Building settlement control method based on recharge process

Abstract

The invention relates to a building settlement control method based on a recharge process, which comprises the following steps: S1, carrying out a single well water pumping test and a group well precipitation test, obtaining hydrogeological parameters of a fracture zone, determining the streaming range of a waterproof curtain of a foundation pit, and delimiting a recharge key control area; s2, a recharge unit is deployed, a deep recharge well is arranged along the periphery of the foundation pit, and a shallow recharge well is arranged on the outer side of the deep layer; and S3, during foundation pit dewatering, the water level outside the pit and building settlement are monitored in real time, and recharge parameters are adjusted through linkage of the automatic control unit. According to the method, through precise layout and dynamic regulation and control of the layered recharge well, multi-parameter monitoring and an LSTM pre-judgment model are combined, targeted water supplementation of deep and shallow stratum differentiation is achieved, and building settlement is effectively inhibited; the settlement risk is remarkably reduced by adjusting recharge parameters and an emergency response mechanism in real time, and the construction safety is improved; an automatic system and closed-loop control optimize resource utilization, both efficiency and environmental protection are considered, and the method is suitable for high-precision settlement prevention and control requirements under complex stratums.

Description

Technical Field

[0001] This invention relates to the field of building settlement control technology, and in particular to a building settlement control method based on a recharge process. Background Technology

[0002] During underground engineering construction, settlement control of surrounding buildings has always been one of the core challenges for engineering safety and environmental protection. With the expansion of urban underground space development, the impact of activities such as excavation and dewatering on ground disturbance is becoming increasingly significant, especially in areas with weak strata or fault fracture zones. A rapid drop in groundwater level can lead to soil instability, a sudden drop in pore water pressure, and consequently, uneven settlement or even structural damage to buildings. To mitigate this problem, recharge technology is widely used in underground engineering. Its core principle is to replenish groundwater to offset the drop in water level caused by dewatering operations and maintain effective stress balance in the soil. Traditional recharge technology typically relies on single-well pumping tests to obtain hydrogeological parameters and combines this with multi-well dewatering tests to simulate actual working conditions. However, its implementation is often limited to shallow strata recharge, lacking targeted design for recharge effects in deep fracture zones. In addition, existing reinjection systems mostly adopt fixed parameter operation mode, which makes it difficult to dynamically adjust the reinjection volume and pressure according to real-time monitoring data. This results in a mismatch between the recharge efficiency and the formation demand, making it impossible to achieve precise water replenishment for different formations at different depths.

[0003] CN113026704A discloses a soil settlement monitoring device for foundation pits, comprising a reference column perpendicular to the ground and a settlement monitoring block. The reference column has multiple pairs of electrical contacts along its height. Any pair of electrical contacts is connected to an external power source and an output device via a wire. The settlement monitoring block has conductive plates matching the electrical contacts. The conductive plates can move vertically within a groove in the reference column and connect the output device to the power source by contacting the electrical contacts. This soil settlement device can be installed in the protected area around the foundation pit and displays the soil settlement status in real time on an external output device in a visual manner, eliminating the need for periodic monitoring of soil settlement. The device automatically triggers an alarm when the settlement reaches an alarm height. The invention also discloses an automatic recharge system that automatically activates when the foundation pit soil settlement monitoring device alarms, recharges water back to the protected area, and prevents disasters.

[0004] Although existing technologies have attempted to improve control effectiveness through methods such as layered deployment of reinjection wells and optimized well spacing, many technical bottlenecks remain to be addressed. For example, in complex strata where fault fracture zones coexist with shallow clay, deep and shallow reinjection units are prone to mutual interference due to differences in hydraulic connectivity, leading to overlapping or omissions in recharge ranges, making it difficult to achieve coordinated control of high-permeability fracture zones and low-permeability clay. Simultaneously, existing monitoring systems primarily focus on acquiring single parameters such as water level and settlement, lacking real-time feedback on formation mechanical parameters (such as elastic modulus and pore water pressure). This results in settlement trend prediction lagging behind actual changes, and control measures often only respond passively after settlement occurs, hindering proactive intervention. Furthermore, for the dynamic control of reinjection parameters, existing methods largely rely on empirical formulas or linear models, making it difficult to accurately quantify the nonlinear relationship between precipitation depth, rate, and surrounding settlement. This leads to insufficient control precision, failing to meet the millimeter-level settlement control requirements for buildings in highly sensitive areas. The existence of these problems severely restricts the application effect of recharge technology under complex geological conditions, and there is an urgent need for a technical solution that can integrate multi-dimensional monitoring data and realize differentiated recharge and dynamic control between deep and shallow layers.

[0005] Furthermore, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making this invention, but due to space limitations, not all details and contents were listed in detail. However, this does not mean that the present invention does not possess the features of these prior art. On the contrary, the present invention already possesses all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a building settlement control method based on recharge technology, thereby solving at least some of the aforementioned technical problems. This invention effectively solves the challenge of building settlement control caused by hydraulic interference between deep and shallow strata during underground engineering construction by constructing a recharge unit based on multi-parameter monitoring and dynamic control. Addressing the problem that traditional recharge technologies often result in overlapping or missed recharge ranges in strata where fault fracture zones and shallow clay layers coexist, this invention uses systematic parameter acquisition through single-well pumping tests and group-well dewatering tests, combined with detailed stratigraphic exploration using ground-penetrating radar and seismic refraction wave methods, to clarify key mechanical parameters such as permeability coefficients and elastic moduli of the fracture zone and shallow strata. Based on this, a layered recharge well deployment strategy is adopted, with deep recharge wells deployed along the outer perimeter of the foundation pit and embedded in the core area of ​​the fracture zone, while shallow recharge wells are moved 2-3 meters outside the deep wells, forming a physical isolation between deep and shallow recharge units and avoiding disturbance of the shallow strata by high-pressure recharge water. By employing differentiated selection of steel and PVC pipes and graded wrapping technology for filters, deep recharge wells achieve efficient water replenishment in sandy gravel aquifers under pressures of 0.5–0.8 MPa, while shallow recharge wells precisely replenish clay pore water at low flow rates (0.1–0.3 MPa). The synergistic effect of these two methods can simultaneously restore the effective stress balance between deep and shallow formations. Furthermore, a multi-parameter monitoring network combining fiber optic strain sensors and pore water pressure gauges is introduced. Combined with an LSTM neural network model to model the nonlinear correlation between elastic modulus, pore water pressure, and settlement trend, a 6–12 hour forward-looking prediction of building settlement is achieved. When monitoring data shows that the settlement rate approaches 80% of the design limit, the system automatically triggers graded adjustment of recharge parameters: the deep recharge pressure is increased by 0.1 MPa to enhance fracture permeability, and the shallow recharge volume is dynamically adjusted by 5%–10% to maintain stable clay moisture content. Simultaneously, a closed-loop feedback mechanism is used to correct the adjustment amplitude in real time, ensuring dynamic matching between water replenishment efficiency and formation needs. This technical solution effectively suppresses interference between shallow and deep layers through the coupling of physical isolation design and intelligent prediction model, improves the accuracy of replenishment and the timeliness of regulation, achieves millimeter-level settlement control under complex geological conditions, and significantly reduces the risk of construction impacting the structural safety of surrounding buildings.

[0007] This invention discloses a method for controlling building settlement based on reinjection technology, which includes the following steps: S1. Conduct single-well pumping tests and group-well dewatering tests to obtain hydrogeological parameters of the fracture zone, clarify the flow range of the foundation pit water-stop curtain, and delineate the key control area for recharge. S2. Deploy the recharge unit, with deep recharge wells arranged around the outer perimeter of the foundation pit and shallow recharge wells arranged on the outer side of the deep layer. S3. During the dewatering of the foundation pit, the water level outside the pit and the settlement of the building are monitored in real time, and the recharge parameters are adjusted in conjunction with the automatic control unit.

[0008] According to a preferred embodiment, step S1 includes a test preparation stage, in which well locations are arranged based on the survey results. The well locations for single-well pumping tests are selected to represent the core area of ​​the fractured zone and the typical characteristics of shallow strata, while avoiding areas with dense underground pipelines and the influence range of building foundations. The well groups for group-well dewatering tests are evenly distributed around the perimeter of the planned excavation pit. During well formation, the well formation process is selected according to the geological characteristics. Spiral drilling is used for shallow clay and sand layers, and percussion drilling is used for gravel layers. After well formation, the well is washed using an air compressor until the water in the well is clear, and then pumping equipment and monitoring instruments are installed.

[0009] According to a preferred embodiment, the single-well pumping test includes two stages: steady-flow pumping and unsteady-flow pumping. In the steady-flow pumping stage, a submersible pump is started to pump water at a constant flow rate. Water level data and pumping volume are recorded intermittently for both the test well and the observation well. A stable state is determined when the difference in drawdown measured multiple times does not exceed a first drawdown difference threshold and the pumping volume fluctuation is less than a pumping volume fluctuation threshold. In the unsteady-flow pumping stage, after the steady-flow pumping ends, the pumping equipment remains in the same operating state, and water level changes over time are continuously recorded. The permeability coefficient of the fractured zone and shallow formation is calculated through the single-well pumping test. The Dubouy formula is used for the steady-flow pumping stage, and the Theis formula is used for the unsteady-flow pumping stage. The static water level, maximum drawdown, and radius of influence parameters of the test well are recorded.

[0010] According to a preferred embodiment, the group well dewatering test simulates the actual dewatering conditions during the excavation of the foundation pit. The group of wells is divided into several groups, and each group of wells pumps water simultaneously. During the test, water level monitoring points are set up around the foundation pit and within the affected area, and settlement monitoring points are set up near the foundations of surrounding buildings. The groundwater level distribution, pumping volume, and ground settlement data are recorded at intervals under different pumping durations. The group well dewatering test obtains data on the shape of the groundwater level drawdown funnel under the action of the well group, the distribution of water inflow from each well, and the flow around the water-stop curtain.

[0011] According to a preferred embodiment, step S1 includes a parameter analysis stage: data obtained from single-well pumping tests and group-well dewatering tests are organized and analyzed, and curves of groundwater level change with time and distance are plotted. Based on the geological distribution characteristics, parameters such as the permeability coefficient, radius of influence, static water level, unit inflow, and permeability coefficient and compressibility coefficient of shallow clay and sand layers are determined. Based on the above parameters, a correlation model of foundation pit dewatering and surrounding water level change is established. This correlation model uses foundation pit dewatering depth and dewatering rate as independent variables, and groundwater level drop and ground subsidence at a certain point in the surrounding area as dependent variables.

[0012] According to a preferred embodiment, step S2 includes a well location planning stage: the well location planning is carried out with the key control area for recharge as the core, combined with the excavation range of the foundation pit, the geological distribution characteristics and the location of surrounding buildings. First, the baseline of the outer perimeter of the foundation pit is determined, and the deep recharge wells are continuously laid out along the baseline of the outer perimeter of the foundation pit. The well spacing is set according to the differentiated level of the key control area for recharge. The shallow recharge wells are laid out parallel to the outside of the deep recharge wells, and the well spacing of the shallow recharge wells is consistent with the well spacing of the deep recharge wells in the corresponding area.

[0013] According to a preferred embodiment, step S2 includes an equipment installation stage: installing corresponding monitoring instruments, control valves, and connecting pipelines according to the reinjection pressure and flow control requirements of deep and shallow reinjection wells. Specifically, the equipment installation for deep reinjection wells includes: sequentially installing a pressure gauge, an electromagnetic flowmeter, and a shut-off valve on the reinjection pipeline at the wellhead; connecting all deep reinjection wellhead pipelines to the deep main reinjection pipeline via branch pipes; connecting the deep main reinjection pipeline to a high-pressure centrifugal pump, with a check valve installed at the outlet of the high-pressure centrifugal pump. The equipment installation for shallow reinjection wells includes: sequentially installing a turbine flowmeter and a ball valve on the reinjection pipeline at the wellhead; connecting all shallow reinjection wellhead pipelines to the shallow main reinjection pipeline via branch pipes; connecting the shallow main reinjection pipeline to a low-pressure centrifugal pump, with a check valve installed at the outlet of the low-pressure centrifugal pump.

[0014] According to a preferred embodiment, step S3 includes a monitoring unit collaborative operation phase, which includes external pit water level monitoring and building settlement monitoring. The external pit water level monitoring covers the area affected by recharge around the foundation pit. The monitoring points are deployed according to the principle of uniform distribution and key density: monitoring points are deployed at intervals in non-key control areas along the outer perimeter baseline of the foundation pit, and the monitoring points are deployed more densely in the fault fracture zone and the first-level key control area. All external pit water level monitoring points use submersible automatic water level gauges, and the water level monitoring data is uploaded to the automatic control unit in real time. Building settlement monitoring involves setting up monitoring points based on the structural characteristics and settlement-sensitive parts of surrounding buildings. Settlement monitoring points are set up at the four corners of the building walls, the middle of the wall base, and the mid-span of the foundation beam. Settlement monitoring uses electronic levels, and the monitoring frequency is dynamically adjusted according to the stage of foundation pit dewatering.

[0015] According to a preferred embodiment, step S3 further includes adding a function for monitoring formation mechanical parameters and predicting settlement trends, and constructing a closed-loop control system. The monitoring of formation mechanical parameters is achieved by a multi-parameter monitoring module to simultaneously acquire hydrological and mechanical dual parameters. The multi-parameter monitoring module includes a fiber optic strain sensor and a pore water pressure gauge that are communicatively connected to the automatic control unit. The settlement trend prediction is based on an LSTM neural network prediction model. The basic data required for model construction comes from the single-well pumping test and group-well dewatering test in step S1. The soil elastic modulus and pore water pressure data under different test stages and conditions are collected simultaneously. After preprocessing the collected raw data to remove abnormal data, it is divided into a training set and a validation set according to the proportion. The input layer neurons of the LSTM neural network prediction model correspond to the real-time monitored elastic modulus and pore water pressure.

[0016] According to a preferred embodiment, when executing the advance control logic, the closed-loop control system can automatically call the LSTM neural network prediction model in real time, input the elastic modulus and pore water pressure data collected in the current monitoring period into the model, obtain the predicted settlement trend values ​​of each building settlement monitoring point in the future, and compare them with the preset building settlement design limits; when the predicted future settlement trend value of a certain building settlement monitoring point will exceed the preset threshold, the advance control command is triggered, the automatic control unit determines the recharge control area corresponding to the settlement monitoring point and retrieves the current recharge parameters of the area. Attached Figure Description

[0017] Figure 1 This is a flowchart of the steps of a preferred embodiment of a building settlement control method provided by the present invention; Figure 2 This is a schematic diagram illustrating the implementation process of step S1 in a preferred embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the implementation process of step S2 in a preferred embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the implementation process of step S3 in a preferred embodiment of the present invention. Detailed Implementation

[0018] The following is a detailed explanation with reference to the accompanying drawings.

[0019] like Figure 1 As shown, this invention discloses a method for controlling building settlement based on recharge technology, which includes the following steps: S1. Conduct single-well pumping tests and group-well dewatering tests to obtain hydrogeological parameters of the fracture zone, clarify the flow range of the foundation pit water-stop curtain, and delineate the key control area for recharge. S2. Deploy the recharge unit, with deep recharge wells arranged around the outer perimeter of the foundation pit and shallow recharge wells arranged on the outer side of the deep layer. S3. During the dewatering of the foundation pit, the water level outside the pit and the settlement of the building are monitored in real time, and the recharge parameters are adjusted in conjunction with the automatic control unit.

[0020] Preferably, such as Figure 2 As shown, step S1 is the preliminary test stage, which aims to obtain key hydrogeological parameters of the fault fracture zone and surrounding strata through systematic pumping tests, clarify the flow range of the foundation pit cutoff curtain, and provide data support for the design and dynamic control of subsequent recharge units. Specifically, it includes stages such as test preparation, single-well pumping test, group well dewatering test, parameter analysis, and delineation of key control areas for recharge.

[0021] Preferably, a detailed site survey can be conducted during the test preparation stage to clarify the stratigraphic distribution characteristics, the orientation, distribution range, and depth of the fault fracture zone within the test area. Simultaneously, underground pipelines, structures, and other obstacles within the site should be identified to prevent damage during the test. For example, during the detailed site survey, a systematic exploration can be carried out using drilling and geophysical exploration methods: First, based on the geological survey data of the project area, the exploration scope is initially delineated. This scope can cover the proposed excavation pit and the surrounding area at least twice the excavation depth of the pit. Second, exploration points are laid out in a grid pattern, with increased density in areas where the fault fracture zone is presumed to be distributed and in areas with significant changes in shallow strata. The spacing between points in denser areas is 5-8 meters, and in non-dense areas, it is 10-15 meters. For drilling, rotary drilling technology is used, and the drilling depth must penetrate the fault fracture zone and reach at least 3 meters into the underlying stable strata. During drilling, core samples or soil samples were continuously collected, and the stratigraphic names, lithological characteristics (such as the degree of fracture development in the fracture zone, the properties of the infill material, the water content and void ratio of the shallow clay), layer thickness, and burial depth were recorded in segments according to depth. For geophysical exploration, ground-penetrating radar was used to scan the shallow strata, and the seismic refraction wave method was used to detect the strike, dip angle, and extension range of the fault fracture zone. The geophysical exploration results were cross-validated with the drilling data. Finally, all exploration data were integrated to draw engineering geological profile maps, fault fracture zone distribution maps, and shallow strata contour maps, thereby clarifying the stratigraphic distribution characteristics, specific parameters of the fault fracture zone, and physical and mechanical properties of the shallow strata in the test area.

[0022] Preferably, well locations can be arranged based on the exploration results. The well locations for single-well pumping tests can be selected at locations that represent the core area of ​​the fracture zone and the typical characteristics of shallow strata, preferably avoiding areas with dense underground pipelines and the influence range of building foundations. The location that represents the core area of ​​the fracture zone refers to the area with the largest fracture zone thickness, the most developed fractures, and the best connectivity, as determined by the previous site exploration. This area is usually characterized by a high degree of core fragmentation, sandy filling material, and water inrush or leakage during drilling. The location that represents the typical characteristics of shallow strata refers to the area where shallow clay is continuously distributed, has a stable thickness, and its physical and mechanical properties (such as water content and compressibility coefficient) are within the average range of this type of strata. This area should avoid areas with local disturbances (such as disturbances from existing building foundations or underground pipeline construction). The well group arrangement for group dewatering tests can simulate the actual dewatering conditions of the foundation pit. The well locations are evenly distributed along the outer perimeter of the foundation pit to be excavated. The well spacing can be determined according to the estimated influence radius, for example, it can be set to 8~15m, and it is ensured that the well group can fully cover the foundation pit and the surrounding affected area. During well construction, the construction techniques for single wells and well groups can be selected based on the formation characteristics. For shallow clay and sand layers, auger drilling can be used, while for gravel layers, percussion drilling can be used. Furthermore, PVC pipes can be used for shallow sections, while steel pipes can be used for deep, fractured zones. The well pipe diameter can be determined according to the test type. For example, the preferred diameter for a single-well pumping test is φ800mm, and the observation well diameter is φ400mm. For well group dewatering tests, the well diameter is uniformly φ400mm. The bottom of the well pipe must be embedded at least 1m into stable formation to ensure stability during pumping. After well completion, well washing can be carried out. The well is washed using an air compressor to remove sediment and mud until the water in the well is clear. After well washing, pumping equipment and monitoring instruments are installed. The pumping equipment can be a submersible pump, and its head and flow rate can be determined according to the estimated water inflow. The monitoring instruments include a water level gauge, a flow meter, and a settlement observation instrument, which are used to monitor changes in groundwater level, pumping volume, and surrounding ground settlement during the test.

[0023] Preferably, the single-well pumping test is divided into two stages: steady-flow pumping and unsteady-flow pumping. In the steady-flow pumping stage, the submersible pump is started to pump water at a constant flow rate. The initial flow rate can be set according to the well's output capacity. During pumping, the water level data and pumping volume of the test well and observation well are recorded at intervals. When the difference in water level drawdown measured multiple times (e.g., 3 times) does not exceed the first water level drawdown difference threshold (e.g., 0.05m) and the pumping volume fluctuation is less than the pumping volume fluctuation threshold (e.g., 5%), it is determined that a steady state has been reached. The duration of the steady state is not less than 8 hours. In the unsteady-flow pumping stage, after the steady-flow pumping ends, the pumping equipment is kept running unchanged, and the water level change data over time is continuously recorded for not less than 24 hours. If any abnormal situation such as a sudden rise or fall in water level occurs during this period, pumping can be suspended, and the well pipe can be checked for damage or blockage. After troubleshooting, the test can be restarted. The permeability coefficients of fractured zones and shallow formations can be calculated through single-well pumping tests. The Dubouy formula is used for steady-flow pumping, while the Theis formula is used for unsteady-flow pumping. Simultaneously, parameters such as the static water level, maximum drawdown, and radius of influence of the test well are recorded. The static water level refers to the water level elevation when the well is stable before the start of the pumping test. The maximum drawdown is the difference between the lowest water level observed during pumping and the static water level. The radius of influence is determined by the water level change data from observation wells; that is, when the drawdown of an observation well is less than 0.01m, the distance from that observation well to the test well is the radius of influence. Specifically, the Dubouy formula used in the steady-flow pumping stage can be selected according to the type of test well (unconfined or confined well), as follows: Unconfined well (aquifer is a phreatic layer): , Confined well (aquifer is a confined zone): .

[0024] Where Q is the inflow rate of the pumping well, K is the permeability coefficient, H is the thickness of the unconfined aquifer, M is the thickness of the confined aquifer, R is the radius of influence, r is the radius of the pumping well, h is the drawdown in the pumping well, and h1 is the vertical thickness from the water level in the unconfined well to the impermeable base after pumping stabilizes (satisfying h = H - h1). In the calculation, first determine the type of test well (unconfined well or confined well), then select the Q and h values ​​for the steady flow stage, and the H (or M) and r values ​​obtained from the previous exploration, combined with the R value obtained from the observation well, and substitute them into the formula to solve for the K value. The Theis formula used in the unsteady flow pumping stage is: s = (Q / (4πT))W(u), where s is the drawdown at the observation point, Q is the inflow rate of the pumping well, T is the hydraulic conductivity, W(u) is the Theis well function, and u = r 2S / (4Tt), where r is the distance from the observation point to the pumping well, S is the water storage coefficient, and t is the pumping time; during the calculation, the values ​​of s and t at different times in the unsteady flow stage and the known value of r are selected, and the correspondence between W(u) and u is determined by trial and error or table lookup method, and then T and S are solved, and the value of K is calculated according to T=KM (confined water) or T=KH (unconfined water).

[0025] Preferably, the group well dewatering test can simulate the actual dewatering conditions during the foundation pit excavation process. The group of wells is divided into several groups, and each group of wells pumps water synchronously. The pumping flow rate can be determined according to the designed dewatering depth of the foundation pit and the number of wells, ensuring that the groundwater level within the foundation pit area drops to the design required depth during the pumping process. During the test, water level monitoring points are set up around the foundation pit and within the affected area, with a spacing of 5-10m. In areas near the fracture zone, the spacing can be increased to 3-5m. At the same time, settlement monitoring points are set up near the foundations of surrounding buildings to record the groundwater level distribution, pumping volume, and ground settlement data at different pumping durations. The test duration is determined based on the stability of the water level. The test can be stopped when the groundwater level fluctuation is less than the groundwater level fluctuation threshold (e.g., 0.1m) and the ground settlement rate is less than the ground settlement rate threshold (e.g., 0.1mm / d) for 24 consecutive hours. By conducting a group well dewatering test, data such as the shape of the groundwater level drawdown funnel under the action of the well group, the distribution of water inflow from each well, and the flow around the cutoff curtain can be obtained. Among them, the shape of the groundwater level drawdown funnel refers to the curved shape data formed by the change of groundwater level with distance from the center of the foundation pit or pumping well under the action of group well dewatering. Its core is the correspondence between the groundwater level drawdown at different locations and the distance from that location to the dewatering center. The water inflow of each well refers to the pumping volume data of each pumping well per unit time in the group well dewatering test. The flow around the curtain is determined by analyzing the drawdown curve outside the curtain. When the difference between the drawdown at a point outside the curtain and the drawdown at the corresponding point inside the curtain is less than the second drawdown difference (such as 0.5m), that point is within the flow around the curtain, thus clarifying whether the flow around the curtain exceeds one times the excavation depth of the foundation pit.

[0026] Preferably, in the parameter analysis stage, data obtained from single-well pumping tests and group-well dewatering tests are organized and analyzed to plot groundwater level variation curves over time and distance. Combined with stratigraphic distribution characteristics, key parameters such as the permeability coefficient, radius of influence, static water level, and unit inflow of the fractured zone, as well as parameters such as the permeability coefficient and compressibility coefficient of shallow clay and sand layers, are determined. Based on these parameters, a correlation model between foundation pit dewatering and surrounding water level changes is established to reflect the quantitative relationship between different dewatering depths and rates and the decrease in surrounding groundwater level and ground subsidence. Coefficients in the model are determined through regression analysis and other methods to ensure that the model's fit is not less than 0.9. Furthermore, this correlation model is a multiple linear regression model, using foundation pit dewatering depth (H) and dewatering rate (v) as independent variables, and the decrease in groundwater level (Δh) and ground subsidence (Δs) at a surrounding point as dependent variables, to quantitatively describe the relationship between different dewatering parameters and surrounding hydrological and deformation responses. The model establishment process is as follows: First, multiple sets of data obtained from the well-based precipitation experiment are compiled. Each set of data includes precipitation depth, precipitation rate, water level drop at multiple surrounding monitoring points at the corresponding time, and ground subsidence. Second, the data is preprocessed to remove abnormal data caused by equipment failure or sudden interference, ensuring the validity of the data. Then, typical monitoring points (such as monitoring points 5m, 10m, and 15m away from the edge of the foundation pit) are selected to establish models. For each monitoring point, the water level drop Δh is used as the dependent variable, and H and v are used as independent variables. Linear regression is performed using the least squares method to obtain Δh = a1H + b1v + c1, where a1 and b1 are regression coefficients, and c1 is a constant term. Similarly, the ground subsidence Δs is used as the dependent variable to obtain Δs = a2H + b2v + c2, where a2 and b2 are regression coefficients, and c2 is a constant term. Finally, the significance of the model is verified through analysis of variance to ensure the model's coefficient of determination R. 2 The value should be no less than 0.9 to ensure the model's fitting accuracy. The quantitative relationships are specifically reflected through the regression equations mentioned above. For example, when the precipitation depth increases by 1m, the water level drop at a certain monitoring point increases by a1m, and the ground subsidence increases by a2mm; when the precipitation rate increases by 0.5m / d, the water level drop at that monitoring point increases by 0.5b1m, and the ground subsidence increases by 0.5b2mm. Through these quantitative relationships, the changes in surrounding water level and subsidence can be predicted based on actual precipitation parameters, providing a calculation basis for the dynamic control of subsequent recharge parameters.

[0027] Preferably, the key control area for recharge can be delineated by combining the bypass range of the cutoff curtain, the shape of the groundwater level drawdown funnel, and the distribution of surrounding buildings. The area within the bypass range where the water level drop exceeds the first drop (e.g., 2m) and the area where buildings are located within a preset range (e.g., 10m) from the edge of the foundation pit are designated as the first-level key control area. The area outside the bypass range but affected by the water level drawdown funnel and where the water level drop is between the first and second drop (e.g., 1~2m) is designated as the second-level control area. The remaining areas are general control areas. The arrangement of recharge wells can be densified in the first-level key control area to ensure effective water replenishment to the area.

[0028] Preferably, such as Figure 3 As shown, step S2 is the deployment stage of the recharge unit. Based on the hydrogeological parameters of the fault fracture zone, the division results of the key control area for recharge, and the correlation model of pit dewatering and surrounding water level changes obtained in the previous test stage, the step aims to complete the precise layout of deep and shallow recharge wells, standardized well construction, installation of supporting monitoring and control equipment, and system debugging. This ensures that the recharge unit can provide targeted water replenishment to the fault fracture zone (high permeability strata) and shallow clay (low permeability strata) respectively, and avoid hydraulic interference caused by deep and shallow recharge. This provides stable and reliable hardware support for subsequent dynamic operation and regulation. This stage may include well location planning, well construction technology, and installation of supporting equipment.

[0029] Preferably, the well location planning should focus on the key control area for recharge, taking into account the excavation range of the foundation pit, the geological distribution characteristics, and the location of surrounding buildings, and should be carried out according to the principles of on-demand layout and precise coverage. First, a baseline around the perimeter of the foundation pit should be determined. This baseline is parallel to the excavation edge of the foundation pit and is 1-2 meters away from the outer edge of the excavation edge. This distance can avoid disturbing the foundation pit support structure (such as water-stop curtain and piles) during the construction of recharge wells, while ensuring that the recharge range can effectively cover the area affected by the foundation pit dewatering. For deep recharge wells, they are continuously laid out along the baseline around the perimeter of the foundation pit. The well spacing is set differently according to the key control area level: In the first-level key control area (the area within the flow bypass, the water level drop exceeds 2m, and the area where buildings are located within 10m of the edge of the foundation pit), the well spacing is 3-5m. This spacing is determined based on the distribution law of the surface settlement funnel obtained from previous tests (the maximum longitudinal settlement point is about 10m from the foundation pit, and about 5m laterally). This allows the water replenishment range of the deep recharge wells to form an effective superposition in the settlement peak area, ensuring that the water loss from the fracture zone is fully replenished; In the second-level control area (the area outside the flow bypass but affected by the water level drop funnel, with a water level drop between 1-2m), the well spacing is 6-8m; In the general control area (other areas besides the first and second-level control areas), the well spacing is 8-10m. The differentiated spacing setting achieves efficient utilization of water replenishment resources and avoids resource waste. For shallow recharge wells, their layout can form a coordinated and non-interfering arrangement with deep recharge wells. Specifically, they are arranged parallel to the deep recharge wells at a distance of 2-3m outside the deep recharge wells. This distance was determined through previous hydraulic interference simulation analysis, which can effectively avoid excessive disturbance of the water level in the shallow recharge area by the high-pressure water flow (0.5-0.8MPa) of the deep recharge wells, while ensuring that the shallow recharge can cover the water-loss area of ​​the shallow clay. The well spacing of the shallow recharge wells is consistent with the well spacing of the deep recharge wells in the corresponding area, that is, 3-5m in the primary key control area, 6-8m in the secondary control area, and 8-10m in the general control area, so as to achieve precise matching of the deep and shallow recharge ranges.

[0030] Preferably, the well construction process can adopt targeted construction methods according to the differences in the geological characteristics traversed by deep reinjection wells and shallow reinjection wells, so as to ensure that the well casing is installed stably, has good water filtration performance, and meets the reinjection pressure and flow requirements. The construction process for deep reinjection wells is as follows: The first step is drilling, which uses percussion drilling technology. This technology is suitable for fault pebble layers (hard rock with well-developed fractures). It can break the rock through impact, ensuring the verticality and diameter stability of the borehole. The borehole diameter is 50-80mm larger than the outer diameter of the well casing. The well casing for deep reinjection wells is made of steel pipe (130mm in diameter), so the borehole diameter can be set to 180-210mm. The borehole depth can be consistent with the excavation depth of the foundation pit (28m), and it is ensured that the bottom of the well casing is embedded in the fault fracture zone for no less than 5m. The embedding depth can be calculated based on the burial depth and thickness of the fracture zone determined in the preliminary site survey. For example, if the burial depth of the fracture zone is 20m and the thickness is 8m, the borehole depth can reach 28m (the excavation depth of the foundation pit), and the bottom of the well casing is embedded in the fracture zone to a depth of 8m (from the burial depth of 20m to the bottom of the 28m borehole) to ensure that the deep reinjection water can directly enter the core area of ​​the fracture zone and spread rapidly through its good hydraulic channels. The second step is well casing fabrication and installation. The wall thickness of the steel pipe used in deep reinjection wells can meet the high-pressure reinjection requirements (0.5~0.8MPa). Therefore, seamless steel pipes with a wall thickness of 5~8mm are selected to avoid deformation or breakage of the well casing under high pressure. Filter holes are fabricated in the pipe section corresponding to the fault fracture zone (i.e., the 5m section embedded in the fracture zone and the section corresponding to the 2~3m overlying strata). The filter hole diameter is 8~10mm, and the hole spacing is 10~15mm, evenly distributed in a quincunx pattern. This quincunx arrangement ensures that the filter holes are evenly distributed around the pipe, improving water intake efficiency. After the filter holes are fabricated... Two layers of filter screens are wrapped around the outside of the pipe section. The first layer is a nylon filter screen (0.2~0.3mm pore size) to block fine particles of mud and sand in the fracture zone. The second layer is a steel wire filter screen (0.5~0.8mm pore size) to support the nylon filter screen and block coarse rock fragments. The two layers of filter screens are fixed with stainless steel cable ties with a spacing of 150~200mm to prevent the filter screens from falling off. The well pipe is hoisted by a crane during installation. During hoisting, the well pipe is kept vertical to avoid collision with the borehole wall and causing the borehole to collapse. A bottom sealing steel plate (8~10mm thick) can be installed at the bottom of the well pipe to prevent sediment from the bottom of the borehole from entering the well pipe. The third step is backfilling and cementing. After the well casing is installed in place, the annular space between the well casing and the borehole wall is backfilled in layers. The backfill material is graded sand and gravel (particle size 5~10mm). This particle size of sand and gravel has good permeability and can serve as a filter layer, while also fixing the well casing. During the backfilling process, a method of backfilling and vibration is adopted, with a vibration frequency of 50~60Hz to ensure that the backfill is dense and to avoid voids. The backfilling height extends from the bottom of the borehole to 1m below the surface. The annular space from 1m below the surface to the surface is filled with clay (moisture content 20%~25%) in layers and compacted, with a compaction degree of not less than 90%, to prevent surface rainwater and sewage from seeping into the well and polluting the groundwater, while also preventing the reinjection water from overflowing from the surface.

[0031] Preferably, the well construction process for shallow reinjection wells is basically the same as that for deep reinjection wells, but the process parameters need to be adjusted for shallow clay (low permeability and soft soil): The first step of drilling adopts the spiral drilling process, which can use the spiral drill rod to discharge clay out of the hole and avoid hole collapse. The diameter of the hole is 30-50mm larger than the outer diameter of the shallow reinjection well pipe. The shallow reinjection well pipe uses PVC pipe (diameter of 110mm), so the diameter of the hole can be set to 140-160mm; the drilling depth is 10-15m, and the specific depth can be determined according to the thickness of the shallow clay determined in the early exploration. For example, if the thickness of the shallow clay is 12m, the drilling depth is set to 12m to ensure that the well pipe penetrates the entire shallow clay and sand layer and enters the underlying silty clay layer by 0.5-1m at the bottom, thereby improving the stability of the well pipe. The second step is well casing processing and installation. The wall thickness of the PVC pipe used for shallow reinjection wells must meet the low-pressure reinjection requirements (0.1~0.3MPa), and rigid PVC pipes with a wall thickness of 4~6mm are selected. Filter holes are processed in the section of the well casing corresponding to the shallow clay layer (i.e., the section from 1m below the surface to the bottom of the hole). The diameter of the filter holes is 6~8mm, and the spacing between the holes is 15~20mm, distributed in a quincunx pattern. A layer of nylon filter mesh (0.1~0.2mm pore size) is wrapped around the outside of the filter holes. Because the shallow clay particles are fine, a finer filter mesh is required to block mud and sand. The filter mesh is fixed with plastic cable ties with a spacing of 200~250mm. A PVC bottom plug is installed at the bottom of the well casing and fixed with glue to prevent sediment from entering. The third step is backfilling and cementing. The annular space between the well casing and the borehole wall is filled with fine sand (particle size 1~3mm). Fine sand has moderate permeability and can meet the needs of shallow, low-flow reinjection. Backfilling is done manually in layers, with each layer being 200~300mm thick and lightly compacted. The backfilling height is 0.5m below the surface. From 0.5m below the surface to the surface, plain soil (moisture content 18%~22%) is used and compacted, with a compaction degree of not less than 85%.

[0032] Preferably, the installation of supporting equipment can be based on the reinjection pressure and flow control requirements of deep and shallow reinjection wells, with corresponding monitoring instruments, control valves, and connecting pipelines installed respectively to ensure that reinjection parameters can be monitored and adjusted in real time. The installation of supporting equipment for deep reinjection wells may include: installing a pressure gauge, a flow meter, and a shut-off valve sequentially on the reinjection pipeline at the wellhead. The pressure gauge should be a high-pressure resistant type to accurately monitor the reinjection pressure, avoiding excessive pressure leading to excessive expansion of fractures in the fault zone or excessive pressure affecting water replenishment efficiency. The flow meter should be an electromagnetic flow meter to monitor the deep reinjection flow in real time, and the electromagnetic flow meter is unaffected by small amounts of silt in the water, ensuring high measurement stability. The shut-off valve should be made of stainless steel with a pressure resistance rating of not less than 1.0 MPa, used to control the start-up and shutdown of the deep reinjection well and to regulate the flow rate. The stainless steel material prevents corrosion and extends service life. The installation of supporting equipment for shallow reinjection wells includes: installing a flow meter and a ball valve sequentially on the reinjection pipeline at the wellhead. The flow meter is a turbine flow meter, which is suitable for shallow low-flow reinjection needs. The turbine flow meter is small in size and easy to install. The ball valve is made of PVC material with a pressure resistance rating of not less than 0.6MPa. It is used to control the start and stop of the shallow reinjection well and to regulate the flow. The PVC material is the same as the well pipe material of the shallow reinjection well, which can reduce the difference in pipeline connection.

[0033] Preferably, the wellhead pipelines of all deep reinjection wells are connected to the main deep reinjection pipe via branch pipes. The diameter of the main reinjection pipe can be determined according to the number of deep reinjection wells. For example, when there are 10 to 15 deep reinjection wells, the diameter of the main reinjection pipe is 200 mm, and when there are 15 to 20 wells, 250 mm is selected. This ensures that the water flow velocity in the main reinjection pipe is controlled at 1.5 to 2.0 m / s to avoid excessive pressure loss in the pipeline due to excessive flow velocity. The main deep reinjection pipe is connected to the deep reinjection pump, which is a high-pressure centrifugal pump. Its head must meet the deep reinjection pressure requirements, and its flow rate must match the total design flow rate of all deep reinjection wells. A check valve is installed at the pump outlet to prevent backflow of reinjection water from damaging the pump. All shallow reinjection wells have their wellhead pipes connected to the shallow main reinjection pipe via branch pipes. The diameter of the main reinjection pipe is determined based on the number of shallow reinjection wells; for example, 10-15 shallow reinjection wells correspond to a 150mm diameter main reinjection pipe, and 15-20 wells correspond to a 200mm diameter main reinjection pipe. The shallow main reinjection pipe is connected to the shallow reinjection pump, which is a low-pressure centrifugal pump suitable for the low-pressure reinjection requirements of shallow areas. A check valve is also installed at the pump outlet.

[0034] Preferably, such as Figure 4As shown, step S3 is the dynamic operation and control stage. It aims to achieve precise and targeted water replenishment to the fault fracture zone (high permeability strata) and shallow clay (low permeability strata) by relying on the hydrogeological parameters of the fault fracture zone obtained in the previous test stage and the established correlation model of foundation pit dewatering-surrounding water level changes throughout the entire foundation pit dewatering operation. By monitoring the water level outside the pit and the settlement data of buildings in real time, the automatic control unit will adjust the recharge parameters (recharge volume and recharge pressure) of the deep recharge well and the shallow recharge well in linkage. This ensures that the settlement of surrounding buildings is controlled within the design limit, while avoiding hydraulic interference between deep and shallow recharge units, thus ensuring the safety of foundation pit construction and surrounding buildings. This stage may include four phases: collaborative operation of monitoring units, graded control of recharge parameters, triggering of emergency response mechanism, and routine maintenance of the system.

[0035] Preferably, the coordinated operation of the monitoring units can ensure the real-time, accuracy, and completeness of water level monitoring and building settlement monitoring data, providing a reliable basis for parameter adjustment. Among them, the external water level monitoring unit can cover the area affected by recharge around the foundation pit. The monitoring points are laid out according to the principle of uniform distribution and key densification: along the baseline of the outer perimeter of the foundation pit (1-2m away from the excavation edge of the foundation pit), an external water level monitoring point is laid out every 5m in non-key control areas. In the fault fracture zone distribution area and the first-level key control area (the area within the flow bypass, the area where the water level drop exceeds 2m, and the area where the building is located within 10m of the edge of the foundation pit), the spacing between monitoring points is increased to 2-3m to ensure accurate capture of water level changes in the fracture zone area. All external water level monitoring points use submersible automatic water level gauges. During installation, the probe of the automatic water level gauge must be lowered to 1-2m below the ground static water level to avoid probe exposure due to water level fluctuations. At the same time, the probe position is fixed with a stainless steel bracket to prevent displacement caused by water flow impact. Water level monitoring data can be uploaded to the automatic control unit in real time through a wired transmission module. The automatic control unit stores the received data in real time and marks anomalies. If a monitoring point has three consecutive data loss transmissions or the data fluctuation exceeds 0.2m, an equipment fault alarm is immediately triggered, prompting staff to check the line or repair the water level gauge.

[0036] Preferably, the building settlement monitoring unit can set up monitoring points according to the structural characteristics and settlement-sensitive parts of the surrounding buildings. Specifically, settlement monitoring points can be set up at the four corners of the building walls, the middle of the wall base, and the mid-span of the foundation beam. The number of monitoring points set up for each independent building is not less than 4. For adjacent buildings ≤5m away from the edge of the foundation pit, an additional monitoring point is set up every 3m along its long side, so that the number of monitoring points for a single building is not less than 6, ensuring that the settlement differences of various parts of the building can be fully reflected. Settlement monitoring can be performed using a high-precision electronic level. The monitoring frequency is dynamically adjusted according to the dewatering stage of the foundation pit: for the first 7 days after the start of dewatering (the initial stage of dewatering, when water level changes drastically), the monitoring frequency can be set to once every 6 hours; after the dewatering rate stabilizes (daily drop in water level outside the pit ≤ 0.5m for 3 consecutive days) and the recharge unit is operating normally, the monitoring frequency can be adjusted to once every 24 hours; if the settlement rate of the building exceeds 0.3mm / d or the water level outside the pit drops suddenly (drop exceeding 0.8m within 1 hour) during the monitoring process, the monitoring frequency can be immediately increased to once every 1 hour to ensure timely detection of abnormal deformation trends. Settlement monitoring data adopts a dual-verification mode of manual reading + automatic recording. After each reading is completed, staff can check the difference between the two readings on-site. If the difference exceeds ±0.1mm, a remeasurement is required; after confirming the validity of the data, the data is uploaded to the automatic control unit and stored in association with the water level monitoring data of the same period to form a water level-settlement corresponding database, providing data support for the adjustment of recharge parameters.

[0037] Preferably, to address the issue of delayed settlement control during the recharge process in the foundation pit, in addition to the water level monitoring and building settlement monitoring functions operating collaboratively by the monitoring units, ground mechanical parameter monitoring and settlement trend prediction functions are further added. This constructs a complete closed-loop control system that enables real-time monitoring, trend prediction, and advance control. This system is deeply integrated with the existing water level monitoring and building settlement monitoring systems, forming a multi-dimensional monitoring and control network covering hydrological parameters and ground mechanical parameters. This ensures proactive control of the settlement of surrounding buildings, avoids the risk of settlement exceeding limits due to passive control, and coordinates with stages such as the graded control of recharge parameters and emergency response mechanisms to jointly ensure the safety of foundation pit construction and surrounding buildings.

[0038] Preferably, the multi-parameter monitoring module can simultaneously acquire hydrological and mechanical parameters. The deployment of fiber optic strain sensors must be strictly combined with the baseline of the outer perimeter of the foundation pit (1-2m from the excavation edge) and the division of key control areas for recharge: In non-key control areas, one fiber optic strain sensor is deployed for every two water level monitoring points outside the pit along the baseline of the outer perimeter of the foundation pit, and the sensor spacing matches the spacing of the water level monitoring points outside the pit, i.e., the sensor spacing in non-key control areas is 10-15m; In the fault fracture zone distribution area and the first-level key control area (areas with a water level drop of more than 2m within the flow range and areas where buildings are located within 10m of the edge of the foundation pit), the sensor deployment density is increased, with one fiber optic strain sensor deployed for every water level monitoring point outside the pit, and the sensor spacing after densification is 2-3m, to ensure accurate capture of the elastic modulus changes of the strata in the fracture zone area and around the buildings. The placement of pore water pressure gauges can be based on the stratigraphic distribution characteristics determined in the preliminary site survey. They can be buried in the transitional strata between shallow clay and deep fractured zones. The specific depth can be adjusted according to the thickness of the shallow clay and the burial depth of the fractured zone, for example, it can be set to 15~20m. The placement of each pore water pressure gauge must correspond to the building settlement monitoring point. That is, at least one pore water pressure gauge should be placed within 3m of each building settlement monitoring point. If the building is a rectangular structure with a long side exceeding 10m, an additional pore water pressure gauge should be added every 5m along its long side to ensure a one-to-one correspondence between mechanical parameters and settlement data, providing an accurate correlation data basis for subsequent model analysis.

[0039] Preferably, both the fiber optic strain sensor and the pore water pressure gauge can be connected to the automatic control unit via a wired transmission module. The transmission cable is an armored cable to prevent cable damage during construction and to ensure that the monitoring data is uploaded to the automatic control unit in real time and stably.

[0040] Preferably, the basic data required for the construction of the predictive model comes from the single-well pumping test and group-well dewatering test in the early experimental stage. During the test, in addition to acquiring parameters such as the permeability coefficient, radius of influence, static water level, groundwater level drawdown cone shape, water inflow distribution of each well, and flow around the cutoff wall in the fractured zone and shallow strata as required, soil elastic modulus and pore water pressure data are collected simultaneously under different test stages and conditions: In the steady-flow pumping stage of the single-well pumping test, the water level data and pumping volume of the test well and observation well are recorded at intervals during the pumping process, and the elastic modulus value monitored by the fiber optic strain sensor and the pore water pressure value monitored by the pore water pressure gauge are recorded simultaneously. The pore water pressure value was recorded at the same interval as the water level data. During the unsteady flow pumping phase of the single-well pumping test, the water level change data over time was continuously recorded, and the elastic modulus and pore water pressure data were recorded every hour until the end of the test. In the group well dewatering test, the groundwater level distribution, pumping volume, and ground settlement data were recorded at intervals according to the requirement of recording the data at different pumping durations. The elastic modulus and pore water pressure data were recorded simultaneously, and the recording frequency was consistent with the ground settlement data recording frequency. This ensured that at least 50 complete sets of correlation data sets of "permeability coefficient-elastic modulus-pore water pressure-water level drop-settlement" were obtained to provide sufficient sample data for model construction.

[0041] Preferably, the collected raw data is preprocessed. First, abnormal data caused by equipment failure (such as sensor offline or data transmission interruption) or sudden interference (such as vibration of construction machinery or temporary changes in pumping flow) is removed. After the data preprocessing is completed, it is divided into a training set and a validation set in a 7:3 ratio. The training set is used for parameter training of the LSTM (Long Short-Term Memory) neural network prediction model, and the validation set is used for model accuracy verification.

[0042] Preferably, the LSTM neural network prediction model is structured as follows: the input layer has two neurons, corresponding to the real-time monitored elastic modulus (E) and pore water pressure (u), respectively; the hidden layer has three layers, each with 128 neurons, and the ReLU activation function is used to enhance the model's ability to fit nonlinear data; the output layer has one neuron, corresponding to the predicted settlement trend value for the next 6-12 hours. During model training, the Adaptive Moment Estimation (Adam) optimization algorithm is used. The initial learning rate is set to 0.001. As the number of training iterations increases, when the prediction error on the validation set does not decrease significantly for 20 consecutive iterations, the learning rate is automatically halved. The number of iterations is set to 1000. When the prediction error on the validation set is less than 0.1 mm for 50 consecutive iterations, training is stopped and the model parameters are saved. After model training is completed, the coefficient of determination R is used... 2 Evaluate the model's accuracy and ensure the model's coefficient of determination R. 2The accuracy should be no less than 0.9 to meet the accuracy requirements for engineering applications. If the model accuracy does not meet the requirements, additional experimental data should be collected and the model retrained until the model accuracy meets the requirements.

[0043] Preferably, when the advance control logic is executed, the automatic control unit can call the LSTM neural network prediction model in real time, input the elastic modulus (E) and pore water pressure (u) data collected during the current monitoring period into the model, obtain the predicted settlement trend values ​​of each building settlement monitoring point for the next 6-12 hours, and compare them with the preset building settlement design limits (the design limits are determined according to the structural safety level, service life and relevant specifications of surrounding buildings). When the automatic control unit detects that the predicted settlement trend value of a certain building settlement monitoring point for the next 12 hours will exceed 80% of the design limit, it immediately triggers the advance control command. First, through the area positioning function of the automatic control unit, it determines the recharge control area (first-level key control area or second-level control area) corresponding to the settlement monitoring point, and retrieves the current deep recharge pressure, shallow recharge volume and other recharge parameters of the area.

[0044] Preferably, for parameter adjustment of deep reinjection wells: the automatic control unit can send a control signal to the pressure regulating valve at the outlet of the deep reinjection pump to increase the deep reinjection pressure by 0.1 MPa. The increased deep reinjection pressure must be strictly controlled within the upper limit of 0.5~0.8 MPa. If the increased pressure exceeds this upper limit, it will automatically adjust to only increase the deep reinjection pressure to the upper limit to avoid excessive expansion of the fault fracture zone or well casing rupture due to excessive pressure. At the same time, the high-pressure gauge at the wellhead of the deep reinjection well will feed back the adjusted reinjection pressure data to the automatic control unit in real time. The automatic control unit will monitor the pressure data in real time. If the pressure fluctuates abnormally (fluctuation amplitude exceeds ±0.05 MPa), the pressure adjustment will be immediately suspended, the pressure regulating valve will be checked for malfunction, and the control command will be continued after the malfunction is eliminated.

[0045] Preferably, for parameter adjustment of shallow reinjection wells: the automatic control unit can send an up-frequency signal to the variable frequency drive of the shallow reinjection pump to increase the shallow reinjection volume by 10%. The increased shallow reinjection volume needs to be verified in conjunction with the requirement that the shallow formation water level fluctuation is less than 0.5m. If the increase may cause the shallow water level fluctuation to exceed 0.5m, the increase ratio of the shallow reinjection volume will be adjusted to 5% to avoid water accumulation in the shallow clay due to excessive reinjection volume, which could lead to soil softening or disturbance. The turbine flow meter at the wellhead of the shallow reinjection well will feed back the adjusted reinjection flow data to the automatic control unit in real time to ensure that the shallow reinjection volume adjustment meets expectations.

[0046] Preferably, during the execution of the control command, the automatic control unit can increase the monitoring frequency of the area, simultaneously monitoring the elastic modulus, pore water pressure, external water level, and building settlement data. If two consecutive monitoring results show that the elastic modulus has recovered to more than 95% of the initial stable value of the area (i.e., the elastic modulus value before the start of foundation pit dewatering), the pore water pressure has recovered to the pressure level of the corresponding external water level design value, and the predicted settlement trend value for the next 12 hours has fallen below 70% of the design limit, then the automatic control unit will initiate the recharge parameter callback program. The callback rate is set as follows: the deep recharge pressure decreases by 0.02 MPa per hour, and the shallow recharge volume decreases by 2 MPa per hour. The rate of increase in subsidence will be maintained until the recharge parameters return to normal levels before regulation, avoiding new ground disturbances caused by sudden changes in recharge parameters. If four consecutive monitoring tests show that the predicted subsidence trend value has not fallen below 70% of the design limit, the initial regulation is deemed ineffective. The automatic control unit will initiate a second regulation after one hour, increasing the deep recharge pressure by 0.05 MPa (still within the upper pressure limit) and the shallow recharge volume by 5%. Simultaneously, the pre-deployed backup monitoring equipment (such as backup water level gauges and backup settlement meters) in the area will be activated to double-verify the monitoring data and ensure its reliability until the predicted subsidence trend value meets safety requirements. In addition, for key control areas, if the predicted subsidence trend value is still not effectively controlled after the second regulation, the automatic control unit will trigger an early warning of the emergency response mechanism, sending a warning signal to staff to prepare for the activation of the first-level emergency response, further ensuring the safety of surrounding buildings.

[0047] Preferably, the graded adjustment of reinjection parameters can be executed by an automatic control unit. Based on a correlation model of pit dewatering and surrounding water level changes, combined with real-time monitoring data, the automatic control unit can precisely adjust the parameters of deep and shallow reinjection wells according to different scenarios to ensure that the water replenishment effect matches the formation requirements. For parameter adjustment of deep reinjection wells, the automatic control unit can set two core triggering conditions: one is the water level drop in the fractured zone area, and the other is the water level elevation outside the pit. When the automatic control unit calculates that the water level drop at any water level monitoring point within the fault fracture zone exceeds 2m, it immediately triggers a deep reinjection volume control command. This command is first transmitted to the variable frequency drive (VFD) of the deep reinjection pump. By increasing the output frequency of the VFD, the pump speed is increased, thereby increasing the deep reinjection volume. Simultaneously, the high-pressure gauge at the wellhead of the deep reinjection well feeds back the reinjection pressure data to the automatic control unit in real time. If the pressure data exceeds the upper limit of the deep reinjection pressure, the automatic control unit automatically sends a signal to the electric shut-off valve on the deep reinjection pipeline to reduce the valve opening and lower the reinjection pressure, preventing excessive expansion of the fracture zone or well pipe rupture due to excessive pressure. Once the water level drop in the fracture zone area falls below 1.5m, the automatic control unit gradually reduces the frequency of the VFD to restore the deep reinjection volume to normal levels.

[0048] Preferably, when the automatic control unit detects that the water level outside the pit has dropped to -4 to -12m (close to the elevation of the fractured zone rock surface), it can trigger a coordinated control command for deep recharge pressure and shallow recharge volume: on the one hand, by adjusting the pressure regulating valve at the outlet of the deep recharge pump, the deep recharge pressure is increased to enhance the infiltration capacity of the recharge water into the fractured zone fissures and quickly replenish the water lost from the fissures; on the other hand, a frequency reduction signal is simultaneously sent to the frequency converter of the shallow recharge pump to reduce the shallow recharge volume. This adjustment logic can prevent water accumulation in the shallow clay due to excessive recharge volume, which could lead to soil softening or disturbance. For parameter control of shallow reinjection wells, the core triggering condition is the fluctuation range of shallow formation water level: when the automatic control unit detects that the water level fluctuation at any water level monitoring point in the shallow formation (within a depth range of 10~15m) is less than 0.5m, it determines that the water content of the shallow formation is stable and there is no need to maintain the original reinjection volume. At this time, the automatic control unit sends a command to the variable frequency drive of the shallow reinjection pump to reduce the output frequency, thereby reducing the shallow reinjection volume. At the same time, the turbine flow meter at the wellhead of the shallow reinjection well provides real-time feedback of flow data to ensure that the reinjection pressure remains within the design range after the shallow reinjection volume is adjusted. During the adjustment of the shallow reinjection volume, the automatic control unit can monitor the pressure data of the deep reinjection well in real time. If the deep reinjection pressure fluctuation is found to exceed 0.1MPa, the adjustment of the shallow reinjection volume is immediately suspended until the deep reinjection parameters stabilize before continuing to execute, to prevent interference caused by hydraulic connection between the deep and shallow reinjection units.

[0049] Preferably, when monitoring data exceeds the normal control range, the automatic control unit can quickly trigger emergency measures to curb the abnormal development of settlement. The core triggering condition for the emergency response is the building settlement rate. When the automatic control unit detects that the single-point settlement rate of any building settlement monitoring point exceeds 0.5 mm / d for three consecutive times (each time at 1-hour intervals), it immediately triggers a Level I emergency response: The automatic control unit first locks the reinjection control area corresponding to the settlement monitoring point, sends a signal to the variable frequency drives of all deep reinjection wells in the area to temporarily increase the reinjection volume, and simultaneously increases the frequency of water level monitoring and settlement monitoring in the area to track the control effect in real time; if continuous monitoring for 12 hours shows that the single-point settlement rate has not yet dropped below the upper limit of the stable settlement rate, the automatic control unit initiates a Level II emergency response: it opens the pre-deployed backup deep reinjection wells in the area, and... The distance between the recharge well and the main deep recharge well should be maintained at 3-5m. The well depth, well pipe material, filter hole parameters and supporting equipment should be consistent with those of the main recharge well. After the backup recharge well is opened, the automatic control unit will control the total deep recharge volume in this area to within 140% of the original total recharge volume to avoid the recharge volume being too high and causing "short circuit" water replenishment in the fracture zone (i.e., the recharge water is lost from other channels before it diffuses). When the settlement monitoring data shows that the settlement rate of this single point is stable below 0.2mm / d for 6 consecutive hours, the automatic control unit will gradually release the emergency status: first, close the backup recharge well, and then restore the recharge volume of the main deep recharge well to the normal control level at a rate of 5% every 2 hours to prevent the sudden change of parameters from causing new settlement fluctuations.

[0050] Preferably, in addition to handling abnormal situations such as blockage of recharge wells and sudden drop in water level outside the pit, the automatic control unit can also deal with abnormal situations such as blockage of recharge wells and sudden drop in water level outside the pit. When the flow meter of a recharge well shows that the flow rate is more than 30% lower than the normal value, the automatic control unit determines that the well is blocked, immediately sends an alarm signal to the control terminal, and automatically closes the inlet valve of the well to avoid affecting the normal operation of other recharge wells. Personnel need to arrive at the site within 4 hours and use the air compressor well washing method (consistent with the well washing process after the initial well construction, working pressure 0.6~0.8MPa) to clean the blocked recharge well, remove the mud and sand blockage in the well pipe filter screen and filter holes, and reopen the valve after cleaning. The flow rate can only be put into use after it is monitored to be restored to more than 90% of the normal value. If a sudden drop in water level outside the pit is detected (the drop exceeds 1m within 1 hour), the automatic control unit immediately starts the emergency water replenishment mode. After the water level outside the pit stabilizes, the water is gradually restored at a rate of 0.2m per hour to avoid large-scale consolidation and settlement of the surrounding soil caused by the sudden drop in water level.

[0051] Preferably, routine system maintenance can be achieved through a full-cycle maintenance mechanism established by the automatic control unit, which regularly inspects and calibrates the monitoring equipment, reinjection equipment, and pipeline system. For monitoring equipment: Submersible automatic water level gauges need to be calibrated weekly using a standard water level gauge comparison method. A calibrated standard water level gauge is slowly lowered into the monitoring well, and the water level data is read and compared with the monitoring data from the automatic water level gauge. If the error exceeds ±0.05m, the sensor of the automatic water level gauge needs to be adjusted; if adjustment is ineffective, the equipment should be replaced. High-precision electronic levels need to be calibrated monthly by measuring the reference point at a known elevation to verify the instrument's measurement error and ensure accuracy is maintained within ±0.1mm. For reinjection equipment: The operating status of deep and shallow reinjection pumps should be checked daily, including motor housing temperature, operating noise and vibration amplitude. If any abnormality is found, the machine should be stopped immediately and the motor bearings, seals and other components should be checked. The pressure gauges and flow meters on the reinjection pipeline should be calibrated weekly. The pressure gauges should be compared with standard pressure gauges and the flow meters should be calibrated using the standard volume method. If the error exceeds ±1%, the instruments should be adjusted or replaced. For the pipeline system: monthly, conduct a sealing check at the interface between the deep main reinjection pipe and the shallow main reinjection pipe. Apply soapy water to the interface gaps and observe whether air bubbles are generated. If there is any air or water leakage, the sealing gasket at the interface needs to be replaced (rubber gasket for deep reinjection and PVC gasket for shallow reinjection), and the flange bolts need to be retightened. Every two weeks, the filter screens of the deep and shallow reinjection wells need to be cleaned. During cleaning, close the wellhead valve, remove the outer protective sleeve of the filter screen, and use a high-pressure water gun to rinse the surface of the filter screen to remove the attached mud and sand particles. After rinsing, reinstall the protective sleeve to ensure that the filter screen's water filtration performance is restored.

[0052] It should be noted that the specific embodiments described above are exemplary. Those skilled in the art can devise various solutions inspired by the disclosure of this invention, and these solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and not intended to limit the scope of the claims. The scope of protection of this invention is defined by the claims and their equivalents. This specification contains multiple inventive concepts; terms such as "preferredly," "according to a preferred embodiment," or "optionally" indicate that the corresponding paragraph discloses an independent concept. The applicant reserves the right to file divisional applications based on each inventive concept.

Claims

1. A method for controlling building settlement based on reinjection technology, characterized in that, It includes the following steps: S1. Conduct single-well pumping tests and group-well dewatering tests to obtain hydrogeological parameters of the fracture zone, clarify the flow range of the foundation pit water-stop curtain, and delineate the key control area for recharge. S2. Deploy the recharge unit, with deep recharge wells arranged around the outer perimeter of the foundation pit and shallow recharge wells arranged on the outer side of the deep layer. S3. During the dewatering of the foundation pit, the water level outside the pit and the settlement of the building are monitored in real time, and the recharge parameters are adjusted in conjunction with the automatic control unit.

2. The method according to claim 1, characterized in that, Step S1 includes the test preparation stage, in which well locations are arranged based on the exploration results. The well locations for single-well pumping tests are selected to represent the core area of ​​the fractured zone and the typical characteristics of shallow strata, while avoiding areas with dense underground pipelines and the influence range of building foundations. The well groups for group-well dewatering tests are evenly distributed around the perimeter of the planned excavation pit. During the well construction process, the well construction technology is selected according to the strata characteristics. Spiral drilling is used for shallow clay and sand layers, and percussion drilling is used for gravel layers. After well construction, the well is washed using an air compressor until the water in the well is clear, and then pumping equipment and monitoring instruments are installed.

3. The method according to claim 1 or 2, characterized in that, The single-well pumping test includes two stages: steady-flow pumping and unsteady-flow pumping. In the steady-flow pumping stage, a submersible pump is started to pump water at a constant flow rate. The water level data and pumping volume of the test well and the observation well are recorded at intervals. When the difference in drawdown measured multiple times does not exceed the first drawdown difference threshold and the pumping volume fluctuation is less than the pumping volume fluctuation threshold, it is determined that a steady state has been reached. In the unsteady-flow pumping stage, after the steady-flow pumping ends, the pumping equipment is kept in the same operating state, and the water level change data over time is continuously recorded. The permeability coefficient of the fractured zone and shallow strata is calculated through the single-well pumping test. The Dubouy formula is used for calculation in the steady-flow pumping stage, and the Theis formula is used for calculation in the unsteady-flow pumping stage. The static water level, maximum drawdown, and radius of influence parameters of the test well are recorded.

4. The method according to any one of claims 1 to 3, characterized in that, The well-group dewatering test simulates the actual dewatering conditions during the excavation of the foundation pit. The wells are divided into several groups, and each group of wells pumps water simultaneously. During the test, water level monitoring points are set up around the foundation pit and within the affected area, and settlement monitoring points are set up near the foundations of surrounding buildings. The groundwater level distribution, pumping volume, and ground settlement data are recorded at intervals under different pumping durations. The well-group dewatering test obtains data on the shape of the groundwater level drawdown funnel under the action of the well group, the distribution of water inflow from each well, and the flow around the cutoff wall.

5. The method according to any one of claims 1 to 4, characterized in that, Step S1 includes the parameter analysis stage: data obtained from single-well pumping tests and group-well dewatering tests are organized and analyzed, and curves of groundwater level change with time and distance are plotted. Based on the geological distribution characteristics, parameters such as permeability coefficient, radius of influence, static water level, unit inflow, and permeability coefficient and compressibility coefficient of shallow clay and sand layers are determined. Based on the above parameters, a correlation model of foundation pit dewatering and surrounding water level change is established. This correlation model uses foundation pit dewatering depth and dewatering rate as independent variables, and groundwater level drop and ground subsidence at a certain point in the surrounding area as dependent variables.

6. The method according to any one of claims 1 to 5, characterized in that, Step S2 includes the well location planning stage: The well location planning is based on the key control area for recharge, combined with the excavation range of the foundation pit, the geological distribution characteristics and the location of surrounding buildings. First, the baseline of the outer perimeter of the foundation pit is determined. Deep recharge wells are continuously laid out along the baseline of the outer perimeter of the foundation pit, and the well spacing is set according to the differentiated level of the key control area for recharge. Shallow recharge wells are laid out parallel to the outside of the deep recharge wells, and the well spacing of the shallow recharge wells is consistent with the well spacing of the deep recharge wells in the corresponding area.

7. The method according to any one of claims 1 to 6, characterized in that, Step S2 includes the installation of supporting equipment: Based on the reinjection pressure and flow control requirements of deep and shallow reinjection wells, install corresponding monitoring instruments, control valves, and connecting pipelines. The installation of supporting equipment for deep reinjection wells includes: installing a pressure gauge, an electromagnetic flowmeter, and a shut-off valve sequentially on the reinjection pipeline at the wellhead; connecting all wellhead pipelines of deep reinjection wells to the deep main reinjection pipeline via branch pipes; connecting the deep main reinjection pipeline to a high-pressure centrifugal pump, with a check valve installed at the outlet of the high-pressure centrifugal pump; The installation of supporting equipment for shallow reinjection wells includes: installing a turbine flow meter and a ball valve sequentially on the reinjection pipeline at the wellhead; connecting the wellhead pipelines of all shallow reinjection wells to the shallow main reinjection pipeline via branch pipes; connecting the shallow main reinjection pipeline to a low-pressure centrifugal pump, and installing a check valve at the outlet of the low-pressure centrifugal pump.

8. The method according to any one of claims 1 to 7, characterized in that, Step S3 includes the collaborative operation phase of the monitoring units, which includes monitoring of the water level outside the pit and monitoring of building settlement. The external water level monitoring covers the area affected by recharge around the foundation pit. The monitoring points are laid out according to the principle of uniform distribution and key densification: monitoring points are laid out at intervals in non-key control areas along the baseline of the outer perimeter of the foundation pit, and the monitoring points are densely laid out in the fault fracture zone and the first-level key control area; all external water level monitoring points use submersible automatic water level gauges, and the water level monitoring data is uploaded to the automatic control unit in real time. Building settlement monitoring involves setting up monitoring points based on the structural characteristics and settlement-sensitive parts of surrounding buildings. Settlement monitoring points are set up at the four corners of the building walls, the middle of the wall base, and the mid-span of the foundation beam. Settlement monitoring uses electronic levels, and the monitoring frequency is dynamically adjusted according to the stage of foundation pit dewatering.

9. The method according to any one of claims 1 to 8, characterized in that, Step S3 also includes adding functions for monitoring geological mechanical parameters and predicting subsidence trends, and constructing a closed-loop control system, wherein, The formation mechanical parameters are monitored by a multi-parameter monitoring module to achieve synchronous acquisition of hydrological and mechanical dual parameters. The multi-parameter monitoring module includes a fiber optic strain sensor and a pore water pressure gauge that are connected to the automatic control unit. The settlement trend prediction is based on the LSTM neural network prediction model. The basic data required for model construction comes from the single-well pumping test and group-well dewatering test in step S1. The soil elastic modulus and pore water pressure data are collected simultaneously under different test stages and conditions. After preprocessing the collected raw data to remove abnormal data, it is divided into training set and validation set according to the ratio. The input layer neurons of the LSTM neural network prediction model correspond to the real-time monitored elastic modulus and pore water pressure.

10. The method according to any one of claims 1 to 9, characterized in that, When executing the advance control logic, the closed-loop control system can automatically call the LSTM neural network prediction model in real time, input the elastic modulus and pore water pressure data collected in the current monitoring period into the model, obtain the predicted settlement trend of each building settlement monitoring point in the future, and compare it with the preset building settlement design limit. When the predicted future settlement trend value of a certain building settlement monitoring point exceeds the preset threshold, an advance control command is triggered. The automatic control unit determines the recharge control area corresponding to the settlement monitoring point and retrieves the current recharge parameters of that area.

Images