A Digital Twin Modeling Method Based on the Piling Process

Abstract

The present invention relates to the field of advanced manufacturing and automation technologies, and particularly to a digital twin modeling method based on the pile driving process. This method includes a construction equipment entity: which generates dynamic parameters during actual construction; a data acquisition platform: which is used to realize the sensing, acquisition, transmission, and processing of relevant parameters; a digital twin model: which conducts simulations by fusing relevant geometric / physical models, combines the input parameters, and predicts the subsequent state of the pile driving model in real time, outputting prediction parameters; and an intelligent control system: which visually displays the numerical values of the core parameters of the current pile driving process in real time and outputs the predicted values of some parameters. The present invention can achieve remote real-time monitoring of the core parameters of the pile driving process and simultaneously predict subsequent changes; during actual construction, the construction process can be optimized according to the prediction results output by the digital twin model, improving the safety and efficiency of the pile driving process and ensuring that the various technical indicators during the pile driving process meet national standards.

Description

Technical Field

[0001] The present invention relates to the field of advanced manufacturing and automation technologies, and particularly to a digital twin modeling method based on the pile driving process. Background Art

[0002] Installing offshore wind turbines is an important way to utilize wind energy. Currently, the pile foundation is an important installation form for installing offshore wind turbines. With the continuous increase in the installed capacity of offshore wind turbines and the increasing distance from the shore, higher requirements are put forward for the installation of wind turbines. Only by continuously optimizing the existing pile driving construction process can the installation of offshore wind turbines be completed with high quality, high precision, and high efficiency.

[0003] Currently, the main detection target objects in the pile driving process are the verticality of pile installation and the penetration of the pile. However, the current detection method is to detect once every certain number of hammer blows or pile displacement. If the requirements are met, the construction continues; otherwise, it stops for adjustment. Therefore, the current pile driving process cannot monitor the changes of core parameters in the construction process in real time and continuously, nor can it make dynamic adjustments according to the actual situation of the operation. Problems such as untimely and inaccurate regulation of pile driving equipment may lead to serious construction problems such as sudden changes in penetration, pile inclination, or severe cracks in the pile during the pile driving process, which will not only cause huge property losses but also lead to safety accidents. Therefore, remotely and real-time monitoring the core parameters of the pile driving process and predicting subsequent changes have important practical significance for ensuring the high-precision and high-efficiency completion of the installation of offshore wind turbines. Summary of the Invention

[0004] The object of the present invention is to propose a digital twin modeling method based on the pile driving process for the problems existing in the background art.

[0005] The technical solution of the present invention, a digital twin modeling method based on the pile driving process, includes constructing a data acquisition platform, a digital twin model, and a digital twin model framework for optimizing the construction process driven by twin data; constructing a digital twin model system that can output the changes of core parameters in each pile driving process in real time and remotely visualize and display the data in front of construction personnel; constructing a pile driving process digital twin prediction model system including a pile driving process digital twin accident prevention prediction model and a pile driving process digital twin verticality error prediction and correction model. Through the established models, the pile driving process can be remotely detected, the verticality error of the pile driving process can be output in real time, and the subsequent verticality error can be predicted. The actual pile driving process is optimized through the feedback data of the digital twin system. The overall framework diagram of the digital twin system is as Figure 1 shown.

[0006] Through functional and feature analysis, combined with the motion characteristics, the actual complex hydraulic system is simplified into a hydraulic hammer. Then, by determining the types and functions of the components in the pile driving process and considering the accuracy requirements of actual modeling and simulation, a one-to-one modeling of the dolly, dolly ring, and pile body is carried out. Finally, a simplified three-dimensional model of the pile driving process composed of a hydraulic hammer, dolly, dolly ring, and pile body is constructed, taking into account the requirements of enhancing the accuracy of simulation and reducing the calculation time. The simplified three-dimensional model diagram of the pile driving process is as shown in Figure 2 shown.

[0007] By analyzing the operating characteristics of the pile driving process and comprehensively considering the core data concerned in actual construction, it is determined that the displacement x (including three components) of each component, the velocity v (including three components) of each component, the acceleration a of each component, the vertical contact force F N between each contact surface, the lateral force F f between each contact surface, the hammering efficiency η, and the pile body verticality error μ are the core operating parameters of the pile driving process, and are remotely visualized through the digital twin system and presented to the operator, facilitating on-site construction personnel to detect the pile driving process and quickly judge the construction status.

[0008] Construct a digital twin prediction model system for the pile driving process, including two parts: the digital twin prediction model for preventing construction accidents in the pile driving process and the digital twin prediction and correction model for the verticality error in the pile driving process.

[0009] The acceleration and contact force during the pile driving process can be calculated in advance and compared with the pre-set maximum allowable values to prevent excessive acceleration and contact force during the pile driving process. The flow chart of the digital twin prediction model for preventing construction accidents is as shown in Figure 3 shown.

[0010] Through finite element simulation calculation, the prediction curves of the verticality error and the lateral force are obtained, and as the pile driving process progresses, the prediction curves will also be continuously updated to ensure the safe progress of the pile driving process.

[0011] If the verticality error exceeds the maximum allowable verticality error in the prediction curve, according to the lateral force prediction curve obtained from the simulation results, a lateral force for correcting the verticality error can be given in advance through the pile stabilizing platform in the pile driving auxiliary equipment until the verticality error in the prediction curve is within the maximum allowable verticality error. The flow chart of the digital twin prediction and correction model for the verticality error is as shown in Figure 4 shown.

[0012] Compared with the prior art, the present invention has the following beneficial technical effects:

[0013] 1. A digital twin model for the pile driving process is constructed, which, in cooperation with the data acquisition platform, realizes the real-time interaction of core parameters during the pile driving process, enabling remote visual real-time monitoring of the complex pile driving process.

[0014] 2. In the existing detection method during the pile driving process, the quality index is detected every certain number of hammer blows or pile body displacements, which cannot continuously reflect the pile driving process and is difficult to prevent serious construction accidents such as sudden changes in penetration, pile body inclination, or severe cracks in the pile. The digital twin accident prevention prediction model for the pile driving process constructed in the present invention can continuously display the changes of relevant parameters during the pile driving process, facilitating on-site construction personnel to detect the pile driving process and quickly judge the construction status.

[0015] 3. The digital twin accident prevention prediction model for the pile driving process and the digital twin verticality error prediction and correction model for the pile driving process constructed in the present invention can, based on the actual construction of the pile driving process, combine with the finite element model to predict the subsequent verticality error and lateral force during the construction process. It can adjust the input parameters of the hydraulic hammer in advance according to the prediction results, prevent excessive verticality errors, increase the safety of the pile driving process, and improve the construction efficiency of the pile driving process. Description of the Drawings

[0016] Figure 1 It is the overall system framework diagram of the present invention.

[0017] Figure 2 It is the simplified three-dimensional model of each component of the present invention.

[0018] Figure 3 It is the flow chart of the digital twin accident prevention prediction model of the present invention.

[0019] Figure 4 It is the flow chart of the digital twin verticality error prediction and correction model of the present invention.

[0020] Reference Signs: 1. Hydraulic Hammer; 2. Dolly Ring; 3. Dolly; 4. Pile Body. Detailed Embodiments

[0021] A digital twin modeling method based on the pile driving process proposed by the present invention constructs a digital twin model framework for the pile driving process;

[0022] The method of the present invention based on digital twin technology models the characteristics of each component during the pile driving process, combines the geometric / physical model with the data modeling method, and based on the digital twin technology and the motion characteristics of the pile driving process, determines the flow chart of the digital twin system for the pile driving process as Figure 1 shown, which mainly includes three parts: a data acquisition platform, a digital twin model, and twin data-driven construction process optimization.

[0023] Among them, the data acquisition platform mainly consists of a high-strain pile driving analyzer, GPS, theodolite, and level. The main data collected includes two parts: component attribute parameters and operation parameters, as shown in Table 1.

[0024] Table 1 Data Collected by the Data Acquisition Platform

[0025]

[0026]

[0027] Static data is obtained based on the equipment conditions during the actual pile sinking process and is input as constants into the digital twin model of the pile sinking process. The real-time measured data is obtained through a high-strain pile driving analyzer, GPS, theodolite, level, and other sensors, and is used both to construct the digital twin model and as parameters to verify the model performance.

[0028] The digital twin model is constructed based on the data output by the data acquisition platform and the geometric / physical model. First, based on the parameters such as the geometric dimensions and assembly relationships of each component measured, and then through functional analysis, the functions of each component during the pile sinking process are determined. Finally, a simplified three-dimensional model of each component during the pile sinking process is obtained, as Figure 2 shown.

[0029] Based on the above-determined simplified three-dimensional model, it is imported into ABAQUS, and combined with the geometric and physical models, finite element analysis is performed on the three-dimensional model in ABAQUS.

[0030] The finite element analysis process includes the following series of steps:

[0031] Setting the basic material properties of each component during the pile sinking process, including: density, elastic modulus, Poisson's ratio, stiffness, relevant friction coefficients of contact, damping;

[0032] Setting the assembly relationships of each component, that is, the relative position relationships.

[0033] The component mesh division determines the appropriate number and size of meshes according to the actual working conditions.

[0034] Determining the boundary conditions during the pile sinking process, including according to the actual energy input of the hydraulic hammer 1, and according to determining the initial velocity of the hammer core in the finite element model as the input parameter for driving the model.

[0035] Selecting the final output parameters, including: displacement x of each component (including three components), velocity v of each component (including three components), acceleration a of each component, vertical contact force F between each contact surface N , lateral force F between each contact surface f , hammering efficiency η, and verticality error μ of the pile body 4.

[0036] Construct a digital twin model system for the pile driving process;

[0037] The construction process of the offshore wind turbine pile driving is as follows: while ensuring the verticality of the single pile, start the hydraulic hammer 1, first jog 1 - 2 hammers with a small energy, and after no abnormalities, continue to jog 2 - 3 hammers. Do this 3 - 4 times, stop the hammer and arrange to measure and observe the pile body data to adjust the verticality of the pile body. If there is no change, continue pile driving, and the observation and adjustment frequency is changed to observe and adjust once every 1 - 2 m. When the pile continues to penetrate 10 m into the soil, the observation and adjustment are changed to observe and adjust once every 3 - 4 m. When the pile penetration depth is about 30 m, change to continuous hydraulic pile driving.

[0038] The current pile driving process cannot monitor the changes of the core parameters in the construction process in real time and continuously. However, the digital twin model system for the pile driving process constructed based on the present invention can output the changes of the core parameters in each pile driving process in real time, such as: the contact forces of each component, the penetration degree of the pile, and the verticality error, etc. The data is remotely visualized and presented in front of the construction personnel, which is convenient for the on - site construction personnel to detect the pile driving process and quickly judge the construction status.

[0039] First, with the help of the data collected by the data acquisition platform, it is integrated by digital twin technology to obtain an interactive human - machine interface. At the same time, the actual energy input of the hydraulic hammer 1 collected is input as the initial parameter into the finite - element model, and the data calculated by the finite - element model can be obtained immediately. The calculated data can be used as an auxiliary reference for the data collected by the data acquisition platform to further enhance the reliability of the data. The data interface mainly consists of these two parts.

[0040] Construct a digital twin prediction model system for the pile driving process;

[0041] The digital twin prediction model system for the pile driving process includes a digital twin accident prevention prediction model for the pile driving process and a digital twin verticality error prediction and correction model for the pile driving process.

[0042] The construction method of the digital twin accident prevention prediction model for the pile driving process is as follows:

[0043] First, based on the data collected by the data acquisition platform, the energy input of the hydraulic hammer 1 is used as the input parameter for driving the model. According to Newton's second law F = ma, it can be known that when the mass of each component remains unchanged, the contact force transmitted between each component is proportional to the acceleration of each component. According to the input parameter, the vertical contact force F of each component in the whole process during the pile driving process can be calculated N and the acceleration a of each component.

[0044] Then, from the calculation results in the finite element model, i.e., the accelerations of each component and the vertical contact force. Compare with the pre-set maximum vertical contact force [F N and the maximum acceleration [a]. If the maximum allowable value is not exceeded, the pile driving process proceeds normally. If subsequent predicted values exceed the maximum allowable value, feedback an alarm, stop the pile driving process, and change the input parameters until the predicted value drops below the maximum allowable value.

[0045] The flow chart of the digital twin prediction model for preventing construction accidents during the pile driving process is as Figure 3 shown.

[0046] The construction method of the digital twin verticality error prediction and correction model during the pile driving process is as follows:

[0047] First, based on the finite element model, with the existing observed values as the initial conditions, including:

[0048] The input energy of the pile driving hammer core for the first time (the initial velocity of the hammer core is calculated according to ), the initial verticality error of the pile, and the initial relative positions of each component. Through one finite element simulation, the verticality error μ1 of the pile after the first hammer blow can be obtained.

[0049] Then, based on this new verticality error μ1, the existing finite element simulation model is updated in real time. Given the input energy of the pile driving hammer core for the second time, the predicted verticality error μ2 for the second time can be obtained. At the same time, the lateral force F C2 generated during the second hammer blow can be obtained. Repeat the calculation n times, then μ1, μ2... μ n and the lateral force F C2 , F C3 , ……, F Cn can be obtained. Then, with the number of hammer blows as the abscissa and the verticality error and lateral force as the ordinate, the verticality error prediction curve and the lateral force prediction curve can be obtained.

[0050] If the verticality errors of the prediction curve do not exceed the maximum allowable verticality error, the construction of the next hammer can proceed normally. At the same time, according to the new data collected by the data acquisition platform, update the input parameters in the finite element model and continue to repeat the simulation process until the entire pile driving process ends smoothly.

[0051] If the verticality error exceeds the maximum allowable verticality error in the prediction curve, then according to the lateral force prediction curve obtained from the simulation results, through the pile stabilizing platform during the pile driving process, a lateral force for correcting the verticality error can be given in advance until the verticality error in the prediction curve is below the maximum allowable verticality error.

[0052] It should be noted that finite element simulations can be carried out in advance for typical working conditions and typical pile driving processes, and the corresponding simulation results can be stored as a database. During actual construction, the database can be called for quick matching, saving calculation time and ensuring real-time control of the pile driving process.

[0053] The flow chart of the digital twin verticality error prediction model for the pile driving process is as Figure 4 shown.

[0054] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto, and various changes can be made without departing from the spirit of the present invention within the scope of knowledge of those skilled in the art to which the present invention pertains.

Claims

1. A digital twin modeling method based on the pile driving process, characterized in that, It includes the following specific steps: S1. Build a data acquisition platform to collect, transmit, and process data during the pile driving process; S2. Conduct structural analysis, functional analysis, and motion characteristic analysis on the pile driving process, and build a digital twin model of the pile driving process; S21. Determine the functions of each component during the pile driving process through functional analysis; S22. Determine the operating parameters during the pile driving process through motion analysis; S23. Determine the main components, relative position characteristics, divide independent subsystems for digital twin modeling during the pile driving process through structural analysis, and finally determine the geometric and physical models; Based on the functional analysis and structural analysis in S21 and S23, determine the simplified three-dimensional models of each component during the pile driving process; Among them, the hydraulic hammer that inputs energy during the pile driving process is simplified into a cylindrical hammer core, and its impact energy and frequency are used as input parameters in the system; The follower and follower ring are buffer components between the hammer core and the pile, and their displacements and contact forces are observation variables in the system; The pile is the core component of the pile driving system, and its displacement, vertical contact force, lateral force, and verticality error are observation variables in the system; S3. Based on the digital twin model of the pile driving process, fuse the geometric, physical models, and dynamic mechanism models, and use the characteristic parameters of the displacements, contact forces of the follower, follower ring, and pile, and the verticality error of the pile body as observation variables to realize digital twin modeling of the pile driving process, specifically including displacement characteristics, contact force characteristics, hammering efficiency characteristics, and verticality change characteristics; S4. Build a digital twin prediction model for the pile driving process to optimize the construction process of the pile driving process with twin data; Import the simplified three-dimensional model into ABAQUS, and combine the geometric and physical models determined in step S2 to perform finite element analysis on the three-dimensional model in ABAQUS; The finite element analysis process includes the following steps: Set the basic material properties of each component during the pile driving process, set the assembly relationship of each component, divide the component meshes, determine the boundary conditions of the pile driving process, and select the final output parameters; Among them, the impact energy and frequency of the hydraulic hammer during the actual pile driving process are used as input parameters in the finite element model to drive the model to run, and the finite element model outputs corresponding parameters in real time; Output the energy finally transmitted to the top of the pile during the pile driving process and compare it with the input impact energy; build the digital twin hammering efficiency characteristics of the pile driving process; Output the lateral displacement of the pile and build the digital twin pile body verticality error characteristics of the pile driving process.

2. The digital twin modeling method based on the pile driving process according to claim 1, wherein The displacement, vertical contact force, lateral force, and verticality error parameters of the pile body are used as prediction variables, and the input parameters during the pile driving process are regulated with the prediction variables as indicators.

3. A digital twin modeling method based on the pile driving process according to claim 1, characterized in that, Based on the motion characteristic analysis in S22, determine that the motion characteristic parameters during the pile driving process are the displacements and accelerations of each component, and the displacements and accelerations are observation variables and prediction variables; Among them, the above-mentioned observation variables and input parameters are collected through the data acquisition platform; Among them, the data acquisition platform includes a high-strain pile driving analyzer, GPS, theodolite, and level; The high-strain pile driving analyzer outputs the hammering force and the hammer efficiency of the hydraulic hammer; GPS, theodolite, and level measure the penetration of the pile and the verticality error of the pile body.

4. A digital twin modeling method based on the pile driving process according to claim 1, characterized in that Compare the predicted values of the accelerations of the sub forging and the sub forging ring and the contact force at the top of the pile calculated in the finite element model with the pre-set allowable maximum values. If the maximum value is exceeded, immediately stop the pile driving process and modify the input.

5. A digital twin modeling method based on the pile driving process according to claim 1, characterized in that, Based on the finite element model, with the existing observed values as the input, simulate the verticality error and lateral force of the pile. Based on this new verticality error, give the next input to obtain the subsequent predicted verticality error and lateral force. Repeat the calculation n times, and then the predicted verticality error curve and lateral force prediction curve after n hammer blows can be obtained.

6. A digital twin modeling method based on the pile driving process according to claim 5, characterized in that If the verticality error exceeds the maximum allowable verticality error in the prediction curve, adjust the input of the hydraulic hammer in advance according to the simulation results until the verticality error in the prediction curve is below the maximum allowable verticality error.

7. A digital twin modeling method based on the pile driving process according to claim 5, characterized in that, If the verticality error of the prediction curve does not exceed the maximum allowable verticality error, the construction of the next hammer can be carried out normally. At the same time, update the input parameters in the finite element model according to the new data collected by the data acquisition platform, and continue to repeat the simulation process until the entire pile driving process is successfully completed.

8. A digital twin modeling method based on the pile driving process according to claim 5, characterized in that If the verticality error exceeds the maximum allowable verticality error in the prediction curve, a lateral force for correcting the verticality error can be given in advance through the pile stabilizing platform during the pile driving process according to the lateral force prediction curve obtained from the simulation results until the verticality error in the prediction curve is below the maximum allowable verticality error.

Images