Deformation monitoring method for underground tunnels in ultra-shallow buried and high-flow strata

Abstract

The present invention belongs to the technical field of subway tunnel excavation deformation monitoring, and specifically discloses a subway tunnel excavation deformation monitoring method based on ultra-shallow buried high-flow strata. In the early stage of construction, geological radar detection is carried out on the construction impact range to generate a geological risk distribution map, and then the geological risk distribution map is used to divide the construction impact area into risk levels and implement differentiated layout of monitoring points. On the premise of ensuring the monitoring effect, the deformation capture capability of key areas is improved and the monitoring redundancy of low-risk areas is reduced. During the construction process, the deformation data collected from the monitoring points in each sub-area are dynamically analyzed to identify the jump and stability characteristics of the deformation, determine whether there are monitoring blind spots or redundant areas, and trigger the encryption or migration operations of the monitoring points accordingly, thereby realizing the transformation of the monitoring system from static layout to dynamic evolution, and effectively ensuring the real-time matching between the monitoring point layout and the formation deformation evolution.

Description

Technical Field

[0001] The present invention belongs to the technical field of subway tunnel excavation deformation monitoring, and specifically discloses a subway tunnel excavation deformation monitoring method based on ultra-shallow buried high-flow strata. Background Art

[0002] With the acceleration of urbanization, underground space development plays an important role in alleviating traffic pressure. As a core component of the underground transportation system, subway projects are often limited by the complex environmental conditions in the urban core area. It is difficult to use open-cut construction and must be implemented using the underground excavation method.

[0003] These projects, often used in subway entrances and exits, transfer passages, and underground commercial connections, often face complex geological conditions such as ultra-shallow burials and water-rich, soft strata. These structures are characterized by thin overburden, loose strata, high permeability, and poor self-stability. Construction can easily lead to risks such as surrounding rock deformation, settlement, convergence, and even sudden water inrush and local collapse. Therefore, monitoring ground deformation during construction is a key technical measure to ensure the safe and smooth implementation of these projects.

[0004] Deformation monitoring during tunnel excavation typically relies on deploying monitoring points within the construction area to capture key data on structural deformation and ground response. Currently, mature technical solutions exist to support this type of monitoring, such as the Chinese invention patent CN104564128B, which discloses a deformation monitoring method for shallow tunnel construction. This method, while simultaneously carrying out tunnel excavation and initial support, deploys multiple sets of support status monitoring points on the completed support structure and multiple settlement monitoring points on the surface. By continuously collecting monitoring data, dynamic monitoring of the tunnel support structure's stability and surface settlement is achieved.

[0005] However, in the actual application of this technical solution, the layout of monitoring points is mainly static, uniform or empirical, that is, they are usually arranged at fixed intervals based on engineering experience or specification requirements, without fully considering the differences in geological conditions at different locations in the construction area, resulting in a mismatch between the distribution of monitoring points and the actual risk distribution. This "one-size-fits-all" layout model can easily lead to insufficient density of monitoring points in high-risk areas and redundant layout in low-risk areas, affecting monitoring efficiency and cost control.

[0006] In addition, the underground excavation construction process has significant dynamic evolution characteristics. With the advancement of the excavation face, the construction of support structures and the redistribution of stratum stress, the environment in which the monitoring points are located is also constantly changing. Continuing to use the initial fixed layout method without adjustment will lead to a mismatch between the layout of the monitoring points and the actual deformation development, and will not be able to accurately reflect the true response status of the structure and stratum, thereby weakening the early warning capability of the monitoring system and increasing construction safety risks. Summary of the Invention

[0007] To this end, one purpose of an embodiment of the present application is to provide a method for monitoring deformation of underground excavation in subway tunnels based on ultra-shallow buried and high-flow strata. By differentially arranging monitoring points based on the geological exploration results of the construction area before construction, and dynamically optimizing and adjusting the layout of monitoring points during the construction process, it is ensured that the monitoring system can reflect the structural deformation and stratum response status in real time and accurately, thereby solving the problems mentioned in the background technology.

[0008] The purpose of the present invention can be achieved by the following technical solutions: a method for monitoring deformation of underground excavation in ultra-shallow buried high-flow strata subway tunnels, comprising the following steps: (1) demarcating the construction impact area with the underground excavation construction face as the center, performing geological radar scanning on the construction impact area to extract geological distribution characteristics including stratum interface distribution, water content distribution and rock mass crack distribution, and performing geological unit division and risk factor marking to generate a geological risk distribution map.

[0009] (2) According to the geological risk distribution map, the construction impact area is divided into high-risk sub-areas, transition sub-areas and stable sub-areas, and the initial monitoring network is deployed in each sub-area in a differentiated manner.

[0010] (3) During the tunneling construction process, deformation data of each monitoring point is collected in real time to construct a spatiotemporal synchronous deformation dataset.

[0011] (4) Based on the spatial distribution characteristics of the deformation data set, the monitoring blind spots and monitoring redundancies are identified. This triggers the dense deployment of monitoring points when the monitoring blind spots are identified, and triggers the migration of monitoring points when the monitoring redundancies are identified.

[0012] (5) After the dynamic adjustment of the monitoring network is completed, deformation trend analysis is performed based on the time series deformation data of the monitoring points in each sub-area during the construction process, thereby initiating a graded deformation warning from local to global level.

[0013] Combining all the above technical solutions, the positive effects of the present invention are as follows: 1. In the early stage of construction, the present invention generates a geological risk distribution map by conducting geological radar detection on the construction impact range, and then uses the geological risk distribution map to divide the construction impact area into risk levels and implement differentiated layout of monitoring points, avoiding the waste of resources caused by traditional uniform layout. On the premise of ensuring the monitoring effect, it improves the deformation capture ability of key areas and reduces the monitoring redundancy of low-risk areas, which is conducive to improving the overall monitoring efficiency.

[0014] 2. During the construction process, the present invention dynamically analyzes the deformation data collected from the monitoring points in each sub-area to identify the jump and stability characteristics of the deformation, judge whether there are blind spots or redundant areas for monitoring, and trigger the encryption or migration operations of the monitoring points accordingly, thereby realizing the transformation of the monitoring system from static layout to dynamic evolution, effectively ensuring the real-time matching between the layout of the monitoring points and the evolution of stratum deformation, and providing reliable data support for subsequent deformation warnings.

[0015] 3. The present invention performs trend analysis on the time-series deformation data of monitoring points in each risk sub-area, identifies its deformation rate and combines it with the deformation coverage to trigger a multi-level linkage early warning mechanism from local to global, achieving a step-by-step response from local anomalies to overall abnormal risk warnings, enhancing the logic and accuracy of the early warning system, and helping to take targeted engineering measures according to the warning level, thereby improving safety management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0017] Figure 1 This is a diagram of the steps for implementing the method of the present invention.

[0018] Figure 2 This is a schematic diagram showing the division of risk areas based on the geological risk distribution map in the present invention.

[0019] Figure 3 Schematic diagram of the division of adjacent areas corresponding to monitoring points in the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] See also Figure 1 As shown, the present invention proposes a method for monitoring deformation of underground excavation in ultra-shallow buried high-flow strata for subway tunnels, which includes the following steps: (1) demarcating the construction impact area with the underground excavation construction face as the center, scanning the construction impact area with geological radar to extract geological distribution characteristics including stratum interface distribution, water content distribution and rock mass crack distribution, and performing geological unit division and risk factor marking to generate a geological risk distribution map.

[0022] It is important to know that the tunnel face is where excavation operations are carried out directly. During underground engineering construction, the rock and soil in front of the tunnel face are the first to be affected by construction activities. Therefore, it is the key starting point for assessing the impact of construction. As the excavation work progresses, the position of the tunnel face continues to move forward, which means that the construction impact area is also constantly updated. Taking the tunnel face as the center can ensure that the monitoring range always covers the area most likely to deform at the moment.

[0023] When delineating the construction impact area, the measured data of historical similar engineering projects and engineering experience can be combined to analyze the impact range of construction activities on the surrounding strata, and based on this, a reasonable construction impact radius can be determined, and the construction impact area of the current project can be delineated based on this.

[0024] In the preferred implementation of the above scheme, a geological radar scan is performed on the construction-affected area to extract geological distribution characteristics including stratum interface distribution, water content distribution and rock fracture distribution, including the following: multiple parallel scanning lines are arranged along the axis of the underground tunnel, and each scanning line covers the lateral boundary of the construction-affected area.

[0025] It should be noted that in underground excavation projects, stratum deformation mainly propagates along the direction of tunnel face advancement. The geological scanning method with axial layout and lateral coverage can effectively capture the impact range of construction disturbance on the stratum. Multiple scanning lines can enhance the spatial continuity and reliability of the data, avoiding the one-sidedness caused by a single path.

[0026] Use geological radar equipment to perform high-resolution geological scanning on each scanning line, collect time series data of underground medium reflection waves, and perform image conversion to obtain geological radar images of the construction-affected area.

[0027] The above-mentioned method converts the original time domain signal of radar detection into a visual geological radar image, providing basic data for subsequent stratigraphic boundary identification, lithology discrimination, water content analysis and fracture detection.

[0028] In the generated geological radar image, lines with varying reflection intensities are identified as the boundaries between different strata, and the spatial distribution of the strata is determined based on the positions of these boundaries.

[0029] It is important to understand that differences in physical properties such as mineral composition, density, and water content between different strata result in inconsistencies in their dielectric constants and electromagnetic wave propagation characteristics, which in turn manifest as variations in reflection intensity during geological radar detection. By identifying and tracking reflection intensity in radar images, the locations of stratum interfaces can be effectively determined, thereby revealing the spatial distribution of strata.

[0030] The geological type of the strata is determined based on the analysis of the reflected wave properties of different strata in the geological radar images, and areas with the same stratum distribution and no gaps are classified as the same geological unit.

[0031] It should be pointed out that different geological types have significant differences in dielectric constants due to differences in their physical properties such as mineral composition, water content and density. This difference is manifested in geological radar detection as changes in the reflected waveform characteristics on the radar image, including amplitude strength, frequency, phase continuity, etc.

[0032] For example, sand layers or gravel layers usually have interfaces with high dielectric constant differences, which easily generate strong and continuous reflection signals.

[0033] Due to the high water content in the clay layer, the electromagnetic wave attenuation is obvious and the reflected signal is weak.

[0034] Therefore, by analyzing and identifying the reflection wave characteristics of each stratigraphic unit in the radar image, the geological type of the stratigraphic unit can be identified and classified.

[0035] The above mentioned methods achieve the preliminary zoning of the construction impact area by classifying the areas with the same stratigraphic distribution and no gaps into the same geological unit, so that the same geological unit should have geological consistency without obvious interruptions or transition areas.

[0036] For each divided geological unit, the attenuation properties of the reflected wave in the corresponding geological radar image are analyzed to determine the water content of the area.

[0037] It is worth noting that as the moisture content increases, the conductivity of the medium increases significantly, causing the electromagnetic wave to experience faster energy attenuation during propagation. By quantitatively analyzing the attenuation trend of the reflected wave signal in the geological radar image, the moisture content level in a specific area can be estimated.

[0038] The shape characteristics of the reflected waves are extracted from the geological radar images of each geological unit to identify the cracks and locate the distribution of the cracks.

[0039] It's important to note that fracture structures typically appear as discontinuous, strong reflection signals in geological radar images. These reflection waveforms exhibit characteristic geometric shapes, such as hyperbolic, discontinuous, or scattered patterns. By identifying and extracting the reflection wave's shape characteristics, the spatial distribution of fractures in the stratum can be effectively located.

[0040] The present invention selects the distribution of stratum interfaces, water content distribution and rock fissure distribution as the core geological distribution characteristics in the geological exploration of the construction impact area, mainly based on the following technical considerations: the distribution of stratum interfaces reflects the lithological characteristics of different strata and is the basic basis for judging the stability of the stratum. If there are potential risk strata such as fault fracture zones and weathered rock strata in the stratum, the excavation disturbance may cause the stratum to become unstable, thereby forming deformation hazards within the original stratum structure.

[0041] Moisture content distribution characterizes the distribution of water in underground media and has a significant impact on the physical and mechanical properties of rock and soil. In areas with high water content, rock and soil are prone to softening, rheological changes, and even localized liquefaction. This can lead to a decrease in the bearing capacity of the stratum, inducing geological hazards such as subsidence and sudden water inrush, significantly increasing the risk of deformation during construction.

[0042] The distribution of rock mass cracks reveals the integrity of the rock mass and the degree of structural damage. In areas with densely developed cracks, the surrounding rock strength is significantly reduced, the permeability is enhanced, and the connection between structural planes is weakened, which easily forms slip surfaces or collapse blocks, leading to deformation and instability of the surrounding rock.

[0043] These geological characteristics collectively constitute the primary sources of geological risk during underground excavation. Identifying these elements as key geological distribution features not only helps comprehensively characterize the geological conditions in the construction-affected area but also provides reliable data support for subsequent geological risk zoning.

[0044] In a further preferred implementation of the above scheme, a geological risk distribution map is generated as follows: the stratigraphic geological type information of each divided geological unit is extracted layer by layer and matched with the risk stratigraphic database. If the geological type of a certain layer matches the high-risk stratigraphic database, it is determined that the geological unit has a stratigraphic risk factor.

[0045] It should be added that the risk stratum database used in the present invention integrates a variety of typical high-risk stratum types, including but not limited to silty soil layers, silty sand layers, fault fracture zones, strongly weathered rock layers, etc. These strata have the characteristics of low strength, high water content, structural fragmentation or easy deformation, which can easily cause geological disasters such as landslides, water gushing, and mud bursts during underground engineering construction. The data sources of the database mainly include records of construction accidents caused by specific strata in historical engineering cases and the definition and classification standards of unfavorable strata in relevant industry specifications. Through multi-source information fusion and structured organization, a risk stratum knowledge base with practical engineering value is constructed.

[0046] In practical applications, the geological type of each stratum in each geological unit detected is intelligently compared with the entries in the database to automatically identify whether it contains high-risk strata.

[0047] The water content of each geological unit is compared with the set safe water content threshold. If the water content of a geological unit exceeds the threshold, it is determined that the geological unit has a water risk factor.

[0048] Based on the distribution position and length of cracks in each geological unit, the total length of cracks per unit area is calculated as the crack distribution density, and a safe crack distribution density comparison is set. If the crack distribution density of a geological unit is higher than the safe crack distribution density, it is identified that the geological unit has crack risk factors.

[0049] It should be pointed out that the above-mentioned safe moisture content thresholds and safe crack distribution density can be comprehensively determined based on construction safety specifications and geotechnical engineering investigation standards. These safety thresholds reflect the controlling boundary conditions of stratum stability during underground engineering construction and are important technical parameters for assessing whether the rock and soil may suffer from adverse engineering responses such as deformation due to high moisture content or structural fragmentation.

[0050] Plane modeling is performed based on the geometric boundaries of the construction-affected area, and geological units are identified based on the boundary contours of each geological unit in the constructed model. The identified risk factors are also annotated to generate a geological risk distribution map.

[0051] The above-mentioned geological risk distribution map can intuitively display the risk distribution of each geological unit in the construction impact area, thereby facilitating the efficient and accurate definition and classification of risk areas.

[0052] (2) According to the geological risk distribution map, the construction impact area is divided into high-risk sub-areas, transition sub-areas and stable sub-areas, and the initial monitoring network is deployed in each sub-area in a differentiated manner.

[0053] For an alternative implementation of the above scheme, see Figure 2 As shown, the construction impact area is divided into high-risk sub-areas, transition sub-areas and stable sub-areas according to the geological risk distribution map. The implementation is as follows: Based on the risk factors identified in each geological unit in the geological risk distribution map, the sub-areas are divided according to the following zoning judgment rules: High-risk sub-area: Two or more risk factors exist in a single geological unit at the same time.

[0054] Transitional sub-area: Only one type of risk factor exists in a single geological unit.

[0055] Stable sub-area: No risk factors are identified in the individual geological units.

[0056] The risk zone delineation method described above is based on a comprehensive assessment of multiple independent risk factors. Each risk factor represents a different type of potential threat, and by considering these factors together, a more comprehensive and accurate reflection of the overall risk level of a geological unit can be achieved.

[0057] In a further optional implementation of the above scheme, the initial monitoring network is deployed differentially in each sub-area, as shown in the following process: monitoring points are deployed in a triangular grid in the high-risk sub-area, and the node spacing between the monitoring points is a fixed preset value.

[0058] It should be noted that the triangular grid pattern is used to arrange monitoring points in high-risk sub-areas because the triangular grid layout method can provide higher spatial coverage and redundancy, and setting a smaller node spacing to achieve dense layout helps to improve spatial resolution and the timeliness of data feedback.

[0059] Monitoring points are arranged in a rectangular grid in the transition sub-area, and the spacing between monitoring point nodes is set as a multiple of the spacing between nodes in the high-risk sub-area.

[0060] It should be noted that the rectangular grid layout of monitoring points in the transition sub-area, while slightly inferior to the triangular grid in terms of spatial coverage uniformity and redundancy, offers advantages such as ease of deployment, clear logic, and ease of on-site implementation. The node spacing of monitoring points in this grid is set to an integer multiple of the node spacing in the high-risk sub-area, exemplarily set at 2. By properly controlling this multiple, we can maintain basic monitoring capabilities for potential deformation risks while effectively reducing the density of monitoring points and avoiding excessive investment in monitoring resources.

[0061] Monitoring points are arranged along the channel axis in the stable sub-region, and the node spacing of the monitoring points is set to be a multiple of the node spacing of the transition sub-region.

[0062] It should be noted that monitoring points are arranged along the channel axis in the stable sub-area, and the node spacing is set to a multiple of the node spacing in the transition sub-area. In this way, the linear monitoring network arranged along the axis can maintain the basic monitoring capability of the overall structural stability while reducing the number of monitoring points.

[0063] This invention achieves efficient resource allocation and utilization by adapting monitoring grid layouts to sub-areas with varying risk levels. For example, triangular grids, due to their high coverage and redundancy, are particularly suitable for high-risk areas requiring meticulous monitoring. Rectangular grids, on the other hand, simplify operational processes without compromising critical monitoring effectiveness, making them more suitable for transitional areas. This approach allows for increased monitoring density in high-risk areas to enhance monitoring, while appropriately reducing it in relatively stable areas, thereby optimizing resource allocation.

[0064] (3) During the tunneling construction process, deformation data of each monitoring point is collected in real time to construct a spatiotemporal synchronous deformation dataset.

[0065] In the example of the above solution, the deformation data may be settlement displacement, tilt angle, etc.

[0066] (4) Based on the spatial distribution characteristics of the deformation data set, the monitoring blind spots and monitoring redundancies are identified. This triggers the deployment of encrypted monitoring points when the monitoring blind spots are identified, and triggers the migration of monitoring points when the monitoring redundancies are identified.

[0067] As a way to implement the above solution, identifying monitoring blind spots and monitoring redundancy includes the following: Figure 3 As shown, S1: in each sub-area, each monitoring point is used as the geometric center to diverge to the surrounding areas to obtain adjacent monitoring points, and the local range formed by the adjacent monitoring points is used as the adjacent area of the monitoring point.

[0068] It should be pointed out that the neighboring monitoring points obtained by radiating from each monitoring point are limited to single-level topological expansion, that is, only the first-level neighboring monitoring points that are directly adjacent to the monitoring point in the spatial layout relationship are considered, which helps to clarify the spatial boundaries of the neighboring area.

[0069] S2: For each neighboring area, the mean and standard deviation of the deformation data of all neighboring monitoring points in the area are calculated based on the spatiotemporal synchronous deformation dataset at the same time, and the neighborhood deformation threshold is constructed accordingly.

[0070] Specifically, the neighborhood deformation threshold expression is: ,in represents the neighborhood deformation threshold, represents the mean of deformation data of adjacent monitoring points, represents the standard deviation of deformation data of adjacent monitoring points, It represents the empirical coefficient, controls the sensitivity of judgment, and is usually set to 3.

[0071] The construction of the above neighborhood deformation threshold is based on statistical principles: under the premise of a stable deformation state, the deformation data observed at multiple monitoring points in the adjacent area should exhibit statistical characteristics that approximately follow a normal distribution. Accordingly, the 3σ criterion (i.e., the triple standard deviation principle) is used as the basis for anomaly identification. If the deformation value of a monitoring point deviates from the mean of the deformation data in the neighborhood by more than three standard deviations, it is considered to have significantly deviated from the normal deformation trend and is a statistically significant outlier.

[0072] S3: Compare the deformation data of each monitoring point in each sub-area with the neighborhood deformation threshold of the adjacent area. If the deformation data of a monitoring point is higher than the neighborhood deformation threshold, execute S4-S5; otherwise, execute S6-S7.

[0073] It should be noted that the neighborhood deformation threshold constructed above is used to characterize the overall average deformation level of the area adjacent to the monitoring point. Since a monitoring point and its surrounding areas usually have a certain degree of continuity and consistency in deformation response in geological structure during underground engineering construction, when the deformation observation value of a monitoring point deviates significantly from the average deformation trend of its adjacent areas, it may indicate that the point is in a local abnormal deformation area.

[0074] One potential reason for such deviations may be that the stratum deformation characteristics at the location of the monitoring point are not fully captured by the limited number of neighboring monitoring points. This reflects that the current monitoring density in the area is insufficient and fails to effectively cover the spatial variation details of the stratum deformation, resulting in the inaccurate identification and reflection of the local deformation evolution trend.

[0075] S4: Mark the point as a transition point, record the time when the transition occurs, and mark the sub-region to which the point belongs as a transition sub-region.

[0076] S5: Continue to track the transition in the transition sub-area after the transition occurs, and count the duration of the transition. If the duration of the transition reaches a critical duration threshold, it is determined that a monitoring blind spot exists in the transition sub-area.

[0077] S6: For the monitoring point that does not show a jump, the deformation difference between the point and the monitoring points in the adjacent area is calculated. This operation is intended to quantify the degree of difference in deformation behavior between the monitoring point and the surrounding environment, and to count the proportion of monitoring points in the adjacent area whose deformation difference is within the set allowable error range. If the proportion reaches the preset stability ratio threshold, for example, the stability ratio threshold is 80%, it is considered that the monitoring point and most of the monitoring points in its neighborhood show consistent deformation characteristics, indicating that the stratum state in this local area is relatively uniform and stable. The sub-area where the point is located is marked as a stable sub-area, and the adjacent area where the point is located is defined as a local stable area, and the time when the stability occurs is recorded.

[0078] The allowable error range set above is used to determine whether the deformation difference is acceptable.

[0079] It is important to understand that if the deformation differences of the vast majority of adjacent monitoring points within a monitoring point's vicinity are within the set tolerance, this indicates that the deformation behavior of the monitoring points within the area is highly consistent and reflects similar formation deformation characteristics. This indicates that the deformation pattern in the area is uniform, with no obvious local disturbances or uncoordinated deformation.

[0080] By introducing a judgment method based on deformation difference and a stability ratio threshold, the interference caused by measurement errors or accidental fluctuations at individual monitoring points on overall stability judgment can be effectively reduced, thereby improving the robustness and accuracy of anomaly identification. An area is only considered stable when multiple adjacent monitoring points exhibit statistically consistent deformation trends. This multi-point collaborative analysis mechanism significantly enhances the credibility of local stability identification results.

[0081] S7: Count the duration of stability of the local stable area in the stable sub-area. If the duration of stability reaches the critical duration threshold, it is determined that there is monitoring redundancy in the stable sub-area, that is, the information provided by some monitoring points is repeated and does not affect the overall deformation trend judgment.

[0082] It should be pointed out that formation deformation is a process that develops over time. An anomaly at a single point in time is not sufficient to determine whether there is a systemic problem. Therefore, the concepts of jump duration and stable duration are introduced, and combined with the time dimension to further confirm whether it is a true monitoring blind spot or redundancy.

[0083] By incorporating local spatial statistical models and temporal evolution analysis, this method dynamically identifies and categorizes blind spots and redundant points within the monitoring system. Compared to traditional static deployment methods, this approach offers greater spatial adaptability and temporal sensitivity, enabling real-time optimization and adjustment of the monitoring network, improving the early warning capabilities and resource utilization efficiency of the deformation monitoring system.

[0084] As a further implementation method of the above scheme, the encrypted deployment of monitoring points is triggered when the monitoring blind spot is identified as follows: for the jump point appearing in the sub-area where the monitoring blind spot is located, the deformation difference between the jump point and each adjacent monitoring point in the adjacent area is calculated moment by moment during the jump duration period, and the adjacent monitoring point with the largest deformation difference with the jump point at each moment is defined as a low-correlation monitoring point.

[0085] It is important to know that this low-correlation monitoring point reflects the monitoring location with the greatest difference in deformation trend from the jump point and inconsistent response.

[0086] The frequency distribution of each low-correlation monitoring point during the jump duration is counted, and the low-correlation monitoring point with the highest frequency is selected. Then, new monitoring points are added in the direction of the spatial connection between it and the jump point to enhance the monitoring density in this direction and fill the local monitoring blind spot.

[0087] The low-correlation monitoring point with the highest occurrence frequency extracted above means that there are significant deformation differences between it and the jump point at multiple moments, which is statistically representative.

[0088] It's important to understand that deformation often propagates along weak surfaces in geological structures or along stress-release pathways. Placing new monitoring points along the line connecting the jump point and the low-correlation monitoring points helps capture the primary direction of deformation development. This also improves the spatial resolution of the monitoring data along this path, enhancing the ability to respond to anomalous deformation evolution.

[0089] As a further implementation method of the above scheme, triggering the migration of monitoring points when identifying monitoring redundancy is implemented as follows: for the local stable area of the sub-area where the monitoring redundancy is located, the time series deformation difference between it and each monitoring point in the adjacent area is calculated based on the geometric center monitoring point during the stable duration period.

[0090] The monitoring points whose deformation differences are always within the preset allowable error range are screened out from the time series deformation differences and are defined as high consistency monitoring points.

[0091] It should be understood that within a stable stratum area, the time-series deformation data of multiple adjacent monitoring points are highly correlated due to similar geological conditions and consistent deformation responses. This high consistency indicates that the information provided by some monitoring points is repetitive and redundant, and they are not independent observation points that must be retained.

[0092] From the above-mentioned high-consistency monitoring points, every other monitoring point is selected as a migration object in the order of spatial distribution to carry out monitoring point migration.

[0093] The above-mentioned strategy of selecting every other monitoring point as a migration target is a spatial sparseness strategy. While removing redundant points, key monitoring nodes can still be retained to ensure the overall identifiability of the deformation trend in the area.

[0094] It should be emphasized that the deformation data collected at the monitoring points in this paper assumes that the monitoring equipment is operating normally during analysis, and the data collection process is not affected by equipment failures or measurement errors. Therefore, the monitoring results only reflect the deformation behavior of the formation at the location of the monitoring point.

[0095] (4) After the dynamic adjustment of the monitoring network is completed, deformation trend analysis is performed based on the time series deformation data of the monitoring points in each sub-area during the construction process, thereby initiating a graded deformation warning from local to global level.

[0096] It's important to note that after the initial monitoring network deployment is completed based on the geological risk distribution map, it will need to be adaptively adjusted based on the dynamic evolution of underground excavation construction. If no blind spots or redundant areas are identified during subsequent monitoring, this indicates that the current monitoring network has sufficient coverage and adaptability, and the stable deformation trend analysis phase can be entered.

[0097] In the above scheme, the deformation trend analysis is optimized as follows: within the set observation window, the time series deformation data of each monitoring point in each sub-area is used to draw a deformation curve, and the overall deformation rate is extracted from the deformation curve as the deformation trend of each monitoring point.

[0098] In the above optimization implementation example, the overall deformation rate of the deformation curve can be obtained by linear fitting, that is, linear regression fitting is performed on the deformation-time series data within the observation window. The slope of the fitting line is the overall deformation rate, which quantifies the rate of change of the deformation degree of the monitoring point within the specified observation period.

[0099] The above-mentioned use of the overall deformation rate as the basis for subsequent deformation warning, rather than relying on instantaneous deformation data at a specific time point, is mainly because the underground excavation construction process is a long-term and complex geological activity. The deformation behavior of rock and soil under external loads is usually gradual and continuous, and follows certain evolutionary laws. Deformation data at a single time can only capture the deformation state at the current moment and cannot reflect the dynamic changes in deformation development. Therefore, it does not have comparative value across time periods. The early warning mechanism aims to provide early warning of deformation risks that may occur in the subsequent construction phase, which requires more stable and predictive indicators as a basis. Through in-depth analysis of deformation trends within the set observation window, more reliable trend information can be extracted, providing solid scientific support for early warning of potential deformation risks in future construction processes.

[0100] The above scheme is further optimized to initiate a graded deformation warning from local to global, including the following contents: comparing the deformation trend of each monitoring point in each sub-area with the safety threshold, screening out the monitoring points whose deformation trend reaches the safety threshold and recording them as abnormal monitoring points.

[0101] The safety thresholds mentioned above are based on the maximum allowable deformation rates determined by relevant industry guidelines for construction specifications.

[0102] According to the spatial distribution of abnormal monitoring points, the abnormal area surrounded by these points is delineated, and the proportion of the abnormal area to the area of the sub-area is calculated.

[0103] Get the normalized Euclidean distance between the geometric center of the abnormal area and the center of the corresponding sub-area, and define this as the center deviation distance coefficient of the abnormal area.

[0104] The local abnormal tendency factor is defined as: ,in represents the local abnormal tendency factor, Indicates the area ratio of the abnormal area in the sub-area to which it belongs, Indicates the center deviation distance coefficient of the abnormal area.

[0105] It should be noted that the principle of constructing the above-mentioned local anomaly tendency factor is: the area ratio of the abnormal area in the sub-area to which it belongs reflects the size ratio of the abnormal range in the sub-area. The smaller the area ratio, the more concentrated the abnormal area is locally and the more localized it is. The larger the value, the wider the impact of the abnormality and the more likely it is to enter the stage of overall deformation. The smaller the center deviation distance coefficient of the abnormal area, the closer the abnormal area is to the center of the sub-area, which may reflect the overall deformation trend. The larger the value, the farther away from the center, that is, the closer to the edge, the more localized it is. By and The local anomaly tendency factor obtained by multiplying them reflects the synchronization of the area proportion and the center deviation distance coefficient with the local tendency. When both values are larger, that is, the abnormal area has a smaller spatial coverage and is obviously deviated from the center and tends to the edge, it indicates that the anomaly has a significant local development trend.

[0106] The local abnormal tendency factor obtained according to the above definition is compared with the pre-assigned warning tendency. If the local abnormal tendency factor does not reach the warning tendency, it indicates that the impact range of the abnormality is limited, triggering a local deformation warning, and only taking targeted reinforcement measures in the abnormal area. Otherwise, it indicates that the abnormality has systematic characteristics, and it is necessary to activate the global deformation warning and take overall reinforcement measures for the sub-area.

[0107] This invention constructs a hierarchical early warning mechanism based on local anomaly propensity factors, which has both physical significance and practical engineering value. By setting appropriate warning propensity thresholds, it achieves a step-by-step identification and response to stratum deformation behavior, from local to global. Local deformation early warning focuses on rapid intervention and local control of abnormal areas, while global deformation early warning aims at comprehensive prevention and control of systemic risks. This hierarchical mechanism not only improves the accuracy of risk identification and the timeliness of response.

[0108] The parameters involved in the above formulas are all dimensionless and calculated using their numerical values. The preset parameters in the formulas are set by those skilled in the art according to actual conditions.

[0109] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0110] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0111] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0112] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0113] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for monitoring deformation of underground tunnels in ultra-shallow, high-flow strata, characterized by: The following steps are involved: (1) Delineate the construction impact area with the tunnel face as the center, conduct geological radar scanning on the construction impact area to extract geological distribution characteristics including stratum interface distribution, water content distribution and rock mass fracture distribution, and generate a geological risk distribution map; (2) Divide the construction impact area into high-risk sub-areas, transition sub-areas, and stable sub-areas based on the geological risk distribution map, and deploy the initial monitoring network in each sub-area in a differentiated manner; (3) During the tunneling process, deformation data of each monitoring point is collected in real time to construct a spatiotemporal synchronous deformation dataset; (4) Identify monitoring blind spots and monitoring redundancies based on the spatial distribution characteristics of the deformation data set, thereby triggering the dense deployment of monitoring points when identifying monitoring blind spots and triggering the migration of monitoring points when identifying monitoring redundancies; (5) After the dynamic adjustment of the monitoring network is completed, deformation trend analysis is performed based on the time series deformation data of the monitoring points in each sub-area during the construction process, thereby initiating a graded deformation warning from local to global level.

2. The method for monitoring deformation of underground tunnels in ultra-shallow, high-flow strata according to claim 1, characterized in that: The specific extraction process of the geological distribution characteristics is as follows: Arrange multiple parallel scanning lines along the axis of the underground tunnel, with each scanning line covering the lateral boundary of the construction impact area; Use geological radar equipment to perform high-resolution geological scanning on each scanning line, collect time series data of underground medium reflection waves, and perform image conversion to obtain geological radar images of the construction-affected area; In the generated geological radar image, lines with varying reflection intensities are identified as the boundaries between different strata, and the spatial distribution of the strata is determined based on the positions of these boundaries. The geological type of the strata is determined based on the analysis of the reflected wave properties of different strata in the geological radar image, and the areas with the same stratum distribution and no gaps are classified as the same geological unit; For each divided geological unit, the attenuation properties of the reflected wave in the corresponding geological radar image are analyzed to determine the water content of the area; The shape characteristics of the reflected waves are extracted from the geological radar images of each geological unit to identify the cracks and locate the distribution of the cracks.

3. The method for monitoring deformation of a subway tunnel in an ultra-shallow, high-flow stratum according to claim 2, characterized in that: The generated geological risk distribution map contains the following contents: The geological types of each geological unit are extracted layer by layer and matched with the risk stratum database. If the geological type of a certain layer of a geological unit is successfully matched, it is determined that the geological unit has a stratum risk factor. Compare the water content of each geological unit with the set safe water content threshold. If the water content of a geological unit exceeds the threshold, it is determined that the geological unit has a water risk factor. The crack distribution density is obtained based on the distribution position of cracks in each geological unit and compared with the set safe crack distribution density. If the crack distribution density of a geological unit is higher than the safe crack distribution density, the geological unit is identified as having crack risk factors. Plane modeling is performed based on the geometric boundaries of the construction-affected area, and geological units are identified based on the boundary contours of each geological unit in the constructed model. At the same time, the identified risk factors are marked to generate a geological risk distribution map.

4. The method for monitoring deformation of a subway tunnel in an ultra-shallow, high-flow stratum according to claim 3 is characterized by: The high-risk sub-area, transition sub-area and stable sub-area are divided as follows: Based on the risk factors identified in each geological unit in the geological risk distribution map, the sub-regions are divided according to the following zoning judgment rules; High-risk sub-area: Two or more risk factors exist simultaneously in a single geological unit; Transitional sub-area: only one type of risk factor exists in a single geological unit; Stable sub-area: No risk factors are identified in the individual geological units.

5. The method for monitoring deformation of underground tunnels in ultra-shallow, high-flow strata according to claim 1 is characterized by: The differentiated deployment of the initial monitoring network in each sub-area is described in the following process: Monitoring points are arranged in a triangular grid within the high-risk sub-area, with the node spacing between monitoring points being a fixed preset value; Monitoring points are arranged in a rectangular grid in the transition sub-area, and the spacing between monitoring point nodes is set to be a multiple of the spacing between nodes in the high-risk sub-area; Monitoring points are arranged along the channel axis in the stable sub-region, and the node spacing of the monitoring points is set to be a multiple of the node spacing of the transition sub-region.

6. The method for monitoring deformation of a subway tunnel in an ultra-shallow, high-flow stratum according to claim 1 is characterized by: The identification of monitoring blind spots and monitoring redundancy includes the following: S1: In each sub-area, the neighboring monitoring points are obtained by radiating from each monitoring point as the geometric center. The local area formed by the neighboring monitoring points is regarded as the neighboring area of the monitoring point. S2: For each neighboring area, the mean and standard deviation of the deformation data of all neighboring monitoring points in the area are calculated based on the spatiotemporal synchronous deformation dataset at the same time, and the neighborhood deformation threshold is constructed accordingly; S3: Compare the deformation data of each monitoring point in each sub-area with the neighborhood deformation threshold of the adjacent area. If the deformation data of a monitoring point is higher than the neighborhood deformation threshold, execute S4-S5; otherwise, execute S6-S7. S4: Mark the point as a transition point, record the time when the transition occurs, and mark the sub-region to which the point belongs as a transition sub-region; S5: Continue to track the transition in the transition sub-area after the transition occurs, and count the duration of the transition. If the duration of the transition reaches a critical duration threshold, it is determined that a monitoring blind spot exists in the transition sub-area. S6: For a monitoring point that does not experience a jump, calculate the deformation difference between the point and each monitoring point in the adjacent area, and count the proportion of monitoring points in the adjacent area whose deformation difference is within the set allowable error range. If this proportion reaches the preset stability ratio threshold, mark the sub-area where the point is located as a stable sub-area, and define the adjacent area where the point is located as a local stable area. At the same time, record the time when the stability occurs; S7: Counting the duration of stability of the local stable region in the stable sub-region. If the duration of stability reaches a critical duration threshold, it is determined that monitoring redundancy exists in the stable sub-region.

7. The method for monitoring deformation of a subway tunnel in an ultra-shallow, high-flow stratum according to claim 6, characterized in that: The triggering of the encrypted deployment of monitoring points when identifying the monitoring blind area is implemented as follows: For the jump point in the sub-area where the monitoring blind spot is located, the deformation difference between the jump point and each adjacent monitoring point in the adjacent area is calculated moment by moment during the jump duration. At each moment, the adjacent monitoring point with the largest deformation difference with the jump point is identified and defined as a low-correlation monitoring point. The frequency distribution of each low-correlation monitoring point during the jump duration is counted, and the low-correlation monitoring point with the highest frequency is selected, and then a new monitoring point is added in the direction of the spatial connection between it and the jump point.

8. The method for monitoring deformation of a subway tunnel in an ultra-shallow, high-flow stratum according to claim 6, wherein: The triggering of monitoring point migration when monitoring redundancy is identified is implemented as follows: For the local stable area of the sub-area where the monitoring redundancy is located, the time series deformation difference between it and each adjacent monitoring point in the adjacent area is calculated based on the geometric center monitoring point during the stability duration; The monitoring points whose deformation differences are always within the preset allowable error range are selected from the time series deformation differences and defined as high consistency monitoring points; From the above-mentioned high-consistency monitoring points, every other monitoring point is selected as a migration object in the order of spatial distribution to carry out monitoring point migration.

9. The method for monitoring deformation of a subway tunnel in an ultra-shallow, high-flow stratum according to claim 1, wherein: The deformation trend analysis refers to the following process: The deformation curve is drawn using the time series deformation data of each monitoring point in each sub-area within the set observation window, and the overall deformation rate is extracted from the deformation curve as the deformation trend of each monitoring point.

10. The method for monitoring deformation of a subway tunnel in an ultra-shallow, high-flow stratum according to claim 1, wherein: The initiation of the graded deformation warning from local to global includes the following: Compare the deformation trend of each monitoring point in each sub-area with the safety threshold, and select the monitoring points whose deformation trend reaches the safety threshold as abnormal monitoring points; According to the spatial distribution of abnormal monitoring points, the abnormal area surrounded by these points is delineated, and the proportion of the abnormal area to the area of the sub-area is calculated; Get the normalized Euclidean distance between the geometric center of the abnormal area and the center of the corresponding sub-area, and define this as the center deviation distance coefficient of the abnormal area; The above ratio and center deviation distance coefficient are fused and calculated to obtain the local abnormal tendency factor, which is then compared with the warning tendency. If the warning tendency is not reached, a local warning is activated; otherwise, a global warning is activated.

Images