Uplift pile bearing capacity prediction method and system

Abstract

The invention provides an uplift pile bearing capacity prediction method and system, and belongs to the field of uplift pile bearing capacity prediction. The technical problem to be solved is that the method comprises the following steps: acquiring parameters including pile parameters, soil body parameters and interface parameters; numerical value calibration and sensitivity identification: establishing a three-dimensional / axisymmetric pile-soil finite element example library, obtaining a load-displacement curve and displacement / plastic zone distribution for mechanism identification and evaluation of the influence of the obtained parameters on the net uplift bearing capacity, and identifying a key control quantity; taking test pile / historical engineering data as a true value, and performing regression setting of model parameters according to an objective function RMSE; analyzing, modeling and solving, including assumed conditions of the model, modeling of a failure surface and expression of an equilibrium equation and bearing capacity, and finally obtaining a net uplift bearing capacity analytic expression; checking and outputting the structure; according to the method, the pile-soil coupling effect and the multi-parameter sensitivity are comprehensively considered, so that the prediction precision and the engineering applicability are improved while the safety is ensured.

Description

Technical Field

[0001] This application relates to the field of predicting the bearing capacity of tensile piles, and in particular to a method and system for predicting the bearing capacity of tensile piles. Background Technology

[0002] For structures such as power transmission towers that are subjected to significant uplift loads over long periods, anti-uplift piles are typically used as the primary foundation type in engineering to meet the requirements for uplift bearing capacity and overall stability. Current design methods for assessing the bearing capacity of such foundations largely rely on empirical formulas or standard table values. These methods are simple to calculate, widely applicable, and provide some reference value when on-site testing is lacking.

[0003] However, existing methods for calculating tensile bearing capacity still have the following shortcomings:

[0004] 1) Insufficient consideration of coupling mechanism: Most methods mainly rely on the piecewise linear superposition of pile side friction, which does not fully reflect the nonlinear laws of pile-soil interaction, stress redistribution and side resistance mobilization, and it is difficult to accurately describe the characteristics of bearing capacity growth when the length-to-diameter ratio increases.

[0005] 2) Unsystematic parameter identification: Under the condition of multi-parameter coupling of pile-soil system, there is still a lack of systematic sensitivity analysis and parameter screening mechanism. The influence of key control parameters (such as pile length, pile diameter, interface friction characteristics, soil strength index, etc.) has not been effectively quantified, resulting in insufficient robustness and transferability of the model.

[0006] 3) Large deviation from actual measurement: Engineering practice shows that the tensile bearing capacity calculated by the current method is systematically conservative under various working conditions, which can easily lead to redundancy in materials and construction period, affecting the overall economy and resource utilization efficiency.

[0007] 4) Unclear application boundaries: Existing methods are incomplete in terms of their applicability and correction strategies for different soil types, construction methods and water content conditions, making it difficult to meet the needs of refined prediction under complex hydrogeological and layered foundation conditions.

[0008] Therefore, there is an urgent need for a precise prediction method and implementation system for pull-out bearing capacity that comprehensively considers pile-soil coupling and multi-parameter sensitivity, so as to improve prediction accuracy and engineering applicability while ensuring safety and reducing unnecessary resource waste caused by conservatism. Summary of the Invention

[0009] To address the aforementioned technical problems, this application proposes a method and system for predicting the bearing capacity of pull-out piles.

[0010] The technical solution adopted in this application is: a method for predicting the bearing capacity of tension piles, including the following steps:

[0011] S1: Parameter acquisition, including pile parameters, soil parameters and pile-soil interface parameters;

[0012] S2: Numerical calibration and sensitivity identification, including:

[0013] S2.1: Establish a three-dimensional / axisymmetric pile-soil finite element example library to obtain load-displacement curves and displacement / plastic zone distribution for mechanism identification;

[0014] S2.2: Evaluate the impact of the acquired parameters on the net tensile bearing capacity and identify key control quantities;

[0015] S2.3: Using test pile / historical engineering data as the true value, perform regression tuning of model parameters according to the objective function RMSE;

[0016] S3: Analytical modeling and solution, including the assumptions of the model, modeling of the failure surface, and expression of the equilibrium equation and bearing capacity, and finally deriving the analytical formula of the net tensile bearing capacity;

[0017] S4: Structural verification and output.

[0018] Furthermore, the pile parameters in step S1 include the pile diameter. Pile length Elastic modulus of pile Soil parameters include cohesion internal friction angle ,density Soil elastic modulus The pile-soil interface parameters include the friction angle. ;

[0019] Step S1 specifically includes:

[0020] S1.1: Collect soil mechanics and pile-soil interface parameters , , , , ;

[0021] S1.2: Collect pile body parameters , , Select the hole forming method and record the interface roughness level;

[0022] S1.3: Establish a "local parameter library" as the calculation input and verification baseline.

[0023] Furthermore, in step S2.2, the single-factor variation method combined with root mean square sensitivity is used to evaluate the influence of parameters on net pull-out bearing capacity.

[0024] Furthermore, the key control variables identified in step S2.2 are: , For highly sensitive control quantities; , , For medium-sensitivity control; , It is a low-sensitivity control quantity.

[0025] Furthermore, the assumptions made about the model in step S3 include:

[0026] (1) The soil is taken as isotropic and homogeneous linear elastic-ideal plastic;

[0027] (2) The pile body is linearly elastic; the pile-soil interface is normal hard contact - tangential Coulomb friction is used as the pile-soil interface friction angle δ;

[0028] (3) Axisymmetric simplification: with the bottom of the pile as the origin, z is the axis upward along the pile, z∈[0,L], and z=L is the ground surface.

[0029] Furthermore, the failure surface model includes:

[0030] Surface damage angle: ;

[0031] Depth distribution of local damage angle: ;

[0032] Explicit destruction curve: ;

[0033] In the formula: This represents the attenuation coefficient.

[0034] Furthermore, the equilibrium equations and the expression of bearing capacity include:

[0035] Shear strength increment relationship: ;

[0036] Normal equilibrium of soil wedge per unit thickness: ;

[0037] Tangential resistance increment: ;

[0038] In the formula: Indicates the incremental shear resistance at the soil failure surface. Represents the shear strength per unit length. Indicates the incremental length of the damaged surface; This represents the incremental normal force acting on the soil failure surface. This represents the weight per unit thickness on the damaged surface. This indicates the incremental weight per unit thickness.

[0039] Furthermore, the expression for the net tensile strength is as follows:

[0040] ;

[0041] In the formula: Indicates net tensile strength. Indicates density, This is a unified correction for the pile self-weight equivalent term, facilitating design conversion under different λ conditions, where λ is the length-to-diameter ratio of the pile. , , These are the dimensionless integral coefficients of the tangential friction and cohesion terms along the failure surface, respectively.

[0042] Furthermore, the result verification in step S4 is as follows: compare the analytical results with the test pile / historical project data. If the RMSE exceeds the limit, then automatically calibrate according to the regression parameters in S2.

[0043] The output of step S4 includes a bearing capacity calculation report, a parameter sensitivity report, a comparison table of specifications and test piles, and provides recommendations for safety factors and displacement thresholds.

[0044] A system for predicting the bearing capacity of tension piles includes:

[0045] 1) Data acquisition module: used to collect survey and test data and form a parameter library;

[0046] 2) Parameter fitting module: used to complete sensitivity analysis, regression and update of the damage surface / attenuation coefficient, and minimize RMSE;

[0047] 3) Analytical solution module: used to calculate the net pull-out bearing capacity and provide the safety factor under a specified displacement threshold;

[0048] 4) Verification and evaluation module: used to compare with test pile / standard results, form consistency conclusions and automatic calibration records;

[0049] 5) Results output interface: Used to export calculation tables, surface plots, error distribution plots, and reports.

[0050] The advantages of this application over the prior art are as follows:

[0051] 1) Mechanism is explainable and accuracy is improved: Based on the mechanism of "pile-soil coupling + lateral resistance mobilization", the surface failure angle is established. With deep evolution A unified expression that explicitly gives the destruction surface. Compared to the integral method of bearing capacity, and the traditional empirical method of "linear superposition of piecewise side resistance", this method can accurately reproduce the bearing capacity as a function of length-to-diameter ratio. The non-linear growth of the RMSE can be reduced to about 10%, which is significantly better than the standard median method.

[0052] 2) Few parameters with clear physical meaning: the core only involves , , , , , , , Use conventionally measurable parameters and decompose the load-bearing capacity into friction and cohesion terms. This facilitates accounting and verification.

[0053] 3) Closed-form results and efficient calculation: A closed-form formula for net tensile strength is given. It has a short computational link and is easy to embed into software / programs, making it suitable for rapid comparison and batch calculation in engineering projects.

[0054] 4) Self-calibration and robustness: The introduction of RMSE objective function and single-factor sensitivity (OAAT+RMSS) identification supports automatic regression of attenuation coefficient and interface parameters based on test pile / historical data, improving cross-project transferability and result stability.

[0055] 5) Wide applicability: It is preferred for dry drilling and grouting micropiles in unsaturated foundations such as loess / silt; after calibration of the interface friction angle and mobilization coefficient, it can be extended to mud wall drilling piles, precast piles / anchors and other pull-out components.

[0056] 6) Smooth alignment with standards: Standard values ​​and safety factor recommendations are given under a unified displacement method criterion, which can be directly compared with the calculation results of standards such as DL / T 5219, and used to check or correct the overly conservative nature of the design.

[0057] 7) Significant engineering benefits: While ensuring safety, it reduces material and schedule redundancy caused by "systemic underestimation", supports scheme optimization and life cycle cost control; the system can export "calculation sheets / sensitivity reports / comparison tables" for easy review, filing and verification. Attached Figure Description

[0058] The following description, in conjunction with the accompanying drawings, further illustrates this application:

[0059] Figure 1 This is a flowchart of the method in this application (S1~S4 and interfaces).

[0060] Figure 2 A schematic diagram of the pile-soil finite element / computational model (boundary, mesh, initial geostress equilibrium).

[0061] Figure 3 For parameter sensitivity analysis;

[0062] Figure 4 This is a schematic diagram of the failure state of the pile-soil system.

[0063] Figure 5 for Schematic diagram of the surface and fitted data points;

[0064] Figure 6 A block diagram of the load-bearing capacity calculation system (modules and data flow);

[0065] Figure 7 Error distribution in predicting the pull-out bearing capacity of cast-in-place piles (current specifications vs. this method). Detailed Implementation

[0066] like Figures 1 to 7 As shown, this application provides a method for predicting the uplift bearing capacity of micropiles, applicable to the prediction of net uplift bearing capacity of micropiles in loess foundations and similar soil conditions. This method is based on the core principle of "pile-soil coupling and lateral resistance mobilization law," following a process of parameter acquisition → numerical calibration → analytical modeling → result verification, and outputs a standard value of net uplift bearing capacity. It provides safety factors and design recommendations, and can generate calculation sheets, design parameter sensitivity reports, and comparison conclusions with standards / test piles.

[0067] The parameters involved in this application are explained below:

[0068] Pile parameters: Pile diameter (m), pile length (m), pile elastic modulus (MPa).

[0069] Soil and interface parameters: Soil cohesion (kPa), soil internal friction angle (°), soil density (kN / m) 3 ), soil elastic modulus (MPa), pile-soil interface friction angle (°).

[0070] Target value: Standard value of net tensile bearing capacity .

[0071] The specific steps of this method include:

[0072] S1: Parameter acquisition, including:

[0073] S1.1: Collect soil mechanics and pile-soil interface parameters , , , , ;

[0074] S1.2: Collect pile body parameters , (and obtain the aspect ratio) ), The drilling method was dry drilling, and the interface roughness level was recorded.

[0075] S1.3: Establish a "local parameter library" as the calculation input and verification baseline.

[0076] S2: Numerical calibration and sensitivity identification, including:

[0077] S2.1: Establish a three-dimensional / axisymmetric pile-soil finite element example library to obtain load-displacement curves and displacement / plastic zone distributions for mechanism identification (see...). Figure 2 ).

[0078] S2.2: The net tensile strength was evaluated using the single-factor variation method (OAAT) + root mean square sensitivity (RMSS) method. The impact of these factors helps identify key control variables: , Highly sensitive; , , Medium sensitivity; , The impact is secondary (see Figure 3 ).

[0079] S2.3: Using test pile / historical engineering data as the true value, perform regression tuning of model parameters (attenuation coefficient of failure surface, etc.) according to the objective function RMSE; the form of the objective function is shown in equation (A-11).

[0080] S3: Modeling and solving of the pile-soil coupling analytical model, including:

[0081] S3.1: Under the assumptions of axisymmetry and unit thickness soil wedge, define: with the pile bottom as the origin, the upward axial direction is positive. The radius of the destruction curve is The angle between the local damaged surface and the horizontal is .

[0082] S3.2: Surface Failure Angle and , The unified fitting relationship is shown in equation (A-1) (corresponding to Figure 5 (the curved surface).

[0083] S3.3: The evolution of the local failure angle depth and the explicit formula of the failure surface are shown in equations (A-2) and (A-3), respectively.

[0084] S3.4: Establishing tangential / normal equilibrium along the failure surface yields the incremental and total integral formulas for bearing capacity (Equations (A-4)~(A-9)); combining these, we obtain the analytical formula for net tensile bearing capacity:

[0085] (See equation (A-10)).

[0086] in , , These are the dimensionless integral coefficients of the tangential friction and cohesion terms along the failure surface, respectively (see equation (A-9)).

[0087] S4: Result verification and output, including:

[0088] S4.1: Compare the analytical results with the test pile / historical project data. If the RMSE exceeds the limit, automatically calibrate according to the regression parameters of S2 (see formula (A-11)).

[0089] S4.2: Generate "Bearing Capacity Calculation Sheet, Parameter Sensitivity Report, and Standard / Test Pile Comparison Table", and provide recommendations for safety factor and displacement threshold.

[0090] The analytical model is derived below:

[0091] A.1 Basic Assumptions and Coordinate System

[0092] A.1.1 Assumptions

[0093] (1) The soil is taken as isotropic and homogeneous linear elastic-ideal plastic (Mohr-Coulomb yield).

[0094] (2) The pile body is linearly elastic; the pile-soil interface is normal hard contact-tangential Coulomb friction, and the friction angle is denoted as δ;

[0095] (3) Axisymmetric simplification: with the bottom of the pile as the origin, the direction upward along the pile axis is z (z∈[0,L], z=L at the ground surface);

[0096] A.1.2 Coordinates and Symbols

[0097] (1) Taking the bottom of the pile as the origin, the direction upward along the pile axis is z (z∈[0,L]), and the ground surface z=L;

[0098] (2) The failure curve x = x(z) (axisymmetric radius), and the angle θ(z) between the local failure surface and the horizontal.

[0099] (3) Pile diameter d, pile length L, slenderness ratio λ = L / d;

[0100] (4) Soil cohesion c, internal friction angle Density γ, interfacial friction angle δ.

[0101] A.2 Failure surface model, including:

[0102] A.2.1 Surface Failure Angle

[0103] (A-1)

[0104] Equation (A-1) reveals that the surface failure angle is simultaneously affected by the internal friction angle. Controlled by the slenderness ratio λ: as λ increases, A reasonable reduction is more in line with the mobilization rules of medium and long piles.

[0105] A.2.2 Depth Distribution of Local Failure Angle

[0106] (A-2)

[0107] Attenuation coefficient recommendation (Angles are measured in degrees; the rougher the interface, the larger δ, and the smaller β).

[0108] A.2.3 Explicit Failure Curve

[0109] Integrating over (A-2) and taking x(0) = 0, we get:

[0110] (A-3)

[0111] A.3 Equilibrium Equations and Bearing Capacity Expression

[0112] A.3.1 Shear strength increment relationship

[0113] (A-4)

[0114] A.3.2 Normal equilibrium of soil wedge per unit thickness

[0115] (A-5)

[0116] A.3.3 Increment of tangential resistance

[0117] (A-6)

[0118] A.3.4 Differential Expression of Bearing Capacity

[0119] Combining equations (A-4) to (A-6) with geometric relations, we get:

[0120] (A-7)

[0121] A.3.5 Total Bearing Capacity and Coefficient Decomposition

[0122] (A-8)

[0123] The integrals of A1 and A2 are defined as follows:

[0124] ;

[0125] (A-19)

[0126] A.4 Net tensile strength

[0127] (A-10)

[0128] Note: 1 / (4λ) is a unified correction for the pile self-weight equivalent term, which facilitates design conversion under different λ conditions.

[0129] A.5 Parameter Calibration and Numerical Verification (Engineering Specifications)

[0130] Data source: pile testing / on-site or indoor model tests;

[0131] Parameter to be calibrated: β or equivalent The correction coefficient and the range of values ​​for the interfacial friction angle δ;

[0132] Objective function: Minimize EMSE

[0133] (A-11)

[0134] Termination criteria: RMSE decline rate <1% or reaching the threshold (e.g., 10–12%).

[0135] Output: β and δ after regression, and the applicable λ, The range of c.

[0136] A.6 Scope of Application and Boundary Conditions

[0137] Pile type / construction: Drilled micropiles, dry operation or equivalent rough interface;

[0138] Slenderness ratio: 10 ≤ λ ≤ 40;

[0139] Soil types: loess, silt, and sandy soil; clay soil can be used after S2 / S4 regeneration.

[0140] Interface parameters: 0.6 ≤δ / φ≤1.0;

[0141] Deviation scenarios: If there is a strong non-drainage effect / significant negative pore pressure, etc., it should be corrected separately in S4.

[0142] This application also proposes a system for predicting the bearing capacity of pull-out piles, such as... Figure 6As shown, it includes a data acquisition module, a parameter fitting module, an analytical solution module, a verification and evaluation module, and a result output interface, wherein:

[0143] 1) Data acquisition module: Integrates survey and test data to form a parameter library.

[0144] 2) Parameter fitting module: Completes sensitivity analysis, The regression and update of the surface / attenuation coefficient β are minimized according to equation (A-11) to minimize the RMSE.

[0145] 3) Analytical Solver Module: Calculate P using equations (A-1) to (A-10). nu It also provides a safety factor for a specified displacement threshold.

[0146] 4) Verification and evaluation module: Compare with the test pile / specification (DL / T 5219-2014) results to form a consistency conclusion and automatic calibration record.

[0147] 5) Result Output Interface: Export calculation tables and surface plots ( Figure 5 Error distribution diagram () Figure 3 (and reports, supporting integration with design software.)

[0148] To verify the applicability and prediction accuracy of the analytical model of this application for the pull-out bearing capacity of cast-in-place piles under engineering conditions, the calculation results of the model of this application are compared with the current standard methods, and the measured values ​​are used as the benchmark to evaluate the error.

[0149] Samples and Construction

[0150] Drilled cast-in-place piles, widely used in engineering, were selected. The construction process was as follows: pre-drilling → placing the 8B16 steel reinforcement cage tied on site → pouring C25 concrete to form the pile.

[0151] The pile diameter d = 0.30m; pile length L = 6m, 8m, 10m; length-to-diameter ratio λ = L / d = 20, 26.7, 33.3. This experiment did not use mud slurry for wall protection and was a dry drilling operation.

[0152] Site and geotechnical parameters

[0153] Moisture content =6.35%, density = 13.10 kN·m -3 ,proportion =2.70, void ratio =0.99, liquid limit =32.10%, Plastic Limit =19.20%, Liquid Limit Index =0.23, Plastic Limit Index =12.90%, cohesion =21.60 kPa, internal friction angle =26°, compression modulus =5.80MPa, collapsibility coefficient =0.13. (Parameter sampling and testing are performed according to conventional geotechnical testing methods, used for unified input between the two methods.)

[0154] Comparison Method (Calculation Based on Current Standards)

[0155] The relevant values ​​are referenced from DL / T 5219-2023 "Code for Design of Foundations for Overhead Transmission Lines":

[0156] 1) Standard value of ultimate lateral resistance of soil along pile : Silt-sparse ( - Dry drilling piles, 24-42 kPa (median value 30 kPa);

[0157] 2) Pull-out coefficient Clay / Silt: 0.70–0.80 (median value 0.75);

[0158] 3) Calculation formula: Equal cross-section pile .

[0159] Therefore, we can conclude that: When = 6, 8, 10m, 127.17, 169.56, 211.95 kN.

[0160] This application's model calculation

[0161] Following the analytical formula and solution process provided in this application, and inputting the aforementioned parameters and geometric information, the predicted uplift bearing capacity for the corresponding pile length is obtained. The ultimate bearing capacity is uniformly determined using the displacement method: for gradually changing piles... 40 mm, if 800 mm 0.05 .

[0162] Table 1 shows a comparison of the pull-out bearing capacity of cast-in-place piles using measured values, standard methods, and the methods described in this application.

[0163] Table 1. Actual Measurements vs. Regulatory Law vs. This Application

[0164]

[0165] Based on Table 1, the following conclusions can be drawn:

[0166] Technical effect

[0167] (1) The standard method is conservative and the underestimation worsens with pile length: the three groups were underestimated by approximately 42% / 40% / 58% respectively, with an overall RMSE of 46.78%. This indicates that the median value was used. and The lateral resistance mobilization response of medium and long piles is insufficient, and the slope of bearing capacity increasing with L / d is too small.

[0168] (2) The application has a high degree of fit and reasonable sensitivity to L / d: the error is basically within ±10-15%, and the overall RMSE is 10.02%; it can reproduce the trend of bearing capacity increasing with pile length and maintains high accuracy in long pile sections.

[0169] (3) Technical reasons: This application comprehensively considers the aspect ratio, soil strength and deformation parameters, pile-soil interface conditions, nonlinearity of side resistance mobilization and stress redistribution in the parameterization and calculation framework. Compared with the standard method of approximating the bearing capacity as "segmented side resistance linear superposition", it is more in line with the field conditions.

[0170] (4) According to the regulations, the upper limit is favorable. =42kPa, =0.8 recalculated, the 8-10 m long pile is still lower than the actual measurement, the conclusion remains unchanged.

[0171] Executable steps and interface conventions (and) Figure 1 Consistent)

[0172] 1) Input: , , , , , , .

[0173] 2) Select working conditions: pile type, drilling method (dry operation in this example), displacement threshold and design safety level.

[0174] 3) Solution: Calculate according to formulas (A-1) to (A-10). .

[0175] 4) Verification: Calculate RMSE according to formula (A-11) and compare it with the threshold; trigger parameter self-calibration if necessary.

[0176] 5) Output: Bearing capacity-λ curve, error distribution ( Figure 3 ), curved surface ( Figure 5 ), calculation sheets and comparison tables (including standard values).

[0177] Compared with existing standards / empirical formulas, this application has the following advantages:

[0178] A unified characterization of the relationship between the surface failure angle and the aspect ratio / friction angle (Equation (A-1), see...) Figure 4 The evolution of the failure surface depth is explicitly introduced (Equations (A-2)~(A-3)), and the mobilization coefficient is integrated (Equation (A-9)), thereby correctly reflecting the nonlinear growth of bearing capacity with λ.

[0179] The parameters are concise and have clear physical meanings. Combined with RMSE-driven self-calibration (Equation (A-11)), the overall error is significantly reduced while ensuring traceability.

[0180] Highly applicable to engineering projects: It provides rapid, verifiable, and comparable standard values ​​and design suggestions for the pull-out design of bored micropiles in loess and similar unsaturated soil areas.

[0181] This method is preferably used for predicting the net uplift bearing capacity of dry-drilled cast-in-place piles; however, it is not limited to this. Based on the pile-soil coupling analytical model and parameterized solution process described above, this method is also applicable to mud-walled bored piles, precast (driven) piles, micropiles / anchors, and other uplift-resistant components. After making appropriate corrections or calibrations to the interface friction angle, aspect ratio, mobilization coefficient, and attenuation parameters, uplift bearing capacity prediction results consistent with those of the aforementioned pile types can be obtained.

[0182] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for predicting the load bearing capacity of a pull-out pile, characterized by: Comprising the following steps: S1: parameter acquisition, including pile parameters, soil parameters and pile-soil interface parameters; S2: numerical calibration and sensitivity identification, including: S2.1: Establish a three-dimensional / axisymmetric pile-soil finite element example library to obtain load-displacement curve, displacement / plastic zone distribution for mechanism identification; S2.2: Evaluate the influence of the obtained parameters on the net uplift capacity, and identify the key control quantity; S2.3: Use the test pile / historical engineering data as the true value, and perform model parameter regression and setting according to the objective function RMSE; S3: Analytical modeling and solving, including assumption conditions of the model, failure surface modeling, and balance equation and bearing capacity expression, and finally obtaining the analytical expression of net uplift capacity; S4: Structure checking and output.

2. The method of claim 1, wherein: The pile parameters in step S1 include pile diameter , pile length , pile elastic modulus , the soil parameters include cohesion , internal friction angle , density , soil elastic modulus , and the pile-soil interface parameters include friction angle ; Step S1 specifically includes: S1.1: Collecting soil mechanics and pile-soil interface parameters 、 、  、 、 ; S1.2: Collect pile body parameters 、  、 , select the hole-forming method, and record the interface roughness level; S1.3: Establish a "local parameter library" as the calculation input and checking baseline.

3. The method of claim 2, wherein: In step S2.2, single factor variation method + root mean square sensitivity is used to evaluate the influence of parameters on net uplift capacity.

4. The method of claim 3, wherein: The key control variables identified in step S2.2 are: ,  high sensitive control variables; , ,  medium sensitive control variables; , low sensitive control variables.

5. The method of claim 2, wherein: In step S3, the assumption conditions of the model include: (1) The soil is isotropic, homogeneous, linear elastic-ideal plastic; (2) The pile is linear elastic; the pile-soil interface is normal hard contact-tangential Coulomb friction as the pile-soil interface friction angle δ; (3) Axisymmetric simplification: taking the pile bottom as the origin, z along the pile axis, z∈[0,L], and the ground surface z=L.

6. The method of claim 5, wherein: The failure surface model includes: Surface failure angle: ; Depth distribution of local damage angle: ; The damage curve explicit formula is: ; In the formulae: denotes the attenuation coefficient.

7. The method of claim 5, wherein: The balance equation and bearing capacity expression include: Shear strength increment relationship: ; Unit thickness soil wedge normal equilibrium: ; tangential force increment: ; wherein: denotes the incremental shear resistance on the soil failure surface, denotes the shear strength per unit length, denotes the incremental length of the failure surface; denotes the incremental normal force acting on the soil failure surface, denotes the unit thickness of the self-weight on the failure surface, denotes the incremental self-weight within the unit thickness.

8. The method of claim 7, wherein: The expression of net uplift capacity is as follows: ； In the formula: N is the net uplift capacity, ρ is the density, is the unified correction of the self-weight equivalent term of the pile, facilitating the design conversion under different λ conditions, where λ is the length-diameter ratio of the pile, , , and are the dimensionless integral coefficients of the tangential friction and cohesion along the failure surface, respectively.

9. The method of claim 1, wherein: The result checking in step S4 is: comparing the analytical results with the test pile / historical engineering data, if the RMSE is out of limit, then automatically calibrate according to the regression parameters of S2; The output results in step S4 include bearing capacity calculation book, parameter sensitivity report, specification / test pile comparison table, and safety factor and displacement threshold value suggestion.

10. A system for implementing the method for predicting the load bearing capacity of a uplift pile according to any one of claims 1-9, characterized in that: Comprising: 1) Data acquisition module: used to access survey and test data to form a parameter library; 2) Parameter fitting module: used to complete sensitivity analysis, regression and update of failure surface curve / decay coefficient, and minimize RMSE; 3) Analytical solving module: used to calculate the net uplift capacity and give the safety factor under the specified displacement threshold; 4) Checking and evaluation module: used to compare with test pile / specification results to form consistent conclusions and automatic calibration records; 5) Result output interface: used to export calculation table, surface graph, error distribution graph and report.

Images