Water-rich stratum underground cavity targeted treatment method

Abstract

The invention discloses a water-rich stratum underground cavity targeted treatment method, and relates to the field of underground cavity targeted treatment, and the method comprises the steps: constructing an underground cavity recognition model through improved YOLOv8, and recognizing an underground cavity in a water-rich stratum; a three-dimensional geologic model is constructed, a drilling track pointing to the cavity from the earth surface or the existing structure is designed, targeted drilling is constructed, and a treatment channel is formed; with the cavity as the center, curtain grouting holes are formed, rapid-hardening type double-liquid grout is injected, and a three-dimensional waterproof curtain is formed; underground water in the three-dimensional waterproof curtain is guided and drained through part of targeted drill holes, and desilting is conducted; and self-compacting filling materials are pumped into the cavity through the other targeted drill holes for main body backfilling, and cavity cracks are reinforced by adopting a sleeve valve pipe grouting technology. And the filling and reinforcing effect is verified, and a monitoring system is arranged to monitor the stratum stability. Through a technical path of combining artificial intelligence identification, three-dimensional accurate positioning and active three-dimensional water stop, safe and efficient targeted treatment of the underground cavity of the water-rich stratum is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of underground cavity targeted treatment, and in particular to a method for targeted treatment of underground cavities in water-rich strata. BACKGROUND

[0002] Underground cavities in water-rich strata are major geological hazards faced in the construction and operation of infrastructure such as tunnels, subways, and municipal roads. These cavities may be formed by karst development, soil loss, mineral mining, or historical human activities, and often contain high-pressure underground water, posing a serious threat to engineering safety. Once exposed or leaking or collapsing during construction or use, it can easily cause disastrous accidents such as water bursting, mud gushing, ground subsidence, and even collapse.

[0003] The existing technology for treating underground cavities in water-rich strata has the following problems: 1. Inaccurate identification: mainly relying on high-resolution seismic reflection, geological radar and other geophysical exploration methods, combined with a small amount of drilling for verification, the results depend on manual interpretation, making it difficult to accurately grasp the true shape, size and location of the cavity, and unable to form a reliable three-dimensional model. 2. Poor treatment effect: especially in water-rich strata, conventional grouting materials are easily washed away by water flow, and the drilling layout lacks targetedness, resulting in many blind areas and unstable effects, with high construction risks. 3. Disjointed process: from detection, design to construction, the information transmission between each link is not smooth, relying on experience-based decision-making, lacking systematic intelligent control, and unable to achieve precise targeted treatment.

[0004] Therefore, a method for targeted treatment of underground cavities in water-rich strata is provided to solve the above problems. SUMMARY

[0005] To solve the above problems, the present application provides a method for targeted treatment of underground cavities in water-rich strata, which combines artificial intelligence identification, three-dimensional accurate positioning and active three-dimensional water stopping to achieve safe and efficient targeted treatment of underground cavities in water-rich strata.

[0006] To achieve the above purpose, the present application provides a method for targeted treatment of underground cavities in water-rich strata, comprising the following steps: S1: obtaining geological survey data; S2: analyzing the geological survey data based on an underground cavity identification model to identify the spatial position, geometric shape and size of the underground cavity in the water-rich strata; the underground cavity identification model is constructed based on an improved YOLOv8 algorithm; S3: constructing a three-dimensional geological model containing the cavity, designing a drilling trajectory from the surface or existing structure to the cavity based on the spatial position and shape of the cavity identified in S2, and constructing a targeted drilling along the trajectory to form a treatment channel; S4: Centered on the identified cavity, curtain grouting holes are arranged around its periphery, and rapid-setting double-liquid slurry is injected to form a closed three-dimensional waterproof curtain; S5: Through partial targeted drilling, groundwater in the three-dimensional waterproof curtain is guided and discharged, and part of the silt or loose filling in the cavity is removed; S6: Through the remaining targeted drilling, self-compacting filling material is pumped into the cavity for main backfilling, and sleeve valve pipe grouting technology is used to supplement and reinforce the cracks in the surrounding rock-soil mass of the cavity.

[0007] S7: Construction inspection holes are drilled in the treatment area to verify the filling and reinforcement effect, and a monitoring system is installed to monitor the stability of the stratum.

[0008] Preferably, the geological survey data obtained in S1 includes one or more of high-resolution seismic reflection wave data, geological radar data, and cross-hole CT data.

[0009] Preferably, the underground cavity recognition model specifically includes: A 3D deformable convolution layer is inserted at the 4th layer of the original backbone of YOLOv8, with a convolution kernel size of 3×3×3, which is used to adaptively capture the irregular geometry of underground cavities; A branch for small-scale cavity detection is added to the detection head, with a convolution kernel size of 3×3×3, which is used to improve the recognition ability of small-scale cavities; A physical consistency term is introduced in the loss function to constrain the physical logic consistency between the predicted cavity depth of the model and the wave velocity information in the geological survey data.

[0010] Preferably, the 3D deformable convolution layer specifically includes: A main convolution path and an offset prediction branch are set in parallel; The offset prediction branch is composed of two cascaded 3D standard convolution layers, where the convolution kernel size of the second layer is 3×3×3, and the number of output channels is 81, which is used to generate an offset field consistent with the spatial dimensions of the input feature map, containing scalar offsets of 27 sampling points in three-dimensional space; The main convolution path uses a deformable convolution kernel of 3×3×3, which samples the input feature map based on the offset field output by the offset prediction branch, and calculates the feature values of the sampling points through bilinear interpolation.

[0011] Preferably, the physical consistency term is constructed and calculated as follows: Based on the wave velocity-depth relationship of seismic waves propagating in different media, a physical mapping function d=f(v) is established between the cavity depth d and the seismic wave velocity v; The physical consistency term L_phy is composed of the difference between the model-predicted cavity depth d_pred and the theoretical depth d_phy=f(v_obs) calculated by the physical mapping function from the measured wave velocity v_obs; The physical consistency term L_phy and the target detection loss function L_det of the model jointly participate in the model training, and the total loss function is Wherein, λ is a hyperparameter for balancing the weights of the two terms.

[0012] Preferably, the design of the accurate drilling trajectory in S3 includes simulating the drilling path in the three-dimensional geological model, calculating the trajectory length, evaluating the spatial relationship with the underground obstacles, and optimizing the drilling incidence angle.

[0013] Preferably, the rapid-setting type double slurry in S4 is a cement-silicate double slurry, and the setting time is regulated between 30 seconds and 3 minutes according to the formation water flow and pressure; The construction of the three-dimensional water stop curtain adopts a segmented and intermittent grouting process, and the grouting pressure is dynamically adjusted according to real-time monitoring data.

[0014] Preferably, the self-compacting filling material in S6 is self-compacting concrete. The fractures around the cavity are supplemented and reinforced, specifically by embedding sleeve valve pipes in the targeted drilling holes, and repeatedly, quantitatively and controllably grouting the fractures at different depths and positions through the one-way valves on the sleeve valve pipes, and the grouting material is a cement-based slurry.

[0015] Preferably, step S7 verifies the drilling holes to implement coring, optical imaging and / or permeability test in the holes, quantitatively obtains the integrity, cementation and impermeability parameters of the filling body, and compares the obtained parameters with the preset threshold to determine whether the filling and reinforcement effect is qualified.

[0016] Preferably, the monitoring system arranged in S7 includes stress sensors embedded in the filling body, water pressure gauges arranged on the inside and outside of the curtain, and settlement monitoring points set on the ground surface, and the monitoring system transmits and intelligently analyzes and warns the data in real time through an Internet of Things platform.

[0017] Therefore, the above-mentioned underground cavity targeted treatment method in water-rich strata has the following beneficial effects: (1) The underground cavity identification model based on the improved YOLOv8 algorithm can automatically and accurately identify the spatial position, geometric shape and scale of the cavity in the water-rich strata, effectively overcoming the errors and limitations of traditional manual interpretation, and providing reliable basis for subsequent targeted treatment.

[0018] (2) The present application constructs a three-dimensional geological model based on the recognition result and designs a precise drilling trajectory, so that the targeted drilling can directly reach the core area of the cavity, significantly improves the precision of treatment, and avoids invalid drilling and resource waste caused by blind construction.

[0019] (3) The present application constructs a three-dimensional water stop curtain around the cavity to isolate the interference of external flowing water on the treatment process, greatly reduces the risk of water gushing, and provides safe and stable construction conditions for subsequent internal operation of the cavity.

[0020] (4) The present application uses targeted drilling to implement step-by-step drainage and dredging, self-compacting material backfilling and fissure grouting reinforcement, forming a multi-level treatment system from the cavity interior to the peripheral rock-soil mass, and comprehensively improving the compactness and long-term stability of the treatment area.

[0021] (5) The present application realizes instant verification and continuous tracking of the treatment effect through the layout of construction inspection holes and long-term monitoring systems, ensuring the quality controllability and safety reliability of the treatment project.

[0022] The technical solutions of the present application will be further described in detail below through the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is a flowchart of a kind of underground cavity targeted treatment method in water-rich stratum in the present application; Figure 2 is a water-rich stratum underground cavity in the embodiment of the present application; Figure 3 is a 3D deformable convolution layer structure diagram in the embodiment of the present application; Figure 4 is a branch structure diagram of small-scale cavity detection in the embodiment of the present application; Figure 5 is a sleeve valve pipe structure schematic diagram in the embodiment of the present application; Figure 6 is a stress sensor embedding process schematic diagram in the embodiment of the present application, wherein (a) is a positioning schematic diagram in stress sensor embedding process, (b) is a slotting schematic diagram in stress sensor embedding process, (c) is a embedding schematic diagram in stress sensor embedding process, (d) is a lead schematic diagram in stress sensor embedding process, (e) is a backfilling schematic diagram in stress sensor embedding process, (f) is a ramming schematic diagram in stress sensor embedding process. DETAILED DESCRIPTION

[0024] The following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the present application.

[0025] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the ordinary meanings as understood by a person of ordinary skill in the art to which the present application pertains.

[0026] The terms such as “comprising” or “including” or similar words used in the present application mean that the elements before the word encompass the elements listed after the word, and do not exclude the possibility of also encompassing other elements. The terms “in”, “out”, “up”, “down” and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. When the absolute position of the described object changes, the relative positional relationship may also change accordingly. In the present application, unless otherwise explicitly specified and limited, the term “attached” and the like should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrated; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0027] Embodiments A method for targeted treatment of underground cavities in water-rich strata, as shown in Figure 1 , comprises the following steps: S1: Obtain geological survey data; the geological survey data includes one or more of high-resolution seismic reflection wave data, geological radar data, and cross-hole CT data; the underground cavity is as shown in Figure 2 .

[0028] S2: Based on an underground cavity identification model, analyze the geological survey data to identify the spatial position, geometric shape and scale of the underground cavity in the water-rich strata; the underground cavity identification model is constructed based on an improved YOLOv8 algorithm; The underground cavity identification model specifically includes: A 3D deformable convolution layer is inserted at the 4th layer of the original backbone of YOLOv8, as shown in Figure 3 , with a convolution kernel size of 3x3x3, which is used to adaptively capture the irregular geometric shape of the underground cavity; the offset learning rate of the deformable convolution layer is set to 0.01.

[0029] The 3D deformable convolution layer specifically includes: A main convolution path and an offset prediction branch are arranged in parallel; The offset prediction branch is composed of two cascaded 3D standard convolution layers, wherein the convolution kernel size of the second layer convolution layer is 3x3x3, and the output channel number thereof is 81, so as to generate an offset field which is consistent with the spatial dimension of the input feature map and contains 27 sampling points in the three-dimensional space. The main convolution path adopts a 3x3x3 deformable convolution kernel, samples the input feature map according to the offset field output by the offset prediction branch, and calculates the feature values of the sampling points through bilinear interpolation.

[0030] To solve the problems of missed detection and inaccurate positioning caused by small-scale cavities and weak features, a branch for small-scale cavity detection is added to the detection head, as shown in FIG. 5, wherein the convolution kernel size of the branch is 3x3x3, and the branch is used to improve the recognition ability of small-scale cavities. Figure 4 Specifically, the branch is not connected at the end of the network, but is directly connected to the output end of the shallow layer of the feature pyramid. On the high-resolution feature map, the branch uses a standard convolution layer with a convolution kernel size of 3x3x3 for processing, and a lightweight prediction head is connected after the convolution layer to independently output the detection results (including the boundary box coordinates, confidence and category) for small-scale cavities. Thus, a "divide and conquer strategy" for the detection task is realized, and the special branch is responsible for small-scale targets, avoiding the performance loss caused by the competition of different scale targets for the same set of network weights. Together with the original detection head, the branch forms a multi-scale collaborative detection system, so that the model has full-scale cavity recognition ability.

[0031] A physical consistency term is introduced into the loss function to constrain the model-predicted cavity depth to be consistent with the wave velocity information in the geological survey data in terms of physical logic.

[0032] The physical consistency term is constructed and calculated as follows: Based on the wave velocity-depth relationship of seismic waves propagating in different media, a physical mapping function d=f(v) is established between the cavity depth d and the seismic wave velocity v. The physical consistency term L_phy is composed of the difference between the model-predicted cavity depth d_pred and the theoretical depth d_phy=f(v_obs) calculated by the physical mapping function from the measured wave velocity v_obs. The physical consistency term L_phy and the target detection loss function L_det of the model jointly participate in the model training, and the total loss function is wherein λ is a hyperparameter for balancing the weights of the two terms.

[0033] The original geological survey data from multiple actual engineering projects are used to train the underground cavity recognition model, specifically including: The original data is standardized and pre-processed to form a unified and standardized training sample, mainly including: (1) format unification: converting different sources and formats of original data (such as SEG-Y, DZT, etc.) into a general tensor format (such as NumPy array or PyTorch Tensor) that can be processed by the model. (2) Normalization: normalize the gray value or amplitude value of each data volume to make its distribution range uniform to the interval [0, 1] or [-1, 1], so as to accelerate the model convergence and improve the training stability. (3) Size regularization: all input data are cropped or padded to a unified size (for example, 128x128x128 voxels) to meet the model input requirements. For two-dimensional GPR profiles, pseudo-three-dimensional data volumes are constructed by stacking or interpolation.

[0034] The pre-drilling verification, borehole television, and expert interpretation results are used as the standard to accurately label the underground cavities in the pre-processed three-dimensional data volume. The labeling results are converted into a label file matching the model output format, such as YOLO format or COCO format, to provide the true value for subsequent loss calculation. To improve the generalization ability and robustness of the model and prevent overfitting, strict data augmentation strategies are applied to the training set, including geometric transformation and numerical perturbation. The data set is divided into training set and validation set in the ratio of 7:3 to ensure the fairness of evaluation.

[0035] S3: Construct a three-dimensional geological model containing cavities, based on the spatial location and morphology of the cavities identified in S2, design a drilling trajectory from the surface or existing structure to the cavity, and construct a target drilling along the trajectory to form a treatment channel; The design of accurate drilling trajectory includes simulating the drilling path in the three-dimensional geological model, calculating the trajectory length, evaluating the spatial relationship with underground obstacles, and optimizing the drilling incidence angle.

[0036] Specifically, the key parameters of the underground cavity location, geometry, and scale output by the underground cavity identification model in S2 are integrated into a three-dimensional geological modeling software (such as GOCAD, GMS, FLAC3D, etc.) as core constraints. On this basis, by integrating drilling, geophysical prospecting, and other original survey data, a high-precision, integrated three-dimensional geological model is constructed, which includes the target cavity, surrounding strata, underground hydrological information, and existing underground structures. This model is the digital foundation for all subsequent design and analysis. In the three-dimensional geological model, the drilling trajectory is intelligently planned according to the following design requirements; the optimal three-dimensional path from the surface or existing structures (such as tunnel sidewalls, construction shafts) to the target cavity is calculated, striving for the shortest path to save engineering quantity and cost; the trajectory must accurately avoid all known underground obstacles, such as municipal pipelines, existing pile foundations, confined water pockets, and large fissure zones. According to the main functions of the drilling in subsequent treatment (such as drainage, dredging, grouting), the incident angle and end position are optimized. For example, the endpoint of the grouting hole should be located at the center of the cavity bottom, and the angle should be beneficial for the diffusion of the grout; the drainage hole should be beneficial for water flow collection and export. The trajectory design needs to comply with the process limitations of the drilling equipment, such as the maximum drilling depth and the allowable trajectory bending curvature. The drilling and measurement system and the directional drilling system are used for construction. During drilling, the hole bottom measurement tool transmits the attitude data of the drilling, such as inclination, azimuth, and depth, back to the ground control system in real time. The real-time collected drilling trajectory data are compared with the designed theoretical trajectory in real time. Once there is a deviation, the system immediately calculates the correction scheme and dynamically controls the drilling direction by adjusting the tool face angle of the drill bit, forming a real-time closed-loop control of measurement, comparison, and correction, ensuring that the drilling can strictly follow the designed trajectory. After drilling is completed, the hole television and other equipment can be used to finally confirm the quality of the passage and the situation inside the cavity. Thus, one or more efficient and accurate targeted treatment passages for subsequent drainage, dredging, grouting, and reinforcement are formed.

[0037] S4: A curtain grouting hole is arranged around the identified cavity, and a rapid-setting double-liquid slurry is injected to form a closed three-dimensional waterproof curtain. The rapid-setting double-liquid slurry is a cement-silicate double-liquid slurry, and its setting time is controlled between 30 seconds and 3 minutes according to the stratum water flow and pressure; the construction of the three-dimensional waterproof curtain adopts a segmented and intermittent grouting process, and the grouting pressure is dynamically adjusted according to real-time monitoring data.

[0038] Specifically, the three-dimensional form of the cavity identified by S2 is taken as the center, and one or more rows of curtain grouting holes are scientifically arranged on the peripheral space of the cavity according to the ring or spherical shell concept. The hole arrangement design not only considers the plane position, but also strictly controls the vertical depth and spacing to ensure that the final curtain can completely wrap the target cavity in three-dimensional space and form a complete waterproof bucket. The cement-silicate double liquid slurry is used as the curtain material, and the top-down or bottom-up segmented grouting method is used for each curtain grouting hole. The drilling hole is divided into several independent segments by using the grouting plug, and grouting is carried out segment by segment. It ensures that different depth strata can be fully and uniformly perfused, avoiding the loss of slurry along the thinnest channel. During the grouting process, an intermittent operation mode of grouting-pause-grouting is adopted. During the pause period, the injected slurry is allowed to fully solidify, strength growth, and consolidate the stratum, thereby providing higher starting pressure and better constraint conditions for the next round of grouting. Real-time pressure-flow monitoring is carried out throughout the grouting process. The grouting pressure is not a fixed value, but is dynamically adjusted according to the stratum absorption, previous grouting records, and real-time monitoring data (such as ground uplift, adjacent hole string slurry). Follow the principle of low pressure slow penetration, high pressure splitting. The initial lower pressure ensures permeability, and the pressure is appropriately increased in the later stage to ensure the density and continuity of the curtain.

[0039] S5: Through partial target drilling, guide and drain the groundwater in the three-dimensional waterproof curtain and remove part of the silt or loose filling in the cavity; Specifically, the partial target drilling formed in step S3 is selected as a special drainage channel, and a deep well pump or a submersible pump is lowered to the bottom of the cavity through the selected target drilling. Start the water pump to continuously drain the groundwater in the three-dimensional waterproof curtain closed space and actively lower the water level inside the cavity. During the drainage process, the water output and water quality changes are monitored in real time until the water changes from turbidity to clear, the flow significantly decreases and tends to be stable, indicating that the free water in the cavity has been basically drained.

[0040] After the drainage is completed, this part of the target drilling will be used as a dredging channel. A high-pressure jet pipe is lowered through the drilling hole, and a high-pressure water jet is used to flush, mix and liquefy the silt and soft plastic filling in the cavity, so that it is converted into suspended slurry. Synchronously use the slurry pump or air lifting pump to establish a negative pressure suction system through the drilling hole, and continuously pump out the liquefied slurry mixture to the ground. Follow the cycle operation mode of "flushing-liquefying-suctioning", and clean up in regions and levels until the pumped-out material is mainly clear water and no obvious particulate matter is brought out, indicating that most of the loose material in the cavity has been removed.

[0041] After the dredging operation is completed, the amount of water and slurry discharged can be compared with the volume of the cavity identified by S2 to indirectly evaluate the thoroughness of the dredging.

[0042] S6: Pumping self-compacting filling material into the cavity through the rest of the targeted drill holes to backfill the main body, and using sleeve valve pipe grouting technology to supplement the reinforcement of the cracks in the surrounding rock-soil mass of the cavity.

[0043] The self-compacting filling material is self-compacting concrete. The cracks in the surrounding rock-soil mass of the cavity are supplemented and reinforced, specifically: sleeve valve pipes are embedded in the targeted drill holes, and through the one-way valves on the sleeve valve pipes, repeated, quantitative, and controllable grouting is carried out on cracks of different depths and different positions, and the grouting material is a cement-based slurry.

[0044] Specifically, self-compacting concrete is used as the main backfill material. Its mix proportion is optimized to have high fluidity, segregation resistance, and good self-filling properties, and it can fill the entire irregular cavity without vibration. Through the remaining targeted drill holes reserved in step S3 and not used for dredging, the mixed self-compacting concrete is continuously pumped into the drained and emptied cavity from the ground mixing station through a high-pressure concrete delivery pump. The pumping process is continuous until the full slurry flows smoothly from the reserved exhaust hole or observation hole, indicating that the cavity has been completely filled and compacted. This process forms the core structure that resists ground deformation.

[0045] After the main backfill body is formed, sleeve valve pipe grouting technology is used to supplement the reinforcement to close the cracks in the surrounding rock-soil mass of the cavity and form an integrated anti-seepage reinforcement ring.

[0046] Specifically, the sleeve valve pipe system is laid out: in the pre-designed targeted drill holes, sleeve valve pipes are lowered. The sleeve valve pipe is a special grouting pipe, as shown in Figure 5 The pipe wall is provided with one-way grouting valves at certain intervals, and the outside is protected by a double-layer rubber sleeve. The sleeve shell material is injected between the sleeve valve pipe and the drill hole wall to firmly fix it in the hole. The role of the sleeve shell material is to close the annular gap and force the subsequent grouting pressure to only open the specified sleeve valve, but not to flow randomly from the outside of the pipe. Segmental and controllable grouting operation: using a double-plug grouting core pipe, segmental grouting is carried out from the bottom of the sleeve valve pipe upwards. By blocking both ends of a certain section, independent and accurate grouting of the cracks in the stratum of the corresponding depth is achieved. For each grouting section, the grouting pressure and grouting volume are strictly controlled. According to the real-time adjustment of the geological conditions, it is ensured that the slurry can fully penetrate into the cracks, while avoiding high pressure that may cause splitting of the stratum. Repeated grouting capability: thanks to the one-way valve design on the sleeve valve pipe, if the effect of the initial grouting of a section does not meet the expected result, the section can be repeatedly grouted twice or more times to ensure the reliability and flexibility of the reinforcement.

[0047] The grouting material used is a cement-based slurry. The material has the advantages of wide source, high strength, good durability and low cost. The water-cement ratio and admixture can be dynamically adjusted according to the water absorption rate of the stratum and the reinforcement requirements.

[0048] S7: Constructing inspection holes in the treatment area, verifying the filling and reinforcement effect, and laying a monitoring system to monitor the stability of the stratum.

[0049] By drilling verification holes, core sampling, optical imaging and / or permeability tests are carried out in the holes to quantitatively obtain parameters of the filling body integrity, cementation and impermeability, and compare the obtained parameters with preset threshold values to determine whether the filling and reinforcement effect is qualified.

[0050] The monitoring system laid includes stress sensors buried in the filling body, water pressure gauges laid on the inside and outside of the curtain, and settlement monitoring points set on the ground surface. The monitoring system transmits and intelligently analyzes and warns the data in real time through an Internet of Things platform. As shown in FIG. 8, the specific process of burying the stress sensor is shown. Figure 6

[0051] Therefore, the present application adopts the above-mentioned method for targeted treatment of underground cavities in water-rich strata, through the systematic technical path of artificial intelligence precise identification, three-dimensional model intelligent decision-making, three-dimensional water stop active prevention and control, targeted drilling precise construction and step-by-step treatment reinforcement, a complete treatment closed loop is formed, the technical leap from traditional experience treatment to modern precise treatment is realized, and the safety, reliability and economy of the treatment of underground cavities in water-rich strata are significantly improved, providing effective technical support for the safe development of underground space.

[0052] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application and not to limit it, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that: it can still modify or equivalently replace the technical solutions of the present application, and these modifications or equivalent replacements also cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present application.​

Claims

1. A method for targeted treatment of a subterranean void in a water- enriched formation, comprising: The method comprises the following steps: S1: obtaining geological survey data; S2: analyzing the geological survey data based on an underground cavity identification model to identify the spatial position, geometric shape and scale of the underground cavity in the water-rich stratum; the underground cavity identification model is constructed based on an improved YOLOv8 algorithm; S3: constructing a three-dimensional geological model containing the cavity, designing a drilling trajectory from the ground or existing structure to the cavity based on the spatial position and shape of the cavity identified in S2, and constructing a target drilling along the trajectory to form a treatment channel; S4: taking the identified cavity as the center, arranging curtain grouting holes in the periphery of the cavity and injecting rapid-setting double-liquid slurry to form a closed three-dimensional waterproof curtain; S5: through part of the target drilling, guiding and draining the underground water in the three-dimensional waterproof curtain and removing part of the silt or loose filling in the cavity; S6: through the remaining target drilling, pumping self-compacting filling material into the cavity for main backfilling, and using sleeve valve pipe grouting technology to supplement and reinforce the fissures in the rock-soil mass around the cavity; S7: constructing inspection holes in the treatment area to verify the filling and reinforcement effect, and arranging a monitoring system to monitor the stability of the stratum.

2. The method of claim 1, wherein the method further comprises: The geological survey data obtained in S1 includes one or more of high-resolution seismic reflection wave data, geological radar data, and cross-hole CT data.

3. The method of claim 2, wherein the method further comprises: The underground cavity identification model specifically comprises: A 3D deformable convolution layer is inserted at the 4th layer of the original backbone of YOLOv8, with a convolution kernel size of 3×3×3, which is used to adaptively capture the irregular geometric shape of the underground cavity; A branch for small-scale cavity detection is added to the detection head, with a convolution kernel size of 3×3×3, which is used to improve the recognition ability of small-scale cavities; A physical consistency term is introduced in the loss function to constrain the consistency of the predicted cavity depth and the wave velocity information in the geological survey data in physical logic.

4. The method of claim 3, wherein the method further comprises: The 3D deformable convolution layer specifically comprises: A main convolution path and an offset prediction branch are arranged in parallel; The offset prediction branch is composed of two cascaded 3D standard convolution layers, where the convolution kernel size of the second layer is 3×3×3, and the number of output channels is 81, which is used to generate an offset field consistent with the spatial dimension of the input feature map and containing the scalar offset of 27 sampling points in three-dimensional space; The main convolution path uses a deformable convolution kernel of 3×3×3, which samples the input feature map according to the offset field output by the offset prediction branch and calculates the feature values of the sampling points through bilinear interpolation.

5. The method of claim 4, wherein the method further comprises: The construction and calculation method of the physical consistency term is as follows: Based on the wave velocity-depth relationship of seismic waves propagating in different media, a physical mapping function d=f(v) is established between the cavity depth d and the seismic wave velocity v; The physical consistency term L_phy is composed of the difference between the predicted cavity depth d_pred and the theoretical depth d_phy=f(v_obs) calculated by the physical mapping function from the measured wave velocity v_obs; The physical consistency term L phy and the target detection loss function L det of the model jointly participate in the model training, and the total loss function is where λ is a hyperparameter for balancing the weights of the two terms.

6. The method of claim 2, wherein the method further comprises: The design of the precise drilling trajectory in S3 includes simulating the drilling path in the three-dimensional geological model, calculating the trajectory length, evaluating the spatial relationship with underground obstacles, and optimizing the drilling incidence angle.

7. The method of claim 2, wherein: The S4 medium setting type double slurry is a cement-sodium silicate double slurry, and the setting time is regulated at 30 seconds to 3 minutes according to the formation water flow and pressure; The construction of the three-dimensional water stop curtain adopts a segmented and intermittent grouting process, and the grouting pressure is dynamically adjusted according to real-time monitoring data.

8. The method of claim 1, wherein: The S6 self-compacting filling material is self-compacting concrete; The fractures around the cavity are supplemented and reinforced, specifically, sleeve valve pipes are buried in the targeted drill holes, and through the one-way valves on the sleeve valve pipes, the fractures at different depths and different positions are repeatedly, quantitatively and controllably grouted, and the grouting material is a cement-based slurry.

9. The method of claim 1, wherein: In step S7, a verification hole is drilled, coring, optical imaging and / or permeation test are performed in the hole, the integrity, cementation and impermeability parameters of the filling body are quantitatively obtained, and the obtained parameters are compared with the preset threshold value to determine whether the filling and reinforcement effect is qualified.

10. The method of claim 1, wherein: The monitoring system arranged in S7 includes stress sensors buried in the filling body, water pressure gauges arranged on the inside and outside of the curtain, and settlement monitoring points set on the ground surface, and the monitoring system transmits and intelligently analyzes and warns the data in real time through an Internet of Things platform.

Images