position vector, and the mathematical expression is as follows: （4）； In formula (4): λ i and f i (u) are the eigenvalue and eigenfunction of the autocorrelation function respectively; ξ i ( ψ ) is an independent standard normal random variable of the material parameter random field, where ψ Represents external space coordinates; 2) Settings X (u) is the Beta target random field defined in the probability space, then X (u) can be based on G (u) is obtained through equal probability transformation, and the mathematical expression formula of the transformation relationship is as follows: （5）； In formula (5): Φ(·) and F = F X ( x ; p , q ) are the cumulative distribution functions of the standard Gaussian variable and the target Beta variable, respectively, where p and q They are the two shape parameters of Beta distribution; 3) Change the shape parameter p and q If both are set to 1, the soil saturated permeability random field is uniformly distributed. k (u) can be obtained by the following formula: （6）； In formula (6): k min and k max They are the lower and upper limits of the variation range of soil saturated permeability coefficient respectively.

10. The intelligent inversion method for irregular seepage paths of heterogeneous reservoir slopes driven by multi-source seepage according to claim 8, characterized in that: In S6, the modified INP file and subroutine file are placed in the same file directory, the seepage calculation is submitted through the command stream, the ODB calculation result file is opened, and the custom field variable TEMP is used as the output variable to display the water flow seepage pattern and soil saturation cloud map calculated based on the numerical model of inhomogeneous reservoir bank slope seepage, clarify the evolution law of irregular seepage path inside the slope, and compare it with the corresponding results calculated based on the deterministic slope model.

Images