A slurry shield lagging early warning method based on CFD-DEM coupling
By using the CFD-DEM coupling method, a fluid-particle coupled model was constructed, which solved the problem of early warning of slurry shield tunneling drainage risk, achieved efficient risk management and construction parameter optimization, and reduced project risks and costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA RAILWAY SHISIJU GROUP CORP
- Filing Date
- 2026-05-06
- Publication Date
- 2026-07-31
AI Technical Summary
Existing technologies cannot effectively provide early warning of the risk of sludge shield tunneling delays. They rely on manual experience and lack an understanding of the complex solid-liquid two-phase flow mechanism inside the pipeline, resulting in lag, strong reliance on experience, and inability to locate and trace the source.
The CFD-DEM coupling method is adopted to construct a fluid-particle coupled calculation model, calibrate the DEM particle and CFD fluid parameters, establish an early warning index system, realize real-time monitoring and hierarchical early warning, and combine engineering parameters for simulation and linkage response.
It has achieved early warning of the risk of slurry shield tunneling delay, improved the accuracy and reliability of the warning, reduced the false alarm rate, formed an intelligent proactive control system, optimized construction parameters and reduced downtime losses.
Smart Images
Figure CN122490964A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of shield tunnel construction technology, and specifically relates to an early warning method for slurry shield tunneling based on CFD-DEM coupling. Specifically, this invention integrates computational fluid dynamics and the discrete element method, and by constructing a fluid-particle coupled computational model, achieves refined simulation of the slurry-carrying flow state, calibration of key parameters, and advanced early warning of slurry discharge risks. Background Technology
[0002] When a slurry-balanced shield tunneling machine traverses complex geological formations, the excavated soil (especially cohesive soil and weathered rock) is mixed with slurry in the excavation chamber and then pumped to the surface through the slurry discharge pipeline. During this process, uneven flow velocity distribution, sudden increases in local resistance, and even complete blockage ("sludge stagnation") often occur in the slurry discharge pipeline due to poor particle size distribution, excessively high concentration of excavated soil, aggregation of large particles, or scaling from adsorbed viscous substances. Sludge stagnation directly leads to shield machine shutdown, is difficult to handle, severely delays the construction period, and increases construction costs.
[0003] Currently, on-site assessment of drainage risk relies primarily on manual experience and limited monitoring data, such as threshold alarms for slurry discharge flow rate and pump pressure. These methods have significant shortcomings: (1) Lag: Macroscopic parameters will only change significantly when the blockage has occurred or developed to a certain extent, making it impossible to achieve early warning.
[0004] (2) High dependence on experience: lack of understanding of the complex solid-liquid two-phase flow mechanism inside the pipeline, rough setting of warning threshold, and poor universality.
[0005] (3) Unable to locate and trace the source: It is difficult to determine the specific pipe section where the blockage occurred and the main cause, which makes preventive treatment difficult.
[0006] In recent years, numerical simulation technology has provided a new approach to studying such problems. Computational fluid dynamics can effectively simulate slurry flow fields, and the discrete element method can finely characterize particle motion and contact. However, existing studies often apply these two methods in isolation or use simplified Eulerian two-phase flow models, which cannot accurately describe the influence of particle shape, friction, adhesion, and other characteristics on aggregation and blockage in actual engineering projects. Furthermore, a complete early warning technology system, from parameter calibration to risk quantification, has not yet been established. Therefore, there is an urgent need for a method that can deeply integrate the microscopic mechanical properties of slag particles with the macroscopic flow behavior of slurry, and can achieve quantitative and proactive early warning of slurry shield tunneling drainage risks. Summary of the Invention
[0007] To address the technical problems existing in the background art, the present invention provides an early warning method for slurry shield tunneling based on CFD-DEM coupling.
[0008] This invention employs the following technical solution: a method for early warning of slurry shield tunneling delay based on CFD-DEM coupling, comprising the following steps: Based on the geological survey report and the DEM particle contact parameters of the shield tunneling excavation soil samples, CFD fluid parameters of the engineering mud were calibrated. A CFD-DEM coupled model was constructed based on the shield machine body and slurry discharge pipeline, and the calculation area of the CFD-DEM coupled calculation model was divided to obtain the CFD-DEM bidirectional transient coupling domain, the pure DEM calculation domain, and the exchange domain. Monitoring units are set up at designated locations within the bidirectional transient coupling domain of CFD-DEM, and an early warning index system is quantified and constructed. Operating parameters are injected into the CFD-DEM coupling model, boundary conditions are set, and CFD-DEM coupling dynamic simulation is carried out to collect the evolution data of early warning indicators in each monitoring area in real time. A tiered early warning mechanism is established based on the aforementioned evolution data, and coordinated response measures are implemented according to the early warning results.
[0009] In a further embodiment, the calibration process for the DEM particle contact parameters is as follows: Based on the geological survey report and the soil samples from the tunnel boring machine, the particle size distribution, density and crushing characteristics of the soil layer were determined. The indoor geotechnical test was reproduced in the discrete element method software. The micro-contact parameters were iteratively optimized through inversion analysis to match the simulated macroscopic mechanical properties with the test results. The micro-contact parameters include at least the static friction coefficient between particles, the rolling friction coefficient, the collision recovery coefficient and the surface energy. The same method was used to calibrate the contact parameters between the particles and the inner wall of the shield tunnel slurry discharge pipeline.
[0010] In a further embodiment, the calibration process for the CFD fluid parameters is as follows: Rheological tests are performed on engineering mud to determine the fluid constitutive model and to measure rheological parameters and physicochemical parameters of the mud; wherein the rheological parameters include at least: yield stress, plastic viscosity and consistency coefficient.
[0011] In a further embodiment, the construction process of the CFD-DEM coupled computation model is as follows: Based on the actual structural dimensions and geometric shape of the tunnel boring machine body and slurry discharge pipeline, a corresponding three-dimensional geometric model is established; The three-dimensional geometric model is meshed and preprocessed. The internal space of the slurry discharge pipeline is divided into a CFD-DEM bidirectional transient coupling domain, the cutterhead area is divided into a pure DEM calculation domain, and the connection between the CFD-DEM bidirectional transient coupling domain and the pure DEM calculation domain is set as an exchange domain.
[0012] In a further embodiment, the monitoring unit set is defined as follows: ， in, The total number of monitoring units; correspondingly, the early warning indicator system is composed of the set of monitoring units. The physical quantities corresponding to each monitoring unit in the system include at least the following: Solid volume concentration / Accumulated mass Local average flow velocity Dynamic pressure gradient and particle velocity distribution uniformity .
[0013] In a further embodiment, the operating parameters are the real-time construction parameters or predicted construction parameters of the current tunneling ring; the boundary conditions include: fluid boundary conditions, particle boundary conditions, and wall boundary conditions. The fluid boundary conditions are the mud inlet velocity and the system back pressure parameters. The particle boundary conditions are the slag injection rate and particle size distribution parameters. The wall boundary conditions are particle-wall contact parameters.
[0014] In a further embodiment, the tiered early warning mechanism includes: Level 1 early warning, Level 2 early warning, and Level 3 early warning; The monitoring unit triggers a corresponding level of early warning when any of the following conditions are met: The triggering conditions for the Level 1 warning are as follows: ,or ,or ; The triggering conditions for the Level 2 warning are as follows: ,and ,and And dynamic pressure gradient It shows a continuous upward trend; The triggering conditions for the Level 3 warning are as follows: ,and ,and It exhibits obvious particle accumulation characteristics and dynamic pressure gradient Approaching the system's safe pressure limit; in, , and These represent the percentage changes in solid volume concentration, local average flow velocity, and particle velocity distribution uniformity relative to their baseline values. , , These are percentage thresholds that increase progressively.
[0015] In a further embodiment, the measures include at least: attention-based measures, control measures, and emergency measures; The measures to address the concerns are: activating enhanced monitoring of the corresponding monitoring unit, adjusting mud performance parameters, or reducing the tunneling speed; The control and treatment measures are: adjusting the mud ratio, increasing the slurry discharge flow rate, or backwashing pipeline; The emergency response measures are as follows: immediately stop tunneling, start the emergency slurry drainage system or carry out pipeline unblocking operations.
[0016] In a further embodiment, the method further includes: setting the mud as a continuous phase in the CFD software, introducing the slag particle phase through DEM coupling, and configuring the turbulence model and multiphase flow coupling algorithm according to the mud characteristics. Within the CFD-DEM bidirectional transient coupling domain, the calculation of the bidirectional transient coupling effect between the continuous mud phase and the granular soil phase is realized, including the dragging and entrainment effect of fluid on particles, as well as the reaction and disturbance effect of particles on fluid. Within the pure DEM computational domain, simulations of the movement, crushing, and accumulation of slag particles were conducted. Within the exchange domain, the transfer and interaction of soil particle information between the CFD-DEM bidirectional transient coupling domain and the pure DEM computation domain are realized.
[0017] The beneficial effects of this invention are as follows: The early warning mechanism has evolved from empirical threshold alarms to mechanistic model predictions, achieving true advanced early warning. This invention, through CFD-DEM coupled simulation, reproduces and re-enacts the entire process of sludge formation (such as particle aggregation and sediment growth) in the digital world. It can identify risk trends several minutes or even earlier before physical blockage actually occurs through microscopic indicators (such as a sudden increase in local concentration and deterioration of flow velocity distribution), significantly advancing the early warning window. The model parameters have been upgraded from coarse estimation to multi-source data calibration, significantly improving the fidelity of the simulation and the reliability of the prediction. This invention abandons the practice of using general or assumed parameters and establishes a rigorous parameter calibration process. DEM parameters are calibrated based on mechanical test inversion of real strata and soil samples, taking into account the cutterhead crushing effect; CFD parameters are determined based on mud rheological tests. This makes the behavior of virtual particles and virtual mud in the model highly realistic, ensuring the simulation results have guiding significance for actual construction and significantly reducing false alarms and false negatives. The computational strategy of this invention optimizes global generalized simulation to partitioned coupled computation, balancing computational accuracy and efficiency. It performs only particle mechanics calculations in regions without continuous fluid, and performs fine coupling in critical two-phase flow regions. This avoids unnecessary fluid calculations in non-critical areas, significantly reducing computational resource consumption while ensuring the simulation accuracy of core processes. This makes the method more suitable for rapid analysis and decision-making in engineering settings. Risk management has been upgraded from a single alarm to a closed-loop intelligent response system encompassing tiered, location-based, and coordinated actions. This invention not only provides early warning but also constructs a complete proactive management system. First, risk is tiered (attention, early warning, alarm) based on multiple indicators (flow rate, buildup mass, pressure gradient) to differentiate the severity of risks. Second, by setting monitoring surfaces / volumes, the location of potentially blocked pipe sections can be precisely located. Finally, the system can coordinate with the control system to intelligently trigger or recommend the most targeted response measures (such as targeted flushing, pump speed adjustment, and mud properties modification) based on the warning level and location, forming an intelligent closed loop of simulation prediction, proactive early warning, and precise response, greatly improving response efficiency. Finally, the engineering benefits extend from reducing downtime losses to full-cycle optimization decision-making, creating significant comprehensive value. This method provides a digital test field for optimizing construction parameters, allowing for safe and low-cost testing of slag removal effects under different geological formations, mud formulations, and tunneling parameters in a virtual space, thereby optimizing construction plans and improving tunneling efficiency. Attached Figure Description
[0018] Figure 1 This is a flowchart of the method in Example 1.
[0019] Figure 2 This is a schematic diagram of the CFD-DEM coupling principle in Example 1.
[0020] Figure 3 This is a schematic diagram of the parameter calibration method for low-cohesion soil in Example 1.
[0021] Figure 4 This is a schematic diagram of the parameter calibration method for high-viscosity soil in Example 1.
[0022] Figure 5 This is a schematic diagram of the parameter calibration method between different particles in Example 1.
[0023] Figure 6 This is a schematic diagram of the modeling and coupling region of Example 1.
[0024] Figure 7 This is a schematic diagram of the CFD software output effect of Example 1.
[0025] Figure 8 This is a schematic diagram of the DEM software output effect of Example 1. Detailed Implementation
[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0027] Example 1 This embodiment uses a subway tunnel boring machine project as an example to explain in detail the implementation process of the coupling method of the present invention. Figure 1 As shown in the figure, this embodiment discloses an early warning method for slurry shield tunneling based on CFD-DEM coupling, including the following steps: Based on the geological survey report and the DEM particle contact parameters of the shield tunneling excavation soil samples, CFD fluid parameters of the engineering mud were calibrated. A CFD-DEM coupled model was constructed based on the shield machine body and slurry discharge pipeline, and the calculation area of the CFD-DEM coupled calculation model was divided to obtain the CFD-DEM bidirectional transient coupling domain, the pure DEM calculation domain, and the exchange domain. Monitoring units are set up at designated locations within the bidirectional transient coupling domain of CFD-DEM, and an early warning index system is quantified and constructed. Operating parameters are injected into the CFD-DEM coupling model, boundary conditions are set, and CFD-DEM coupling dynamic simulation is carried out to collect the evolution data of early warning indicators in each monitoring area in real time. A tiered early warning mechanism is established based on the aforementioned evolution data, and coordinated response measures are implemented according to the early warning results.
[0028] In a further embodiment, the calibration process for the DEM particle contact parameters is as follows: Based on the geological survey report and the excavated soil samples from the tunnel boring machine, the particle size distribution, density, and crushing characteristics of the soil layer are determined. For rock formations, the particle crushing effect under cutterhead cutting needs to be considered. The crushing resistance strength of the particles and the particle size distribution after crushing are determined through laboratory crushing tests or empirical formulas.
[0029] Indoor geotechnical tests (such as angle of repose tests, direct shear tests, and adhesion tests) are reproduced in discrete element method (DEM) software. Microscopic contact parameters are iteratively optimized through inversion analysis to match the simulated macroscopic mechanical properties with the experimental results. These microscopic contact parameters include at least: the static friction coefficient between particles, the rolling friction coefficient, the collision recovery coefficient, and surface energy (e.g., using the JKR adhesion model). Figure 3 , Figure 4 and Figure 5 As shown, the contact parameters between the particles and the inner wall of the shield tunnel slurry discharge pipeline were calibrated using the same method.
[0030] Specifically, taking a shield tunneling project in a composite stratum as an example, the calibration process of its DEM particle contact parameters is as follows: (1) Sample acquisition and basic testing: Samples were taken from the current ring soil and sieved to obtain the gradation curve. The natural density was measured.
[0031] (2) Microscopic contact parameter calibration test: Angle of repose test: In the laboratory, the soil excavation is piled up using a cylindrical lifting method, and the angle of repose is measured.
[0032] Direct shear test: The internal friction angle and cohesion of cohesive soil were obtained through indoor direct shear test.
[0033] Inclined surface test: The rolling distance of particles on inclined surfaces with different contact materials (dissimilar soil and tunnel boring machine) is obtained through indoor inclined surface tests.
[0034] EDEM Virtual Calibration: Virtual stacking test benches, virtual direct shear boxes, and virtual inclined plane test benches of the same size are established in EDEM to generate single particles or particle groups with the same gradation as the actual particle size distribution. First, basic physical parameters such as particle density and Poisson's ratio are input. Then, by designing response surface methodology, multiple sets of virtual experiments are conducted to obtain regression curves, thereby obtaining optimized solutions for multiple sets of key contact parameters such as the static friction coefficient, rolling friction coefficient, and JKR surface energy between particles. (3) Iteration and verification: Run a virtual test and measure the angle of accumulation of the virtual particle accumulation. Adjust the above contact parameters through multiple iterations until the error between the virtual test value and the actual response value is within 10%. The parameter set obtained at this time is considered as the calibrated parameters, which can represent the macroscopic accumulation and friction characteristics of the slag in this stratum.
[0035] (4) Wall parameters: The friction coefficient between particles and shield machine materials is calibrated by referring to the friction test data of metal and soil through a similar principle.
[0036] In a further embodiment, the calibration process for the CFD fluid parameters is as follows: Rheological tests are performed on engineering mud to determine the fluid constitutive model and to measure rheological parameters and physicochemical parameters of the mud; wherein the rheological parameters include at least: yield stress, plastic viscosity and consistency coefficient.
[0037] Based on the above examples, the calibration process of CFD fluid parameters is further explained: (1) Sampling and rheological testing of mud: Sampling was taken from the mud circulation system during synchronous construction and rheological testing was performed using a rotational viscometer.
[0038] (2) Constitutive model selection and parameter fitting: observe the relationship between mud shear stress and shear rate from the test data and determine which fluid properties are similar to it, and whether yield stress exists. For example, Herschel-Bulkely fluid and Bingham fluid are selected and fitted to obtain model parameters: yield stress, plastic viscosity and consistency coefficient.
[0039] (3) Model input: Input the above three parameters and the measured mud density as material properties into the mud phase settings in CFD.
[0040] To address the problems of redundant global simulation calculations, low efficiency, computational distortion in non-fluid regions, discontinuous particle transport, and simplified simulation of two-phase flow interactions in existing technologies, the construction process of the CFD-DEM coupled computational model described in this embodiment is as follows: Based on the actual structural dimensions and geometric shape of the tunnel boring machine body and slurry discharge pipeline, a corresponding three-dimensional geometric model is established; it should be noted that the model described in this embodiment is mainly the slurry discharge pipeline system, including the slurry inlet (excavation chamber outlet), slurry pump, straight pipe sections, elbows, reducers, valves, and slurry discharge outlets, etc.
[0041] The three-dimensional geometric model is meshed and preprocessed. The internal space of the slurry discharge pipeline is divided into a CFD-DEM bidirectional transient coupling domain, and the cutterhead area is divided into a pure DEM computational domain. An exchange domain is set at the connection between the CFD-DEM bidirectional transient coupling domain and the pure DEM computational domain. Figure 2 and Figure 6 As shown.
[0042] Furthermore, in the CFD software, the mud is set as a continuous phase, and the slag particle phase is introduced through DEM coupling. A turbulence model and multiphase flow coupling algorithm (such as...) are configured according to the mud characteristics. Model).
[0043] Within the CFD-DEM bidirectional transient coupling domain (inside the chamber and slurry discharge pipeline system), this area is a closed pipeline where the slurry is a continuously transported medium. Therefore, a fine CFD mesh is established and bidirectionally coupled with DEM software to realize the calculation of the bidirectional transient coupling effect between the continuous phase of the slurry and the particle phase of the slag, including the dragging and entrainment effects of the fluid on the particles (such as drag force, pressure gradient force, lift, etc.), as well as the reaction and disturbance effects of the particles on the fluid.
[0044] Within the pure DEM computational domain, the movement, crushing, and accumulation of slag particles are simulated. It should be noted that in this region, the slurry exists as a discontinuous, permeable soil-slurry mixture, and flow is not dominant. This is unlike areas such as the cutterhead panel where there is no continuous slurry flow but significant particle collision and accumulation. Therefore, CFD calculations are not performed in this region; only the DEM is used for simulation. Particle movement within this domain is governed by the movement of mechanical components (such as cutterhead rotation) and gravity; the cutterhead is set according to its actual rotational speed.
[0045] Within the exchange domain, the transfer and interaction of soil particle information between the CFD-DEM bidirectional transient coupling domain and the pure DEM computation domain are realized.
[0046] After the model is built, a bidirectional coupled CFD-DEM solution is performed: Initialization: In CFD, the fluid domain mesh refinement and boundary condition setting of the CFD-DEM bidirectional transient coupling domain are completed. The inlet is a velocity inlet and the outlet is a pressure outlet. In DEM, a particle factory is defined in the pure DEM computational domain of the cutterhead area to simulate the generation of excavated soil at the excavation face. The particle generation rate and initial physical properties are set. Particles can smoothly enter the CFD-DEM bidirectional transient coupling domain through the exchange domain.
[0047] Two-way data exchange loop: 1) CFD→DEM force transfer: at each coupling time step (e.g. Within seconds, the CFD calculates the drag force and other fluid forces acting on each DEM particle based on the current flow field velocity and pressure distribution, and transmits the force vector array to the DEM through the coupling interface; 2) DEM→CFD Momentum Feedback: After receiving the fluid force, the DEM solves the Newton-Euler equations by combining the particle gravity and contact force, and updates the particle position and velocity; it also counts the momentum change of the particles with respect to the fluid in each CFD grid cell and feeds it back to the corresponding CFD grid cell in the form of momentum source terms. 3) CFD flow field update: CFD substitutes the momentum source term into the Navier-Stokes equations to solve and update the velocity and pressure distribution of the entire flow field.
[0048] Time stepping: After completing one data exchange, the CFD and DEM synchronously advance to the next time step, repeating the bidirectional coupling loop until the simulation ends, realizing the complete bidirectional transient coupling physical process of fluid-driven particle motion and particle reaction disturbance of the flow field, such as... Figure 7 and Figure 8 As shown.
[0049] Based on the above description, the working parameters described in this embodiment are the real-time construction parameters or predicted construction parameters of the current tunneling ring; the boundary conditions include: fluid boundary conditions, particle boundary conditions and wall boundary conditions; The fluid boundary conditions are the mud inlet velocity and the system back pressure parameters; the mud inlet velocity should satisfy: , Slurry discharge flow rate (unit: ), Cross-sectional area of the slurry discharge pipeline (unit: ), Mud inlet velocity (unit: ).
[0050] The system back pressure parameters satisfy the following formula: ,in, Standard atmospheric pressure (unit: ), To control the increase in back pressure during construction (unit: ), Back pressure of the slurry system (unit: ).
[0051] The particle boundary conditions are the soil injection rate and particle size distribution parameters.
[0052] Furthermore, the slag injection rate should meet the following requirements: ,in, Injection rate of slag and soil mass (unit: ), Density of slag particles (unit: ), The tunnel boring machine's speed (unit: ), The cross-sectional area of the tunnel boring machine (unit: ), This represents the cutterhead opening ratio.
[0053] The particle size distribution parameters described in this embodiment adopt a graded particle size sequence. It means that, among them, For the first Representative particle size (median particle size within the corresponding particle size range, unit: mm ), This represents the total number of particle size fractions. ; For the first The mass percentage of the particles, and satisfying the normalization condition: .
[0054] The wall boundary conditions are particle-wall contact parameters, including: static friction coefficient. coefficient of kinetic friction With the collision recovery coefficient It is used to characterize the sliding, rolling and collision behavior of particles in pipelines.
[0055] In a further embodiment, multiple monitoring surfaces or statistical areas are set within the computational domain, such as on the cutterhead surface, the bottom of the mud-water chamber air cushion chamber, the downstream side of the sharp bend section, the starting area of the pipe diameter enlargement section, behind the valve, at the pipe interface, and in the middle and rear of the long horizontal pipe section.
[0056] Therefore, the set of monitoring units is defined as follows: ,in, The total number of monitoring units; correspondingly, the early warning indicator system is composed of the set of monitoring units. The physical quantities corresponding to each monitoring unit in the system include at least the following: Solid volume concentration / Accumulated mass Local average flow velocity Dynamic pressure gradient and particle velocity distribution uniformity Furthermore, when tunneling in a single stratum, the packing mass should be selected. For complex formations, i.e., when there is a large density difference, solid volume concentration is used. .
[0057] Specifically, regarding the first monitoring units Solid volume concentration used to reflect the degree of particle aggregation and stacking mass The calculation formulas are as follows: ,in, For monitoring unit The volume of solid particles passing through the monitoring surface. The total volume of the fluid and particles; ,in, For monitoring unit Inner particles quality For monitoring unit The total number of particles inside.
[0058] Local average flow velocity It is used to characterize the state of obstructed flow, and its calculation method is as follows: ;in, For monitoring unit The monitoring area The location is corresponding to the monitoring surface. ,time The axial velocity of the fluid.
[0059] In a further embodiment, the dynamic pressure gradient A key indicator of blockage formation is the pressure drop per unit length of pipe along the flow direction; the dynamic pressure gradient. The calculation formula is: ;in, and Monitoring units The monitoring pressure at both ends of the corresponding pipe section, and This refers to the axial positions at both ends of the pipe section.
[0060] Particle velocity distribution uniformity Defined as the standard deviation or coefficient of variation of particle transport velocity within the monitoring area, it reflects whether particle velocity stratification or stagnation occurs; its calculation formula is: ; In the formula, For monitoring unit Inner particles speed, This represents the average velocity of the particles within the corresponding unit.
[0061] By analyzing data from simulations of a large number of different operating conditions (normal, minor blockage, severe blockage), the threshold values for each early warning indicator under different risk levels were determined.
[0062] The corresponding tiered early warning mechanism includes: Level 1, Level 2, and Level 3 early warning. The monitoring unit triggers a corresponding level of early warning when any of the following conditions are met: The triggering conditions for the Level 1 warning are as follows: ,or ,or The general level is a low-risk warning level.
[0063] The triggering conditions for the Level 2 warning are as follows: ,and ,and And dynamic pressure gradient It shows a continuous upward trend; The triggering conditions for the Level 3 warning are as follows: ,and ,and It exhibits obvious particle accumulation characteristics and dynamic pressure gradient Approaching the system's safe pressure limit; in, , and These represent the percentage changes in solid volume concentration, local average flow velocity, and particle velocity distribution uniformity relative to their baseline values. , , These are percentage thresholds that increase progressively.
[0064] To illustrate further, , , ; , , , The calibration can be determined based on the geological formation and mud characteristics of the project, for example, by taking 10%, 25%, and 40% respectively.
[0065] Specifically, the corresponding measures for handling a Level 1 warning include: activating enhanced monitoring of the corresponding monitoring unit and increasing the frequency of data collection; and, based on the changing trends of monitoring indicators, fine-tuning mud performance parameters (such as increasing mud viscosity) or reducing tunneling speed to alleviate the trend of localized decrease in flow velocity and increase in particle concentration.
[0066] 2) Control and response measures for Level II early warning: The system can automatically or manually adjust the mud ratio to improve the mud's slag carrying capacity; increase the flow rate of the discharge pump to flush potentially blocked sections; if necessary, the bypass pipeline can be opened briefly to reduce the local pressure gradient and inhibit further particle accumulation.
[0067] 3) Emergency response measures for Level 3 warning: Immediately stop tunneling and shut down cutterhead propulsion; start the emergency slurry drainage system and use high-pressure water or air to backwash the pipeline; if backwashing is ineffective, stop the machine and carry out segmented pipeline clearing operations to prevent the blockage from expanding.
[0068] When the warning level reaches level two or three, and the tunnel boring machine has reached the corresponding position, the operators activate the pre-set congestion handling measures. These measures are matched to the congestion type and location predicted by the simulation, for example: For localized particle aggregation: The high-pressure flushing valve near the pipe section is automatically triggered to perform a short-term pulse flush.
[0069] For the overall decrease in flow rate: it is recommended to adjust the frequency of the discharge pump and optimize the rheological properties of the mud (such as by injecting dispersants).
[0070] For the risk of large particles getting stuck: it is recommended to switch to reverse washing mode or adjust the cutter head speed to change the particle input characteristics.
[0071] The system compares the dynamic simulation results of the current operating conditions with the graded early warning indicators. Once the indicators of a specific monitoring area reach a certain threshold, the system immediately issues an early warning message of the corresponding level, which includes: the risk level, the spatiotemporal location where the sludge may accumulate (e.g., 2 meters downstream of the bend, with accumulation expected in 30 seconds), and the main causes (e.g., large particle blockage, insufficient mud carrying capacity), providing a basis for disposal measures.
Claims
1. A slurry shield lagging pre-warning method based on CFD-DEM coupling, characterized in that, Includes the following steps: Based on the geological survey report and the DEM particle contact parameters of the shield tunneling excavation soil samples, CFD fluid parameters of the engineering mud were calibrated. A CFD-DEM coupled model was constructed based on the shield machine body and slurry discharge pipeline, and the calculation area of the CFD-DEM coupled calculation model was divided to obtain the CFD-DEM bidirectional transient coupling domain, the pure DEM calculation domain, and the exchange domain. Monitoring units were set up at specified locations within the bidirectional transient coupling domain of CFD-DEM, and a quantitative early warning indicator system was constructed. The operating parameters are injected into the CFD-DEM coupled model, boundary conditions are set, and CFD-DEM coupled dynamic simulation is carried out to collect the evolution data of early warning indicators in each monitoring area in real time. A tiered early warning mechanism is established based on the aforementioned evolution data, and coordinated response measures are implemented according to the early warning results.
2. The slurry shield lagging pre-warning method based on CFD-DEM coupling according to claim 1, characterized in that, The calibration process for the DEM particle contact parameters is as follows: Based on the geological survey report and the soil samples from the tunnel boring machine, the particle size distribution, density and crushing characteristics of the soil layer were determined. The indoor geotechnical test was reproduced in the discrete element method software. The micro-contact parameters were iteratively optimized through inversion analysis to match the simulated macroscopic mechanical properties with the test results. The micro-contact parameters include at least the static friction coefficient between particles, the rolling friction coefficient, the collision recovery coefficient and the surface energy. The same method was used to calibrate the contact parameters between the particles and the inner wall of the shield tunnel slurry discharge pipeline.
3. The slurry shield lagging pre-warning method based on CFD-DEM coupling according to claim 1, characterized in that, The calibration process for the CFD fluid parameters is as follows: Rheological tests are performed on engineering mud to determine the fluid constitutive model and to measure rheological parameters and physicochemical parameters of the mud; wherein the rheological parameters include at least: yield stress, plastic viscosity and consistency coefficient.
4. The slurry shield lagging pre-warning method based on CFD-DEM coupling according to claim 1, characterized in that, The construction process of the CFD-DEM coupled computation model is as follows: Based on the actual structural dimensions and geometric shape of the tunnel boring machine body and slurry discharge pipeline, a corresponding three-dimensional geometric model is established; The three-dimensional geometric model is meshed and preprocessed. The internal space of the slurry discharge pipeline is divided into a CFD-DEM bidirectional transient coupling domain, the cutterhead area is divided into a pure DEM calculation domain, and the connection between the CFD-DEM bidirectional transient coupling domain and the pure DEM calculation domain is set as an exchange domain.
5. The method for early warning of slurry shield tunneling delay based on CFD-DEM coupling according to claim 1, characterized in that, The set of monitoring units is defined as ， wherein, is the total number of monitoring units; correspondingly, the early warning index system is composed of the set of monitoring units corresponding physical quantities of each monitoring unit, the physical quantities at least including: Solid volume concentration / bulk mass , local average flow velocity , dynamic pressure gradient and particle velocity distribution uniformity .
6. The slurry shield lagging pre-warning method based on CFD-DEM coupling according to claim 1, characterized in that, The operating parameters are the real-time or predicted construction parameters of the current tunneling ring; the boundary conditions include: fluid boundary conditions, particle boundary conditions, and wall boundary conditions. The fluid boundary conditions are the mud inlet velocity and the system back pressure parameters. The particle boundary conditions are the slag injection rate and particle size distribution parameters. The wall boundary conditions are particle-wall contact parameters.
7. The slurry shield lagging pre-warning method based on CFD-DEM coupling according to claim 1, characterized in that, The tiered early warning mechanism includes: Level 1 early warning, Level 2 early warning, and Level 3 early warning; The monitoring unit triggers a corresponding level of early warning when any of the following conditions are met: The trigger condition of the first early warning is: or or ; The triggering conditions for the Level 2 warning are as follows: ,and ,and And dynamic pressure gradient It shows a continuous upward trend; The triggering conditions for the Level 3 warning are as follows: ,and ,and It exhibits obvious particle accumulation characteristics and dynamic pressure gradient Approaching the system's safe pressure limit; in, , and These represent the percentage changes in solid volume concentration, local average flow velocity, and particle velocity distribution uniformity relative to their baseline values. , , These are percentage thresholds that increase progressively.
8. The method for early warning of slurry shield tunneling delay based on CFD-DEM coupling according to claim 1, characterized in that, The measures to be taken include at least: monitoring measures, control measures, and emergency measures; The measures to address the concerns are: activating enhanced monitoring of the corresponding monitoring unit, adjusting mud performance parameters, or reducing the tunneling speed; The control and treatment measures are: adjusting the mud ratio, increasing the slurry discharge flow rate, or backwashing pipeline; The emergency response measures are as follows: immediately stop tunneling, start the emergency slurry drainage system or carry out pipeline unblocking operations.
9. The method for early warning of slurry shield tunneling delay based on CFD-DEM coupling according to claim 4, characterized in that, Also includes: In the CFD software, the mud is set as a continuous phase, and the slag particle phase is introduced through DEM coupling. The turbulence model and multiphase flow coupling algorithm are configured according to the mud characteristics. Within the CFD-DEM bidirectional transient coupling domain, the calculation of the bidirectional transient coupling effect between the continuous mud phase and the granular soil phase is realized, including the dragging and entrainment effect of fluid on particles, as well as the reaction and disturbance effect of particles on fluid. Within the pure DEM computational domain, simulations of the movement, crushing, and accumulation of slag particles were conducted. Within the exchange domain, the transfer and interaction of soil particle information between the CFD-DEM bidirectional transient coupling domain and the pure DEM computation domain are realized.