Foundation pit excavation micro-disturbance construction control system adjacent to existing subway tunnel

Abstract

The present application belongs to the technical field of foundation pit engineering construction, and provides a foundation pit excavation micro-disturbance construction control system adjacent to an existing subway tunnel, comprising: a frictional static pressure calibration module, a temperature drift zero drift compensation module, a splitting characteristic extraction module, a filtering separation estimation module and a comparison output control module. By calibrating the pipeline friction and hydrostatic pressure function, compensating for temperature drift and zero offset, extracting the soil splitting pressure characteristic value, and using Kalman filtering to separate the pure diffusion resistance and the excess pore water pressure dissipation curve, the diffusion resistance is finally compared with the allowable threshold value at each time, and the control instruction is output to adjust the grouting parameters when the limit is exceeded. The present application realizes multi-source decoupling and closed-loop control of the grouting pressure composite signal, effectively improving the micro-disturbance control accuracy and reliability of the grouting construction adjacent to the subway tunnel.

Description

Technical Field

[0001] This invention belongs to the field of foundation pit engineering construction technology, specifically a micro-disturbance construction control system for foundation pit excavation adjacent to existing subway tunnels. Background Technology

[0002] In the micro-disturbance construction control process of foundation pit excavation adjacent to existing subway tunnels, grouting is a key technical means to compensate for soil loss and control tunnel deformation. Current engineering practice typically involves monitoring grouting pressure by installing pressure sensors at the grouting pump outlet, borehole opening, or bottom, and adjusting the grouting rate and final pressure threshold in real time based on the pressure readings. However, the pressure signal actually obtained is not pure soil response data. On the one hand, the signal is mixed with high-frequency pulsations generated by the grouting pump itself, pipeline friction, hydrostatic pressure, and systematic errors caused by sensor temperature changes and zero-point drift. On the other hand, the grouting process is accompanied by the generation and dissipation of excess pore water pressure, as well as pressure abrupt changes when the soil is fractured. These multi-source physical factors superimpose each other, resulting in a complex composite signal from the original pressure signal collected by the sensors. Existing technologies struggle to effectively separate the effective components that truly reflect grout diffusion resistance and soil stability from this composite signal, thus failing to accurately determine whether the grouting process causes excessive disturbance to the adjacent subway tunnel, affecting the reliability and accuracy of micro-disturbance construction control.

[0003] To this end, the present invention provides a micro-disturbance construction control system for foundation pit excavation adjacent to existing subway tunnels. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0005] The technical solution adopted by this invention to solve its technical problem is: a micro-disturbance construction control system for foundation pit excavation adjacent to an existing subway tunnel, comprising the following modules: Friction and static pressure calibration module: Pressure sensors are installed at the grouting pump outlet, orifice and bottom of the hole to analyze the pressure difference between the outlet and the bottom of the hole and to establish the pipeline friction function and the hydrostatic pressure function. Temperature drift and zero drift compensation module: During the grouting stop period, the pressure and corresponding temperature of each sensor are continuously recorded, and the temperature drift coefficient and zero point offset are obtained by linear regression analysis. Splitting feature extraction module: Perform wavelet packet transform on the original pressure signal during the grouting stage to separate the low-frequency trend component, extract the slope change points of the low-frequency trend component, and obtain the soil splitting pressure feature value. The filtering and separation estimation module uses the pipeline friction function, hydrostatic pressure function, temperature drift coefficient and zero offset, and soil splitting pressure characteristic value as a priori parameters. It uses Kalman filtering to perform state estimation on the original pressure signal and outputs the pure diffusion resistance and excess pore water pressure dissipation curves at each time step. The comparison output control module compares the diffusion resistance with the allowable threshold hourly. If the limit is exceeded, the diffusion resistance is output as a grouting control command, and the excess pore water pressure dissipation curve is output simultaneously for formation stability judgment. If the limit is not exceeded, the diffusion resistance is output as a monitoring signal.

[0006] As a further aspect of the present invention: in the frictional static pressure calibration module, the pressure sensor is arranged as follows: The sensor at the outlet of the grouting pump is installed on a straight pipe section at a distance of not less than 0.5 meters from the grout outlet of the pump body; The sensor at the orifice is installed at the front end of the orifice sealing device via a tee connector; The sensor at the bottom of the hole is a miniature flat diaphragm pressure sensor, which is installed in the side hole of the pipe wall within 0.3 meters from the bottom of the hole at the front end of the grouting pipe, with the outer shell flush with the outer wall of the pipe; The three sensor signal lines are connected to the same multi-channel dynamic data acquisition instrument, triggered by the same clock source, with a sampling frequency of not less than 100 Hz and a time synchronization error of less than 1 millisecond.

[0007] As a further aspect of the present invention: in the frictional static pressure calibration module, the process of establishing the hydrostatic pressure function is as follows: Turn off the grouting pump to fill the pipeline with static grout, and record the elevation difference and pressure difference between the outlet and the bottom sensor. Change the liquid level in the hole and record at least three sets of hydrostatic pressure differences corresponding to different elevation differences; use linear regression to fit the function relationship between hydrostatic pressure difference and elevation difference, where the slope corresponds to the product of slurry effective density and gravitational acceleration, and the intercept is the fitting constant.

[0008] As a further aspect of the present invention: in the frictional static pressure calibration module, the process of establishing the pipeline frictional function is as follows: Start the grouting pump and perform grouting at different flow rates. At the same time, record the outlet pressure, the bottom pressure of the hole, and the static pressure difference under the current elevation difference. Calculate the total pressure loss at each flow rate and subtract the static pressure difference to obtain the pipeline friction. Least square fitting was performed on friction and flow data points at at least five different flow rates using a quadratic polynomial model, where the coefficients of the quadratic term represent friction along the flow path and the coefficients of the linear term represent local resistance and laminar flow contribution.

[0009] As a further aspect of the present invention: in the temperature drift and zero drift compensation module, the process of continuously recording the pressure and corresponding temperature of each sensor during the grouting stop period is as follows: Select a period when grouting has stopped and the pipeline is filled with static grout, and continuously record for no less than two hours, simultaneously collecting the pressure readings of each pressure sensor and the temperature at its installation location at a sampling frequency of no less than 0.5 Hz; The pressure and temperature at the same moment are paired; a univariate linear regression is performed independently on each sensor, with temperature as the independent variable and pressure as the dependent variable, to obtain the temperature drift coefficient. Extrapolate the regression line to the reference temperature, calculate the pressure value at the reference temperature, and determine the zero-point offset.

[0010] As a further aspect of the present invention: in the splitting feature extraction module, the process of separating the low-frequency trend component by performing wavelet packet transform on the original pressure signal during the grouting stage is as follows: Select the Daubechies wavelet function, determine the number of decomposition levels based on the sampling frequency and the upper limit of the target trend frequency band, and perform complete wavelet packet decomposition. Calculate the frequency range of each node after wavelet packet decomposition, find the node covering the lowest frequency band, retain the coefficients of the node and set the coefficients of other nodes to zero, and then perform wavelet packet reconstruction to obtain the low-frequency trend component; the frequency range of each node is divided into equal-width frequency bands according to the sampling frequency and the number of decomposition layers.

[0011] As a further aspect of the present invention: in the splitting feature extraction module, the process of extracting the slope abrupt change points of the low-frequency trend component to obtain the soil splitting pressure feature value is as follows: The central difference method is used to numerically differ the low-frequency trend components to obtain the slope sequence. The zero-crossing point in the slope sequence that changes from positive to negative is found. The change in slope before and after the zero-crossing point is calculated. If the change exceeds the set threshold, the zero-crossing point is determined to be the moment when the split occurs. Record the low-frequency trend component pressure value corresponding to the moment of splitting, which is the characteristic value of soil splitting pressure.

[0012] As a further aspect of the present invention: in the filtering separation estimation module, the process of outputting the pure diffusion resistance and excess pore water pressure dissipation curves at each time step is as follows: The total pressure at the bottom of the orifice is considered as the superposition of diffusion resistance and excess pore water pressure. The outlet pressure is converted into the total pressure at the bottom of the orifice using the pipe friction function and hydrostatic pressure function. Then, temperature compensation and zero-point correction are performed using the temperature drift coefficient and zero-point offset to obtain the Kalman filtered observation value. The diffusion resistance and excess pore water pressure are set as two state variables, and initial values ​​are assigned to the state variables. At each sampling time, the state prediction, Kalman gain calculation, state estimation updated with observations, and error covariance are performed in sequence. The updated two state variables are then output as the pure diffusion resistance and excess pore water pressure at the current time.

[0013] As a further aspect of the present invention: in the filtering and separation estimation module, the specific method for converting the outlet pressure into the total pressure at the bottom of the orifice and performing temperature compensation is as follows: Real-time data collection of grouting pump outlet pressure, grouting flow rate, and temperature at each sensor location; The pressure loss is calculated based on the flow rate using the pipeline friction function, and the gravity pressure is calculated based on the elevation difference using the hydrostatic pressure function. The total pressure at the bottom of the hole is obtained by subtracting the pipeline friction loss and hydrostatic pressure from the outlet pressure. The total pressure at the bottom of the borehole is corrected using the temperature drift coefficient and the zero-point offset. Specifically, the temperature drift coefficient multiplied by the difference between the current temperature and the reference temperature is subtracted, and then the zero-point offset is subtracted to obtain the Kalman filter observation.

[0014] As a further aspect of the present invention: in the filtering separation estimation module, the initial values ​​of the state variables and noise parameters of the Kalman filter are set as follows: The initial value of the pure diffusion resistance is taken as the bottom hydrostatic pressure calculated by the hydrostatic pressure function before grouting begins, and the initial value of the excess pore water pressure is taken as zero. If the characteristic value of the soil splitting pressure has been obtained, the initial value of the excess pore water pressure at the moment of splitting is set as the characteristic value of the splitting pressure minus the stable diffusion resistance before splitting. The variance of the observed noise is taken as the average variance of the fluctuation of the observed values ​​during the grouting stop period. The process noise covariance matrix is ​​set according to the permeability and consolidation characteristics of the soil.

[0015] The beneficial effects of this invention are as follows: First, the friction function and hydrostatic pressure function of the pipeline are established through the friction and static pressure calibration module, eliminating the systematic deviations of pipeline friction loss and gravity static pressure. Second, the temperature drift and zero drift compensation module uses linear regression analysis of the grouting stop time to obtain the temperature drift coefficient and zero-point offset, eliminating the influence of ambient temperature and sensor drift on the measured values. Third, the splitting feature extraction module uses wavelet packet transform to automatically identify the splitting time and characteristic pressure value of the soil, overcoming the subjectivity and lag of manual experience judgment. Fourth, the filtering separation estimation module uses calibration parameters... Using data as a priori, Kalman filtering is employed to separate the pure diffusion resistance and excess pore water pressure dissipation curves time-by-time, solving the problem of indistinguishable coupling of multiple sources. Finally, the comparison output control module compares the diffusion resistance with the allowable threshold in real time. When the threshold is exceeded, a step-wise pressure correction is automatically executed, while the pore water dissipation rate is monitored. This constructs a "perception-judgment-execution" closed loop, which actively limits the excessive growth of diffusion resistance while ensuring the grouting effect, preventing severe ground disturbance from threatening the safety of adjacent subway tunnels. This significantly improves the accuracy and automation level of micro-disturbance construction control. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1This is a flowchart of the micro-disturbance construction control system for excavation of a foundation pit adjacent to an existing subway tunnel, as described in an embodiment of the present invention. Figure 2 This is a flowchart illustrating the steps for implementing micro-disturbance control in the micro-disturbance construction control system for excavation of a foundation pit adjacent to an existing subway tunnel, as described in an embodiment of the present invention. Detailed Implementation

[0018] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0019] Example: Please refer to Figure 1-2 As shown in the embodiment of the present invention, the micro-disturbance construction control system for excavation of foundation pit adjacent to an existing subway tunnel includes the following modules: Friction and static pressure calibration module: Pressure sensors are installed at the grouting pump outlet, orifice and bottom of the hole to analyze the pressure difference between the outlet and the bottom of the hole and to establish the pipeline friction function and the hydrostatic pressure function. In the frictional static pressure calibration module, pressure sensors are installed at the grouting pump outlet, orifice, and bottom of the borehole as follows: Grouting pump outlet layout: On the straight pipe section where the grouting pump outlet connects to the grouting pipe, at a distance of not less than 0.5m from the grout outlet of the pump body, a pressure measuring hole is opened in the pipe wall. A diaphragm pressure sensor is installed using threaded or flanged connection. The pressure sensing surface is in direct contact with the grout in the pipe. The measuring range is 1.5 times the maximum grouting pressure. Orifice layout: At the orifice where the grouting pipe enters the formation (i.e., at the front end of the orifice sealing device), install a three-way connector. One way leads to the grouting pipe inside the hole, and the other way is used to install a pressure sensor. The sensor is installed in the same way as the outlet, ensuring that the measuring point is located between the orifice valve and the pipe inside the hole. Bottom hole layout: A lateral hole is opened on the pipe wall at the front end of the grouting pipe (within about 0.3m from the bottom of the grouting hole), and a miniature flat diaphragm pressure sensor is installed. The sensor shell is flush with the outer wall of the grouting pipe, and the signal line is led out along the inner wall of the grouting pipe and connected to the acquisition instrument through the hole sealing device. The three sensor signal lines are connected to the same multi-channel dynamic data acquisition instrument, triggered by the same clock source, with a sampling frequency of not less than 100Hz, ensuring that the time synchronization error is less than 1ms; In the friction and static pressure calibration module, the process of analyzing the pressure difference between the outlet and the bottom of the orifice and establishing the pipeline friction function and the hydrostatic pressure function is as follows: Establish the hydrostatic pressure function: A1. Turn off the grouting pump, fill the pipeline with static grout, and record the elevation difference between the outlet sensor and the bottom sensor. (Unit: meters), and simultaneously read the outlet pressure. With the pressure at the bottom of the hole ; A2, Calculate the hydrostatic pressure difference:

[0020] A3. Change the liquid level in the hole (e.g., by lifting the grouting pipe or injecting / extracting a small amount of grout), and repeatedly record multiple sets (at least 3 sets) of different elevation differences. Corresponding hydrostatic pressure difference ; A4, using linear regression to analyze different elevation differences Corresponding hydrostatic pressure difference By fitting the data, the hydrostatic pressure function is obtained:

[0021] in, The effective density of the slurry obtained by fitting (unit: kg / m³) 3 ), Take 9.8 m / s 2 , for intercept; Establish the pipeline friction function: B1, Start the grouting pump at different flow rates (Unit: L / min) Grouting is performed, and the flow rate is measured by an electromagnetic flow meter or turbine flow meter installed at the pump outlet; B2, for each traffic Simultaneously record export pressure With the pressure at the bottom of the hole At the same time, based on the current elevation difference Calculate hydrostatic pressure difference ; B3, Calculate the total pressure loss under the flow rate:

[0022] B4, subtracting the hydrostatic pressure portion, yields the pipe friction:

[0023] B5, for multiple (no fewer than 5) different traffic conditions , The data points are fitted using least squares. According to fluid mechanics, frictional resistance is usually proportional to the square of the flow rate; therefore, a quadratic polynomial model is used.

[0024] in, The friction coefficient is the coefficient of friction along the path. This is a contribution term for local resistance and potential laminar flow. Understandably, the significance of the friction and static pressure calibration module lies in the following: by using pressure sensors deployed at the grouting pump outlet, orifice, and bottom of the hole, and combining the pressure difference analysis under different flow rates and static conditions, the module establishes the pipeline friction function and the hydrostatic pressure function. The friction and static pressure calibration module solves the systematic deviation problem in the grouting pressure signal caused by pipeline friction loss and gravity static pressure, providing a quantitative basis for accurately converting the pump outlet pressure into the actual bottom pressure. The calibrated function is the basis for eliminating pipeline characteristic interference, ensuring the physical authenticity of all subsequent pressure signals, and is a prerequisite for achieving accurate sensing in micro-disturbance control. Temperature drift and zero drift compensation module: During the grouting stop period, the pressure and corresponding temperature of each sensor are continuously recorded, and the temperature drift coefficient and zero point offset are obtained by linear regression analysis. In the temperature drift and zero drift compensation module, the process of continuously recording the pressure and corresponding temperature of each sensor during the grouting stop period is as follows: Select a time period when grouting has stopped and the pipeline is filled with static grout (such as during nighttime shutdowns or grout replacement intervals). At this time, there is no pump pressure pulsation or grout flow, and the sensor is only affected by changes in ambient temperature and its own zero-point offset. The continuous recording time should be no less than 2 hours, covering a sufficiently large temperature variation range. The following data were collected synchronously at a sampling frequency of not less than 0.5Hz: Pressure readings of each pressure sensor (outlet, orifice, bottom of orifice), and temperature at the installation location of each pressure sensor; The pressure and temperature data from the same sensor at the same time were compared. ,in, The sampling time sequence number; Establish a linear regression model of pressure versus temperature:

[0025] in, Temperature drift coefficient (unit: kPa / ℃). For each sensor, a univariate linear regression is performed independently, including the constant term encompassing zero-point offset and initial static pressure, with temperature as the metric. As the independent variable, pressure For the dependent variable, the least squares method is used to solve. and ; The solution obtained This is the temperature drift coefficient of the sensor, which is used to extrapolate the linear regression model of pressure versus temperature to the reference temperature. (e.g., 20℃), calculate the reference temperature. The pressure value below:

[0026] The zero-point offset is then defined as:

[0027] in, The zero-point pressure that should theoretically be displayed at the reference temperature (usually taken as 0 kPa, i.e., the theoretical reading without external pressure). If the sensor cannot obtain absolute zero pressure after installation, the zero-point offset can be defined relatively as:

[0028] That is, the intercept value of the regression line at the reference temperature. When using it later, the zero point is zeroed out by subtracting this value. The final output for each pressure sensor is the temperature drift coefficient. and zero offset ; Understandably, the significance of the temperature drift and zero drift compensation module lies in: continuously recording the pressure and temperature of each sensor during the grouting stop period, obtaining the temperature drift coefficient and zero-point offset through linear regression, thereby eliminating the influence of ambient temperature changes and the sensor's own zero-point drift on the measured values; the temperature drift and zero drift compensation module solves the systematic errors generated by the sensor with temperature and time, ensuring that the pressure readings remain consistent under different temperatures and operating conditions, and the compensated pressure signal eliminates non-physical disturbances, significantly improving the long-term stability and repeatability of the monitoring data, providing a clean pressure baseline for accurately judging the formation response; Splitting feature extraction module: Perform wavelet packet transform on the original pressure signal during the grouting stage to separate the low-frequency trend component, extract the slope change points of the low-frequency trend component, and obtain the soil splitting pressure feature value. In the splitting feature extraction module, the original pressure signal is obtained by continuously acquiring the bottom hole pressure sensor signal at a sampling frequency of not less than 100Hz during the grouting stage; In the fracture feature extraction module, the process of performing wavelet packet transform on the original pressure signal during the grouting stage to separate the low-frequency trend component is as follows: Choose the Daubechies wavelet function (such as db4 or db6), and set the decomposition level to 3-5 levels. If the level is too low, the low-frequency trend will still contain interference; if the level is too high, it will be over-smoothed and lose the details of abrupt changes. Generally, the sampling frequency should be used as the decomposition level. and the upper limit of the target trend band (Typically <0.5Hz) Determine: Number of decomposition layers For example, sampling frequency =100Hz, =0.5Hz, then calculate the number of decomposition layers. =6, but 4 to 5 layers are actually used; Complete wavelet packet decomposition yields 2L Calculate the frequency range of each node, identify the node covering the lowest frequency band, retain the coefficients of the node covering the lowest frequency band, and set the coefficients of the remaining nodes to zero. Then, perform wavelet packet reconstruction to obtain the low-frequency trend component. ; The method for calculating the frequency range of each node is as follows: Let the sampling frequency of the original signal be... (Unit: Hz), then the effective frequency range of the signal is 0~ / 2 (Nyquist frequency), after After complete wavelet packet decomposition, the signal is uniformly divided into 2... L Each frequency band has an equal width, and the width of each frequency band is:

[0029] Then the k-th node (k=0,1,2,…,2) L -1 The frequency range of ) is: ; In the splitting feature extraction module, the process of extracting the slope abrupt change points of the low-frequency trend component to obtain the characteristic value of soil splitting pressure is as follows: Central difference method for low-frequency trend components Numerical differencing yields the slope sequence: ; When soil splits, the diffusion resistance reaches its peak and then suddenly decreases, manifested as the slope rapidly changing from a positive value (pressure increase) to a negative value (pressure drop), that is, the slope crosses zero and the rate of change is large: Finding slope sequence The zero-crossing point is determined by the change in slope from positive to negative. The magnitude of the change in slope before and after the zero-crossing point is calculated. If the magnitude of the change exceeds a set threshold (e.g., 5 kPa / s), then the zero-crossing point is the moment when the splitting occurs. Record the pressure value at the moment of splitting. This is the characteristic value of soil splitting pressure; Understandably, the significance of the splitting feature extraction module lies in performing wavelet packet transform on the original bottom hole pressure signal during the grouting stage, extracting low-frequency trend components and detecting slope abrupt change points, thereby automatically identifying the time when soil splitting occurs and the corresponding pressure characteristic value. The splitting feature extraction module solves the problems of strong subjectivity and lag in manual experience judgment, and can quickly capture the key turning point of the formation from elastic compression to the generation of cracks during the grouting process. The obtained splitting pressure characteristic value is not only used for the initial state setting of Kalman filtering, but also provides a quantitative indicator for judging whether grouting has caused excessive disturbance to the formation. The filtering and separation estimation module uses the pipeline friction function, hydrostatic pressure function, temperature drift coefficient and zero offset, and soil splitting pressure characteristic value as a priori parameters. It uses Kalman filtering to perform state estimation on the original pressure signal and outputs the pure diffusion resistance and excess pore water pressure dissipation curves at each time step. In the filtering and separation estimation module, the process of outputting the pure diffusion resistance and excess pore water pressure dissipation curves at each time step is as follows: The total pressure at the bottom of the borehole is considered as the superposition of two components: one is the pure diffusion resistance (reflecting the actual diffusion capacity of the grout in the formation), and the other is the excess pore water pressure (reflecting the increase and dissipation of pore water pressure caused by grouting). Three data points are collected in real time: grouting pump outlet pressure, grouting flow rate, and temperature at each sensor. The pipeline friction function (which calculates pressure loss based on flow rate) and hydrostatic pressure function (which calculates gravity pressure based on elevation difference) obtained by the friction and static pressure calibration module are used to convert the outlet pressure into the total pressure at the bottom of the hole: Total pressure at the bottom of the hole = Outlet pressure - Pipeline friction loss - Hydrostatic pressure. The temperature drift coefficient (unit: kPa / ℃) and zero offset (unit: kPa) obtained by the temperature drift and zero drift compensation module are used to perform temperature compensation and zero-point correction on the total pressure at the bottom of the hole: Corrected bottom pressure = total pressure at the bottom of the hole - temperature drift coefficient × (current temperature - reference temperature) - zero offset; The corrected bottom pressure is the actual observed value obtained through Kalman filtering, denoted as Z(i), where i represents the i-th sampling time. Initialize the Kalman filter by setting initial estimates for two state variables: Initial value of pure diffusion resistance: Take the hydrostatic pressure at the bottom of the hole before grouting begins (calculated using the hydrostatic pressure function); Initial value of excess pore water pressure: set to 0 (assuming no excess pore water pressure before grouting). If the soil splitting pressure characteristic value (i.e., the pressure at the moment of splitting) has been obtained through the splitting feature extraction module, then at the moment of splitting, the initial estimated value of excess pore water pressure is set as the splitting pressure characteristic value minus the stable diffusion resistance before splitting (i.e., the pressure change). At the same time, the initial estimation error covariance matrix is ​​set (a 2×2 matrix, with smaller values ​​such as 0.01 and 0.1 on the diagonal and 0 on the off-diagonal). For each new sampling time i, perform the following five sub-steps (C1-C5): C1, Prediction: Based on the estimated value at the previous moment, predict the state at the current moment. Since it is assumed that the state changes slowly, the predicted value is directly equal to the estimated value at the previous moment (i.e., the state transition is an identity). The prediction error covariance is equal to the error covariance at the previous moment plus a process noise covariance matrix. The process noise covariance matrix is ​​set according to the permeability and consolidation characteristics of the soil, for example, 0.01 and 0.1 on the diagonal respectively, reflecting the expected uncertainty of the magnitude of state change. The setting basis is: based on the permeability classification in Appendix F of GB 50487-2008 "Code for Geological Investigation of Water Conservancy and Hydropower Engineering" and the permeability coefficient and consolidation coefficient measured in GB / T50123-2019 "Standard for Geotechnical Testing Methods", the process noise variances q1 (diffusion resistance) and q2 (excess pore water pressure) are set according to the following rules: Clay: Permeability coefficient ≤ 1×10 -6 cm / s, consolidation coefficient ≤1×10 -3 cm 2 / s, q1=0.005, q2=0.05; Silty clay: 1×10 -6 cm / s < permeability coefficient ≤ 1×10 -4 cm / s, q1=0.01, q2=0.1; Sand: 1×10 -4 cm / s < permeability coefficient, q1=0.02, q2=0.2; C2, Calculate the Kalman gain: The Kalman gain is a 2×1 vector used to balance prediction and actual observation. The Kalman gain depends on the prediction error covariance, the observation matrix (in this scheme, it is [1, 1]), and the observation noise variance. The observation noise variance is obtained by statistically analyzing the fluctuations of the observations during the grouting stop period (for example, taking the average of the variances of the observations within the period). The larger the gain, the more confident the current observation is; the smaller the gain, the more confident the prediction is. C3, Update the state estimate: Multiply the difference between the actual observation Z(i) and the predicted observation by the Kalman gain to correct the predicted state and obtain the optimal estimate at the current time. The corrected state estimate is: New state estimate = Predicted state + Kalman gain × (Actual observation - Predicted observation). C4, Update Error Covariance: Based on the Kalman gain and the observation matrix, update the estimated error covariance matrix for prediction at the next time step; C5, take the first state variable estimated at the current moment as the pure diffusion resistance and the second state variable as the excess pore water pressure, record the two outputs at each time step, and obtain the pure diffusion resistance curve and the excess pore water pressure dissipation curve. Understandably, the significance of the filtering and separation estimation module lies in the following: using the frictional hydrostatic pressure function, temperature drift zero-drift compensation parameters, and fracturing characteristic values ​​as priors, it employs Kalman filtering to estimate the state of the corrected bottom hole pressure, separating the pure diffusion resistance and excess pore water pressure dissipation curves at each time step. This module solves the problem of indistinguishable coupling of multiple components (diffusion resistance and pore pressure) in the measured pressure signal, achieving a physically clear state decoupling. The output results eliminate residual noise and dynamic interference, providing a high-fidelity decision-making basis for micro-perturbation control. The comparison output control module compares the diffusion resistance with the allowable threshold hourly. If the threshold is exceeded, the diffusion resistance is output as a grouting control command, and the excess pore water pressure dissipation curve is output simultaneously for formation stability judgment. If the threshold is not exceeded, the diffusion resistance is output as a monitoring signal. In the comparison output control module, the diffusion resistance is compared with the allowable threshold hourly: If the diffusion resistance is less than or equal to the allowable threshold, it is judged as not exceeding the limit; otherwise, if the diffusion resistance is greater than the allowable threshold, it is judged as exceeding the limit. If the limit is not exceeded, the current pure diffusion resistance is output as a monitoring signal. The signal is transmitted to the data acquisition and monitoring control system (SCADA) or grouting recorder at the construction site for real-time display, storage and trend analysis, but the operating parameters of the grouting pump are not actively interfered with. At the same time, the excess pore water pressure dissipation curve is also output as a monitoring signal to help engineers judge whether the formation pore pressure is dissipating normally. If the limit is exceeded, the current pure diffusion resistance is output as a grouting control command, which is directly sent to the frequency converter or proportional control valve of the grouting pump to reduce the grouting rate or pressure setpoint. A specific adjustment strategy could be: when the diffusion resistance exceeds the threshold, the command lowers the grouting pressure target value to 90% of the allowable threshold and gradually stabilizes it using a step-by-step pullback method, simultaneously outputting an excess pore water pressure dissipation curve for formation stability assessment. The system can automatically check whether the excess pore water pressure is within the normal dissipation range (e.g., a decrease rate ≥ 5% per minute). If the dissipation is too slow, an auxiliary warning message is issued, indicating a possible risk of formation blockage or over-grouting. For example, a tiered callback approach could be: Step 1: If the diffusion resistance is detected to exceed the allowable threshold, immediately reduce the grouting pressure target value from the current value to 90% of the allowable threshold; Step 2: Maintain this pressure for grouting for 10 seconds. If the diffusion resistance is still higher than the allowable threshold, reduce it to 80% of the allowable threshold; Step 3: Reduce the pressure by one level (each level by 5% to 10%) at short intervals (e.g., 5-10 seconds) until the diffusion resistance stabilizes below the allowable threshold. After the diffusion resistance has remained below the allowable threshold for a preset time (e.g., 30 seconds), return the pressure to the normal target pressure. It should be noted that the allowable threshold is obtained through historical construction data, that is: when grouting is carried out in non-protected areas (i.e. far away from the subway tunnel), the diffusion resistance value when the stratum shows obvious deformation or splitting is recorded, and 80% of it is taken as the allowable threshold. Understandably, the significance of the comparison output control module lies in: comparing the pure diffusion resistance obtained from filtering and separation with the allowable threshold hourly, and outputting monitoring signals or control commands according to whether the limit is exceeded. When the limit is not exceeded, only data is recorded and construction is not interfered with; when the limit is exceeded, a step-wise pressure correction is automatically executed. At the same time, the excess pore water pressure dissipation curve is used to assist in assessing the formation stability. The comparison output control module constructs a perception-judgment-execution closed loop, which actively limits the excessive growth of diffusion resistance while ensuring the grouting effect, prevents severe formation disturbance from threatening the safety of adjacent subway tunnels, and achieves true micro-disturbance construction control.

[0030] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A micro-disturbance construction control system for foundation pit excavation adjacent to an existing subway tunnel, characterized in that: Includes the following modules: Friction and static pressure calibration module: Pressure sensors are installed at the grouting pump outlet, orifice and bottom of the hole to analyze the pressure difference between the outlet and the bottom of the hole and to establish the pipeline friction function and the hydrostatic pressure function. Temperature drift and zero drift compensation module: During the grouting stop period, the pressure and corresponding temperature of each sensor are continuously recorded, and the temperature drift coefficient and zero point offset are obtained by linear regression analysis. Splitting feature extraction module: Perform wavelet packet transform on the original pressure signal during the grouting stage to separate the low-frequency trend component, extract the slope change points of the low-frequency trend component, and obtain the soil splitting pressure feature value. The filtering and separation estimation module uses the pipeline friction function, hydrostatic pressure function, temperature drift coefficient and zero offset, and soil splitting pressure characteristic value as a priori parameters. It uses Kalman filtering to perform state estimation on the original pressure signal and outputs the pure diffusion resistance and excess pore water pressure dissipation curves at each time step. The comparison output control module compares the diffusion resistance with the allowable threshold hourly. If the limit is exceeded, the diffusion resistance is output as a grouting control command, and the excess pore water pressure dissipation curve is output simultaneously for formation stability judgment. If the limit is not exceeded, the diffusion resistance is output as a monitoring signal.

2. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 1, characterized in that: In the frictional hydrostatic calibration module, the pressure sensor is arranged as follows: The sensor at the outlet of the grouting pump is installed on a straight pipe section at a distance of not less than 0.5 meters from the grout outlet of the pump body; The sensor at the orifice is installed at the front end of the orifice sealing device via a tee connector; The sensor at the bottom of the hole is a miniature flat diaphragm pressure sensor, which is installed in the side hole of the pipe wall within 0.3 meters from the bottom of the hole at the front end of the grouting pipe, with the outer shell flush with the outer wall of the pipe; The three sensor signal lines are connected to the same multi-channel dynamic data acquisition instrument, triggered by the same clock source, with a sampling frequency of not less than 100 Hz and a time synchronization error of less than 1 millisecond.

3. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 1, characterized in that: In the frictional static pressure calibration module, the process of establishing the hydrostatic pressure function is as follows: Turn off the grouting pump to fill the pipeline with static grout, and record the elevation difference and pressure difference between the outlet and the bottom sensor. Change the liquid level in the hole and record at least three sets of hydrostatic pressure differences corresponding to different elevation differences; use linear regression to fit the function relationship between hydrostatic pressure difference and elevation difference, where the slope corresponds to the product of slurry effective density and gravitational acceleration, and the intercept is the fitting constant.

4. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 3, characterized in that: In the friction and static pressure calibration module, the process of establishing the pipeline friction function is as follows: Start the grouting pump and perform grouting at different flow rates. At the same time, record the outlet pressure, the bottom pressure of the hole, and the static pressure difference under the current elevation difference. Calculate the total pressure loss at each flow rate and subtract the static pressure difference to obtain the pipeline friction. Least square fitting was performed on friction and flow data points at at least five different flow rates using a quadratic polynomial model, where the coefficients of the quadratic term represent friction along the flow path and the coefficients of the linear term represent local resistance and laminar flow contribution.

5. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 1, characterized in that: In the temperature drift and zero drift compensation module, the process of continuously recording the pressure and corresponding temperature of each sensor during the grouting stop period is as follows: Select a period when grouting has stopped and the pipeline is filled with static grout, and continuously record for no less than two hours, simultaneously collecting the pressure readings of each pressure sensor and the temperature at its installation location at a sampling frequency of no less than 0.5 Hz; The pressure and temperature at the same moment are paired; a univariate linear regression is performed independently on each sensor, with temperature as the independent variable and pressure as the dependent variable, to obtain the temperature drift coefficient. Extrapolate the regression line to the reference temperature, calculate the pressure value at the reference temperature, and determine the zero-point offset.

6. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 1, characterized in that: In the splitting feature extraction module, the process of separating the low-frequency trend component from the original pressure signal during the grouting stage using wavelet packet transform is as follows: Select the Daubechies wavelet function, determine the number of decomposition levels based on the sampling frequency and the upper limit of the target trend frequency band, and perform complete wavelet packet decomposition. Calculate the frequency range of each node after wavelet packet decomposition, find the node covering the lowest frequency band, retain the coefficients of the node and set the coefficients of other nodes to zero, and then perform wavelet packet reconstruction to obtain the low-frequency trend component; the frequency range of each node is divided into equal-width frequency bands according to the sampling frequency and the number of decomposition layers.

7. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 6, characterized in that: In the splitting feature extraction module, the process of extracting the slope abrupt change points of the low-frequency trend component to obtain the soil splitting pressure feature value is as follows: The central difference method is used to numerically differ the low-frequency trend components to obtain the slope sequence. The zero-crossing point in the slope sequence that changes from positive to negative is found. The change in slope before and after the zero-crossing point is calculated. If the change exceeds the set threshold, the zero-crossing point is determined to be the moment when the split occurs. Record the low-frequency trend component pressure value corresponding to the moment of splitting, which is the characteristic value of soil splitting pressure.

8. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 1, characterized in that: In the filtering and separation estimation module, the process of outputting the pure diffusion resistance and excess pore water pressure dissipation curves at each time step is as follows: The total pressure at the bottom of the orifice is considered as the superposition of diffusion resistance and excess pore water pressure. The outlet pressure is converted into the total pressure at the bottom of the orifice using the pipe friction function and hydrostatic pressure function. Then, temperature compensation and zero-point correction are performed using the temperature drift coefficient and zero-point offset to obtain the Kalman filtered observation value. Diffusion resistance and excess pore water pressure are set as two state variables, and initial values ​​are assigned to the state variables. At each sampling time, state prediction, Kalman gain calculation, state estimation updated with observations, and error covariance are performed in sequence. The updated two state variables are then output as the pure diffusion resistance and excess pore water pressure at the current time.

9. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels as described in claim 8, characterized in that: In the filtering and separation estimation module, the specific method for converting the outlet pressure into the total pressure at the bottom of the orifice and performing temperature compensation is as follows: Real-time data collection of grouting pump outlet pressure, grouting flow rate, and temperature at each sensor location; The pressure loss is calculated based on the flow rate using the pipeline friction function, and the gravity pressure is calculated based on the elevation difference using the hydrostatic pressure function. The total pressure at the bottom of the hole is obtained by subtracting the pipeline friction loss and hydrostatic pressure from the outlet pressure. The total pressure at the bottom of the borehole is corrected using the temperature drift coefficient and the zero-point offset. Specifically, the temperature drift coefficient multiplied by the difference between the current temperature and the reference temperature is subtracted, and then the zero-point offset is subtracted to obtain the Kalman filter observation.

10. The micro-disturbance construction control system for excavation of foundation pits adjacent to existing subway tunnels according to claim 9, characterized in that: In the filtering and separation estimation module, the initial values ​​of the state variables and noise parameters for the Kalman filter are set as follows: The initial value of the pure diffusion resistance is taken as the hydrostatic pressure at the bottom of the hole calculated by the hydrostatic pressure function before grouting begins, and the initial value of the excess pore water pressure is taken as zero. If the characteristic value of the soil splitting pressure has been obtained, the initial value of the excess pore water pressure at the moment of splitting is set as the characteristic value of the splitting pressure minus the stable diffusion resistance before splitting. The variance of the observed noise is the average variance of the fluctuations in the observed values ​​during the grouting stop period; The process noise covariance matrix is ​​set according to the permeability and consolidation characteristics of the soil.

Images