Shield tunneling machine tunneling speed and ground surface settlement prediction method based on mechanism and data driving
By establishing a force model and nonlinear finite element simulation for cutting soil by cutting knife, combining actual measured data and noise reduction processing, a prediction model for improving particle swarm algorithm optimization support vector machine is established, and the force detection of cutting knife in shield construction is solved The problem of difficulty and reliability of surface settlement prediction achieves real-time prediction of high-precision and low resource consumption.
Patent Information
- Application Number
- CN202411020329.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-28
- Publication Date
- 2024-11-12
AI Technical Summary
The existing shield construction is difficult to detect the cutting force, which makes it difficult to obtain cutting force. The existing surface settlement prediction methods lack material mechanism and mechanical theoretical support, the prediction results are poorly interpreted, and lack reliability and scalability. Relying on a single simulation software causes high computing resources to consume and real-time predictions cannot be achieved.
Based on the mechanism and data-driven method, the stress model of the cutting soil is established through geological exploration, and nonlinear finite element simulation is used to perform nonlinear finite element simulation, combined with actual measured data, and the cutting force calculation formula is verified, and construction data is collected for noise reduction processing. Establish a dual-drive prediction model for improving particle swarm algorithm optimization support vector machine to achieve accurate prediction of excavation speed and surface settlement.
It significantly improves the accuracy of excavation speed and surface settlement prediction, shortens prediction time, saves computing resources, and provides more reliable and real-time construction support.
Smart Images

Figure CN118940579A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a shield machine excavation speed and surface settlement prediction method based on mechanism and data drive, belonging to the technical field of shield tunnel development. Background Art
[0002] The shield method has been widely used in urban underground projects due to its advantages such as high safety and construction not being affected by weather.
[0003] The cutting force of the shield cutter is the key load of the shield cutter head. Existing equipment is difficult to directly detect the force of the cutter during the excavation process, which makes it difficult and costly to obtain the cutting force in actual construction. Therefore, how to design a set of accurate calculation formulas for cutting force is of great significance to improving the accuracy of subsequent excavation speed and surface settlement prediction.
[0004] Existing methods often predict possible surface settlement of shield construction based on actual recorded construction parameters and geological parameters obtained from exploration. The prediction results are poorly interpretable, and the input-output relationship is a "black box". It lacks the support of material science mechanism and mechanical theory, and lacks reliability and scalability. At present, single simulation software is often used to predict the excavation speed and surface settlement. When the input model parameters of this method are consistent with the actual situation, the results obtained are highly consistent with the detection data of the actual project. However, the method of prediction based on simulation alone has obvious defects. Each calculation consumes a lot of time and computing resources, which is not conducive to the staff to achieve real-time prediction during the construction process.
[0005] Therefore, those skilled in the art are in urgent need of providing a new method for predicting surface settlement during shield construction to solve the above technical problems. Summary of the invention
[0006] The purpose of the present invention is to provide a shield machine excavation speed and surface settlement prediction method based on mechanism and data drive to solve the problems existing in the above-mentioned prior art.
[0007] In order to achieve the above object, the present invention provides the following technical solutions:
[0008] A shield machine excavation speed and surface settlement prediction method based on mechanism and data drive includes the following steps;
[0009] S1. After geological exploration, starting from rock and soil mechanics, considering the mechanical properties of the soil, a force model for the cutter cutting the soil is established. Figure 2 As shown, the calculation formula of the horizontal cutting force of the cutter is further derived;
[0010] S2. Use ABAQUS software to simulate the soil cutting process of the cutter using the nonlinear finite element method, and compare and verify the accuracy of the cutting force calculation formula with the measured data of the shield machine. Figure 3 As shown in Figure 2, the average error between the theoretical cutting force and the simulated cutting force is Figure 4 As shown;
[0011] S3, collect the data of cutter head speed, main thrust and screw conveyor speed at the actual shield machine construction site, and perform noise reduction on them. The effects of different data after noise reduction are as follows: Figure 5-7 As shown. The cutting force is further used as an input parameter, and the construction data such as shield machine are introduced into the dual-drive prediction model as input parameters of the dual-drive prediction model. The tunneling speed and surface settlement prediction model of the support vector machine optimized by the improved particle swarm algorithm is established. The specific process is as follows Figure 8 As shown, the excavation speed and surface settlement can be predicted.
[0012] Preferably, step S1 comprises the following sub-steps:
[0013] S11, analyzing the motion characteristics of the shield machine cutter and obtaining the motion parameter equation of any cutter;
[0014] S12. Based on the motion parameter equation of the cutter, a cutter cutting soil mechanism model is constructed;
[0015] S13. Perform a decoupling analysis of the cutter cutting force and the cutter disc torque, and construct a cutter disc torque decoupling model.
[0016] Preferably, step S2 comprises the following sub-steps:
[0017] S21. According to the geological exploration situation, the soil constitutive model is constructed, and the linear Drucker-Prager (DP) model in ABAQUS is used to simulate the constitutive relationship of the cutting tool affecting the soil, and the soil material parameters are set;
[0018] S22, establish a three-dimensional model of the tool, set the three-dimensional size of the tool, define the tool as a rigid body in ABAQUS software, and set the conductivity;
[0019] S23, establish tool and soil constraints, create rigid body constraints at the reference node on the tool, and surface-to-surface contact is used to describe the contact between the tool and the soil. Apply the base surface of the fixed constraint and the front and rear surfaces of the symmetry constraint to the soil model. Set the top, right and left surfaces as free surfaces;
[0020] S24, perform simulation contact settings, use the Abaqus / Explicit dynamic display algorithm to ignore the convergence problem in the calculation, reduce the calculation complexity of the contact, and use the penalty function method in the contact algorithm between the tool and the soil;
[0021] S25, conduct simulation analysis based on actual exploration data and apply ALE adaptive mesh technology so that the deformation of these soil meshes can more coherently describe the cutting process;
[0022] Preferably, step S3 comprises the following sub-steps:
[0023] S31, collecting three construction data of shield machine cutter head speed, screw conveyor speed and main thrust as shield construction control parameters; the cutter head torque and the shield cutter cutting force calculated in the previous text are defined as shield construction state parameters; the excavation speed and surface settlement are defined as shield construction performance indicators;
[0024] S32, performing noise reduction processing on the tunneling state parameters based on variational mode decomposition and detrending fluctuation analysis algorithm;
[0025] S33, the correlation of shield machine excavation parameters is analyzed. After analysis, it is found that the correlation between excavation speed and screw conveyor is above 0.9, and the linear correlation is high. SVM is used for prediction;
[0026] S34. Surface subsidence is predicted using an improved particle swarm algorithm.
[0027] The present invention can greatly improve the subsequent excavation speed and the accuracy of surface settlement prediction, save time and save computing resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 This is a diagram of the earth pressure balance shield machine cutter cutting mechanism model and the mechanism and data-driven tunneling speed prediction method of the present invention.
[0029] Figure 2 This is a flow chart of surface settlement and tunneling speed prediction of the present invention.
[0030] Figure 3 It is a schematic diagram of the three-dimensional model of the cutter and soil of the present invention.
[0031] Figure 4 This is a comparison chart of the theoretical cutting force and the simulated cutting force of the present invention.
[0032] Figure 5 This is a comparison chart of the average error between the theoretical cutting force and the simulated cutting force of the present invention.
[0033] Figure 6 This is a comparison diagram of the noise reduction of the cutter head speed data of the present invention.
[0034] Figure 7 This is a comparison diagram of the noise reduction of the screw conveyor data of the present invention.
[0035] Figure 8This is a comparison diagram of main thrust data noise reduction of the present invention.
[0036] Fig. 9 This is the IPSO-SVR prediction flow chart of the present invention. DETAILED DESCRIPTION
[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.
[0038] Embodiment 1:
[0039] This embodiment discloses a method for predicting shield machine excavation speed and surface settlement based on mechanism and data drive. The overall flow chart is as follows: Figure 1 As shown, the following steps are included:
[0040] S1. After geological exploration, starting from rock and soil mechanics, considering the mechanical properties of the soil, a force model for the cutter cutting the soil is established. Figure 2 As shown, the calculation formula of the horizontal cutting force of the cutter is further derived;
[0041] S2. Use ABAQUS software to simulate the soil cutting process of the cutter using the nonlinear finite element method, and compare and verify the accuracy of the cutting force calculation formula with the measured data of the shield machine. Figure 3 As shown in Figure 2, the average error between the theoretical cutting force and the simulated cutting force is Figure 4 As shown;
[0042] S3, collect the data of cutter head speed, main thrust and screw conveyor speed at the actual shield machine construction site, and perform noise reduction on them. The effects of different data after noise reduction are as follows: Figure 5-7 As shown. The cutting force is further used as an input parameter, and the construction data such as shield machine are introduced into the dual-drive prediction model as input parameters of the dual-drive prediction model. The tunneling speed and surface settlement prediction model of the support vector machine optimized by the improved particle swarm algorithm is established. The specific process is as follows Figure 8 As shown, the excavation speed and surface settlement can be predicted.
[0043] Specific:
[0044] (1) Step S1 is to establish a force model of the cutter cutting the soil, and further derive a calculation formula for the horizontal cutting force of the cutter, which specifically includes the following sub-steps:
[0045] S11, analyzing the motion characteristics of the shield machine cutter and obtaining the motion parameter equation of any cutter;
[0046] q i (t) = R i cos(ωt+κ i )
[0047] p i (t) = -R i sin(ωt+κ i )
[0048] r i (t) = r i +vt where: i is the cutter number, indicating the cutter at different positions; q i (t) is the coordinate value of the cutter on the q axis; p i (t) is the coordinate value of the cutter on the p axis; r i (t) is the coordinate value of the cutter on the r axis; R i is the cutter installation radius (m); κ i is the initial phase angle of the cutter (°); t is the excavation time (s).
[0049] S12. Based on the motion parameter equation of the cutter, a cutter cutting soil mechanism model is constructed;
[0050] The cutting depth per rotation of the cutter disc is expressed as
[0051] Where v is the advancement speed of the earth pressure balance shield, and n is the cutterhead speed;
[0052] The cutting depth of a single tool is:
[0053] Where β is the angle between two adjacent tools in the circumferential direction,
[0054] The shear area of the tool during cutting is:
[0055] Where w is the tool width; α is the tool forward tilt angle,
[0056] The earth pressure applied on tool i is:
[0057] σ i =K0γ(HL i sinθ i ), where θ i is the angle of the tool relative to the horizontal plane,
[0058] Then the shear strength of the soil around tool i is:
[0059] Where c is soil cohesion,
[0060] Therefore, the cutting force of cutter i can be expressed as:
[0061] F ci =τ i ·A
[0062] S13. Perform a decoupling analysis of the cutter cutting force and the cutter disc torque, and construct a cutter disc torque decoupling model.
[0063] The torque of the cutterhead is determined based on the diameter of the shield machine. The formula currently widely used is:
[0064] T=αD 3
[0065] First, calculate the positive friction torque. When the cutter head rotates, the friction torque is generated as shown in the formula
[0066] p=K0γH, where γ is the soil bulk density; H is the depth of the covering layer; K0 is the earth pressure coefficient related to the soil properties.
[0067] Assume that the front of the cutterhead is perpendicular to the horizontal line and ignore the additional load on the ground. Consider the front action point, define the angle of the point relative to the horizontal center as θ, and define the radial distance between the point and the cutterhead center as r. By integration, the friction torque on the front of the cutterhead is obtained as:
[0068] Where f is the dynamic friction coefficient. In fact, considering the influence of the cutter head opening rate, the above formula is corrected to:
[0069] Where η is the cutter head opening rate. Usually, when the cutter head speed is low and the cutter head speed is less than 2 r / min, the cutter head viscous friction is negligible compared with the Coulomb friction.
[0070] Then the friction moment of the cutter disc side is calculated. The friction moment of the cutter disc side is caused by the soil pressure. The soil pressure consists of two parts: the vertical component p1 and the lateral component p2. The formula is expressed as:
[0071]
[0072] Therefore, the friction torque generated by the vertical pressure and the lateral pressure can be calculated by integrating
[0073] Where W is the width of the cutter head,
[0074] The total friction torque applied to the side of the cutter head is
[0075]
[0076] Then the mixed torque in the soil bin is calculated. There are many cutters distributed in an orderly manner on the cutterhead, and the total cutting torque is composed of the torque applied to each cutter.
[0077] Based on the calculated cutting force formula, the cutting torque formula is as follows
[0078] In the formula, F ci is the cutting force of the tool, L i is the distance between tool i and the center of the cutter head,
[0079] The cutting torque of the cutter disc is:
[0080]
[0081] (2) Step S2 is the modeling and simulation of the soil cutting mechanism of the shield machine cutter, which specifically includes the following sub-steps:
[0082] S21. According to the geological exploration situation, the soil constitutive model is constructed, and the linear Drucker-Prager (DP) model in ABAQUS is used to simulate the constitutive relationship of the cutting tool affecting the soil, and the soil material parameters are set;
[0083] S22, establish a three-dimensional model of the tool, set the three-dimensional size of the tool, define the tool as a rigid body in ABAQUS software, and set the conductivity;
[0084] S23, establish tool and soil constraints, create rigid body constraints at the reference node on the tool, and surface-to-surface contact is used to describe the contact between the tool and the soil. Apply the base surface of the fixed constraint and the front and rear surfaces of the symmetry constraint to the soil model. Set the top, right and left surfaces as free surfaces;
[0085] S24, perform simulation contact settings, use the Abaqus / Explicit dynamic display algorithm to ignore the convergence problem in the calculation, reduce the calculation complexity of the contact, and use the penalty function method in the contact algorithm between the tool and the soil;
[0086] S25. Perform simulation analysis based on actual exploration data and apply ALE adaptive meshing technology so that the deformation of these soil meshes can more coherently describe the cutting process.
[0087] (3) Step S3 is to improve the particle swarm algorithm to optimize the support vector machine excavation speed and surface settlement prediction model, which specifically includes the following sub-steps:
[0088] S31. Collect the tunneling parameters of the shield machine. The cutter head speed, screw conveyor speed and main thrust are the shield construction control parameters. The cutter head torque and the shield cutter cutting force calculated above are defined as the shield construction state parameters. The tunneling speed and surface settlement are defined as the shield construction performance indicators:
[0089] S32, performing noise reduction processing on the tunneling state parameters based on variational mode decomposition and detrending fluctuation analysis algorithm;
[0090] S33. The Pearson correlation coefficient formula is used to analyze the correlation of shield machine excavation parameters. After analysis, it is found that the correlation between excavation speed and screw conveyor is above 0.9, and the linear correlation is high. SVM is used for prediction;
[0091] S34. The remaining excavation performance indicators are predicted using an improved particle swarm algorithm.
[0092] The present invention sets an adaptive inertia weight factor w, which can change with the increasing number of iterations. The specific formula is:
[0093] In the formula, w min Represents the initial inertia weight value, w max Indicates the inertia weight value at the maximum number of iterations, T represents the maximum number of iterations, and t represents the current number of iterations.
[0094] The steps to improve the IPSO algorithm are as follows:
[0095] First, initialize the population, including the size of the population, the position and speed of the initial particle, and calculate the fitness value of each particle to determine the individual optimal particle and the global optimal particle. Assume that in a D-dimensional search space, there are N particles forming a community, where the i-th particle is represented as a D-dimensional vector. The formula is expressed as:
[0096] X i =(x i1 ,x i2 ,…,x iD ),i=1,2,…,N
[0097] The velocity formula of the i-th particle is expressed as:
[0098] V i =(v i1 ,v i2 ,…,v iD ,),i=1,2,…,N
[0099] At the same time, record the optimal solution p for each particle best and the optimal solution g for the entire population best .
[0100] Then update the speed and position of each particle and calculate the fitness value. If the fitness value of each particle is better than the historical fitness value, update the position and fitness value of the particle; if the optimal fitness value in the entire population is better than the current global optimal fitness value, update the position and fitness value of the global optimal particle. The speed and position update formula of the i-th particle is:
[0101] v (i+1)d =wv id +c1r1(p id -x id )+c2r2(p gd -x id )
[0102] x (i+1)d =x id +v (i+1)d In the formula, p id is the individual known optimal solution, p gd is the known optimal solution of the population, w is the inertia weight, c1, c2 are learning factors, and r1, r2 are random numbers in the range of [0, 1].
[0103] Finally, determine whether the algorithm convergence conditions are met. If so, stop the iteration and output the results; otherwise, jump to step 2 to continue execution.
[0104] The IPSO-SVR algorithm can be used to predict the tunneling speed and surface settlement during the shield machine construction process. The algorithm implementation process is as follows: Fig. 9 Table 1 shows the performance index of the excavation speed prediction of the IPSO-SVR algorithm of the present invention. Table 2 shows the performance index diagram of the surface settlement prediction of the IPSO-SVR algorithm of the present invention.
[0105] Table 1
[0106]
[0107] Table 2
[0108]
[0109] As shown in Tables 1 and 2, the present invention can greatly improve the subsequent excavation speed and the accuracy of surface settlement prediction, save time and save computing resources.
[0110] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for predicting shield machine excavation speed and surface settlement based on mechanism and data drive, characterized in that: The following steps are involved: S1. After geological exploration, starting from rock and soil mechanics and considering the mechanical properties of the soil, a force model of the cutter cutting the soil is established, and a calculation formula for the horizontal cutting force of the cutter is further derived; S2. Use ABAQUS software to simulate the soil cutting process of the cutter using the nonlinear finite element method, and compare and verify the accuracy of the cutting force calculation formula with the measured data of the shield machine; S3. Collect data such as the cutter head speed, main thrust, and screw conveyor speed at the actual shield machine construction site, perform noise reduction processing on them, and further use the cutting force as an input parameter. It is introduced into the dual-drive prediction model together with the construction data of the shield machine as the input parameter of the dual-drive prediction model, and an improved particle swarm algorithm is used to optimize the support vector machine to establish a prediction model of tunneling speed and surface settlement, so as to realize the prediction of tunneling speed and surface settlement.
2. The shield machine excavation speed and surface settlement prediction method based on mechanism and data drive according to claim 1 is characterized in that: The content of step S1 is the established force model of the cutter cutting the soil, and further derives the calculation formula of the horizontal cutting force of the cutter, which specifically includes the following sub-steps: S11, analyzing the motion characteristics of the shield machine cutter and obtaining the motion parameter equation of any cutter; S12. Based on the motion parameter equation of the cutter, a cutter cutting soil mechanism model is constructed; S13. Perform a decoupling analysis of the cutter cutting force and the cutter disc torque, and construct a cutter disc torque decoupling model.
3. The shield machine excavation speed and surface settlement prediction method based on mechanism and data drive according to claim 1 or 2, characterized in that: Step S2 is about modeling and simulation of the soil cutting mechanism of the shield machine cutter, which specifically includes the following sub-steps: S21. According to the geological exploration situation, the soil constitutive model is constructed, the linear Drucker-Prager model in ABAQUS is used to simulate the constitutive relationship of the cutting tool affecting the soil, and the soil material parameters are set; S22, establish a three-dimensional model of the tool, set the three-dimensional size of the tool, define the tool as a rigid body in ABAQUS software, and set the conductivity; S23, establishing constraints between the tool and the soil, creating rigid constraints at the reference node on the tool, surface-to-surface contact is used to describe the contact between the tool and the soil, applying a base surface of a fixed constraint and a front and rear surface of a symmetry constraint to the soil model, and setting the top, right, and left surfaces as free surfaces; S24, perform simulation contact settings, use the Abaqus / Explicit dynamic display algorithm to ignore the convergence problem in the calculation, reduce the calculation complexity of the contact, and use the penalty function method in the contact algorithm between the tool and the soil; S25. Perform simulation analysis based on actual exploration data and apply ALE adaptive meshing technology so that the deformation of these soil meshes can more coherently describe the cutting process.
4. The shield machine excavation speed and surface settlement prediction method based on mechanism and data drive according to claim 3 is characterized in that: Step S3 is to establish a prediction model of tunneling speed and surface settlement based on the support vector machine optimized by the improved particle swarm algorithm, which specifically includes the following sub-steps: S31. Collect the tunneling parameters of the shield machine. The cutter head speed, screw conveyor speed and main thrust are the shield construction control parameters. The cutter head torque and the shield cutter cutting force calculated above are defined as the shield construction state parameters. The tunneling speed and surface settlement are defined as the shield construction performance indicators: S32, performing noise reduction processing on the tunneling state parameters based on variational mode decomposition and detrending fluctuation analysis algorithm; S33, the correlation of shield machine excavation parameters is analyzed. After analysis, it is found that the correlation between excavation speed and screw conveyor is above 0.9, and the linear correlation is high. SVM is used for prediction; S34. Surface subsidence is predicted using an improved particle swarm algorithm.
