Pipe launching yellow mud grouting multi-parameter linkage closed-loop management and control method and system

Abstract

The application discloses a pipe jacking originating yellow mud grouting multi-parameter linkage closed-loop management and control method and system, relates to the field of tunnel construction technology, and comprises the following steps: dividing a reinforcing area into multiple management and control areas according to strength distribution, and setting corresponding hierarchical management and control benchmarks; synchronously collecting multi-source operation data of cross-project equipment and completing standardized cleaning to form a unified time sequence standardized data set; sequentially executing grouting parameter self-adaptive adjustment suitable for the reinforcing area, mud-water and grouting coupling collaborative regulation and deviation correction action and supplementary grouting collaborative regulation according to preset priorities; performing real-time evaluation on the integrity of the plugging body through a pre-constructed multi-field coupling damage evolution model of the plugging body; and integrating operation data and effect monitoring data of the whole period of the originating section. The application integrates three types of actions, i.e., mud seepage, deviation extrusion and slurry strength growth, into a unified evaluation framework by establishing a multi-field coupling damage evolution model of the plugging body, and realizes real-time calculation of the residual effective thickness and breakthrough probability of the plugging body cycle by cycle.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction technology, and in particular to a multi-parameter linkage closed-loop control method and system for starting grouting in yellow mud during pipe jacking. Background Technology

[0002] The slurry-balanced pipe jacking method, with its advantages of minimal construction disturbance and wide adaptability to geological formations, has become the mainstream construction method for underground pipelines, drainage tunnels, and other underpass projects in urban areas. The initial stage is the core risk node in the entire pipe jacking construction process, and the effectiveness of the portal sealing directly determines the safety and stability of the initial operation. Especially in sensitive conditions where the project is adjacent to existing buildings and structures and has stringent requirements for controlling ground deformation, the control precision, response speed, and risk prediction capability of slurry grouting sealing directly affect the safety of the surrounding environment and the smooth progress of the project. With the increasing intensity of underground space development in urban core areas, pipe jacking projects generally face objective conditions such as uneven end reinforcement effects, strong environmental constraints, and dynamically changing construction conditions, placing higher demands on the refined and intelligent management of portal grouting sealing.

[0003] The existing grouting control technology for pipe jacking has undergone several stages of development: In the early stages, it relied entirely on the construction experience of on-site technicians to manually adjust parameters such as grouting pressure and grouting volume. The control accuracy was greatly affected by the experience level of the personnel, and the stability was insufficient. Subsequently, single-system automated grouting control devices were gradually developed, which can achieve constant pressure or constant volume grouting based on preset parameters, reducing the reliance on manual labor to a certain extent. In recent years, risk warning schemes combined with environmental monitoring data have been gradually applied. By collecting monitoring data such as mud and water pressure and surface settlement, alarms and emergency responses are triggered when the values ​​exceed preset thresholds. Related zoning control and multi-parameter linkage ideas have also been gradually introduced into the field of grouting construction control, and the overall control level is showing a gradual upward trend.

[0004] However, existing technologies still have many obvious shortcomings. For example, the control benchmarks mostly adopt a unified standard, which does not fully consider the spatial distribution differences of the end reinforcement strength, and the zoning adaptability is insufficient, making it difficult to match the heterogeneous characteristics of the reinforcement area. The grouting system, mud-water system, and pipe jacking posture system operate relatively independently, lacking a multi-system collaborative linkage mechanism, making it difficult to effectively deal with the sealing risks under the coupling effect of multiple factors. In addition, risk control is mostly a threshold-triggered post-action, which only tracks the quantitative change of physical quantities and cannot identify the qualitative changes that lead to instability and failure. It also lacks a time-based pre-intervention logic, resulting in a lag in risk handling.

[0005] To address the aforementioned technical issues, there is an urgent need for a multi-parameter linkage closed-loop control method and system for grouting of yellow mud during pipe jacking. Summary of the Invention

[0006] Therefore, it is necessary to provide a multi-parameter linkage closed-loop control method and system for grouting of yellow mud at the beginning of pipe jacking to address the above-mentioned technical problems.

[0007] In a first aspect, the present invention provides a multi-parameter linkage closed-loop control method for grouting of yellow mud at the beginning of pipe jacking, comprising:

[0008] S1. Based on existing engineering parameters, the reinforced area is divided into multi-level control areas according to the intensity distribution, and corresponding hierarchical control benchmarks are set to establish quantitative mapping rules for multi-set parameter linkage.

[0009] S2. Based on a unified data interaction interface and time series benchmark, multi-source operation data across engineering equipment is collected synchronously and standardized cleaning is performed to form a standardized dataset with unified time series.

[0010] S3. Based on the hierarchical control benchmark and quantitative mapping rules, combined with the operational data in the standardized dataset, the adaptive adjustment of grouting parameters, the coordinated control of mud and water coupling and grouting, and the coordinated control of correction action and grouting are executed sequentially according to the preset priority to complete the coordinated control of grouting.

[0011] S4. During the grouting collaborative management process, the integrity of the sealing body is evaluated in real time through a pre-constructed multi-field coupled damage evolution model of the sealing body, the failure precursor characteristics of the sealing body are identified and the remaining safe time is calculated, and the quantitative mapping rule pre-correction and graded risk disposal are performed in stages.

[0012] S5. Integrate the operational data and monitoring feedback data accumulated throughout the entire cycle of grouting collaborative control and correction treatment, quantify the control effect from multiple dimensions, and reverse-calibrate the benchmark parameters of the graded control benchmark and the multi-field coupled damage evolution model of the sealing body based on the review results. Then, the iteratively optimized calibration results are deposited into the engineering knowledge base for the cold start of the initial benchmark of similar projects, so as to realize closed-loop control.

[0013] Secondly, the present invention also provides a multi-parameter linkage closed-loop control system for grouting of yellow mud at the beginning of pipe jacking, the system comprising:

[0014] The control area division module is used to divide the reinforced area into multi-level control areas according to the intensity distribution based on existing engineering parameters, set corresponding hierarchical control benchmarks, and establish quantitative mapping rules for multi-set parameter linkage.

[0015] The data acquisition and processing module is used to synchronously acquire multi-source operating data across engineering equipment based on a unified data interaction interface and time series benchmark, and to perform standardized cleaning to form a standardized dataset with a unified time series.

[0016] The collaborative dynamic adjustment module is used to perform adaptive adjustment of grouting parameters adapted to the reinforcement zone, coordinated control of mud and water and grouting coupling, and coordinated control of correction action and grouting replenishment according to preset priority, based on hierarchical control benchmarks and quantitative mapping rules, combined with the operational data in the standardized dataset, and to complete the coordinated control of grouting.

[0017] The multi-level risk response module is used to assess the integrity of the sealing body in real time during the grouting collaborative management process through a pre-built multi-field coupled damage evolution model of the sealing body, identify the failure precursor characteristics of the sealing body and calculate the remaining safe time, and perform tiered pre-correction of quantitative mapping rules and tiered risk disposal.

[0018] The feedback closed-loop calibration module is used to integrate the operational data and monitoring feedback data accumulated throughout the entire cycle of grouting collaborative management and correction treatment. It quantifies and reviews the management effect from multiple dimensions, and reverse-calibrates the benchmark parameters of the graded management benchmark and the multi-field coupled damage evolution model of the sealing body based on the review results. The iteratively optimized calibration results are then stored in the engineering knowledge base for cold start of the initial benchmark of similar projects, thus realizing closed-loop management.

[0019] The beneficial effects of this invention are as follows:

[0020] 1. This invention breaks through the industry's conventional understanding of treating the sealing body as a static structure. By establishing a multi-field coupled damage evolution model for the sealing body, it incorporates three types of effects—mud-water seepage, correction compression, and grout strength growth—into a unified evaluation framework, enabling real-time calculation of the remaining effective thickness and breakdown probability of the sealing body on a cycle-by-cycle basis. Simultaneously, using hierarchical control benchmarks and quantitative mapping rules as a link, it organically links the control logic of the grouting system, mud-water system, and jacking system, upgrading each system from independent operation to collaborative control. This solves the fundamental problem of uncontrollable coupling risks caused by fragmented control of multiple systems at the system level.

[0021] 2. This invention uses spatial interpolation technology to convert discrete core sampling strength data into a continuous three-dimensional reinforcement strength distribution field. Based on strength thresholds, the end reinforcement area is divided into a strong reinforcement zone, a normal reinforcement zone, and a weak overlap zone. Differentiated upper and lower limits for grouting pressure, and benchmark values ​​for single-hole grouting volume and grout absorption rate are set for each zone. Based on this, three sets of quantitative mapping rules are established: the linkage between grout absorption rate and grouting pressure, the linkage between slurry pressure and grouting intensity, and the linkage between correction stroke and grouting volume. These enable the control parameters to adaptively adjust according to real-time operating conditions, achieving refined control with one standard per zone and one parameter per hole, effectively balancing the adequacy of sealing weak areas with the safety of heave in strong reinforcement zones.

[0022] 3. This invention introduces a joint judgment mechanism of the first-order and second-order rates of change of health indicators. By capturing the critical characteristics of the sealing body transitioning from stable deterioration to accelerated deterioration, it achieves early identification of failure precursors. Based on this, the risk judgment dimension is upgraded from spatial judgment of whether the thickness threshold is below a certain threshold to judgment of how much time is left before failure. The urgency of the remaining safety time is used as the basis for graded intervention, shifting risk management from passive response to proactive prediction, and providing sufficient time redundancy for construction safety under highly sensitive conditions. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0024] Figure 1 This is a flowchart of a multi-parameter linkage closed-loop control method for grouting in yellow mud at the beginning of pipe jacking according to an embodiment of the present invention;

[0025] Figure 2 This is a system principle block diagram of a multi-parameter linkage closed-loop control system for grouting in yellow mud during pipe jacking according to an embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of a construction project according to an embodiment of the present invention;

[0027] Figure 4 This is a schematic cross-sectional view of the starting end reinforcement zone structure according to an embodiment of the present invention;

[0028] Figure 5 This is a schematic diagram of the pipe jacking section from working shaft 7B# to working shaft 14# according to an embodiment of the present invention;

[0029] Figure 6 This is a schematic diagram of the elevation of the formation reinforcement at the beginning of the working well according to an embodiment of the present invention (from left to right: elevation of the formation reinforcement at the beginning of the working well, schematic diagram of the jet grouting arrangement, and schematic diagram of the jet grouting inspection hole arrangement).

[0030] The diagram shows the following modules: 1. Control area division module; 2. Data acquisition and processing module; 3. Collaborative dynamic adjustment module; 4. Multi-level risk response module; 5. Feedback closed-loop calibration module. Detailed Implementation

[0031] 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.

[0032] Please see Figure 1 This paper provides a multi-parameter linkage closed-loop control method for grouting in yellow mud at the beginning of pipe jacking, including:

[0033] S1. Based on existing engineering parameters, the reinforced area is divided into multi-level control zones according to the intensity distribution, and corresponding hierarchical control benchmarks are set to establish quantitative mapping rules for multi-set parameter linkage.

[0034] Specifically, this step is the benchmark calibration stage before construction. All parameters are initialized before the pipe jacking operation begins, and a hierarchical control benchmark table and a quantitative mapping rule table are output as the initial decision-making basis for subsequent dynamic control. All parameters are stored in the benchmark database of the control system after structured verification, supporting real-time access and dynamic correction during construction.

[0035] In the description of this invention, based on existing engineering parameters, the reinforced area is divided into multi-level control zones according to the intensity distribution, and corresponding hierarchical control benchmarks are set. A quantitative mapping rule for multi-set parameter linkage is established, including:

[0036] S11. Collect and unify the stratigraphic and engineering foundation parameters, and store them in the benchmark database after structured verification, as the input basis for hierarchical control and quantitative mapping rule calibration; the engineering foundation parameters include reinforcement data, equipment parameters, environmental constraint data and predetermined grouting construction parameters.

[0037] Specifically, the stratigraphic parameters are extracted from the formal engineering geological survey report, and the compression modulus, Poisson's ratio, permeability coefficient, internal friction angle, and cohesion of each soil layer are recorded in layers, with the data accuracy consistent with the survey report.

[0038] The reinforcement data was obtained from a third-party test report on the horizontal jet grouting pre-reinforcement at the end, including the plane coordinates, core depth, unconfined compressive strength, permeability coefficient, and core integrity coefficient of each core sampling point. The core sampling points were arranged in a grid with equal spacing of 1.5m × 1.5m within the plane of the end reinforcement area. The testing method complies with the relevant requirements of the "Code for Testing of Building Foundations" JGJ340.

[0039] The equipment parameters are extracted from the pipe jacking machine's manufacturer's manual and on-site commissioning report, including the cutterhead's rated torque, the total stroke of the correction cylinder, the maximum thrust of a single cylinder, the design pressure range of the mud and water chamber, the rated flow rate of the inlet and outlet mud pumps, and the arrangement of the correction cylinders.

[0040] Environmental constraint data is extracted from the special safety assessment report of sensitive structures, including the structure's plane coordinates, foundation type, foundation depth, design value of superstructure load, allowable cumulative settlement, allowable differential settlement, allowable settlement rate, and allowable tilt value.

[0041] The predetermined grouting construction parameters are extracted from the grouting-specific construction plan approved by the supervisor, including the benchmark value of the single-hole design grouting pressure, the benchmark value of the single-hole design grouting volume, the design bearing threshold of the sealing body, the planar position of the grouting holes around the tunnel and the benchmark grouting volume of a single hole, and the design initial setting time of the sealing body.

[0042] Structured verification includes two aspects: field integrity verification and numerical rationality verification. If a field is missing, the data is marked as abnormal and a manual completion prompt is given. If the value exceeds the reasonable range of the project, it is marked as abnormal, such as negative grouting pressure or strength value exceeding the reasonable range of the stratum. The data can only be stored in the benchmark database after being reviewed and confirmed by technical personnel.

[0043] S12. Based on the core sample strength test results of the pre-reinforced end, a continuous reinforcement strength distribution field is generated by spatial interpolation. The initial end reinforcement area is divided into multi-level control areas according to the reinforcement strength effect, and the zoning matching and labeling of all grouting holes is completed. The control areas include strong reinforcement area, normal reinforcement area and weak overlap area.

[0044] Specifically, this step uses discrete core sampling intensity data as a basis and transforms the discrete point data into a continuous three-dimensional intensity distribution field through a spatial interpolation algorithm. This solves the problem that the number of core sampling points is limited and cannot directly reflect the spatial differences in the overall reinforcement intensity. Finally, the area is divided into three levels of control zones according to the intensity threshold, providing a spatial basis for subsequent differentiated control.

[0045] During implementation, a local construction coordinate system was established with the center of the portal steel ring as the origin, the Z-axis along the tunnel axis, the X-axis horizontally, and the Y-axis vertically. The coordinates of all core sampling points and grouting holes were unified under this coordinate system. Ordinary kriging was used for interpolation, with the three-dimensional coordinates of the core sampling points as the independent variable and the unconfined compressive strength of the core sample as the dependent variable. A spherical semi-variogram model was constructed, and the nugget value, abutment value, and range parameters were fitted using the least squares method. The three-dimensional space of the end reinforcement zone was discretized into a 0.5m × 0.5m × 0.5m cubic grid, and the interpolated strength value was calculated grid by grid to generate a three-dimensional reinforcement strength distribution cloud map.

[0046] The grading thresholds are determined based on the engineering characteristics of the reinforcement at the starting end of the pipe jacking project. For example, a grid strength value ≥ 1.5 MPa indicates a strong reinforcement zone, corresponding to areas with dense strata and low grout absorption; a strength value of 1.0-1.5 MPa indicates a normal reinforcement zone, corresponding to areas where the reinforcement effect meets design requirements; and a strength value < 1.0 MPa indicates a weak overlap zone, corresponding to areas with high grout absorption and high sealing risk, such as pile overlaps and stratum interfaces. After spatially overlaying and matching all grouting hole locations with the strength cloud map grid, the reinforcement zone level of each hole location is marked and associatedly stored in the baseline database.

[0047] S13. Using the design value of the predetermined grouting construction parameters as the global benchmark value, the control threshold of the divided multi-level control areas is adjusted proportionally to generate the control benchmark corresponding to each reinforcement area; the control benchmark includes the upper and lower limits of grouting pressure, the benchmark grouting volume of a single hole, and the benchmark value of grout absorption rate.

[0048] Specifically, the global benchmark value is directly adopted from the approved grouting special plan design value as the control standard for normal reinforcement areas; in strong reinforcement areas, due to the dense strata and weak grout absorption capacity, excessive grouting pressure can easily cause strata uplift, so the pressure and grouting volume benchmarks are adjusted downwards proportionally; in weak overlap areas, due to the loose strata and strong grout absorption capacity, insufficient grouting can easily lead to sealing failure, so the grouting volume benchmarks are adjusted upwards proportionally and additional grouting trigger conditions are added.

[0049] In one embodiment, the reinforced zone is configured with: an upper limit for grouting pressure of 85% of the global benchmark value, a single-hole benchmark grouting volume of 90% of the global benchmark value, and a grout absorption rate benchmark value of 80% of the global benchmark value; all parameters in the normal reinforced zone are set to the global benchmark value; the weak overlap zone is configured with: an upper limit for grouting pressure of 90% of the global benchmark value, a single-hole benchmark grouting volume of 120% of the global benchmark value, and an additional grouting trigger threshold: when the single-hole grouting volume reaches 120% of the benchmark value and the grouting pressure does not reach 80% of the upper limit, the grouting process is automatically triggered. The lower limit for grouting pressure in all areas is uniformly set to 60% of the global benchmark value. It should be noted that the values ​​in this embodiment are merely illustrative and do not constitute a limitation of the present invention. Actual values ​​can be calibrated according to the working conditions. The same applies to the embodiments below, and will not be described in detail again.

[0050] All control benchmarks for all control areas must pass environmental constraint verification. This involves using the layered summation method to calculate the theoretical ground uplift under the corresponding grouting parameters. If the theoretical uplift exceeds 70% of the allowable uplift of sensitive structures, the upper limit of grouting pressure for that area will be further reduced until the environmental constraint requirements are met.

[0051] S14. Based on multi-level control areas and corresponding control benchmarks, establish multiple sets of quantitative mapping rules fixed in the form of linear functions. The quantitative mapping rules are used to correlate grout absorption rate with grouting pressure, mud-water pressure with grouting intensity, and correction stroke with grouting volume.

[0052] Specifically, all three sets of mapping rules are based on regression calibration using measured data from historical projects under the same working conditions. The correspondence between parameters is solidified in the form of linear functions, allowing for direct programming execution without the need for manual fuzzy judgment. Each mapping rule corresponds one-to-one with the hierarchical control benchmark, and different reinforcement zones can utilize appropriate proportional coefficients. These three sets of mapping rules correspond to three physical mechanisms: formation grout absorption adaptation, slurry load coupling, and correction gap compensation, respectively, collectively forming a quantitative execution basis for cross-system linkage.

[0053] In the description of this invention, based on multi-level control areas and corresponding control benchmarks, multiple sets of quantization mapping rules fixed in the form of linear functions are established, including:

[0054] S141. Using the benchmark grout absorption rate corresponding to each reinforced area as a reference, pre-calibrate the first proportional relationship between the deviation of the grout absorption rate and the adjustment amount of the grouting pressure. When the real-time grout absorption rate is higher than the benchmark grout absorption rate of the corresponding control area, the grouting pressure is reduced according to the first proportional relationship; when the real-time grout absorption rate is lower than the benchmark grout absorption rate of the corresponding control area, the grouting pressure is increased according to the first proportional relationship.

[0055] Specifically, the physical meaning of the first proportional relationship is the step size of the grouting pressure adjustment corresponding to the deviation of the grout absorption rate. The calibration basis is the grouting diffusion characteristics of the stratum in the end reinforcement zone. That is, a high grout absorption rate indicates that the stratum's grout absorption capacity is stronger than expected, and lowering the pressure can prevent the grout from spreading too far and causing grout leakage; a low grout absorption rate indicates that the stratum is dense, and raising the pressure can ensure that the grout fully fills the pores.

[0056] In one embodiment, the calibration method for the proportional coefficient is as follows: Grouting data from at least three historical projects under the same geological conditions are selected, and the correlation between the grout absorption rate deviation and the corresponding optimal pressure adjustment is statistically analyzed. The proportional coefficient is obtained through linear regression. Under engineering constraints, every 10% deviation in grout absorption rate corresponds to a pressure adjustment of 0.01 MPa. At this step size, the change in additional stress is sufficient to alter the grout diffusion behavior without causing formation damage. The single adjustment amplitude does not exceed 20% of the pressure difference between the upper and lower limits of the corresponding area to prevent pressure oscillation. During construction, this coefficient can be dynamically adjusted through the pre-correction in S4 and the iterative optimization in S5.

[0057] S142. Using the design value of slurry pressure as a reference, pre-calibrate the second proportional relationship between the increase in slurry pressure and the increase in grouting volume per hole around the tunnel; determine the increase in grouting volume per hole around the tunnel according to the second proportional relationship based on the increase in real-time slurry pressure relative to the design value.

[0058] Specifically, the physical meaning of the second proportional relationship is the increase in grouting intensity corresponding to a unit increase in slurry pressure. The calibration basis is the seepage resistance and bearing capacity characteristics of the plugging body: an increase in slurry pressure will increase the water pressure load borne by the plugging body, and grouting is needed to improve the seepage resistance of the plugging body in order to maintain pressure balance.

[0059] In one embodiment, the proportionality coefficient is calibrated as follows: based on the geometric dimensions of the plug (thickness 0.32m, outer diameter 4.14m), the impermeability strength of the grout, and the formation permeability coefficient, the theoretical grouting increment is calculated through seepage mechanics calculations, and then corrected by combining engineering measured data to obtain the final proportionality coefficient; under engineering constraints, for every 0.01MPa increase in slurry pressure, the corresponding grouting volume per borehole increases by 5% of the benchmark value. This value ensures that the grouting volume is sufficient to offset the seepage risk and will not cause formation heave. The coefficient value must ensure that the grouting volume is sufficient to offset the seepage risk caused by the increase in slurry pressure, while preventing formation heave due to excessive grouting.

[0060] S143. Pre-calibrate the third proportional relationship between the single-sided stroke difference of the correction cylinder and the increase in slurry replenishment amount per hole on the correction side. Based on the real-time detected single-sided stroke difference, determine the increase in slurry replenishment amount per hole on the correction side according to the third proportional relationship.

[0061] Specifically, the physical meaning of the third proportional relationship is the amount of grout filling in the gap corresponding to the unit increment of the correction stroke. The calibration basis is the geometric characteristics of the pipe jacking machine's correction: the difference in correction stroke corresponds to the deflection angle of the machine head. The deflection will increase the circumferential gap between the correction side pipe section and the stratum, and the gap needs to be filled by grout filling to maintain the integrity of the sealing body.

[0062] In one embodiment, the calibration method of the proportional coefficient is as follows: based on the outer diameter of the pipe section of 4.14m, the ring width of 2.5m and the porosity of the formation, the gap volume increment corresponding to the deflection angle is obtained through geometric calculation, converted into grouting volume, and then corrected with engineering measured data to obtain the final proportional coefficient; under engineering constraints, for every 20mm increase in the correction stroke, the grouting volume of a single hole increases by 10% of the benchmark value to match the building gap volume increment caused by the deflection.

[0063] S144. Generate a quantitative mapping rule table by setting the judgment thresholds and adjustment ratios for the established first proportional relationship, second proportional relationship, and third proportional relationship.

[0064] Specifically, the quantitative mapping rule table is stored in a structured two-dimensional table format. Each row corresponds to a reinforcement zone level, and each column corresponds to a set of mapping rule ratio coefficients and trigger thresholds. The rule table is stored in the benchmark database. During construction, the control module can directly look up the table to call the corresponding parameters without recalculation. After the benchmark parameters are corrected, the rule table is updated synchronously and overwrites the old version to ensure that the called parameters are always the latest values.

[0065] S2. Based on a unified data interaction interface and timing benchmark, multi-source operation data across engineering equipment is collected synchronously and standardized cleaning is performed to form a standardized dataset with unified timing.

[0066] Specifically, this step is the data infrastructure layer. By using an edge computing gateway, the data barriers between four independent equipment types—the grouting control system, the pipe jacking machine PLC control system, the slurry conveying control system, and the environmental monitoring system—are broken down. This enables the synchronous collection, cleaning, and standardized conversion of multi-source data, providing a unified and reliable data source for subsequent dynamic control and risk prediction. All data is accompanied by a unified timestamp to ensure the temporal consistency of data from multiple systems.

[0067] In one embodiment, the system interface adopts an edge gateway and multi-protocol adaptation architecture, requiring no modification to the control logic of the original equipment. Data reading can be achieved solely through the communication interface, without affecting the normal operation of the original equipment. The grouting control system and mud-water conveying control system use the Modbus-RTU protocol to interface via an RS-485 bus. Grouting pressure corresponds to register address 40001, instantaneous flow rate to 40002, and cumulative grouting volume to 40003. The data frame baud rate is 9600bps, with 8 data bits and 1 stop bit. The pipe jacking machine PLC control system uses the OPCUA protocol to interface via industrial Ethernet, with the main controller being a Siemens S7-1200 and a subscription period of 100ms. The environmental monitoring system uses the MQTT protocol to interface via a 4G wireless network, with data encapsulated in JSON format. All connected devices achieve clock synchronization through an NTP network time server.

[0068] The data acquisition frequency is configured according to the system type. For example, the acquisition frequency for grouting, pipe jacking, and slurry systems is set to 0.1Hz. The environmental monitoring system's regular acquisition frequency is once every 10 minutes, automatically increasing to once every 30 seconds during early warning periods. The data cleaning process includes four steps: outlier removal, zero-point drift correction, unit format standardization, and missing value handling. Outliers are determined using the 3σ criterion, taking the data from the previous 10 consecutive periods as a sample. Values ​​deviating from the mean by three times the standard deviation are removed. This window length matches the fluctuation characteristics of pipe jacking construction data to avoid misjudging normal fluctuations or missing real anomalies. Zero-point calibration is automatically performed every 24 hours to offset temperature drift and zero-point drift errors. All physical quantities are uniformly converted to the International System of Units (SI), and the values ​​are uniformly formatted as single-precision floating-point numbers. Single missing points are filled using linear interpolation. If more than three consecutive points are missing, the data is marked as abnormal, triggering a sensor fault warning. Data during this period is not included in the control calculation.

[0069] S3. Based on the hierarchical control benchmark and quantitative mapping rules, and combined with the operational data in the standardized dataset, the grouting parameters adapted to the reinforcement zone are adaptively adjusted, the mud and water and grouting are coupled and coordinated, and the correction action and grouting are coordinated and coordinated, in order to complete the grouting coordinated control.

[0070] Specifically, this step is the core layer of dynamic control. The control priority is as follows: environmental constraints are the highest, coupled control is the second highest, and zonal adaptation is the lowest. When there is a risk of exceeding environmental limits, environmental safety is given priority. Coupled control of mud and water and correction has a higher priority than zonal adaptation adjustment of a single system. The three types of control logic correspond to three physical mechanisms: formation inhomogeneity, fluid pressure transmission, and geometric gap changes, respectively, forming a progressive control system: S31 solves the spatial adaptation problem, S32 solves the dynamic coupling problem, and S33 solves the geometric compensation problem, together covering all the core risks of the initial portal sealing.

[0071] In the description of this invention, based on the hierarchical control benchmark and quantitative mapping rules, and combined with the operational data in the standardized dataset, the dynamic adjustment of grouting parameters adapted to the reinforcement zone, the coupled and coordinated control of mud and water and grouting, and the coordinated control of correction actions and grout replenishment are executed sequentially according to preset priorities to complete the dynamic adjustment of grouting parameters, including:

[0072] S31. Using the graded control benchmark as a constraint and the current cumulative grouting volume and real-time grouting pressure of the grouting hole as inputs, dynamically match the control benchmark according to the reinforcement zone to which the current grouting hole belongs, and adaptively adjust the grouting pressure in real time according to the mapping relationship between grout absorption rate and grouting pressure.

[0073] Specifically, before construction, the current grouting hole number is entered using a barcode scanner, and the system automatically calls the control benchmark of the reinforcement area to which the hole belongs; during construction, the cumulative grouting volume is updated once every collection cycle, the real-time grout absorption rate and the grout absorption rate deviation are calculated, and the pressure adjustment amount is obtained by substituting into the first proportional relationship; the adjusted pressure value needs to be checked to see if it exceeds the upper and lower pressure limits of the corresponding area. If it exceeds, the limit value is taken as the final set value.

[0074] Boundary constraint verification includes two parts: pressure limit verification and environmental constraint verification. For example, if the settlement rate of the current sensitive structure is ≥0.15mm / d, pressure increases are prohibited, and only decreases are allowed to prioritize the safety of the structure. The final pressure setpoint is sent to the grouting pump frequency converter via the Modbus protocol, automatically adjusting the output pressure to complete a parameter fine-tuning.

[0075] S32. With the design bearing threshold of the sealing body and the rated torque of the cutterhead as constraints, and with the real-time pressure of the mud-water chamber and the real-time torque of the cutterhead as inputs, the two-way linkage control of the mud-water system and the grouting control is performed based on the mapping relationship between mud-water pressure and grouting intensity.

[0076] Specifically, this step includes two sub-logic: linkage between slurry pressure and sealing strength, and linkage between cutterhead torque and grouting rhythm. These sub-logic address the risks of sealing body rupture and grout leakage into the grouting chamber, respectively, thereby achieving bidirectional coordination between the slurry system and the grouting system.

[0077] The mud-water pressure linkage adopts a graded triggering mechanism: the sliding average value of the mud-water chamber pressure is used as the judgment input to avoid false triggering due to instantaneous fluctuations; when the pressure reaches 75% of the bearing threshold, the first-level linkage is executed, and the flow rate of the grout pump is reduced to suppress the pressure rise; when the pressure reaches 85% of the bearing threshold, the second-level linkage is executed, and the grouting and reinforcement sealing body around the tunnel is opened simultaneously.

[0078] The cutterhead torque linkage is used to break the vicious cycle of increased torque, worsened cutter clogging, and further increased torque: the 10-second sliding average value of the cutterhead torque is taken as the judgment input. When the torque reaches 80% of the rated value, the grouting interval is extended and the standby pressure is reduced to allow the grout to have sufficient initial setting time and prevent uncured grout from being carried into the mud-water chamber; the upper limit of the grouting interval extension shall not exceed 1 / 3 of the initial setting time of the sealing body to prevent pipeline blockage.

[0079] In one embodiment, the 30-second sliding average of the slurry chamber pressure is used as the judgment input to avoid false triggering due to instantaneous fluctuations. When the average pressure of three consecutive sampling cycles reaches 75% of the bearing threshold, a first-level linkage is executed, reducing the operating frequency of the grout pump by 5Hz (corresponding to a flow rate reduction of approximately 10%), and simultaneously initiating closed-loop control of the slurry pressure, setting the pressure target value to 70% of the bearing threshold, and controlling the fluctuation within the ±0.02MPa range through PID regulation. When the average pressure of five consecutive sampling cycles reaches 85% of the bearing threshold, a second-level linkage is executed, opening the grouting holes around the tunnel for grouting based on the first-level linkage, with the grouting volume calculated according to the second proportional relationship, and the grouting pressure set to 80% of the current grouting pressure. After the pressure drops below 0.7 times the bearing threshold and stabilizes for 10 minutes, the parameters are gradually restored in steps of 1Hz every 5 minutes.

[0080] Based on the cutterhead's rated torque of 1988 kN·m, the warning value is set to 80% of the rated value, i.e., 1590 kN·m, and the normal recovery value is set to 60% of the rated value, i.e., 1193 kN·m. The 10-second sliding average of the cutterhead torque is used as the judgment input. When the warning value is reached for two consecutive collection cycles, the single-hole grouting interval is extended to 1.2 times the baseline value. Simultaneously, the grouting pump standby pressure is reduced from 2.0 MPa to 1.5 MPa to allow the injected grout sufficient initial setting time, forming an initial structural strength of ≥0.1 MPa, preventing uncured grout from being carried into the mud-water chamber. The upper limit of the grouting interval extension does not exceed 1 / 3 of the designed initial setting time of the sealing body to prevent pipeline blockage. After the torque falls below the recovery value and stabilizes for 5 minutes, the parameters are gradually adjusted back.

[0081] S33. Using the stroke values ​​of multiple sets of correction cylinders as input, the collaborative control of attitude deviation and grouting is achieved based on the mapping relationship between correction stroke and grouting amount.

[0082] Specifically, the stroke values ​​of the four sets of correction cylinders (upper left, upper right, lower left, and lower right) are read in each acquisition cycle, and the average stroke difference between the left and right sides is calculated: the average stroke of the left side minus the average stroke of the right side. A positive value indicates that the stroke of the left side is greater and the machine head corrects to the right, while a negative value indicates that the stroke of the right side is greater and the machine head corrects to the left.

[0083] In one embodiment, when the absolute value of the stroke difference between two consecutive acquisition cycles is ≥60mm, it is determined that the trigger threshold has been reached, corresponding to a correction angle of about 1.4°. Lateral extrusion may have caused the sealing body to crack. The grouting hole around the hole on the correction side is automatically opened, the grouting volume is calculated according to the third ratio, and the lateral gap generated by the correction is filled by low-pressure slow injection. The grouting pressure is 70% of the reference pressure.

[0084] The slurry replenishment is terminated when the cumulative slurry replenishment amount reaches the calculated value, or when the correction stroke difference falls back to within 30mm and remains stable for 2 collection cycles; after slurry replenishment stops, it automatically returns to the normal zoning adaptation and control logic.

[0085] S4. During the grouting collaborative management process, the integrity of the sealing body is evaluated in real time through a pre-constructed multi-field coupled damage evolution model of the sealing body, the failure precursor characteristics of the sealing body are identified and the remaining safe time is calculated, and the quantitative mapping rule pre-correction and graded risk disposal are performed in stages.

[0086] Specifically, this step is the core of proactive risk management, breaking through the traditional reactive mode of monitoring and handling after exceeding limits. It uses a multi-field coupled damage evolution model of the sealing body to assess the health status of the sealing body in real time. First, it identifies the precursors of failure through the second-order rate of change, and then calculates the remaining safe time through trend fitting. Proactive intervention is carried out in stages according to the urgency of time. The risk handling priority is higher than that of conventional dynamic control, which can proactively eliminate hidden dangers before the risk actually occurs, and significantly improve the safety redundancy of highly sensitive working conditions.

[0087] In the description of this invention, the integrity of the sealing body is evaluated in real time through a pre-constructed multi-field coupled damage evolution model, the precursor characteristics of sealing body failure are identified and the remaining safe time is calculated, and the quantitative mapping rule pre-correction and graded risk management are performed in a tiered manner, including:

[0088] S41. Based on the multi-field coupling effect of mud-water seepage, correction compression and slurry strength growth, a multi-field coupling damage evolution model of the sealing body is established.

[0089] Specifically, the multi-field coupled damage evolution model for the sealing body is constructed based on the specific failure mechanism of the sealing body for the tunnel entrance of the pipe jacking project. It considers three types of effects: scouring and deterioration caused by mud and water seepage, shear damage caused by correction and compression, and strength enhancement caused by the increase in grout age. It also introduces a coupling amplification factor to characterize the damage superposition effect under the combined action of multiple fields, which is different from the conventional single-factor linear superposition analysis. The input of the multi-field coupled damage evolution model for the sealing body is real-time construction parameters, and the output is the remaining effective thickness of the sealing body and the probability of breakdown, which can be dynamically calculated on a cycle-by-cycle basis.

[0090] In the description of this invention, a multi-field coupled damage evolution model of the sealing body is established based on the multi-field coupling effects of mud-water seepage, correction compression, and slurry strength growth, including:

[0091] S411. Obtain the initial geometric and material parameters of the sealing body used to seal the circumferential gap between the steel ring of the tunnel portal and the stratum. The geometric parameters include the thickness and spatial range of the sealing body, and the material parameters include the design impermeability and design shear strength of the grout after initial setting.

[0092] Specifically, the initial geometric parameters are extracted from the grouting design scheme, where the thickness of the sealing body is the design grouting thickness, and the spatial range of the sealing body corresponds to the circumferential grouting area between the portal steel ring and the stratum. The material parameters are taken from the grout mix ratio test report, including the impermeability strength grade and shear strength index under standard curing conditions, which serve as the initial material benchmark for calculating the multi-field coupled damage evolution model of the sealing body.

[0093] S412. Establish the correspondence between mud and water pressure and the expansion rate of seepage channels inside the sealing body; the higher the mud and water pressure, the faster the seepage channel expansion rate and the greater the rate of decrease in the effective anti-seepage thickness of the sealing body.

[0094] Specifically, this correspondence is based on the piping expansion theory of seepage mechanics, with the ratio of mud-water pressure to the impermeability of the plugging body as the independent variable and the seepage channel expansion rate as the dependent variable; the greater the pressure difference, the stronger the seepage scouring effect, the faster the channel expands, and the faster the effective impermeability thickness of the corresponding plugging body decreases.

[0095] During implementation, the channel expansion rate under different pressures can be determined through indoor seepage tests. A proportional coefficient, representing the weakening coefficient of the slurry pressure on the sealing body, is obtained through linear fitting. After initial calibration using the above method, this coefficient is dynamically optimized during construction through a parameter correction mechanism driven by the prediction deviation in S47. Specifically, when the deviation between the predicted and measured values ​​of the remaining effective thickness of the sealing body continuously exceeds a preset confidence interval, the coefficient is determined to deviate from the true value under the current working conditions. The system automatically adjusts the coefficient upwards or downwards based on the sign and degree of deviation, with a single correction step not exceeding 10% of the current parameter value. After correction, the prediction deviation is continuously monitored until the correction is confirmed to be effective. S413: Establish the correspondence between the stroke difference of the correction cylinder and the lateral compressive strain of the sealing body; the larger the stroke difference, the greater the local compressive strain of the sealing body and the greater the reduction rate of the effective thickness of the sealing body.

[0096] Specifically, this correspondence is based on the local compressive deformation theory of materials mechanics. The deflection angle of the machine head is calculated by converting the difference in the correction stroke, and then the local compressive strain is calculated by combining the elastic modulus of the sealing body. The greater the strain, the higher the degree of compression and compaction of the sealing body, the more obvious the reduction in the effective anti-seepage thickness, and the easier it is to generate micro-cracks that intensify seepage.

[0097] S414. Establish the correlation between the time elapsed after grouting and the strength growth rate of the sealing material; the strength growth rate gradually increases to a peak value during the initial setting stage of the grout and then tends to stabilize.

[0098] Specifically, this correspondence is based on the hydration and hardening law of cement-based grout and is described by the age-dependent strength growth function: after grouting, as time goes by, the hydration reaction of the grout continues, the strength gradually increases, the growth rate accelerates after initial setting, and tends to stabilize after final setting; the corresponding strength growth is equivalent to the increase in the effective thickness of the sealing body.

[0099] S415. The remaining effective thickness of the plugging body is equal to the initial thickness minus the seepage equivalent thickness attenuation, minus the product of the compression equivalent thickness reduction and the coupling amplification factor, and then added to the strength equivalent thickness enhancement to construct a multi-field coupled damage evolution model of the plugging body.

[0100] Specifically, the coupling amplification factor is greater than 1, which is used to characterize the damage superposition effect under the combined action of seepage deterioration and compression damage: the microcracks generated by compression provide channels for seepage and accelerate seepage expansion; seepage scouring will further deteriorate the sealing structure and aggravate the impact of compression damage. The damage caused by the combined action of the two is greater than the linear superposition of the individual actions.

[0101] The initial value of the coupling amplification factor is based on engineering experience under the same working conditions, ranging from 1.1 to 1.5. The specific value can be calibrated through indoor multi-field coupling tests.

[0102] Among them, the remaining effective thickness of the sealing body The calculation formula is:

[0103] ;

[0104] In the formula, for The remaining effective thickness of the sealing body at any given time, in meters; The initial design thickness of the sealing body is expressed in meters (m). for The cumulative reduction in equivalent thickness caused by seepage of mud and water at any given time, in meters; for The cumulative reduction in equivalent thickness caused by constant correction and extrusion, in meters; is the seepage-compression coupling damage amplification factor, which is dimensionless and takes a value greater than 1. It is used to characterize the damage superposition amplification effect under the combined action of multiple fields. for The cumulative equivalent thickness increase corresponding to the increase in slurry strength at any given time, expressed in meters (m).

[0105] It should be noted that, It is obtained by integrating the seepage channel expansion rate over time. The difference in travel distance for correction is calculated using geometric relationships. It was calculated based on the strength growth curve of the slurry over time. The physical meaning is that when seepage degradation and compression damage coexist, the microcracks generated by compression accelerate seepage propagation, and seepage erosion further deteriorates the sealing structure, thus exacerbating the impact of compression damage. The combined effect of these two factors results in damage greater than the linear superposition of their individual effects. This is a quantitative characterization of the superposition amplification effect.

[0106] S42. Based on the preset acquisition cycle, the real-time pressure of the mud and water chamber, the real-time stroke difference of the correction cylinder, and the grouting completion timestamp are converted into the time elapsed after grouting and substituted into the multi-field coupled damage evolution model of the sealing body to calculate the remaining effective thickness of the sealing body at the current moment, and calculate the probability of sealing body breakdown based on the ratio of the remaining effective thickness to the safe thickness threshold.

[0107] Specifically, the preset collection cycle is consistent with the data collection frequency and can be adjusted according to the engineering risk level; the elapsed time after grouting is obtained by subtracting the grouting completion timestamp of the corresponding hole position from the current time, accurate to the minute.

[0108] The breakdown probability is calculated as follows: Breakdown probability = 1 - Remaining effective thickness / Safe thickness threshold, with a value ranging from 0 to 1. The closer the value is to 1, the higher the risk of the seal body breaking down; a value of 0 indicates complete safety. The safe thickness threshold is determined based on the ratio of the seal body's design bearing capacity threshold to the maximum slurry pressure. To ensure the reliability of the seal, the safe thickness should not be less than 50% of the design thickness.

[0109] S43. Based on the remaining effective thickness of the plug body calculated cycle by cycle, construct a health index for the plug body and solve its first-order and second-order rates of change. When the three indices simultaneously meet the preset instability criteria, it is determined that the plug body has shown signs of failure and the risk level is automatically upgraded by one level.

[0110] Specifically, this step captures the precursors of accelerated deterioration before the sealing body becomes unstable by using the multi-level change rate of health status. This corresponds to the critical state of the seepage channel transitioning from stable expansion to unstable expansion. Compared with simple thickness threshold prediction, this step can identify risks 1-2 construction cycles earlier, further advancing the risk control time point.

[0111] Among them, health index The calculation formula is:

[0112] ;

[0113] In the formula, For the first The health index of the closure body in each collection cycle reflects the degree of leeway of the remaining effective thickness of the closure body relative to the safety threshold; For the first The remaining effective thickness of the sealing body calculated in each acquisition cycle, in meters; The safe thickness threshold of the sealing body, in meters, is determined by the ratio of the design bearing capacity threshold of the sealing body to the maximum slurry pressure.

[0114] The formula for calculating the first-order rate of change is:

[0115] ;

[0116] In the formula, For the first The first-order rate of change of the health of the occluded body in each collection cycle, in units of 1 / s, reflects the rate of deterioration of the integrity of the occluded body over time. For the first Health index values ​​for each data collection period; For the first Health index values ​​for each data collection period; The data collection period is measured in seconds (s).

[0117] The formula for calculating the second-order rate of change is:

[0118] ;

[0119] In the formula, For the first The second-order rate of change of the health of the occluded body in each collection cycle, in units of 1 / s², reflects the changing trend of the rate of deterioration of the occluded body. For the first The first-order rate of change for each acquisition cycle is expressed in units of 1 / s.

[0120] In the description of this invention, based on the remaining effective thickness of the plugging body calculated cycle by cycle, a health index for the plugging body is constructed and its first-order and second-order rates of change are solved. When all three indices simultaneously meet the preset instability criteria, it is determined that the plugging body has shown signs of failure, and the risk level is automatically upgraded by one level, including:

[0121] S431. The ratio of the remaining effective thickness of the closure body at the current moment to the preset safe thickness threshold is defined as the health index of the closure body.

[0122] Specifically, the health index is a dimensionless value with a range greater than 0. A value of 1 indicates that the remaining effective thickness is exactly equal to the safe thickness threshold, a value greater than 1 indicates that it is in a safe state, and a value less than 1 indicates that it has entered the risk zone. This index can intuitively reflect the relative health status of the sealing body and facilitates comparison across working conditions.

[0123] S432. Based on the health index of the continuous acquisition cycle, the first-order rate of change of the health index is calculated by the difference method to reflect the deterioration rate of the integrity of the blockage.

[0124] Specifically, the first-order rate of change is calculated by dividing the difference in health between two adjacent collection cycles by the collection cycle duration, with the unit being 1 / second. A negative value indicates a continuous decline in health and a continuous deterioration of the blockage, and the larger the absolute value, the faster the deterioration. A positive value indicates an improvement in health and an improvement in the blockage status.

[0125] S433. Based on the first-order rate of change of the continuous acquisition cycle, the second-order rate of change of the health index is calculated using the difference method to reflect the changing trend of the deterioration rate of the blockage body.

[0126] Specifically, the second-order rate of change is calculated by dividing the difference between the first-order rates of change of two adjacent acquisition cycles by the acquisition cycle duration, with the unit being 1 / second². A positive value indicates that the rate of deterioration is accelerating, the blockage body is in a state of accelerated deterioration, and it is a typical precursor to instability and failure. A negative value indicates that the rate of deterioration is slowing down, and the risk is converging.

[0127] S434. When all failure criteria are met simultaneously, the sealing body is deemed to have shown signs of failure. The failure criteria include: the health index is continuously lower than the preset health threshold, the first-order rate of change is less than zero, and the second-order rate of change is greater than zero.

[0128] Specifically, the default health threshold is set to 1.0, meaning that monitoring for signs of failure begins once the blockade body enters the risk zone; a threshold of less than 1.0 means that the condition is met for three consecutive data collection cycles to avoid misjudgments caused by instantaneous data fluctuations.

[0129] When all three conditions are met, it indicates that the blockage body is not only in a risk range and continuously deteriorating, but the rate of deterioration is also accelerating. The corresponding seepage channel is about to enter the unstable and connected stage. At this time, the failure precursor judgment is triggered, the risk level is automatically upgraded by one level, and higher-level control measures are initiated in advance.

[0130] S44. Based on the time series of the remaining effective thickness of the plug body obtained by cycle-by-cycle calculation, perform trend fitting, calculate the remaining safe time of the plug body, and classify the breakdown risk level according to the duration interval of the remaining safe time.

[0131] Specifically, this step transforms risk assessment from a spatial dimension of whether the thickness is below the threshold to a temporal dimension of how long until failure. The urgency of the remaining safety time is used as the basis for classification, so that the intensity of control measures is precisely matched with the urgency of the risk, avoiding over-control affecting construction efficiency or under-control leaving safety hazards.

[0132] In the description of this invention, trend fitting is performed based on the time series of the remaining effective thickness of the plug body obtained by periodic calculation to calculate the remaining safe time of the plug body, and the breakdown risk level is divided according to the duration interval of the remaining safe time, including:

[0133] S441. Arrange the remaining effective thickness values ​​of the sealing body calculated in each acquisition cycle according to the acquisition time to construct a time series of the remaining effective thickness of the sealing body.

[0134] Specifically, the time series uses the most recent 30 consecutive minutes of historical data by default, corresponding to a total of 180 data points in a 10-second collection period; the data points are updated on a rolling basis with the collection period, and the oldest historical data is removed each time a new data point for the current period is added to ensure that the sequence length is fixed.

[0135] S442. Using the acquisition time of the time series of the remaining effective thickness of the sealing body as the independent variable and the remaining effective thickness of the sealing body as the dependent variable, the first-order linear least squares method is used to perform trend fitting. The slope and intercept of the time series are obtained by fitting to determine the direction of change of the integrity of the sealing body.

[0136] Specifically, first-order linear least squares fitting obtains the linear equation that best reflects the overall trend of change by minimizing the sum of squared errors between the measured value and the fitted line; the sign of the slope reflects the direction of thickness change, and the magnitude of the absolute value reflects the rate of change; linear fitting can ensure computational efficiency and meet the response requirements of real-time control.

[0137] S443. When the slope obtained from the fitting is less than zero, the difference between the current remaining effective thickness and the safe thickness threshold is divided by the absolute value of the slope to calculate the remaining safe time for the sealing body. When the slope is greater than or equal to zero, the remaining safe time is recorded as infinity.

[0138] Specifically, the unit of remaining safe time is the number of acquisition cycles, which can be converted into seconds or minutes; when the slope is greater than or equal to zero, it indicates that the thickness of the sealing body is stable or continues to increase, and there is no risk of breakdown, so the remaining safe time is infinite.

[0139] This calculation is based on the assumption that the current rate of deterioration remains constant, and it reflects the estimated time until the sealing fails, providing a time basis for proactive intervention. The prediction period covers two future construction cycles by default, corresponding to jacking two pipe sections for a total advance of 5 meters, matching the construction step length in the initial stage.

[0140] The formula for calculating the remaining safety time is as follows:

[0141] ;

[0142] In the formula, The remaining safe time for the closure body is expressed in the number of data collection cycles. It represents the estimated time required for the closure body to deteriorate from its current state to the safe thickness threshold at the current rate of deterioration. This represents the remaining effective thickness of the sealing body during the current data acquisition cycle, in meters (m). The safe thickness threshold for the sealing body, in meters (m); The slope of the remaining effective thickness of the closure body is obtained by first-order linear least squares fitting of the time series, with units of m / acquisition period, reflecting the overall trend and rate of change of the closure body thickness.

[0143] S444. Based on the preset risk level mapping table, divide the corresponding breakdown risk level according to the remaining safe time.

[0144] Specifically, the risk level mapping table has two preset risk levels: a safe state when the remaining safe time is greater than 10 collection cycles; a level 1 breakdown risk when the remaining safe time is greater than 3 collection cycles and less than or equal to 10 collection cycles; and a level 2 breakdown risk when the remaining safe time is less than or equal to 3 collection cycles.

[0145] The grading thresholds are set based on the duration of the on-site emergency response chain: 10 collection cycles correspond to a reasonable response window for parameter fine-tuning, and 3 collection cycles correspond to the shortest response time for emergency shutdown and disposal, which is matched with the construction collection frequency and on-site disposal procedures.

[0146] If a failure warning has been triggered, the risk level will be automatically increased by one level based on the remaining safe time, ensuring that risk management has sufficient redundancy.

[0147] S45. When a risk is identified as Level 1, the proportional coefficients of various quantitative mapping rules shall be adjusted in advance based on the urgency of the remaining safety time, and the control benchmark shall be adjusted in advance before the risk actually occurs.

[0148] Specifically, Level 1 risk belongs to the risk incubation stage. By fine-tuning the proportional coefficient of the quantitative mapping rule, the intensity of grouting control is appropriately strengthened, and the risk is mitigated without affecting the normal construction rhythm. The revised mapping rule takes effect immediately and acts simultaneously on the dynamic control logic of S3 to achieve a closed loop of pre-control of prediction, correction and control.

[0149] The magnitude of the correction is positively correlated with the urgency of the risk: the shorter the remaining safe time, the greater the magnitude of the correction. The magnitude of a single correction shall not exceed 20% of the original proportional coefficient to avoid control oscillations caused by sudden parameter changes. The corrected benchmark value must be verified by environmental constraints simultaneously. If it exceeds the environmental constraint red line, the constraint limit value shall be used as the standard, and the level 2 risk disposal shall be automatically triggered.

[0150] For example, the specific execution actions are set as follows: reduce the grouting pressure of the entire section by 10%, reduce the jacking speed from the benchmark value by 15%, increase the frequency of environmental monitoring to once every 30 seconds, and push early warning information to the on-site duty terminal.

[0151] S46. When the risk is determined to be level 2, the risk management is carried out in a tiered manner based on the calculated probability of the blockage breaking down. The priority of the management is higher than the baseline pre-correction and all conventional control logic.

[0152] Specifically, Level 2 risk is considered a high-risk state. At this time, emergency response measures should be implemented immediately, and the instructions should directly cover all routine control and benchmark correction actions. The response measures include: suspending grouting operations on the high-risk side, initiating low-pressure reinforcement grouting around the tunnel, and setting the grouting pressure to 50% of the benchmark value; simultaneously reducing the pressure of the slurry chamber to the lower limit of the design value, suspending the jacking operation, and maintaining a stable slurry balance; triggering the on-site audible and visual alarm, pushing emergency alarm information to the terminals of the project manager and technical manager, and automatically retrieving the preset emergency grouting plan.

[0153] Once the risk has fallen back to a safe range, the parameters should be gradually adjusted in the order of first restoring the jacking speed and then restoring the grouting pressure, with an interval of no less than 10 minutes between each adjustment to avoid secondary disturbances caused by sudden parameter changes.

[0154] S47. In the next acquisition cycle, compare the predicted value of the remaining effective thickness of the sealing body with the recalculated measured value and calculate the prediction deviation. When the prediction deviation exceeds the preset confidence interval for a consecutive preset number of times, it is determined that the parameters of the multi-field coupled damage evolution model of the sealing body are mismatched, and the damage coefficient and coupling amplification coefficient of the multi-field coupled damage evolution model of the sealing body are corrected in real time.

[0155] Specifically, this step is a model self-calibration mechanism within a single project. It dynamically optimizes the model parameters of the multi-field coupled damage evolution model of the sealing body by predicting deviations, so that the model can continuously adapt to the current strata and construction conditions and ensure prediction accuracy. This step is a real-time micro-calibration. The calibration parameters only apply to the current project and do not directly update the project knowledge base, thus forming a hierarchical distinction from the full-cycle large calibration of S5.

[0156] In the description of this invention, when the prediction deviation exceeds a preset confidence interval for a consecutive preset number of times, it is determined to be a parameter mismatch in the multi-field coupled damage evolution model of the occlusion body. Real-time correction of the damage coefficient and coupling amplification coefficient of the multi-field coupled damage evolution model of the occlusion body includes:

[0157] S471. Starting from the current collection period, take a number of consecutive preset prediction deviation values ​​from the previous collection period to form a statistical sample, calculate the mean and standard deviation of the statistical sample, and construct the prediction deviation confidence interval for the current collection period based on the calculation results of the mean and standard deviation.

[0158] Specifically, the default sample size is 10 consecutive periods of prediction bias; the confidence interval is constructed at ±2 standard deviations, corresponding to a 95% confidence level; if the current prediction bias falls outside the confidence interval, it indicates that the prediction result has deviated significantly.

[0159] S472. Compare the prediction deviation of the current acquisition cycle with the prediction deviation confidence interval. If the prediction deviation is higher than the upper limit of the confidence interval for a certain number of consecutive preset times or lower than the lower limit of the confidence interval for a certain number of consecutive preset times, it is determined that the parameters of the multi-field coupled damage evolution model of the occlusion body are mismatched.

[0160] Specifically, the default number of consecutive preset times is set to 3 to avoid triggering incorrect corrections due to a single random deviation; a continuous high number of times indicates that the model is too conservative and the prediction risk is higher than the actual risk; a continuous low number of times indicates that the model is too aggressive and the prediction risk is lower than the actual risk, and parameter correction is required in both cases.

[0161] S473. When parameter mismatch is determined, the model parameters of the multi-field coupled damage evolution model of the plugging body are corrected according to the sign and degree of prediction deviation. The model parameters include the weakening coefficient of slurry pressure on the plugging body, the damage coefficient of the corrective lateral extrusion on the plugging body, the coupling amplification coefficient, and the time constant of the slurry strength growth curve.

[0162] Specifically, the correction direction is determined by the sign of the prediction deviation: if the predicted value is consistently too high, the damage type coefficient is lowered and the intensity growth coefficient is raised; if the predicted value is consistently too low, the damage type coefficient is raised and the intensity growth coefficient is lowered. The correction step size is proportional to the absolute value of the prediction deviation, and the single correction step size shall not exceed 10% of the current parameter value to avoid excessive single correction causing model oscillation.

[0163] S474. The corrected model parameters will be continuously monitored for prediction deviation over a preset number of future data collection periods. If the prediction deviation falls back to the prediction deviation confidence interval and remains within the preset number of data collection periods, the correction will be confirmed as effective.

[0164] Specifically, the verification period is set to 10 acquisition cycles by default. If the deviation remains stable within the confidence interval during the verification period, the corrected parameters are retained. If the deviation still exceeds the interval, the above correction process is repeated until the model accuracy meets the requirements.

[0165] S5. Integrate the operational data and monitoring feedback data accumulated throughout the entire cycle of grouting collaborative control and correction treatment, quantify the control effect from multiple dimensions, and reverse-calibrate the benchmark parameters of the graded control benchmark and the multi-field coupled damage evolution model of the sealing body based on the review results. Then, the iteratively optimized calibration results are deposited into the engineering knowledge base for the cold start of the initial benchmark of similar projects, so as to realize closed-loop control.

[0166] Specifically, this step is a full-cycle iterative optimization layer, which is executed after the initial section (first 50m) of construction is completed. It evaluates the control effect through quantitative review, reverse-calibrates the control benchmark and benchmark parameters, and structures and precipitates the optimized parameters to realize the reusability and continuous accumulation of engineering experience, so that the control accuracy increases spirally with the number of projects. This step is a full-cycle large-scale calibration, and the optimization results are stored in the engineering knowledge base, which can be used for the cold start of new projects under the same conditions.

[0167] The quantitative review was conducted from three dimensions: control precision, linkage effectiveness, and risk prevention and control. The control precision dimension included the deviation rate of grouting parameters and the adaptability of grout absorption rate to assess the compatibility between the benchmark parameters and the formation. The linkage effectiveness dimension included the linkage response delay and the sealing effect improvement rate to assess the actual effect of multi-system linkage. The risk prevention and control dimension included the trend prediction hit rate and false alarm rate to assess the accuracy of advance prediction.

[0168] Benchmark calibration employs deterministic rules, calibrating explicit controllable benchmark parameters and reference parameters. The process is transparent and interpretable, allowing engineers to directly verify and intervene. After calibration, parameters must be backtested using historical data to ensure effective optimization. The engineering knowledge base uses structured entry storage, supporting new projects to quickly recall initial benchmarks through similarity matching, enabling cold starts.

[0169] In the description of this invention, operational data and monitoring feedback data accumulated throughout the entire cycle of grouting collaborative management and corrective treatment are integrated to quantify the management effect from multiple dimensions. Based on the review results, the benchmark parameters of the graded management benchmark and the multi-field coupled damage evolution model of the sealing body are calibrated in reverse. The iteratively optimized calibration results are then stored in the engineering knowledge base for cold start of the initial benchmark of similar projects, realizing closed-loop management including:

[0170] S51. Collect the control and execution data and monitoring feedback data of the entire cycle of the initial stage, and quantify the evaluation indicators from three dimensions: control accuracy, linkage effectiveness and risk prevention and control, and form an effect evaluation report.

[0171] Specifically, the control accuracy dimension indicators include: parameter deviation rate = |actual operating average value - initial baseline value| / initial baseline value × 100%, and the average deviation rate and maximum deviation rate are calculated separately; slurry absorption rate adaptability = actual average slurry absorption rate / baseline slurry absorption rate × 100%, the ideal value is 100%, and the greater the deviation, the worse the baseline adaptability.

[0172] Effectiveness metrics for linkage: Linkage response delay = the time difference between the moment the trigger condition is met and the moment the control command is executed, and the average delay and maximum delay are calculated; Blocking effect improvement rate = (leakage of the tunnel portal before grouting - leakage of the tunnel portal after grouting) / leakage of the tunnel portal before grouting × 100%, which quantifies the actual effect of linkage grouting.

[0173] Risk control indicators: Prediction accuracy = Number of times a risk actually occurred and was predicted in advance / Total number of times a risk occurred × 100%; False alarm rate = Number of times a risk was predicted but did not actually occur / Total number of predictions × 100%.

[0174] S52. Based on the review results, the baseline parameters of the hierarchical control benchmark, the quantitative mapping parameters, and the multi-field coupled damage evolution model of the closure body are corrected in reverse according to deterministic rules to complete the full-cycle closed-loop optimization.

[0175] Specifically, for the calibration of the control benchmark for the reinforced area, for example, if the grout absorption rate of a certain reinforced area is greater than 120%, it means that the actual grout absorption capacity is significantly higher than expected. The benchmark grouting volume per hole in that area will be increased by 5%, and the upper limit of the grouting pressure will be increased by 3% accordingly. If the compatibility is less than 80%, the benchmark grouting volume per hole will be decreased by 5%, and the upper limit of the grouting pressure will be decreased by 3% accordingly. The adjustment range at one time shall not exceed 10% to avoid excessive fluctuations in control.

[0176] For the calibration of quantitative mapping relationships, the actual effect of parameter adjustment during the statistical linkage process is calculated. For example, if the actual change in the slurry absorption rate corresponding to a 10% pressure adjustment is less than 8%, it indicates that the adjustment step size is too small. The unit adjustment step size of the first proportional relationship is increased by 20%. If the change is greater than 12%, the step size is decreased by 20%. The other two sets of mapping rules are calibrated according to the same logic.

[0177] For the calibration of the multi-field coupled damage evolution model of the closure body and its trend prediction logic, the fitting weight allocation and sliding window length are optimized based on the prediction hit rate and false alarm rate. If the false alarm rate is too high, the weight of settlement and thickness factors is increased, and the sliding window is extended. If the hit rate is too low, the weight of construction parameters is increased to improve the sensitivity to construction disturbances. The baseline damage coefficient and coupling amplification coefficient of the multi-field coupled damage evolution model of the closure body are optimized simultaneously. After calibration, the model needs to be backtested with historical data to ensure improved accuracy.

[0178] S53. The iteratively optimized control parameters, the baseline parameters of the multi-field coupled damage evolution model of the sealing body, and the corresponding working condition features are stored in a structured manner. The cold start of new projects is achieved through similarity matching, and the knowledge base is dynamically updated according to the time-weighted strategy.

[0179] Specifically, each knowledge base entry contains three types of structured information: working condition tags, including stratum type, end reinforcement method, sensitive structure level, pipe jacking machine type, and tunnel overburden thickness; control parameter set, including graded control benchmarks, three sets of quantitative mapping relationship parameters, trend prediction parameters (fitting window length and slope judgment threshold of S44), and benchmark parameters of the multi-field coupled damage evolution model of the sealing body; and effect score, including control accuracy score, linkage effect score, and risk prevention and control score, with higher comprehensive scores indicating higher priority.

[0180] After a new project inputs its working condition tags, the system uses a cosine similarity algorithm to calculate the similarity with historical entries. Entries with a similarity ≥ 85% are selected, and an initial control benchmark is calculated using a weighted average of comprehensive scores. This benchmark is directly used for pre-construction benchmark calibration, eliminating the need for starting from scratch and enabling a cold start for the project. For entries with the same working condition, a time-weighted update strategy is employed, with the latest project data accounting for 60% of the weight and historical data accounting for 40%, ensuring continuous optimization of the knowledge base as projects accumulate, and a spiral improvement in control capabilities.

[0181] Please see Figure 2 The present invention also provides a multi-parameter linkage closed-loop control system for grouting of yellow mud at the beginning of pipe jacking, the system comprising:

[0182] The control area division module 1 is used to divide the reinforced area into multi-level control areas according to the intensity distribution based on the existing engineering basic parameters, set the corresponding hierarchical control benchmarks, and establish quantitative mapping rules for the linkage of multiple sets of parameters.

[0183] Data acquisition and processing module 2 is used to synchronously acquire multi-source operating data across engineering equipment based on a unified data interaction interface and time series benchmark, and complete standardized cleaning to form a standardized dataset with unified time series.

[0184] The collaborative dynamic adjustment module 3 is used to perform adaptive adjustment of grouting parameters adapted to the reinforcement zone, coordinated control of mud and water and grouting coupling, and coordinated control of correction action and grouting replenishment according to preset priority, based on the hierarchical control benchmark and quantitative mapping rules and combined with the operation data in the standardized dataset, thereby completing the coordinated control of grouting.

[0185] The multi-level risk response module 4 is used to assess the integrity of the sealing body in real time during the grouting collaborative management process through a pre-constructed multi-field coupled damage evolution model of the sealing body, identify the failure precursor characteristics of the sealing body and calculate the remaining safe time, and perform tiered pre-correction of quantitative mapping rules and tiered risk disposal.

[0186] The feedback closed-loop calibration module 5 is used to integrate the operational data and monitoring feedback data accumulated throughout the entire cycle of grouting collaborative control and correction treatment. It quantifies the control effect from multiple dimensions, reverse-calibrates the benchmark parameters of the graded control benchmark and the multi-field coupled damage evolution model of the sealing body based on the review results, and deposits the iteratively optimized calibration results into the engineering knowledge base for the cold start of the initial benchmark of similar projects, so as to realize closed-loop control.

[0187] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0188] This embodiment is applied to the 7B# to 14# pipe jacking section of the 110 kV and above transmission line relocation project at Zengcheng South Station of a certain city power supply bureau. Figure 2 and Figure 4 As shown. This section is a 110kV power cable tunnel. The newly built pipe jacking tunnel starts from working shaft 7B# and ends at working shaft 14#, with a length of approximately 220.8m. It is constructed using a slurry-balanced rock pipe jacking machine. The tunnel overburden thickness is 10m to 13m. The pipe section lining has an outer diameter of 4.14m, an inner diameter of 3.5m, a thickness of 0.32m, and a ring width of 2.5m. The minimum curve radius in the plane is 800m, and the maximum design longitudinal slope is 5‰. The clear dimensions inside the starting working shaft 7B# are 12.5m (length) × 8.9m (width) × 16.775m (height), and the outer diameter of the steel ring at the starting portal is 4468mm. The starting end is pre-reinforced with horizontal high-pressure jet grouting piles, with the reinforced area measuring 6m (length) × 8.14m (width) × 8.14m (height). Figure 6 As shown, the soil reinforcement at the end of the working well adopts a combination of double-pipe jet grouting piles and reinforced concrete cast-in-place piles, and the jet grouting holes and inspection holes are arranged according to the design spacing.

[0189] There are two adjacent 110kV high-voltage power poles along this section: the 110kV Lidong A / B line #001 pole is located at 3DLK+062.908, with a minimum horizontal distance of 4.80m from the outer edge of the tunnel, and its base is 1.73m higher than the tunnel top surface; the 110kV Lisan line #001 pole is located at 3DLK+079.700, with a minimum horizontal distance of 3.91m from the outer edge of the tunnel, and its base is 0.75m higher than the tunnel top surface. Both poles are energized and the design allows for foundation settlement of no more than 3mm, classifying them as extremely sensitive structures.

[0190] Before construction, according to the engineering geological survey report, the soil layers within the tunnel axis area, from top to bottom, consisted of miscellaneous fill, silty clay, and completely weathered granite. Based on the core sampling report of the pre-reinforced end area, core sampling points were arranged in a 1.5m × 1.5m grid. The structure of the end reinforcement area is as follows: Figure 3 As shown, from top to bottom, the structure consists of a reinforced concrete retaining wall, a sloped excavation face, a jet grouting pile reinforcement zone, and cast-in-place piles. Core sampling points are arranged in a grid within the jet grouting pile reinforcement zone. The unconfined compressive strength test results for the core samples range from 0.6 MPa to 2.3 MPa. A three-dimensional reinforcement strength distribution field with a 0.5 m × 0.5 m grid is generated using ordinary Kriging interpolation. The reinforcement zone is divided into three areas based on strength thresholds of 1.5 MPa and 1.0 MPa: a strong reinforcement zone (strength ≥ 1.5 MPa, approximately 25%), a normal reinforcement zone (strength 1.0-1.5 MPa, approximately 55%), and a weak overlap zone (strength < 1.0 MPa, approximately 20%). The planar coordinates of all grouting holes are superimposed on the reinforcement strength distribution field, and the grade of each grouting hole is pre-marked. Based on the established grouting construction plan (single-hole design grouting pressure benchmark value 0.25MPa, single-hole design grouting volume benchmark value 2.5m³, and sealing body design bearing threshold 0.45MPa), a graded control benchmark is established for three types of reinforcement zones: For the strong reinforcement zone, the upper limit of grouting pressure is 85% of the benchmark value, and the single-hole benchmark grouting volume is 90% of the benchmark value; for the weak overlap zone, the upper limit of grouting pressure is 90% of the benchmark value, and the single-hole benchmark grouting volume is 120% of the benchmark value; the normal reinforcement zone follows the benchmark values. The lower limit of grouting pressure for all zones is uniformly set at 60% of their respective upper limits.

[0191] During construction, the edge data gateway was deployed inside the 7B# starting working shaft. The grouting control system uses the Modbus-RTU protocol to connect the grouting pump frequency converter, pressure transmitter, and electromagnetic flowmeter, with a data acquisition frequency of 0.1Hz. The pipe jacking machine PLC control system uses the OPC UA protocol to connect to the main controller, subscribing to variables such as cutterhead torque, jacking speed, and the stroke of four sets of correction cylinders, with a subscription period of 100ms. The slurry conveying control system uses the Modbus-RTU protocol to connect the inlet and outlet slurry pump frequency converters and the slurry silo pressure transmitter, with a data acquisition frequency of 0.1Hz. The environmental monitoring system uses the MQTT protocol to connect the hydrostatic level and tilt sensor, with a data acquisition frequency of once every 10 minutes. All systems are synchronized via NTP time synchronization, with a timestamp error not exceeding 100ms.

[0192] When grouting operations enter the weak overlap zone, the system calls the benchmark parameters of the hole's grade, calculates the grout absorption rate deviation in real time, and automatically adjusts the grouting pressure according to the first proportional relationship. The adjusted target pressure is then sent to the grouting pump frequency converter for execution after boundary verification. When the slurry chamber pressure continuously reaches 75% of the design bearing threshold of the sealing body, the system automatically triggers the first-level linkage, reduces the grout pump frequency, and starts the slurry pressure closed-loop control; when the pressure continuously reaches 85%, the second-level linkage is triggered, and skip-hole grouting is started at the hole perimeter. When the cutterhead torque continuously reaches 80% of the rated value, the system automatically extends the single-hole grouting interval and reduces the grouting pump standby pressure to allow the grout sufficient initial setting time. When the stroke difference on one side of the correction cylinder continuously exceeds 60mm, the system automatically opens the grouting hole on the correction direction side for grouting.

[0193] During the jacking process, the system continuously performs integrity assessments of the sealing body. Taking a typical monitoring moment as an example: the current slurry chamber pressure is 0.30 MPa, the correction stroke difference is 45 mm, and 2.5 hours have passed since grouting. Substituting these parameters into the multi-field coupled damage evolution model of the sealing body, the calculated remaining effective thickness of the sealing body is 0.283 m, the breakdown probability is 0, and the sealing body is in a safe state. The health index is 2.95. A negative first-order rate of change indicates continuous deterioration of the sealing body, while a positive second-order rate of change indicates an accelerating rate of deterioration. However, since the health index has not consistently fallen below 1.0, the failure precursor judgment conditions are not met, and the system only records monitoring data without triggering intervention. If, at some future moment, the health index drops below 1.0 and all three failure precursor conditions are met, the system automatically calculates the remaining safe time and performs pre-emptive corrections or emergency measures according to the urgency of the remaining safe time.

[0194] After the initial 50m section of construction was completed, the system automatically collected full-cycle data and performed a review and calibration. The grout absorption rate of each reinforced zone was within a reasonable range, the linkage response delay met the engineering requirements, and the trend prediction effect was good. Based on the review results, the system performed reverse calibration on the graded control benchmark and quantitative mapping parameters according to deterministic rules. After the calibrated parameters were reviewed and confirmed, they were stored in the engineering knowledge base as the initial benchmark source for subsequent projects with similar strata and reinforcement methods.

[0195] The above embodiments are only used to illustrate the working principle and specific implementation of the present invention, and the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0196] In summary, by utilizing the above-mentioned technical solutions of this invention, the present invention breaks through the industry's conventional understanding of treating the sealing body as a static structure. By establishing a multi-field coupled damage evolution model for the sealing body, it incorporates three types of effects—mud-water seepage, correction compression, and grout strength growth—into a unified evaluation framework, enabling real-time calculation of the remaining effective thickness and breakdown probability of the sealing body on a cycle-by-cycle basis. Simultaneously, using hierarchical control benchmarks and quantitative mapping rules as a link, it organically links the control logic of the grouting system, mud-water system, and jacking system, upgrading each system from independent operation to collaborative control. This solves the fundamental problem of uncontrollable coupling risks caused by fragmented multi-system control at the system level.

[0197] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

Claims

1. A multi-parameter linkage closed-loop control method for grouting in yellow mud at the start of pipe jacking, characterized in that, include: S1. Based on existing engineering parameters, the reinforced area is divided into multi-level control areas according to the intensity distribution, and corresponding hierarchical control benchmarks are set to establish quantitative mapping rules for multi-set parameter linkage. S2. Based on a unified data interaction interface and time series benchmark, multi-source operation data across engineering equipment is collected synchronously and standardized cleaning is performed to form a standardized dataset with unified time series. S3. Based on the hierarchical control benchmark and quantitative mapping rules, combined with the operational data in the standardized dataset, the adaptive adjustment of grouting parameters, the coordinated control of mud and water coupling and grouting, and the coordinated control of correction action and grouting are executed sequentially according to the preset priority to complete the coordinated control of grouting. S4. During the grouting collaborative management process, the integrity of the sealing body is evaluated in real time through a pre-constructed multi-field coupled damage evolution model of the sealing body, the failure precursor characteristics of the sealing body are identified and the remaining safe time is calculated, and the quantitative mapping rule pre-correction and graded risk disposal are performed in stages. S5. Integrate the operational data and monitoring feedback data accumulated throughout the entire cycle of grouting collaborative control and correction treatment, quantify the control effect from multiple dimensions, and reverse-calibrate the benchmark parameters of the graded control benchmark and the multi-field coupled damage evolution model of the sealing body based on the review results. Then, the iteratively optimized calibration results are deposited into the engineering knowledge base for the cold start of the initial benchmark of similar projects, so as to realize closed-loop control.

2. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 1, characterized in that, Based on existing engineering parameters, the reinforced area is divided into multi-level control zones according to intensity distribution, and corresponding hierarchical control benchmarks are set. A quantitative mapping rule for multi-parameter linkage is established, including: S11. Collect and unify the stratigraphic and engineering foundation parameters, and store them in the benchmark database after structured verification, as the input basis for hierarchical control and quantitative mapping rule calibration; the engineering foundation parameters include reinforcement data, equipment parameters, environmental constraint data and predetermined grouting construction parameters; S12. Based on the core sample strength test results of the pre-reinforced end, a continuous reinforcement strength distribution field is generated by spatial interpolation. The initial end reinforcement area is divided into multi-level control areas according to the reinforcement strength effect, and the partition matching and labeling of all grouting holes is completed. The control areas include strong reinforcement area, normal reinforcement area and weak overlap area. S13. Using the design value of the predetermined grouting construction parameters as the global benchmark value, the control threshold of the divided multi-level control areas is adjusted proportionally to generate the control benchmark corresponding to each reinforcement area; the control benchmark includes the upper and lower limits of grouting pressure, the benchmark grouting volume of a single hole, and the benchmark value of grout absorption rate. S14. Based on the multi-level control area and the corresponding control benchmark, establish multiple sets of quantitative mapping rules solidified in the form of linear functions; the quantitative mapping rules are respectively used to correlate grout absorption rate with grouting pressure, mud-water pressure with grouting intensity, and correction stroke with grouting volume.

3. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 2, characterized in that, The establishment of multiple sets of quantization mapping rules, solidified in the form of linear functions, based on multi-level control areas and corresponding control benchmarks includes: S141. Using the benchmark grout absorption rate corresponding to each reinforcement zone as a reference, pre-calibrate the first proportional relationship between the deviation of the grout absorption rate and the adjustment amount of the grouting pressure; when the real-time grout absorption rate is higher than the benchmark grout absorption rate of the corresponding control area, adjust the grouting pressure according to the first proportional relationship; when the real-time grout absorption rate is lower than the benchmark grout absorption rate of the corresponding control area, adjust the grouting pressure according to the first proportional relationship. S142. Using the design value of slurry pressure as a reference, pre-calibrate the second proportional relationship between the increase in slurry pressure and the increase in grouting volume per hole around the tunnel; determine the increase in grouting volume per hole around the tunnel according to the second proportional relationship based on the increase in real-time slurry pressure relative to the design value. S143. Pre-calibrate the third proportional relationship between the single-sided stroke difference of the correction cylinder and the incremental amount of slurry added to a single hole on the correction side; determine the incremental amount of slurry added to a single hole on the correction side according to the third proportional relationship based on the real-time detected single-sided stroke difference. S144. Generate a quantitative mapping rule table by setting the judgment thresholds and adjustment ratios for the established first proportional relationship, second proportional relationship, and third proportional relationship.

4. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 1, characterized in that, The dynamic adjustment of grouting parameters, based on hierarchical control benchmarks and quantitative mapping rules, and combined with operational data in a standardized dataset, is performed sequentially according to preset priorities. This includes adaptive adjustment of grouting parameters to suit the reinforcement zone, coupled and coordinated control of mud-water and grouting, and coordinated control of correction actions and grout replenishment. S31. Using the graded control benchmark as a constraint and the current cumulative grouting volume and real-time grouting pressure of the grouting hole as inputs, dynamically match the control benchmark according to the reinforcement zone to which the current grouting hole belongs, and adaptively adjust the grouting pressure in real time according to the mapping relationship between grout absorption rate and grouting pressure. S32. With the design bearing threshold of the sealing body and the rated torque of the cutterhead as constraints, and with the real-time pressure of the mud and water chamber and the real-time torque of the cutterhead as inputs, the two-way linkage control of the mud and water system and the grouting control is performed based on the mapping relationship between mud and water pressure and grouting intensity. S33. Using the stroke values ​​of multiple sets of correction cylinders as input, the collaborative control of attitude deviation and grouting is achieved based on the mapping relationship between correction stroke and grouting amount.

5. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 1, characterized in that, The process involves real-time assessment of the integrity of the sealing body using a pre-constructed multi-field coupled damage evolution model, identification of pre-failure characteristics, calculation of remaining safe time, and tiered execution of quantitative mapping rule pre-correction and tiered risk management, including: S41. Based on the multi-field coupling effect of mud-water seepage, correction extrusion and slurry strength growth, a multi-field coupling damage evolution model of the sealing body is established. S42. Based on the preset acquisition cycle, the real-time pressure of the mud and water chamber, the real-time stroke difference of the correction cylinder and the grouting completion timestamp are converted into the time after grouting and substituted into the multi-field coupled damage evolution model of the sealing body to calculate the remaining effective thickness of the sealing body at the current moment, and calculate the probability of sealing body breakdown based on the ratio of the remaining effective thickness to the safe thickness threshold. S43. Based on the remaining effective thickness of the plugging body calculated cycle by cycle, construct a health index of the plugging body and solve its first-order rate of change and second-order rate of change. When the three indicators simultaneously meet the preset instability criteria, it is determined that the plugging body has a failure precursor and the risk level is automatically upgraded by one level. S44. Based on the time series of the remaining effective thickness of the plug body obtained by cycle-by-cycle calculation, perform trend fitting, calculate the remaining safe time of the plug body, and classify the breakdown risk level according to the duration interval of the remaining safe time. S45. When a risk is identified as Level 1, adjust the proportional coefficients of various quantitative mapping rules in advance based on the urgency of the remaining safety time, and adjust the control benchmark in advance before the risk actually occurs. S46. When the risk is determined to be level 2, the risk management is carried out in a graded manner based on the calculated probability of the current blockage body breaking down, and the management priority is higher than the baseline pre-correction and all conventional control logic. S47. In the next acquisition cycle, compare the predicted value of the remaining effective thickness of the plug with the recalculated measured value and calculate the prediction deviation. When the prediction deviation exceeds the preset confidence interval within a preset number of consecutive times, it is determined that the parameters of the multi-field coupled damage evolution model of the plug are mismatched, and the damage coefficient and coupling amplification coefficient of the multi-field coupled damage evolution model of the plug are corrected in real time.

6. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 5, characterized in that, The multi-field coupled damage evolution model of the sealing body, based on the multi-field coupling effects of mud-water seepage, correction compression, and slurry strength growth, includes: S411. Obtain the initial geometric parameters and material parameters of the sealing body used to seal the circumferential gap between the steel ring of the tunnel portal and the stratum; the geometric parameters include the thickness of the sealing body and the spatial range of the sealing body, and the material parameters include the design impermeability strength and design shear strength after the grout has initially set; S412. Establish the correspondence between mud and water pressure and the expansion rate of seepage channels inside the sealing body; S413. Establish the correspondence between the stroke difference of the correction cylinder and the lateral extrusion strain of the sealing body; S414. Establish the relationship between the time elapsed after grouting and the growth rate of the strength of the sealing material; S415. The remaining effective thickness of the plugging body is equal to the initial thickness minus the seepage equivalent thickness attenuation, minus the product of the compression equivalent thickness reduction and the coupling amplification factor, and then added to the strength equivalent thickness enhancement to construct a multi-field coupled damage evolution model of the plugging body.

7. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 5, characterized in that, Based on the remaining effective thickness of the plugging body calculated cycle by cycle, a health index for the plugging body is constructed, and its first-order and second-order rates of change are solved. When all three indicators simultaneously meet the preset instability criteria, it is determined that the plugging body has shown signs of failure, and the risk level is automatically upgraded by one level, including: S431. The ratio of the remaining effective thickness of the closure body at the current moment to the preset safe thickness threshold is defined as the health index of the closure body. S432. Based on the health index of the continuous acquisition cycle, the first-order rate of change of the health index is calculated by the difference method to reflect the deterioration rate of the integrity of the blockage. S433. Based on the first-order rate of change of the continuous acquisition cycle, the second-order rate of change of the health index is calculated using the difference method to reflect the changing trend of the deterioration rate of the occlusion body. S434. When all failure determination conditions are met simultaneously, it is determined that the sealing body has shown signs of failure. The failure determination conditions include: the health index is continuously less than the preset health threshold, the first-order rate of change is less than zero, and the second-order rate of change is greater than zero.

8. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 5, characterized in that, The remaining effective thickness time series of the plugging body, calculated on a cycle-by-cycle basis, is trend-fitted to calculate the remaining safe time of the plugging body. The breakdown risk level is then classified according to the duration interval of the remaining safe time, including: S441. Arrange the remaining effective thickness values ​​of the sealing body calculated in each acquisition cycle according to the acquisition time to construct a time series of the remaining effective thickness of the sealing body; S442. Using the acquisition time of the time series of the remaining effective thickness of the sealing body as the independent variable and the remaining effective thickness of the sealing body as the dependent variable, the first-order linear least squares method is used to perform trend fitting. The slope and intercept of the time series are obtained by fitting to determine the direction of change of the integrity of the sealing body. S443. When the slope obtained from the fitting is less than zero, the difference between the current remaining effective thickness and the safety thickness threshold is divided by the absolute value of the slope to calculate the remaining safe time of the sealing body; when the slope is greater than or equal to zero, the remaining safe time is recorded as infinity. S444. Based on the preset risk level mapping table, divide the corresponding breakdown risk level according to the remaining safe time.

9. The multi-parameter linkage closed-loop control method for grouting in the starting stage of pipe jacking according to claim 5, characterized in that, When the prediction deviation exceeds a preset confidence interval for a consecutive preset number of times, it is determined to be a parameter mismatch in the multi-field coupled damage evolution model of the closure body. Real-time correction of the damage coefficient and coupling amplification coefficient of the multi-field coupled damage evolution model of the closure body includes: S471. Starting from the current acquisition period, take a preset number of prediction deviation values ​​from the previous acquisition period to form a statistical sample, calculate the mean and standard deviation of the statistical sample, and construct the prediction deviation confidence interval for the current acquisition period based on the calculation results of the mean and standard deviation. S472. Compare the prediction deviation of the current acquisition cycle with the prediction deviation confidence interval. If the prediction deviation is higher than the upper limit of the confidence interval for a consecutive preset number of times or lower than the lower limit of the confidence interval for a consecutive preset number of times, it is determined that the parameters of the multi-field coupled damage evolution model of the occlusion body are mismatched. S473. When parameter mismatch is determined, the model parameters of the multi-field coupled damage evolution model of the plugging body are corrected according to the sign and degree of the prediction deviation. The model parameters include the weakening coefficient of mud-water pressure on the plugging body, the damage coefficient of the corrective lateral extrusion on the plugging body, the coupling amplification coefficient, and the time constant of the slurry strength growth curve. S474. The corrected model parameters will be continuously monitored for prediction deviation over a preset number of future data collection periods. If the prediction deviation falls back to the prediction deviation confidence interval and remains within the preset number of data collection periods, the correction will be confirmed as effective.

10. A multi-parameter linkage closed-loop control system for grouting in jacking-starting yellow mud, used to implement the multi-parameter linkage closed-loop control method for grouting in jacking-starting yellow mud as described in any one of claims 1-9, characterized in that, The system includes: The control area division module is used to divide the reinforced area into multi-level control areas according to the intensity distribution based on existing engineering parameters, set corresponding hierarchical control benchmarks, and establish quantitative mapping rules for multi-set parameter linkage. The data acquisition and processing module is used to synchronously acquire multi-source operating data across engineering equipment based on a unified data interaction interface and time series benchmark, and to perform standardized cleaning to form a standardized dataset with a unified time series. The collaborative dynamic adjustment module is used to perform adaptive adjustment of grouting parameters adapted to the reinforcement zone, coordinated control of mud and water and grouting coupling, and coordinated control of correction action and grouting replenishment according to preset priority, based on hierarchical control benchmarks and quantitative mapping rules, combined with the operational data in the standardized dataset, and to complete the coordinated control of grouting. The multi-level risk response module is used to assess the integrity of the sealing body in real time during the grouting collaborative management process through a pre-built multi-field coupled damage evolution model of the sealing body, identify the failure precursor characteristics of the sealing body and calculate the remaining safe time, and perform tiered pre-correction of quantitative mapping rules and tiered risk disposal. The feedback closed-loop calibration module is used to integrate the operational data and monitoring feedback data accumulated throughout the entire cycle of grouting collaborative management and correction treatment. It quantifies and reviews the management effect from multiple dimensions, and reverse-calibrates the benchmark parameters of the graded management benchmark and the multi-field coupled damage evolution model of the sealing body based on the review results. The iteratively optimized calibration results are then stored in the engineering knowledge base for cold start of the initial benchmark of similar projects, thus realizing closed-loop management.

Images