Machine learning algorithm-based spiral pile core stiff composite pile forming process and use method

Abstract

The invention relates to the technical field of spiral piles, and discloses a spiral pile core stiff composite pile forming process based on a machine learning algorithm and a use method, and the spiral pile core stiff composite pile forming process specifically comprises the following steps: carrying out finite element analysis; performing machine learning to obtain a machine learning model related to a machine learning parameter M (x); the trained machine learning model related to the machine learning parameters M (x) is imported into a computer, and the computer is connected with a spiral pile core stiff composite pile driver; after on-site soil information and the anti-pulling bearing capacity needed by a project are input on a construction site, the structural size of the screw anchor needed by the project is displayed on a computer screen, the anti-pulling bearing capacity of the grouting screw anchor is reflected in real time through the formula F = M (x) S (x) ktT, and the anti-pulling bearing capacity of the grouting screw anchor is obtained through combination of finite element analysis and a machine learning model. Accurate prediction and real-time adjustment of the uplift bearing capacity of the screw pile are achieved. Waste of manpower and material resources can be reduced, construction performance is improved, the intelligent level and construction safety are enhanced, and remarkable technical advantages and application value are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of screw piles, and in particular relates to a process for forming a screw pile core rigid composite pile based on a machine learning algorithm and a method for using the same. Background Art

[0002] The spiral pile core rigid composite pile replaces the traditional rigid pile core with a spiral pile, and uses the grouting process to compensate for the disturbance of the blade to the soil. At the same time, it retains the advantage of the convenient installation of the spiral pile, improves the pile-making process and the pile-soil contact, and enhances the construction efficiency and bearing capacity of the rigid composite pile. It has been widely used in the fields of construction, ports, power grids, highway and railway engineering.

[0003] However, the pull-out bearing capacity of spiral core reinforced composite piles is still calculated using traditional methods. However, the soil layer information at each piling point is different. There are many options for the anchor diameter of the spiral anchor plate at the front end of the spiral core reinforced composite pile, the number of anchor plates, the anchor plate spacing, the diameter of the anchor rod, the strength of the cement mortar, and the friction coefficient of the steel pipe cement soil contact surface. As a result, calculating the pull-out bearing capacity of the grouting spiral pile in the project requires a lot of manpower and material resources.

[0004] Machine learning is widely used in today's society to predict, classify and solve complex mathematical problems in civil engineering. Its characteristics are used to optimize the calculation of the pull-out bearing capacity of grouting screw piles in engineering.

[0005] Machine learning calculation of the pull-out bearing capacity of grouting spiral piles is usually calculated using the following formula in scientific research.

[0006] F=M(x)S(x)k t T;

[0007] F: pull-out bearing capacity of grouting screw pile; M(x): machine learning parameter; S(x): bearing capacity improvement coefficient of screw pile after grouting; k t : Torque coefficient; T: Torque provided by the pile driver;

[0008] Therefore, in order to solve the problems existing in the existing spiral pile core rigid composite pile construction process, it is necessary to propose a new spiral pile core rigid composite pile construction process so that the pull-out bearing capacity of the spiral pile core rigid composite pile can be accurately and in real time known. Summary of the Invention

[0009] The present invention aims to solve the technical problem in the prior art that construction relies on experience and standard formulas.

[0010] To achieve the above object, the present invention provides the following technical solutions:

[0011] A spiral pile core rigid composite pile construction process based on a machine learning algorithm specifically includes the following steps:

[0012] Step 1: Conduct finite element analysis: First, establish a geometric model of the screw pile, soil, and grouting area to reflect the dimensions and layout of the actual project. Next, define the physical properties of the screw pile, soil, and grouting material, such as the elastic modulus, shear modulus, and friction coefficient. Set boundary conditions and load application methods to simulate the foundation conditions and the stress on the pile during construction, and apply corresponding pressure to the grouting process. Next, perform meshing, using a fine mesh in key areas to improve calculation accuracy. Finally, select an appropriate analysis type, such as static, dynamic, or nonlinear analysis, to ensure that the model can accurately simulate the mechanical behavior of the screw pile under various working conditions.

[0013] Through the above steps, finite element analysis provides accurate data support for machine learning, improving the accuracy and efficiency of pile design and construction processes.

[0014] Step 2: Perform machine learning to obtain a machine learning model for the machine learning parameter M(x);

[0015] Step 3: Import the trained machine learning model of the machine learning parameter M(x) into the computer, and connect the computer to the spiral pile core rigid composite pile driver;

[0016] Step 4. After inputting the soil information and the required pull-out bearing capacity at the construction site, the computer screen will display the required screw anchor structure dimensions (anchor diameter, number of screw anchors, spacing between anchors, diameter of anchor rods), cement mortar strength, and torque provided by the machine.

[0017] Step 5: Use the formula F = M(x)S(x)k t T reflects the pull-out bearing capacity of the grouting screw anchor in real time.

[0018] Preferably, the driving and grouting of spiral piles are achieved using a spiral core rigid composite pile driver and a grouting system. The spiral core rigid composite pile driver has multiple functions, including torsion pile driving, grouting, and real-time torque measurement. The spiral core rigid composite pile driver includes a base steel frame, a lifting power assembly, a sliding platform, and a torque power assembly. The grouting system includes a sealable pressurized cement drum and an air compressor.

[0019] The spiral pile core rigid composite pile driver connects various functional components through the base steel frame. The lifting power component is fixed to the top of the base steel frame. The lower end of the lifting power component is rigidly welded to the sliding platform. The torsion power component is fixed to the front upper end of the sliding platform. The upper end of the torsion power component is connected to the slurry pipe, the lower end is connected to the grouting spiral anchor, and the side end is connected to the information transmission cable and the control computer.

[0020] The side of the grouting pipe away from the torsion power assembly is connected to a sealable pressurized cement barrel for mixing cement mortar. The other side of the sealable pressurized cement barrel is connected to an air compressor for pressurizing the sealable pressurized cement barrel through an air pipe. The cement mortar being mixed in the sealable pressurized cement barrel is pressed into the grouting pipe through the air compressor and then pressed into the interior of the grouting screw anchor along the grouting pipe. Then, it is ejected into the soil through the grouting holes on the anchor body of the grouting screw anchor.

[0021] The lifting power assembly drives the grouting screw anchor to rise and fall, and the torsion power assembly drives the grouting screw anchor to rotate. The lifting power assembly and the torsion power assembly cooperate to realize the drilling and installation of the grouting screw anchor. The sliding platform realizes the drilling and installation of the grouting screw anchor through the cooperation of the lifting power assembly and the torsion power assembly. At the same time, the grouting pipe can carry out grouting work inside the grouting screw anchor from the top.

[0022] Preferably, the torsion power assembly is provided with spiral blades cooperating with the grouting spiral anchor to break through the soil and drill down, thereby achieving the effect of breaking through the soil and drilling down.

[0023] The grouting pipe carries out grouting work on the inside of the grouting spiral anchor from the upper part of the grouting spiral anchor; the grouting spiral anchor is provided with a grouting channel, and the grouting channel is connected with the grouting hole.

[0024] Preferably, the control computer sends control signals and information collection feedback signals to the spiral pile core rigid composite pile driver through an information transmission cable, and the grouting spiral anchor performs constant speed drilling and installation based on the control signals of the control computer.

[0025] Preferably, the torsion power assembly provides torque to rotate the grouting spiral anchor. The torsion power assembly includes a torque sensor for measuring the torque used in real time. The spiral pile core rigid composite pile driver is connected to the control computer through a data transmission line, so that the control computer receives the torque data measured by the spiral pile core rigid composite pile driver in real time.

[0026] Preferably, the control computer has a built-in trained machine learning model, which includes training results on the machine learning parameters M(x) to reflect and predict the pull-out bearing capacity of the grouting spiral pile in real time.

[0027] As a preferred method, during the entire construction process, the spiral pile core rigid composite pile driver collects torque and grouting volume data in real time and transmits it to the control computer. The control computer uses a machine learning algorithm to analyze the data, adjust the construction parameters in real time, optimize the construction plan, and adjust the parameters of the construction process in real time according to the analysis results of the machine learning model to improve construction quality and efficiency.

[0028] Learning parameters are obtained through a large number of random models, including the anchor diameter of the spiral anchor plate at the front end of the spiral pile, the number of anchor plates, the spacing between anchor plates, the diameter of the anchor rod, the strength of the cement mortar, the friction coefficient of the steel pipe cement soil contact surface, soil information, the torque provided by the machine, and the pull-out bearing capacity in each case, providing sufficient data samples for the next step of machine learning.

[0029] As a preference, the root mean square error (RMSE), mean absolute error (MAE), and R-squared (R 2 ) Three indicators verify the effectiveness of machine learning;

[0030]

[0031] Where yi is the measured data, To predict data, is the average value of the data set of yi, and N is the size of the data set. The RMSE and MAE index values ​​are low, and the index R 2 The value of is close to 1, indicating that the model has good performance.

[0032] Preferably, machine learning is performed on the data samples provided by the finite element analysis to train a machine learning model about the machine learning parameter M(x).

[0033] As a preference, based on a large number of stochastic models of the finite element model, the pullout bearing capacity of the grouting screw pile is predicted according to the input data, and the prediction results are fed back to the construction team to adjust the construction plan or calculate the optimal anchor plate configuration;

[0034] The control computer will use the real-time numerical value, machine learning parameters M(x), the bearing capacity improvement coefficient s(x) after screw pile grouting, and the torque coefficient k t And the torque T provided by the pile driver, through the formula F=M(x)S(x)k t T predicts the real-time pull-out bearing capacity of the grouting spiral pile and displays it on the computer screen in real time.

[0035] Compared with the prior art, the technical effects and advantages of the present invention are:

[0036] This machine learning-based spiral core reinforced composite pile construction process and method utilizes machine learning to achieve precise predictions and real-time adjustments. Traditional methods rely on empirical experience or standard formulas to calculate the pullout bearing capacity of spiral piles, but geological conditions vary from project site to project site, leading to inaccurate predictions. By incorporating a machine learning algorithm, this invention can adjust the model in real time based on actual construction data, accurately predicting pullout bearing capacity. This reduces the waste of manpower and material resources associated with traditional methods and improves calculation accuracy and efficiency.

[0037] The combination of screw piles and grouting improves construction efficiency and performance: This invention combines screw piles with grouting, retaining the convenient construction advantages of screw piles while simultaneously reducing soil disturbance caused by the blades through grouting. This improvement optimizes pile-soil contact, improves overall construction efficiency, and significantly enhances the load-bearing capacity of the screw pile, particularly in terms of pullout resistance.

[0038] A real-time monitoring and feedback system enhances the intelligent level of construction: This invention uses a computer to monitor key parameters such as torque and grouting volume during the piling process in real time. This system, combined with a trained machine learning model, provides feedback and adjusts construction plans based on this real-time data. This intelligent system makes the construction process more precise and safer, while reducing unnecessary waste and significantly improving the intelligent and automated level of construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is the overall flow chart of the present invention;

[0040] Figure 2 This is a machine learning flowchart of the present invention;

[0041] Figure 3 It is a construction flow chart of the present invention;

[0042] Figure 4 It is a structural schematic diagram of the construction equipment of the present invention.

[0043] In the figure: 1. Base steel frame; 2. Lifting power assembly; 3. Torque power assembly; 4. Air compressor; 5. Sealable pressurized cement barrel; 6. Air pipe; 7. Slurry pipe; 8. Sliding platform; 9. Control computer; 10. Information transmission cable; 11. Grouting screw anchor; 12. Grouting hole; 13. Spiral blade. DETAILED DESCRIPTION

[0044] 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] The following is combined with Figure 1-4 To further explain this application,

[0046] The present application discloses a method for forming a spiral core rigid composite pile based on a machine learning algorithm, which specifically includes the following steps:

[0047] Step 1: Conduct finite element analysis: First, establish a geometric model of the screw pile, soil, and grouting area to reflect the dimensions and layout of the actual project. Next, define the physical properties of the screw pile, soil, and grouting material, such as the elastic modulus, shear modulus, and friction coefficient. Set boundary conditions and load application methods to simulate the foundation conditions and the stress on the pile during construction, and apply corresponding pressure to the grouting process. Next, perform meshing, using a fine mesh in key areas to improve calculation accuracy. Finally, select an appropriate analysis type, such as static, dynamic, or nonlinear analysis, to ensure that the model can accurately simulate the mechanical behavior of the screw pile under various working conditions.

[0048] Through the above steps, finite element analysis provides accurate data support for machine learning, improving the accuracy and efficiency of pile design and construction processes.

[0049] Step 2: Perform machine learning to obtain a machine learning model for the machine learning parameter M(x);

[0050] Step 3: Import the trained machine learning model of the machine learning parameter M(x) into the computer, and connect the computer to the spiral pile core rigid composite pile driver;

[0051] Step 4. After inputting the soil information and the required pull-out bearing capacity at the construction site, the computer screen will display the required screw anchor structure dimensions (anchor diameter, number of screw anchors, spacing between anchors, diameter of anchor rods), cement mortar strength, and torque provided by the machine.

[0052] Step 5: Use the formula F = M(x)S(x)k t T reflects the pull-out bearing capacity of the grouting screw anchor 11 in real time.

[0053] Specifically, the driving and grouting of spiral piles are achieved using a spiral core rigid composite pile driver and a grouting system. The spiral core rigid composite pile driver has multiple functions, including torsion pile driving, grouting, and real-time torque measurement. The spiral core rigid composite pile driver includes a base steel frame 1, a lifting power assembly 2, a sliding platform 8, and a torque power assembly 3. The grouting system includes a sealable pressurized cement drum 5 and an air compressor 4.

[0054] The spiral pile core rigid composite pile driver is connected to various functional components through the base steel frame 1. The lifting power component 2 is fixed to the top of the base steel frame 1. The lower end of the lifting power component 2 is rigidly welded to the sliding platform 8. The torsion power component 3 is fixed to the front upper end of the sliding platform 8. The upper end of the torsion power component 3 is connected to the slurry delivery pipe 7, the lower end is connected to the grouting spiral anchor 11, and the side end is connected to the information transmission cable 10 connected to the control computer 9;

[0055] The side of the grouting pipe 7 away from the torsion power assembly 3 is connected to a sealable pressurized cement barrel 5 for mixing cement mortar. The other side of the sealable pressurized cement barrel 5 is connected to an air compressor 4 for pressurizing the sealable pressurized cement barrel 5 through an air pipe. The cement mortar being mixed in the sealable pressurized cement barrel 5 is pressed into the grouting pipe 7 by the air compressor 4 and is pressed into the interior of the grouting screw anchor 11 along the grouting pipe 7. Then, it is ejected into the soil through the grouting holes 12 on the anchor body of the grouting screw anchor 11.

[0056] The lifting power assembly 2 drives the grouting screw anchor 11 to rise and fall, and the torsion power assembly 3 drives the grouting screw anchor 11 to rotate. The lifting power assembly 2 and the torsion power assembly 3 cooperate to realize the drilling and installation of the grouting screw anchor 11. The sliding platform 8 realizes the drilling and installation of the grouting screw anchor 11 through the cooperation of the lifting power assembly 2 and the torsion power assembly 3. At the same time, the grouting pipe 7 can carry out grouting work on the inside of the grouting screw anchor 11 from the upper part of the grouting screw anchor 11.

[0057] Specifically, the torsion power assembly 3 is provided with a spiral blade 13 that cooperates with the grouting spiral anchor 11 to break through the soil and drill down, thereby achieving the effect of breaking through the soil and drilling down.

[0058] The grouting pipe 7 carries out grouting work on the interior of the grouting spiral anchor 11 from the upper part of the grouting spiral anchor 11 ; the grouting spiral anchor 11 is provided with a grouting channel, which is connected to the grouting hole 12 .

[0059] Specifically, the control computer 9 sends a control signal and a signal collection feedback signal to the spiral pile core rigid composite pile driver through the information transmission cable 10 , and the grouting spiral anchor 11 performs constant speed drilling and installation based on the control signal of the control computer 9 .

[0060] Specifically, the torsion power assembly 3 provides torque to rotate the grouting screw anchor 11. The torsion power assembly 3 includes a torsion sensor for measuring the torque used in real time. The spiral pile core rigid composite pile driver is connected to the control computer 9 through a data transmission line, so that the control computer 9 receives the torque data measured by the spiral pile core rigid composite pile driver in real time.

[0061] Specifically, the control computer 9 has a built-in trained machine learning model, which includes the training results of the machine learning parameters M(x) to reflect and predict the pull-out bearing capacity of the grouting spiral pile in real time.

[0062] Specifically, during the entire construction process, the spiral pile core rigid composite pile driver collects torque and grouting volume data in real time and transmits it to the control computer. The control computer uses machine learning algorithms to analyze the data, adjust construction parameters in real time, and optimize the construction plan. Based on the analysis results of the machine learning model, the control computer adjusts the parameters of the construction process in real time to improve construction quality and efficiency.

[0063] Learning parameters are obtained through a large number of random models, including the anchor diameter of the spiral anchor plate at the front end of the spiral pile, the number of anchor plates, the spacing between anchor plates, the diameter of the anchor rod, the strength of the cement mortar, the friction coefficient of the steel pipe cement soil contact surface, soil information, the torque provided by the machine, and the pull-out bearing capacity in each case, providing sufficient data samples for the next step of machine learning.

[0064] Specifically, the root mean square error RMSE, mean absolute error MAE and R-squared (R 2 ) Three indicators verify the effectiveness of machine learning;

[0065]

[0066] Where yi is the measured data, To predict data, is the average value of the data set of yi, and N is the size of the data set. The RMSE and MAE index values ​​are low, and the index R 2 The value of is close to 1, indicating that the model has good performance.

[0067] Specifically, machine learning is performed on the data samples provided by the finite element analysis to train a machine learning model about the machine learning parameter M(x).

[0068] Specifically, based on a large number of stochastic models in the finite element model, the pullout bearing capacity of the grouting screw pile is predicted based on the input data, and the prediction results are fed back to the construction team to adjust the construction plan or calculate the optimal anchor plate configuration;

[0069] The control computer will use the real-time numerical values, machine learning parameters M(x), the bearing capacity improvement coefficient S(x) after screw pile grouting, and the torque coefficient k t And the torque T provided by the pile driver, through the formula F=M(x)S(x)k t T predicts the real-time pull-out bearing capacity of the grouting spiral pile and displays it on the computer screen in real time.

[0070] A method for using a spiral pile core rigid composite pile construction process based on a machine learning algorithm, specifically comprising the following steps:

[0071] Preparation for the screw pile driving process: Before construction, detailed preparation work must be done on the piling equipment and related hardware to ensure the smooth progress of the entire process.

[0072] The specific steps are as follows:

[0073] First, the preparation of the spiral core rigid composite pile driver is crucial. This includes the installation and commissioning of the base steel frame 1, the lifting power assembly 2, the torsion power assembly 3, the control computer 9, and the grouting screw anchor 11. The pile driver is connected to the control computer 9 via an information transmission cable 10 to ensure stable data transmission. The grouting screw anchor 11 is bolted to the bottom of the pile driver to ensure a secure installation. The grouting screw anchor 11 is connected to the torsion power assembly 3, and the spiral blade 13 is used to break the ground and drill down. At the same time, the grouting hole 12 is ensured to be unobstructed, preparing for the subsequent grouting process.

[0074] Before construction, ensure that the sealable pressurized cement barrel 5 is connected to the grouting pipe 7, and use the air compressor 4 to provide pressure to smoothly press the mixed cement mortar into the grouting screw anchor 11. The sliding platform 8 moves precisely under the control of the lifting power assembly 2 to ensure that the screw pile is drilled to the set depth.

[0075] Before construction, soil information of the construction site is collected and input into the control computer 9. Based on the actual needs of the site, the parameters of the pile driver, such as the torque of the pile driving, the drilling speed and the grouting volume, are adjusted to ensure the accuracy of the process.

[0076] Geometric modeling and finite element analysis: Before construction, finite element analysis is used to establish the geometric models of the spiral pile, soil, and grouting area, and to define the physical properties of the spiral pile, soil, and grouting materials (such as elastic modulus, shear modulus, and friction coefficient). By setting boundary conditions and load application methods, the pile-soil interaction and pull-out bearing capacity in actual construction are simulated. When dividing the mesh, fine meshes are used in key areas to ensure calculation accuracy. Finally, based on static, dynamic, or nonlinear analysis, accurate analysis results are obtained to provide data support for subsequent machine learning.

[0077] Machine Learning Model Training: To optimize the screw pile construction process, a machine learning model is trained based on actual construction data. Field data, including the front end of the screw pile, the diameter of the grouting screw anchor 11, the number and spacing of the spiral blades 13, the strength of the cement mortar in the sealable pressurized cement drum 5, and the grouting pressure provided by the air compressor 4, is collected and transmitted to the control computer 9 via sensors.

[0078] During the training process, the model is continuously optimized through a large amount of construction data, and the root mean square error (RMSE), mean absolute error (MAE) and R-squared (R 2) and other indicators to evaluate its accuracy. After model training is completed, it is embedded in the control computer 9. During on-site construction, the control computer 9 automatically generates optimal construction parameters based on soil information and pull-out bearing capacity requirements. It also sends real-time control signals to the pile driver via the information transmission cable 10, adjusting the torque provided by the torsion power assembly 3 and the grouting speed, ensuring the intelligent and automated construction process.

[0079] During the entire construction process, the control computer 9 receives on-site data through the information transmission cable 10, provides real-time feedback on the progress of grouting and piling, adjusts parameters to cope with on-site changes, and ensures the optimization of the construction plan.

[0080] During construction, the control computer 9 monitors key piling parameters, such as torque, drilling speed, and grouting volume, based on the model's real-time predictions. If deviations occur during construction, the model immediately adjusts, providing corrective suggestions and optimizing the construction plan. Furthermore, the model continuously learns from new data to improve its prediction accuracy.

[0081] The advantages of the spiral pile core rigid composite pile construction process combined with machine learning algorithms over traditional processes are mainly reflected in the following aspects:

[0082] Accurate prediction and real-time adjustment: Traditional construction processes rely on experience and standard formulas, while machine learning algorithms can adjust models based on real-time data from actual construction to make more accurate predictions of pull-out capacity, thereby optimizing construction plans and improving construction efficiency and accuracy.

[0083] Reduce waste of manpower and material resources: Real-time data analysis through machine learning algorithms can reduce unnecessary construction attempts and material usage, thereby reducing costs.

[0084] Improved construction performance: The application of machine learning algorithms can optimize the pile-soil contact, reduce the disturbance of the blades on the soil through grouting technology, and improve the bearing capacity of the spiral pile, especially in terms of pullout resistance.

[0085] Improved intelligence: The introduction of a real-time monitoring and feedback system makes the construction process more intelligent, automatically adjusting construction parameters such as torque and grouting speed to ensure construction quality.

[0086] Strong adaptability: Machine learning algorithms can self-adjust according to different geological conditions and construction environments to provide more adaptable construction solutions.

[0087] Construction safety: The intelligent system can monitor key parameters during construction in real time, detect potential problems in a timely manner, prevent accidents, and improve construction safety.

[0088] The real-time monitoring and feedback system uses a sensor network to collect various construction site data, including key parameters such as torque, grouting volume, drilling speed, and soil parameters. These sensors monitor various physical quantities during the construction process in real time and transmit the data to the control computer 9. After receiving the sensor data, the control computer 9 rapidly processes and analyzes it. Using pre-trained machine learning models, this data is used to assess and predict various conditions and potential risks during the construction process in real time.

[0089] In summary, the application of machine learning algorithms has significantly improved the prediction accuracy, efficiency, performance, intelligence level and safety of spiral pile core rigid composite pile construction technology compared with traditional technologies.

[0090] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A process for forming a spiral pile core rigid composite pile based on a machine learning algorithm, characterized in that: The specific steps include: Step 1: Conduct finite element analysis: First, establish the geometric model of the screw pile, soil and grouting area to reflect the size and layout in the actual project; then, define the physical properties of the screw pile, soil and grouting materials, set boundary conditions and load application methods, simulate the foundation conditions and the stress of the pile during construction, and apply corresponding pressure to the grouting process; then, perform grid division, and use fine grids in key areas to improve calculation accuracy; finally, select the analysis type to ensure that the model can simulate the mechanical behavior of the screw pile under various working conditions; Step 2: Perform machine learning to obtain a machine learning model for the machine learning parameter M(x); Step 3: Import the trained machine learning model of the machine learning parameter M(x) into the computer, and connect the computer to the spiral pile core rigid composite pile driver; Step 4: After inputting the soil information and the pull-out bearing capacity required for the project at the construction site, the required screw anchor structure size for this project will be displayed on the computer screen; Step 5: Use the formula F = M(x)S(x)k t T reflects the pull-out bearing capacity of the grouting spiral anchor in real time.

2. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 1, characterized in that: The driving and grouting of the spiral piles are achieved by a spiral pile core rigid composite pile driver and a grouting system; the spiral pile core rigid composite pile driver comprises a base steel frame (1), a lifting power component (2), a sliding platform (8), and a torsion power component (3); the grouting system comprises a sealable pressurized cement barrel (5) and an air compressor (4); The lifting power assembly (2) is fixed on the top of the base steel frame (1), the lower end of the lifting power assembly (2) is rigidly welded to the sliding platform (8), the torsion power assembly (3) is fixed on the front upper end of the sliding platform (8), the upper end of the torsion power assembly (3) is connected to the grouting pipe (7), the lower end is connected to the grouting screw anchor (11), and the side end is connected to the information transmission cable (10) connected to the control computer (9); The side of the grouting pipe (7) away from the torque power assembly (3) is connected to a sealable pressurized cement barrel (5) for mixing cement mortar, and the other side of the sealable pressurized cement barrel (5) is connected to an air compressor (4) for pressurizing the sealable pressurized cement barrel (5) through an air pipe (6). The cement mortar being mixed in the sealable pressurized cement barrel (5) is pressed into the grouting pipe (7) and into the interior of the grouting screw anchor (11) along the grouting pipe (7), and then sprayed into the soil through the grouting hole (12) on the anchor body of the grouting screw anchor (11); The lifting power assembly (2) drives the grouting spiral anchor (11) to lift and lower, and the torsion power assembly (3) drives the grouting spiral anchor (11) to rotate. The lifting power assembly (2) and the torsion power assembly (3) cooperate to realize the drilling and installation of the grouting spiral anchor (11).

3. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 2, characterized in that: The torsion power assembly (3) is provided with a spiral blade (13) cooperating with the grouting spiral anchor (11) to break through the soil and drill down; The grouting pipe (7) performs grouting work on the interior of the grouting spiral anchor (11) from the upper part of the grouting spiral anchor (11); a grouting channel is provided on the grouting spiral anchor (11), and the grouting channel is connected to the grouting hole (12).

4. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 2, characterized in that: The control computer (9) sends a control signal and a signal collection feedback signal to the spiral pile core rigid composite pile driver via an information transmission cable (10), and the grouting spiral anchor (11) performs constant speed drilling and installation based on the control signal of the control computer (9).

5. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 2, characterized in that: The torque power assembly (3) provides torque to rotate the grouting screw anchor (11). The torque power assembly (3) includes a torque sensor for measuring the torque used in real time. The spiral pile core rigid composite pile driver is connected to the control computer (9) via a data transmission line, so that the control computer (9) receives the torque data measured by the spiral pile core rigid composite pile driver in real time.

6. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 2, characterized in that: The control computer (9) has a built-in trained machine learning model, which includes the training results of the machine learning parameters M(x) to reflect and predict the pull-out bearing capacity of the grouting spiral pile in real time.

7. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 6, characterized in that: The spiral pile core rigid composite pile driver collects torque and grouting volume data in real time and transmits it to the control computer. The control computer uses machine learning algorithms to analyze the data, adjust construction parameters in real time, optimize the construction plan, and adjust the parameters of the construction process in real time according to the analysis results of the machine learning model; Learning parameters are obtained through a large number of random models, including the anchor diameter of the spiral anchor plate at the front end of the spiral pile, the number of anchor plates, the spacing between anchor plates, the diameter of the anchor rod, the strength of cement mortar, the friction coefficient of the steel pipe cement soil contact surface, soil information, the torque provided by the machine, and the pull-out bearing capacity in each case, providing sufficient data samples for the next step of machine learning.

8. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 6, characterized in that: By using the root mean square error RMSE, mean absolute error MAE and R 2 Three indicators verify the effectiveness of machine learning; Where yi is the measured data, Forecast data, is the mean of the dataset of yi, and N is the size of the dataset.

9. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 6, characterized in that: Perform machine learning on the data samples provided by finite element analysis to train a machine learning model about the machine learning parameter M(x).

10. The process for forming a spiral pile core rigid composite pile based on a machine learning algorithm according to claim 6, characterized in that: Based on a large number of stochastic models of the finite element model, the pullout bearing capacity of the grouting screw pile is predicted according to the input data, and the prediction results are fed back to the construction team to adjust the construction plan or calculate the optimal anchor plate configuration; The control computer will use the real-time numerical values, machine learning parameters M(x), the bearing capacity improvement coefficient S(x) after screw pile grouting, and the torque coefficient k t And the torque T provided by the pile driver, through the formula F = M (x) S (x) k t T predicts the real-time pull-out bearing capacity of the grouting spiral pile and displays it on the computer screen in real time, where F is the pull-out bearing capacity of the grouting spiral pile; M(x) is the machine learning parameter; S(x) is the bearing capacity improvement coefficient of the spiral pile after grouting; k t is the torque coefficient; T is the torque provided by the pile driver.

Images