Data-mechanism dual-driven method and system for predicting earth pressure of fill structure

Abstract

The present application belongs to the technical field of earth pressure calculation of fill structure, and specifically discloses a data-mechanism dual-driven fill structure earth pressure prediction method and system. The method comprises the following steps: obtaining test earth pressure and test covering soil parameters according to fill structure earth pressure test; inputting the test earth pressure into a three-dimensional earth pressure model of the fill structure top covering soil to obtain test stress width of the fill structure and test lateral earth pressure coefficient of the soil body; constructing a first prediction model between the test stress width and the test covering soil parameters and a second prediction model between the test lateral earth pressure coefficient and the test covering soil parameters; obtaining real-time covering soil parameters and inputting the real-time covering soil parameters into the first prediction model and the second prediction model to obtain predicted stress width and predicted lateral earth pressure coefficient; inputting the predicted stress width and the predicted lateral earth pressure coefficient into the three-dimensional earth pressure model to obtain real-time predicted earth pressure. The present application can enhance the interpretability and accuracy of the prediction results of the prediction model.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of earth pressure calculation of fill structure, and discloses a data-mechanism dual-driven fill structure earth pressure prediction method and system. BACKGROUND

[0002] The typical structure in the fill structure is the earth-covered horizontal tank. The earth-covered horizontal tank is highly concerned in the field of storage of dangerous chemicals in petrochemical industry due to its high safety performance, strong stability, and advantages of effectively reducing the hazards of explosion shock wave and radiation. The earth covering technology forms a protective layer by covering the soil, realizes constant temperature corrosion prevention, and reduces evaporation loss, and becomes an important development direction of storage equipment of major hazards. However, the earth-covered horizontal tank technology is still in its infancy, and the related research is focused on the design field. There is a significant gap in the research on the earth pressure response in the settlement process, and the theoretical system needs to be improved to ensure the safety of the project.

[0003] The earth-covered horizontal tank is prone to uneven settlement of the foundation under the action of self-weight and external load, which leads to the destruction of the tank structure and the risk of leakage of dangerous chemicals. The existing researches mostly refer to the earth pressure theory of buried pipelines or tunnels, but there are significant differences between the diameter, material properties and mechanical response of the earth-covered horizontal tank and the above structures, and the direct application of the theory has limitations. Although the traditional mechanical theory can reflect the interaction between soil and structure, it often ignores the three-dimensional soil arching effect and the dynamic evolution of parameters based on simplified assumptions, resulting in insufficient prediction accuracy. In addition, the key parameters of the mechanical theory are difficult to obtain through field monitoring, which further limits the engineering applicability of the mechanical theory. SUMMARY

[0004] The purpose of the present application is to provide a data-mechanism dual-driven fill structure earth pressure prediction method and system to solve the technical problem of insufficient prediction accuracy of the existing fill structure pressure prediction method.

[0005] The first aspect of the present application provides a data-mechanism dual-driven fill structure earth pressure prediction method, comprising:

[0006] Step 1: obtaining the test earth pressure and test earth covering parameters according to the fill structure earth pressure test.

[0007] Step 2: inputting the test earth pressure into the three-dimensional earth pressure model of the fill structure top earth covering to obtain the test stress width of the fill structure and the test lateral earth pressure coefficient of the soil.

[0008] Step 3: constructing a first prediction model between the test stress width and the test earth covering parameters and a second prediction model between the test lateral earth pressure coefficient and the test earth covering parameters.

[0009] Step 4, obtaining real-time covering soil parameters, and inputting the real-time covering soil parameters into the first prediction model and the second prediction model respectively to obtain a predicted stress width and a predicted lateral pressure coefficient.

[0010] Step 5, inputting the predicted stress width and the predicted lateral pressure coefficient into the three-dimensional soil pressure model to obtain a real-time predicted soil pressure.

[0011] Preferably, when the fill structure is a vertical slope fill structure, before step 1, it further comprises:

[0012] Obtaining the mechanical parameters of the covering soil at the top of the fill structure, the burial depth of the top of the fill structure, and the angle between the vertical slope slip surface and the vertical direction.

[0013] According to the mechanical parameters of the covering soil at the top of the fill structure, the burial depth of the top of the fill structure, and the angle between the vertical slope slip surface and the vertical direction, a three-dimensional soil pressure model between the stress width, the lateral pressure coefficient, and the soil pressure is constructed.

[0014] Preferably, when the fill structure is an inclined slope fill structure, before step 1, it further comprises:

[0015] Obtaining a first parameter, the first parameter including the diameter of the fill structure and the base length of the inclined slope reduction triangular load.

[0016] Obtaining a second parameter, the second parameter including the mechanical parameters of the covering soil at the top of the fill structure, the burial depth of the top of the fill structure, and the angle between the vertical slope slip surface and the vertical direction.

[0017] According to the first parameter and the second parameter, a three-dimensional soil pressure model between the stress width, the lateral pressure coefficient, and the soil pressure is constructed.

[0018] Preferably, the mechanical parameters include the internal friction angle of the soil, the specific weight of the soil, and the cohesion of the soil.

[0019] Preferably, when the fill structure is a vertical slope fill structure, the test covering soil parameters and the real-time covering soil parameters both include the settlement length, the settlement height, the settlement width, and the slope angle.

[0020] Preferably, when the fill structure is an inclined slope fill structure, the test covering soil parameters and the real-time covering soil parameters both include the slope distance, the covering soil thickness, the settlement length, the settlement height, the settlement width, and the slope angle.

[0021] Preferably, the first prediction model is a model constructed based on a single gene genetic programming algorithm.

[0022] Preferably, the second prediction model is a model constructed based on a multi-gene genetic programming algorithm.

[0023] The second aspect of the present application provides a data-mechanism dual-driven fill structure soil pressure prediction system, which adopts the data-mechanism dual-driven fill structure soil pressure prediction method and comprises:

[0024] An experimental parameter acquisition module is configured to acquire experimental soil pressure and experimental soil covering parameters according to a fill structure soil pressure test.

[0025] An intermediate parameter determination module is configured to input the experimental soil pressure into a three-dimensional soil pressure model of fill structure top soil covering to obtain a test stress width of the fill structure and a test lateral pressure coefficient of the soil body.

[0026] A modeling module is configured to construct a first prediction model between the test stress width and the test soil covering parameters and a second prediction model between the test lateral pressure coefficient and the test soil covering parameters.

[0027] A parameter input module is configured to acquire real-time soil covering parameters, input the real-time soil covering parameters into the first prediction model and the second prediction model respectively, and obtain a predicted stress width and a predicted lateral pressure coefficient.

[0028] A prediction module is configured to input the predicted stress width and the predicted lateral pressure coefficient into the three-dimensional soil pressure model to obtain real-time predicted soil pressure.

[0029] Compared with the prior art, the data-mechanism dual-driven fill structure soil pressure prediction method and system have the following beneficial effects:

[0030] The present application derives a three-dimensional soil pressure model of fill structure top soil covering based on the Terzaghi soil pressure theory and the EEMUA 190 soil pressure theory, and determines real-time predicted soil pressure in combination with a first prediction model and a second prediction model constructed based on a genetic programming algorithm. It can be seen that the present application integrates mechanical theory in the modeling link, enhances the explainability of the first prediction model and the second prediction model constructed based on the genetic programming algorithm, and can dynamically update parameter values when facing different settlement conditions, thereby improving the accuracy of soil pressure prediction. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 The flowchart of the data-mechanism dual-driven fill structure soil pressure prediction method of the present application embodiment.

[0032] Figure 2 The soil pressure test structure diagram corresponding to the vertical slope horizontal tank under the action of uneven settlement of the foundation in the present application embodiment.

[0033] Figure 3The parameter inversion flowchart in the embodiment of the present application.

[0034] Figure 4 (a) in FIG. 8 is a fitting result of the predicted stress width of the vertical slope tank corresponding to the training data; and (b) in FIG. 8 is a fitting result of the predicted stress width of the vertical slope tank corresponding to the test data.

[0035] Figure 5 (a) in FIG. 9 is a fitting result of the predicted stress width of the inclined slope tank corresponding to the training data; and (b) in FIG. 9 is a fitting result of the predicted stress width of the inclined slope tank corresponding to the test data.

[0036] Figure 6 (a) in FIG. 8 is a fitting result of the predicted lateral pressure coefficient of the vertical slope tank corresponding to the training data; and (b) in FIG. 8 is a fitting result of the predicted lateral pressure coefficient of the vertical slope tank corresponding to the training data.

[0037] Figure 7 (a) in FIG. 9 is a fitting result of the predicted lateral pressure coefficient of the inclined slope tank corresponding to the test data; and (b) in FIG. 9 is a fitting result of the predicted lateral pressure coefficient of the inclined slope tank corresponding to the test data.

[0038] Figure 8 The earth pressure fitting result of the vertical slope tank in the embodiment of the present application.

[0039] Figure 9 The earth pressure fitting result of the inclined slope tank in the embodiment of the present application. DETAILED DESCRIPTION

[0040] In the following description, specific details are set forth, such as a particular system architecture, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. However, persons skilled in the art will understand that the present application can be practiced without such specific details. In other instances, well-known structures, devices, circuits, and methods have not been described in detail in order to avoid obscuring the present application.

[0041] A first aspect of the embodiment of the present application provides a data-mechanism dual-driven earth pressure prediction method for a fill structure, as shown in FIG. 1, comprising the following steps. Figure 1

[0042] Step 1: Obtain the test earth pressure and test earth parameters according to the fill structure earth pressure test.

[0043] ​The filling structure of the embodiment of the present application can be a vertical side slope filling structure or an inclined side slope filling structure, and the soil pressure test of the filling structure under the uneven settlement of the foundation is carried out for the vertical side slope filling structure and the inclined side slope filling structure respectively, so that the test soil pressure corresponding to the test covering parameter of each type of filling structure is obtained. The monitoring data of 6 measuring points of the filling structure can be selected in the embodiment of the present application.

[0044] The filling structure of the embodiment of the present application can be a covering horizontal tank, and as shown in the example, Figure 2 When the horizontal tank is a vertical side slope horizontal tank, the test covering parameters include the settlement length, the settlement height, the settlement width and the slope angle, and then the soil pressure test of the horizontal tank under the uneven settlement of the foundation is carried out, so that the test soil pressure of the vertical side slope horizontal tank under different settlement lengths, settlement heights, settlement widths and slope angles can be obtained. When the horizontal tank is an inclined side slope horizontal tank, the test covering parameters include the slope distance, the covering thickness, the settlement length, the settlement height, the settlement width and the slope angle, and then the soil pressure test of the horizontal tank under the uneven settlement of the foundation is carried out, so that the test soil pressure of the inclined side slope horizontal tank under different slope distances, covering thicknesses, settlement lengths, settlement heights, settlement widths and slope angles can be obtained.

[0045] Step 2, input the test soil pressure into the three-dimensional soil pressure model of the top covering soil of the filling structure, to obtain the test stress width of the filling structure and the test lateral pressure coefficient of the soil body.

[0046] The embodiment of the present application combines the Terzaghi soil pressure theory and the EEMUA 190 soil pressure theory, considers the difference in soil arching effect between the settlement area and the non-settlement area, deduces the superposition formula of the transverse cross-section negative arching effect and the longitudinal cross-section positive arching effect, decomposes the soil pressure of the top covering soil of the filling structure into the stress superposition of the transverse cross-section and the longitudinal cross-section, introduces the dynamic change of the stress width and the lateral pressure coefficient in the transverse cross-section calculation, and adopts the limit state lateral pressure coefficient in the longitudinal cross-section; then, the three-dimensional soil pressure model of the vertical side slope and the three-dimensional soil pressure model of the inclined side slope are established for the vertical side slope filling structure and the inclined side slope filling structure respectively, and the three-dimensional soil pressure model of the inclined side slope introduces the slope distance and the slope angle to correct the soil pressure distribution.

[0047] When the filling structure is a vertical side slope filling structure, before step 1, the mechanical parameters of the top covering soil of the filling structure, the top buried depth of the filling structure and the angle between the vertical side slope sliding surface and the vertical direction are obtained, and then the three-dimensional soil pressure model between the stress width, the lateral pressure coefficient and the soil pressure is constructed according to the mechanical parameters of the top covering soil of the filling structure, the top buried depth of the filling structure and the angle between the vertical side slope sliding surface and the vertical direction. The mechanical parameters include the internal friction angle of the soil body, the specific gravity of the soil body and the cohesion of the soil body.

[0048] Exemplarily, a three-dimensional earth pressure model of the vertical side slope filling structure is constructed according to the following formula (1).

[0049]

[0050] wherein, σ is the earth pressure, in kN / m 2 ; M = -2c cos β, c is the cohesion of the soil, in kN / m 2 , β is the angle between the slip surface and the vertical direction, specifically the angle between the slip surface of the vertical side slope and the vertical direction, in °; F = 2[tan β + K(tan α cos β - sin β)], K is the lateral pressure coefficient of the cross section, α is the internal friction angle of the soil, in °; A = B + 2Htan β, B is the force width of the filling structure, in m, H is the top buried depth of the filling structure, in m; E = -2tan β; γ is the unit weight of the soil, in kN / m 3 ; h is the height of the upper part of the micro-unit from the soil surface, in m.

[0051] When the filling structure is an inclined side slope filling structure, before step 1, the following steps are further included: obtaining first parameters, the first parameters including the diameter of the filling structure and the base length of the inclined side slope reduction triangular load; obtaining second parameters, the second parameters including the mechanical parameters of the top soil of the filling structure, the top buried depth of the filling structure and the angle between the slip surface of the inclined side slope and the vertical direction; constructing a three-dimensional earth pressure model between the force width, the lateral pressure coefficient and the earth pressure according to the first parameters and the second parameters. The mechanical parameters include the internal friction angle of the soil, the unit weight of the soil and the cohesion of the soil.

[0052] Exemplarily, a three-dimensional earth pressure model of the inclined side slope filling structure is constructed according to the following formula (2).

[0053]

[0054] wherein, σ is the earth pressure, in kN / m 2 ; M = -2c cos β, c is the cohesion of the soil, in kN / m 2 , β is the angle between the slip surface and the vertical direction, specifically the angle between the slip surface of the inclined side slope and the vertical direction, in °; F = 2[tan β + K(tan α cos β - sin β)], K is the lateral pressure coefficient of the cross section, α is the internal friction angle of the soil, in °; A = B + 2Htan β, B is the force width of the filling structure, in m, H is the top buried depth of the filling structure, in m; E = -2tan β; γ is the unit weight of the soil, in kN / m 3; h is the height of the upper part of the unit from the surface of the soil layer, in meters; D is the diameter of the filling structure, in meters; L2 is the base length of the inclined slope reduction triangular load, in meters, L2 = B + 1 / 3H - L1, L1 is the distance from the top of the filling structure to the inclined slope, in meters; i is the inclination angle of the inclined slope, in degrees.

[0055] Further, the test earth pressure is input into the corresponding three-dimensional earth pressure model of the topsoil of the filling structure, and the test stress width of the filling structure and the test lateral pressure coefficient of the soil corresponding to each test earth pressure are obtained through programming parameter inversion. The flow of the parameter inversion is shown in Figure 3 .

[0056] Specifically, the topsoil of the filling structure is divided into a plurality of uniform regions J, and the test stress width and the test lateral pressure coefficient corresponding to the test earth pressure of each region are determined one by one.

[0057] In the case of J = 1, first, the basic parameters (when it is a vertical slope filling structure, the basic parameters are the corresponding topsoil parameters and mechanical parameters, the topsoil depth of the filling structure, and the angle between the vertical slope slip surface and the vertical direction, etc.; when it is an inclined slope filling structure, the basic parameters are the corresponding topsoil parameters, mechanical parameters, the topsoil depth of the filling structure, the angle between the vertical slope slip surface and the vertical direction, the diameter of the filling structure, and the base length of the inclined slope reduction triangular load, etc.) are input. Then, the test earth pressure of the region J is input, and the specific gravity γ J of the longitudinal section is solved. It is further judged whether the test earth pressure of the region J is greater than 0, and if yes, the cross section is a negative arch, at this time, the test lateral pressure coefficient K

[0058] K J = K max , wherein K max is the upper limit threshold of the test lateral pressure coefficient, because the value of K J is determined, the test stress width B J of the region J cross section is solved by using the corresponding three-dimensional earth pressure model. If B J < B max (B max is the upper limit threshold of the test stress width), B J and K J are output, the test stress width and the test lateral pressure coefficient of the next region are calculated, and this process is repeated until the last region. If B J ≥ B max , the cross section is a negative arch, B J = B max , and B J is input into the corresponding three-dimensional earth pressure model to re-solve the cross section K JWhen 0 < K J < K max , output B J and K J , and perform the calculation of the test stress width and the test lateral pressure coefficient of the next region until the last region. When K J does not belong to the above range, the cross-section normal arch is B J =B max , K J is solved again and the judgment is performed again.

[0059] When the test earth pressure of the region J is less than or equal to 0, the cross-section normal arch is B J =B max , because the value of B J is determined, the test lateral pressure coefficient K J of the region J cross-section can be solved by using the corresponding three-dimensional earth pressure model. When 0 < K J < K max , output B J and K J , and perform the calculation of the test stress width and the test lateral pressure coefficient of the next region until the last region. When K J does not belong to the above range, the cross-section normal arch is B J =B max , K J is solved again and the judgment is performed again.

[0060] Step 3, constructing a first prediction model between the test stress width and the test covering soil parameters and a second prediction model between the test lateral pressure coefficient and the test covering soil parameters.

[0061] The embodiment of the present application constructs a training data set and a test data set by using the test covering soil parameters and the corresponding test stress width obtained by inversion, trains the first prediction model by using the training data set, and tests the effectiveness of the first prediction model trained by using the test data set.

[0062] The first prediction model of the embodiment of the present application is a model constructed based on a single gene genetic algorithm.

[0063] Exemplarily, when the filling structure is a vertical side slope filling structure, the first prediction model between the test stress width and the test covering soil parameters is shown in formula (3).

[0064]

[0065] In the formula, x1 is the position of the measuring point, x2 is the settlement length (m), x3 is the settlement height (m), x4 is the settlement width (m), x5 is the slope angle (°), and B is the test stress width (m).

[0066] When the fill structure is an inclined slope fill structure, the first prediction model between the test stress width and the test covering soil parameters is shown in formula (4).

[0067]

[0068] In the formula, x1 is the position of the measuring point; x2 is the settlement length (m); x3 is the settlement height (m); x4 is the settlement width (m); x5 is the slope angle (°); x6 is the slope distance (m); x7 is the covering soil thickness (m); and B is the test stress width (m).

[0069] The embodiment of the present application utilizes the test covering soil parameters and the corresponding test lateral pressure coefficients obtained by inversion to construct a training data set and a test data set, utilizes the training data set to train a second prediction model, and utilizes the test data set to test the effectiveness of the second prediction model after training.

[0070] The second prediction model of the embodiment of the present application is a model constructed based on a multi-gene genetic algorithm. The distribution characteristics based on the lateral pressure coefficient K are selected: as shown in formula (5), K is affected by the high-dimensional nonlinear coupling of the covering soil parameters, and presents a strong nonlinear response characteristic. The single-gene genetic algorithm is prone to overfitting and local convergence problems when processing such a complex model, and the multi-gene genetic algorithm can decompose it into multiple gene codes for independent optimization and weighted output, significantly improving the compatibility of the model to nonlinear characteristics, and being more suitable for modeling of such high-dimensional nonlinear parameters.

[0071] Exemplarily, when the fill structure is a vertical slope fill structure, the second prediction model between the test lateral pressure coefficient and the test covering soil parameters is shown in formula (5).

[0072]

[0073] In the formula, x1 is the position of the measuring point; x2 is the settlement length (m); x3 is the settlement height (m); x4 is the settlement width (m); x5 is the slope angle (°); and K is the test lateral pressure coefficient.

[0074] When the fill structure is an inclined slope fill structure, the second prediction model between the test lateral pressure coefficient and the test covering soil parameters is shown in formula (6).

[0075]

[0076] In the formula, x1 is the position of the measuring point; x2 is the settlement length (m); x3 is the settlement height (m); x4 is the settlement width (m); x5 is the slope angle (°); x6 is the slope distance (m); x7 is the covering soil thickness (m); and K is the test lateral pressure coefficient.

[0077] Step 4, obtaining real-time covering soil parameters, and inputting the real-time covering soil parameters into the first prediction model and the second prediction model respectively to obtain a predicted stress width and a predicted lateral pressure coefficient.

[0078] In the embodiment of the present application, when the fill structure is a vertical slope fill structure, the real-time covering soil parameters include a settlement length, a settlement height, a settlement width and a slope angle.

[0079] When the fill structure is an inclined slope fill structure, the real-time covering soil parameters include a slope distance, a covering soil thickness, a settlement length, a settlement height, a settlement width and a slope angle.

[0080] Step 5, inputting the predicted stress width and the predicted lateral pressure coefficient into the three-dimensional soil pressure model to obtain a real-time predicted soil pressure.

[0081] The present application derives a three-dimensional soil pressure model of the top covering soil of the fill structure on the basis of the Terzaghi soil pressure theory and the EEMUA 190 soil pressure theory, and determines a real-time predicted soil pressure in combination with the first prediction model and the second prediction model constructed based on the genetic programming algorithm. It can be seen that the present application integrates the mechanical theory in the modeling link, enhances the explainability of the first prediction model and the second prediction model constructed based on the genetic programming algorithm, and can dynamically update the parameter values when facing different settlement conditions, thereby improving the accuracy of soil pressure prediction and having strong generalization ability.

[0082] The second aspect of the embodiment of the present application provides a data-mechanism dual-driven fill structure covering soil pressure prediction system based on the above-mentioned prediction method, which comprises a test parameter acquisition module, an intermediate parameter determination module, a modeling module, a parameter input module and a prediction module.

[0083] The test parameter acquisition module is used to acquire test soil pressure and test covering soil parameters according to a fill structure soil pressure test.

[0084] The intermediate parameter determination module is used to input the test soil pressure into the three-dimensional soil pressure model of the top covering soil of the fill structure to obtain a test stress width of the fill structure and a test lateral pressure coefficient of the soil body.

[0085] The modeling module is used to construct a first prediction model between the test stress width and the test covering soil parameters and a second prediction model between the test lateral pressure coefficient and the test covering soil parameters.

[0086] The parameter input module is used to acquire real-time covering soil parameters, and input the real-time covering soil parameters into the first prediction model and the second prediction model respectively to obtain a predicted stress width and a predicted lateral pressure coefficient.

[0087] The prediction module is used to input the predicted stress width and the predicted lateral pressure coefficient into the three-dimensional soil pressure model to obtain a real-time predicted soil pressure.

[0088] The present application mines the parameter mapping relationship in the test data and inverses the parameter evolution law through the genetic programming algorithm, and constructs an explicit prediction model of the test stress width and the test lateral pressure coefficient; then the genetic programming algorithm is coupled with the traditional three-dimensional mechanics theory to form a soil pressure prediction method and system with the advantages of mechanical mechanism and data driving, which significantly improves the prediction accuracy and interpretability, and provides a new idea for the safety evaluation of earth filling structure.

[0089] In order to clearly illustrate the advantages of the method of the present application, taking the earth filling structure as a lying tank for example, the prediction results of the method of the present application and the prediction results of the traditional prediction model based on genetic programming algorithm are compared, and the results are shown in Tables 1 to 2 and Figure 4 to Figure 9

[0090] Table 1 Comparison of soil pressure prediction results of vertical slope lying tank

[0091]

[0092] Table 2 Comparison of soil pressure prediction results of inclined slope lying tank

[0093]

[0094] Among the vertical slope lying tank, the soil pressure measurement point position, the earth thickness, and the lying tank diameter are the top three influencing parameters of the genetic programming algorithm model parameters; among the inclined slope lying tank, the soil pressure measurement point position, the slope distance, and the earth thickness are the top three influencing parameters of the genetic programming algorithm model parameters. The above sensitivity analysis results are used to optimize the model input parameters, eliminate redundant variables, and improve the prediction efficiency.

[0095] As shown in Tables 1 and 2, under the conditions of vertical slope lying tank and inclined slope lying tank, the soil pressure prediction results of the present application show more consistent trends with the test data and smaller prediction error values than the prediction results of the traditional prediction model based on genetic programming algorithm, and have better generalization ability and prediction accuracy.

[0096] The above is only a few embodiments of the present application, and does not limit the present application in any form. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any skilled person in the art can make some changes or modifications to the above disclosed technical content without departing from the scope of the technical solution of the present application, which are equivalent to equivalent embodiments, and all belong to the scope of the technical solution.​

Claims

1. A data-mechanism dual-driven fill structure earth pressure prediction method, characterized by, The application relates to a data-mechanism dual-driven fill structure soil pressure prediction method. Step 1: obtaining test soil pressure and test covering soil parameters according to a fill structure soil pressure test; Step 2: inputting the test soil pressure into a three-dimensional soil pressure model of top covering soil of the fill structure to obtain test stress width of the fill structure and test lateral pressure coefficient of the soil body through parameter inversion; Step 3: constructing a first prediction model between the test stress width and the test covering soil parameters and a second prediction model between the test lateral pressure coefficient and the test covering soil parameters; Step 4: obtaining real-time covering soil parameters and inputting the real-time covering soil parameters into the first prediction model and the second prediction model to obtain predicted stress width and predicted lateral pressure coefficient; Step 5: inputting the predicted stress width and the predicted lateral pressure coefficient into the three-dimensional soil pressure model to obtain real-time predicted soil pressure. When the fill structure is a vertical slope fill structure, the three-dimensional soil pressure model of the vertical slope fill structure is constructed according to formula (1) as follows: wherein σ is the earth pressure, in kN / m 2 ; M = -2c cos β, c is the cohesion of the soil, in kN / m 2 ; β is the angle between the slip surface and the vertical direction, specifically the angle between the vertical direction and the slip surface of the side slope, in °; F = 2[tan β + K(tan α cos β - sin β)], K is the lateral pressure coefficient of the cross section, α is the internal friction angle of the soil, in °; A = B + 2H tan β, B is the force bearing width of the filling structure, in m, H is the top buried depth of the filling structure, in m; E = -2 tan β; γ is the unit weight of the soil, in kN / m 3 ; h is the height of the upper part of the micro-unit from the surface of the soil layer, in m; When the fill structure is an inclined slope fill structure, the three-dimensional soil pressure model of the inclined slope fill structure is constructed according to formula (2) as follows: In the formula, D is the diameter of the filling structure, in meters; L2 is the base length of the inclined slope reduction triangle load, in meters, L2=B+1 / 3H-L1, L1 is the distance from the top of the filling structure to the inclined slope, in meters; i is the inclination angle of the inclined slope, in °.

2. The data-mechanism dual-driven fill structure earth pressure prediction method according to claim 1, characterized by, When the fill structure is a vertical slope fill structure, the test covering soil parameters and the real-time covering soil parameters both include settlement length, settlement height, settlement width and slope angle.

3. The data-mechanism dual-driven fill structure earth pressure prediction method according to claim 1, characterized by, When the fill structure is an inclined slope fill structure, the test covering soil parameters and the real-time covering soil parameters both include slope distance, covering soil thickness, settlement length, settlement height, settlement width and slope angle.

4. The data-mechanism dual-driven fill structure earth pressure prediction method according to claim 1, characterized by, The first prediction model is a model constructed based on a single gene genetic programming algorithm.

5. The data-mechanism dual-driven fill structure earth pressure prediction method according to claim 1, characterized by, The second prediction model is a model constructed based on a multi-gene genetic programming algorithm.

6. A data-mechanism dual-driven fill structure earth pressure prediction system, characterized by, The data-mechanism dual-driven fill structure soil pressure prediction method according to any one of claims 1-5 comprises: a test parameter acquisition module configured to obtain test soil pressure and test covering soil parameters according to a fill structure soil pressure test; an intermediate parameter determination module configured to input the test soil pressure into a three-dimensional soil pressure model of top covering soil of the fill structure to obtain test stress width of the fill structure and test lateral pressure coefficient of the soil body; a modeling module configured to construct a first prediction model between the test stress width and the test covering soil parameters and a second prediction model between the test lateral pressure coefficient and the test covering soil parameters; a parameter input module configured to obtain real-time covering soil parameters and input the real-time covering soil parameters into the first prediction model and the second prediction model to obtain predicted stress width and predicted lateral pressure coefficient; a prediction module configured to input the predicted stress width and the predicted lateral pressure coefficient into the three-dimensional soil pressure model to obtain real-time predicted soil pressure.

Images