Pile sinking soil body response prediction regulation and control method based on physical model

Abstract

The invention discloses a pile sinking soil body response prediction regulation and control method based on a physical model, and relates to the technical field of geotechnical mechanics pile foundation engineering, and the method comprises the following steps: constructing an experimental model, arranging a sensor, and operating a loading device to enable the experimental model to sink a pile; according to the model pile sinking process, the soil body response is monitored in real time, and monitoring data are obtained; inverting the dynamic response of the soil body through the PINN inversion model, and coupling the DEM and the FEM to simulate and predict the response of the soil body; and risk control is carried out through a CEI early warning decision-making method. The device and the method are used for solving the problems of incomplete data acquisition, inaccurate simulation and lack of early warning in a traditional open tubular pile soil squeezing model test.

Description

Technical Field

[0001] This invention relates to the field of geotechnical engineering for pile foundations, and more specifically, to a method for predicting and controlling the soil response of pile driving based on a physical model. Background Technology

[0002] Open-face pipe piles are widely used in power facility foundation engineering. However, the soil squeezing effect caused by the pile driving process can easily lead to soil displacement, stress redistribution, and soil plugging around the pile, thereby threatening the safety of adjacent underground pipelines and equipment foundations. Mild soil squeezing may cause pipeline deformation and loosening of joints, while severe cases may lead to a chain of engineering accidents such as pipeline rupture, equipment foundation displacement, or even structural instability. In order to deeply understand and accurately predict the laws of this soil squeezing effect and its potential risks, scaled-down physical model tests are an important research method.

[0003] Existing methods for predicting the soil displacement effect of pipe piles mostly rely on a single technical approach, making it difficult to simultaneously achieve high-precision real-time monitoring, complex dynamic simulation, and safety early warning. These methods suffer from incomplete data collection, inaccurate simulation, and lack of early warning.

[0004] To address the above problems, this invention proposes a solution. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method for predicting and controlling the soil response of pile driving based on a physical model. The present invention solves the problems of incomplete data acquisition, inaccurate simulation, and lack of early warning in traditional open-pipe pile soil displacement model tests by using multi-sensor fusion, advanced numerical simulation, and physical constraint inversion.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The method for predicting and controlling the soil response of pile driving based on a physical model includes the following steps: constructing an experimental model and deploying sensors, and operating a loading device to drive the experimental model into a pile; monitoring the soil response in real time and acquiring monitoring data according to the pile driving process of the model; performing simulation prediction of the soil response by inverting the inverted soil body through PINN and coupling DEM and FEM; and performing risk control through the CEI early warning decision method.

[0007] In a preferred embodiment, the experimental model materials include soil sample preparation, model piles, and model boxes.

[0008] In a preferred embodiment, the soil sample preparation specifically involves: drying the soil sample until its quality stabilizes, and then crushing it using a soil crusher to obtain a crushed soil sample; the crushed soil sample is then sieved to obtain a uniform soil sample with similar density.

[0009] In a preferred embodiment, the preparation of the model pile specifically involves: fabricating steel cages according to different specifications, placing the steel cages into PVC templates; pouring grout into the PVC templates, curing, and demolding to obtain the open-face pipe pile model pile.

[0010] In a preferred embodiment, the experimental model is constructed as follows: a top-opening model box is made of steel, and a uniform soil sample is filled in layers according to a preset layer thickness. A cross-shaped support is set on the top of the model box, and a central ring is set based on the cross-shaped support. The model pile is placed above the soil sample through the support ring.

[0011] In a preferred embodiment, the operation method of the loading device includes static pressure method, specifically as follows: setting up a reaction force device in conjunction with an electric push rod to perform static pressure pile driving.

[0012] In a preferred embodiment, the operation method of the loading device further includes a hammering method, specifically as follows: a cylindrical steel block is used, guided by a guide rod, to perform guided free-fall hammering.

[0013] In a preferred embodiment, the sensor deployment specifically includes DAS fiber optic cable laying, piezoelectric ceramic array installation, and magnetostrictive displacement chain implantation.

[0014] In a preferred embodiment, the step of monitoring the soil response in real time and acquiring monitoring data according to the model pile driving process specifically includes: tracking the axial strain of the pile in real time and acquiring strain data; capturing soil damage signals in real time and classifying the soil damage types; and constructing the soil displacement field through a Gaussian kernel function based on the displacement data of the magnetically encoded nodes to analyze the soil deformation and soil response during the pile driving process.

[0015] In a preferred embodiment, the dynamic response of the soil is inverted using the PINN inversion model and coupled with DEM and FEM to simulate and predict the soil response. Risk control is performed using the CEI early warning decision method, specifically: a PINN inversion model is constructed by combining physical constraint equations, and the dynamic response of the soil is inverted using real-time collected monitoring data; DEM-FEM coupled simulation technology is used to simulate the interaction between the soil and the pile during pile driving, and the soil response during pile driving is accurately predicted; the bearing capacity influence index is calculated, and the interaction between the pile and the soil is monitored in real time. When the bearing capacity influence index exceeds the set range, the system automatically takes early warning measures.

[0016] The technical effects and advantages of the physical model-based method for predicting and controlling soil response during pile driving in this invention are as follows: 1. This invention integrates multi-source sensing technologies such as DAS optical fiber, piezoelectric ceramic array and magnetostrictive displacement chain, and combines PINN physical constraint inversion and DEM-FEM coupled simulation to achieve transparent monitoring of the dynamic response of the pile-soil system across the entire cross section during the pile driving process for the first time. This breaks through the blind spots of traditional point measurement and makes the evolution of soil damage visible throughout the entire process.

[0017] 2. This invention uses a combination of grouting material and wire mesh reinforcement cage to create model piles, and constructs a "monitoring-prediction-control" closed-loop system based on the CEI bearing capacity index. Through real-time magnetic coding displacement calibration, soil type adaptive weight matching, and sensor orientation optimization, it achieves proactive early warning of pile driving risks, advances the accident intervention point to the damage initiation stage, and promotes the leap from passive response to proactive intervention in safety control. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the process of predicting and controlling the soil response of pile driving based on a physical model according to the present invention. Detailed Implementation

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention 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. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0020] Example 1, Figure 1 The present invention provides a method for predicting and controlling the soil response during pile driving based on a physical model, comprising the following steps: S1. Construct an experimental model, install sensors, and operate the loading device to drive the experimental model into the pile. S2, based on the model pile driving process, monitor the soil response in real time and acquire monitoring data; S3 uses the PINN inversion model to simulate the dynamic response of the soil body, and couples DEM and FEM to perform simulation prediction of soil response; S4 uses the CEI early warning decision-making method for risk control.

[0021] This invention ensures physical similarity by using a combination of grouting material and wire mesh reinforced cage model piles, and integrates multi-source sensing technologies such as DAS optical fiber, piezoelectric ceramic array, and magnetostrictive displacement chain. Combined with PINN physical constraint inversion and DEM-FEM coupled simulation, it achieves transparent monitoring of the dynamic response of the pile-soil system across the entire cross-section of the pile-driving process for the first time, breaking through the blind spots of traditional measurements. Furthermore, based on the CEI bearing capacity index, a "monitoring-prediction-control" closed-loop system is constructed. Through real-time magnetically encoded displacement calibration, soil type adaptive weight matching, and sensor orientation optimization, proactive early warning of pile driving risks is achieved, significantly advancing the accident intervention point to the damage initiation stage, and promoting the leap from passive response to proactive intervention in safety control.

[0022] S1. Construct an experimental model, install sensors, and operate the loading device to drive the experimental model into the pile.

[0023] In this embodiment, the experimental model materials include soil sample preparation, model piles, and model boxes.

[0024] In this embodiment, the soil sample preparation specifically involves: The soil samples were dried until their quality stabilized, and then crushed using a soil crusher to obtain crushed soil samples. The crushed soil sample was sieved to obtain a homogeneous soil sample with similar density.

[0025] In this embodiment, sand, clay, and loam were used as fill soil samples. After the soil samples were collected, they were dried and baked: a representative sample was weighed daily, placed in an oven and heated for 6 hours, and then weighed again, until there was no significant difference between two consecutive weighing results. Subsequently, a soil crusher was used to crush the soil samples to eliminate large particles and ensure that the soil samples were finely and uniformly crushed.

[0026] Furthermore, by using sieves with different parameters for layered screening and filling, a uniform soil layer with similar relative density is obtained.

[0027] In this embodiment, the preparation of the model pile is specifically as follows: Prepare steel cages according to different specifications and place the steel cages into PVC templates; Grouting material is used to pour PVC templates, and after curing and demolding, an open-pit pipe pile model pile is obtained.

[0028] In this embodiment, based on the commonly used specifications of PHC pipe piles in actual engineering, three types of pipe piles with outer diameters of 400mm, 500mm, and 600mm are simulated respectively. The corresponding scaled-down models are outer diameters of 40mm, 50mm, and 75mm, and wall thicknesses of 15mm, 18mm, and 25mm respectively (see Table 1).

[0029] The model pile uses a mixture of grout and sand, supplemented with a reinforcing cage made of thin iron wire (2.3mm in diameter). To ensure the verticality of the reinforcing cage, a reinforcing skeleton is built inside a PVC pipe. After the skeleton is built, the PVC pipe is removed to obtain the finished reinforcing cage.

[0030] The reinforcing cage was placed into the template of the model pile, and grout with a water-cement ratio of 15% was poured. After pouring, it was cured for 28 days. After curing, the inner and outer PVC pipes were removed to obtain the open model pile. Table 1 shows the dimensions of the three types of model pipe piles:

[0031] In this embodiment, the experimental model is constructed as follows: A steel-framed model box with an open top is used to fill a uniform soil sample in layers, with each layer being filled according to a preset layer thickness. A cross-shaped support is installed on the top of the model box, and a central ring is set based on the cross-shaped support. The model pile is passed through the support ring and placed above the soil sample.

[0032] In this embodiment, the model test uses a cubic model box with dimensions of 0.8m×0.5m×0.6m (length×width×height). The box is made of steel structure with an open top to facilitate the driving of pipe piles from top to bottom.

[0033] Soil samples were filled using a layered filling method, with each layer set to a thickness of 10cm, and construction was carried out in accordance with quality control principles.

[0034] Furthermore, to limit the tilting of the pile body during the pile driving process, a cross-shaped fixing bracket is installed on the top of the model box. A circular ring with a height of 10cm and an inner diameter slightly larger than the outer diameter of the model pile is placed in the center of the bracket. After the model pipe pile is positioned by passing it through the circular ring, it is placed on the surface of the foundation soil.

[0035] In this embodiment, the operation method of the loading device includes the static pressure method, as follows: A reaction force device is set up in conjunction with an electric push rod for static pressure pile driving.

[0036] In this embodiment, static pressure pile driving is achieved using an electric pusher. The pusher is 655mm long, has a jacking stroke of 500mm, a maximum thrust of 1000N, and a speed of 5mm / s. The electric pusher is installed on top of the reaction frame and pushes the pipe pile model into the foundation soil from top to bottom.

[0037] In this embodiment, the operation method of the loading device further includes a hammering method, as detailed below: A cylindrical steel block is used, guided by a guide rod, for guided free-fall hammering.

[0038] In this embodiment, the reaction frame is constructed of angle steel and steel plates, with dimensions of 120cm × 60cm × 200cm (length × width × height). Wherein: Four 2m angle steel bars marked in blue serve as reaction columns; Six 0.6m angle steels marked in yellow serve as reaction longitudinal beams; Four 1.2m angle steels marked in red serve as reaction beams; Four holes are drilled in the middle of the top steel plate for fixing the electric push rod.

[0039] The lower part of the reaction column is fixed to the four corners of the model box, and the top of the push rod is fixed to the reaction frame steel plate, forming a complete static pressure loading system.

[0040] The hammer-driven pile driving system uses a free-fall loading device. The steel hammer is a cylinder with a diameter of 10cm, a height of 10cm, and a mass of 12kg. To ensure the stability of the steel hammer during its fall, a hole (9mm in diameter) is drilled in its center, and a guide rod with a diameter of 8mm and a length of 1.6m is fitted onto it to achieve vertical hammer loading.

[0041] In this embodiment, the sensor deployment specifically includes DAS fiber optic cable laying, piezoelectric ceramic array installation, and magnetostrictive displacement chain implantation.

[0042] In this embodiment, DAS fiber optic cable laying is performed by chiseling a spiral groove (50mm pitch, 2mm depth) on the outer wall of the pile body, laying a single-mode fiber optic cable (tensile strength ≥1GPa), and connecting the two ends of the fiber optic cable to a φ-OTDR demodulator to establish a strain-wave velocity conversion model.

[0043] The specific formula for converting fiber strain to shear wave velocity is as follows:

[0044] In the formula, For shear wave velocity, Shear modulus The strain change measured by DAS For material density, It is Young's modulus (elastic modulus). It is Poisson's ratio.

[0045] In this embodiment, the piezoelectric ceramic array is installed as follows: eight piezoelectric ceramic sensors are arranged in a ring at the pile end, with the sensor normal directions distributed according to the vertices of a regular icosahedron (ensuring the direction cosine matrix is ​​full rank). The system is configured to capture the acoustic signal of soil rupture (50kHz–1MHz) to achieve centimeter-level positioning of the soil displacement failure source.

[0046] The specific formula for locating acoustic emission signals is as follows:

[0047] In the formula, For sensors and The signal arrival time difference The longitudinal wave velocity of the soil. The coordinates of the acoustic emission source are... For sensors coordinate, For sensors coordinate.

[0048] In this embodiment, the magnetostrictive displacement chain is implanted: a hole is drilled at a distance of 1.5 times the pile diameter from the pile circumference, and magnetic coding nodes are buried in layers (spaced 20cm apart). An external magnetic field excites the nodes to generate a displacement vector field.

[0049] The specific formula for measuring nodal displacement is as follows:

[0050] In the formula, The distance of the node displacement. This is the magnetic-force coupling coefficient (calibrated value). The rate of change of magnetic flux density This is the sampling interval.

[0051] S2 monitors the soil response in real time and acquires monitoring data based on the model's pile driving process.

[0052] In this embodiment, the real-time monitoring of the soil response and acquisition of monitoring data based on the model pile driving process specifically includes: Real-time tracking of the axial strain of the pile body to obtain strain data; Real-time capture of soil damage signals and classification of soil damage types; Based on the displacement data of the magnetically encoded nodes, the displacement field of the soil is constructed using a Gaussian kernel function to analyze the soil deformation and response during pile driving.

[0053] In this embodiment, the real-time tracking of the axial strain of the pile body to obtain high-precision strain data specifically involves: the fiber optic system acquiring the phase change of Rayleigh scattered light at a sampling rate of 1000Hz and plotting a dynamic cloud map of shear stiffness.

[0054] The phase change of Rayleigh scattered light is obtained at a sampling rate of 1000Hz, and the specific formula is as follows:

[0055] In the formula, For the phase change of Rayleigh scattered light, This refers to the axial position of the optical fiber. Using time as the coordinate, The wavelength of the laser. The refractive index of light, For axial dynamic strain, This is the integration path variable.

[0056] In this embodiment, the real-time capture of soil damage signals and the classification of soil damage types specifically involve: The piezoelectric ceramic sensor continuously receives acoustic emission signals. When the signal amplitude exceeds a preset threshold, it is determined to be a valid event. The start time of the event (the moment when the signal first exceeds the threshold) and the end time (the moment when the signal falls back below the threshold) are recorded, and energy integration calculation and main frequency analysis are performed. Based on the dominant frequency analysis, soil damage is classified as follows: Plastic deformation determination: The cause of the damage is the reorganization and slippage of soil particles; Structural cracking: If The cause of the damage is the propagation of cracks in the concrete / rock.

[0057] The energy integral formula is as follows:

[0058] In the formula, For acoustic emission event energy, For time-domain signals, The start time of the event. This is the end time of the event.

[0059] The specific formula for clock speed analysis is as follows:

[0060] In the formula, The dominant frequency of the signal. For Fourier transform operators, It is a time-domain signal.

[0061] In this embodiment, the system reconstructs the displacement field of the soil based on the displacement data of the magnetically encoded nodes, and comprehensively analyzes the deformation and response of the soil during pile driving. Specifically: The magnetic displacement chain generates a full-section displacement field every 5cm of penetration, and the soil movement trajectory is reconstructed by the Gaussian kernel function.

[0062] The formula for the Gaussian kernel function is as follows:

[0063] In the formula, For Gaussian kernel function, To calculate the Euclidean distance from a point to a node, This is the kernel width parameter.

[0064] The specific formula for the Gaussian kernel function interpolation displacement field reconstruction method is as follows:

[0065] In the formula, The target point displacement vector, For spatial coordinates, This represents the total number of displacement monitoring nodes. For nodes The magnitude of the displacement, For nodes The displacement direction is a unit vector. For radial basis functions, To calculate the point-to-node distance Euclidean distance, Let the coordinates of any field point be... For nodes The coordinates.

[0066] S3 uses the PINN inversion model to simulate the dynamic response of the soil and couples it with DEM and FEM to predict the soil response.

[0067] S4 uses the CEI early warning decision-making method for risk control.

[0068] In this embodiment, the dynamic response of the soil is inverted using the PINN inversion model, and coupled with DEM and FEM to simulate and predict the soil response; risk control is performed using the CEI early warning decision-making method, specifically as follows: By combining physical constraint equations, a PINN inversion model is constructed, and the dynamic response of the soil is inverted using real-time collected monitoring data. The DEM-FEM coupled simulation technology is used to simulate the interaction between the soil and the pile during the pile driving process and to accurately predict the soil response during the pile driving process. By calculating the bearing capacity influence index, the interaction between the pile and the soil is monitored in real time. When the bearing capacity influence index exceeds the set range, the system automatically takes early warning measures.

[0069] In this embodiment, the strain tensor, acoustic emission energy density, and magnetic displacement gradient of the distributed fiber acoustic sensor (DAS) are collected in real time and used as inputs to the neural network to construct a multi-physics coupling inversion model based on the physical information neural network. In this model, the constraint physical equation adopts the momentum conservation equation in incremental form and outputs key parameters.

[0070] The incremental form of the momentum conservation equation is as follows:

[0071] In the formula, The density of the soil. For displacement increment, This is the second derivative of the displacement increment with respect to time. This is the resultant force caused by changes in stress distribution. This is the equivalent external disturbance source term.

[0072] Incremental expression for nonlinear constitutive relation:

[0073] In the formula, For stress increment, For the strain increment, This represents the current state of emergency. The first derivative of the stiffness tensor with respect to strain is... Generate real-time updated nonlinear stiffness tensors for soil for neural networks.

[0074] In this embodiment, to further improve the simulation accuracy of the pile driving disturbance field, a multi-scale coupled simulation framework based on the discrete element-finite element method (DEM-FEM) is introduced, and the average particle displacement in the DEM domain is transferred to the FEM node to perform synchronous updates of the displacement fields in the two domains.

[0075] The specific formula for the DEM-FEM coupling equation is as follows: Discrete Element Domain (DEM):

[0076] Finite element field (FEM):

[0077] In the formula, For the first The mass of a discrete element particle For the first The position vector of a discrete element particle For the first The acceleration vector of a discrete element particle. For time variables, For the action in the Contact force vector on each particle; The strain-displacement matrix, For Cauchy stress tensor, External load vector, This is the integration region.

[0078] The formula for the mean displacement is as follows:

[0079] In the formula, The average displacement acting on the boundary of the FEM domain. This represents the total number of discrete particles participating in coupling at the interface. For the first The position vector of each particle.

[0080] By calculating the bearing capacity influence index, the interaction between the pile and the soil is monitored in real time. When the bearing capacity influence index exceeds the set range, the system automatically takes early warning measures.

[0081] In this embodiment, in order to achieve proactive risk control, a dynamic early warning mechanism based on the bearing capacity influence index (CEI) is established to reflect the cumulative impact of pile driving disturbance on the bearing capacity of the pile tip and pile side. When CEI>0.8, the pile driving rate is reduced to 50%; when CEI>1.0, pile driving is suspended and the pre-hole decompression is started; when CEI>1.2, bentonite slurry is injected to improve the soil.

[0082] The specific formula for calculating the bearing capacity influence index is as follows:

[0083] In the formula, As a bearing capacity influence index, This is a weighting coefficient used to adjust the contribution of changes in side friction resistance to the total bearing capacity. This is a weighting coefficient used to adjust the contribution of changes in end resistance to the total bearing capacity. For a certain time The increase in side friction resistance at time t, For a certain time The increment of end resistance at time , The initial side friction resistance, For initial end resistance, The maximum score at the current moment. This is the time integration variable, representing any moment within the time window.

[0084] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0085] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0086] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0087] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0088] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0089] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting and controlling soil response during pile driving based on a physical model, characterized in that, Includes the following steps: Construct an experimental model, install sensors, and operate the loading device to drive the experimental model into the pile. Based on the model pile driving process, the soil response is monitored in real time, and monitoring data is acquired; By using the PINN inversion model to simulate the dynamic response of the soil body, and coupling DEM and FEM, the soil response is predicted through simulation. Risk control is carried out through the CEI early warning decision-making method.

2. The method for predicting and controlling the soil response of pile driving based on a physical model according to claim 1, characterized in that, The experimental model materials include soil sample preparation, model piles, and model boxes.

3. The method for predicting and controlling the soil response of pile driving based on a physical model according to claim 2, characterized in that, The soil sample preparation is specifically as follows: The soil samples were dried until their quality stabilized, and then crushed using a soil crusher to obtain crushed soil samples. The crushed soil sample was sieved to obtain a homogeneous soil sample with similar density.

4. The method for predicting and controlling soil response during pile driving based on a physical model according to claim 3, characterized in that, The preparation of the model pile is specifically as follows: Prepare steel cages according to different specifications and place the steel cages into PVC templates; Grouting material is used to pour PVC templates, and after curing and demolding, an open-pit pipe pile model pile is obtained.

5. The method for predicting and controlling soil response during pile driving based on a physical model according to claim 4, characterized in that, The experimental model is constructed as follows: A steel-framed model box with an open top is used to fill a uniform soil sample in layers, with each layer being filled according to a preset layer thickness. A cross-shaped support is installed on the top of the model box, and a central ring is set based on the cross-shaped support. The model pile is passed through the support ring and placed above the soil sample.

6. The method for predicting and controlling soil response during pile driving based on a physical model according to claim 7, characterized in that, The operation method of the loading device includes the static pressure method, as follows: A reaction force device is set up in conjunction with an electric push rod for static pressure pile driving.

7. The method for predicting and controlling soil response during pile driving based on a physical model according to claim 6, characterized in that, The operation method of the loading device also includes a hammering method, as follows: A cylindrical steel block is used, guided by a guide rod, for guided free-fall hammering.

8. The method for predicting and controlling soil response during pile driving based on a physical model according to claim 7, characterized in that, The sensor deployment specifically includes DAS fiber optic cable laying, piezoelectric ceramic array installation, and magnetostrictive displacement chain implantation.

9. The method for predicting and controlling the soil response of pile driving based on a physical model according to claim 8, characterized in that, The process of pile driving based on the model involves real-time monitoring of the soil response and acquisition of monitoring data, specifically as follows: Real-time tracking of the axial strain of the pile body to obtain strain data; Real-time capture of soil damage signals and classification of soil damage types; Based on the displacement data of the magnetically encoded nodes, the displacement field of the soil is constructed using a Gaussian kernel function to analyze the soil deformation and response during pile driving.

10. The method for predicting and controlling the soil response of pile driving based on a physical model according to claim 9, characterized in that, The dynamic response of the soil is inverted using the PINN inversion model and coupled with DEM and FEM to simulate and predict the soil response; risk control is carried out using the CEI early warning decision method, specifically: By combining physical constraint equations, a PINN inversion model is constructed, and the dynamic response of the soil is inverted using real-time collected monitoring data. The DEM-FEM coupled simulation technology is used to simulate the interaction between the soil and the pile during the pile driving process and to accurately predict the soil response during the pile driving process. By calculating the bearing capacity influence index, the interaction between the pile and the soil is monitored in real time. When the bearing capacity influence index exceeds the set range, the system automatically takes early warning measures.

Images