A hierarchical early warning method, device and medium for subway tunnel segment displacement

Abstract

The application discloses a kind of hierarchical early warning method, equipment and medium for subway tunnel segment displacement, belong to data early warning technical field, for solving in the segment displacement early warning of existing subway tunnel, single early warning factor has limitation, there is also threshold static solidification and the technical problem of insufficient correlation analysis of multi-source monitoring data.Method includes: the data monitoring processing of target subway tunnel is under the influence factor of precipitation subway tunnel segment displacement state;Multi-source monitoring index data is based on the correlation analysis of tunnel segment displacement precursor characteristics, and multidimensional correlation index data set is obtained;Risk threshold trigger condition is judged to multi-source monitoring index data set, and the primary risk signal for indicating tunnel segment displacement risk is obtained;The risk level correction processing based on reference value deviation and multidimensional correlation index data is carried out to preliminary comprehensive risk level, and the final early warning level is obtained.

Description

Technical Field

[0001] This application relates to the field of data early warning technology, and in particular to a graded early warning method, equipment and medium for displacement of subway tunnel segments. Background Technology

[0002] In recent years, extreme rainfall events have become increasingly frequent, leading to increasingly prominent problems such as tunnel segment uplift, structural displacement, and even leakage and cracking in subway tunnels. Studies have shown that the uplift of shield tunnel segments in water-rich strata is generally 20-70 mm, and can even exceed 100 mm if not properly controlled. Heavy rainfall causes a sharp rise in groundwater levels and a significant increase in pore pressure at the basement level—the cumulative rainfall during a continuous rainfall period can cause a 5.5-fold increase in water level and a 7.1-fold increase in pore pressure. In permeable strata, the tunnel segments are immersed in water, and the enormous buoyancy causes them to float, seriously threatening the safety of subway operations.

[0003] Currently, there are several shortcomings in the monitoring and early warning systems for subway tunnel segment displacement. Existing technologies mostly rely on single displacement or water level monitoring data for early warning, failing to incorporate multiple factors such as rainfall intensity, geological conditions, seepage direction, and segment structural condition into a unified early warning analysis framework. This leads to significant discrepancies between warning results and actual conditions. Current early warning methods typically use fixed thresholds (e.g., triggering an early warning when displacement exceeds a certain value), neglecting the differences in segment displacement response patterns under varying geological conditions and rainfall scenarios, thus failing to achieve precise, tiered, and scenario-specific early warnings. Furthermore, current technologies utilize monitoring data at the level of independent analysis of single data types, lacking a mechanism for correlated analysis between multi-source monitoring data, making it difficult to capture precursory information about segment displacement from interconnected data changes.

[0004] Therefore, there is an urgent need for a technical solution that can comprehensively consider the coupling effects of multiple factors and has dynamic hierarchical early warning capabilities. Summary of the Invention

[0005] This application provides a graded early warning method, device and medium for subway tunnel segment displacement, which is used to solve the following technical problems: In the existing early warning of subway tunnel segment displacement, the single early warning factor has limitations, and there are also problems such as static fixed threshold and insufficient correlation analysis of multi-source monitoring data.

[0006] The embodiments of this application adopt the following technical solutions: On one hand, this application provides a graded early warning method for subway tunnel segment displacement, including: monitoring and processing data on the displacement state of subway tunnel segments under precipitation influence factors through a monitoring network deployed in the target subway tunnel to obtain a multi-source monitoring index dataset; performing correlation analysis based on the precursor characteristics of tunnel segment displacement on the multi-source monitoring index dataset to obtain a multi-dimensional correlation index dataset; determining the triggering conditions of risk thresholds on the multi-source monitoring index dataset to obtain a primary risk signal representing the risk of tunnel segment displacement; wherein the primary risk signal includes: a primary rainfall intensity risk signal, a primary water level change risk signal, a primary segment stress risk signal, a primary vertical displacement risk signal, and a primary seepage rate risk signal; determining a preliminary comprehensive risk level based on different combinations of the primary risk signals; performing risk level correction processing based on the deviation data of the tunnel segments and the multi-dimensional correlation index dataset to obtain a final early warning level; and generating corresponding emergency response recommendations based on the final early warning level.

[0007] This application's embodiments achieve cross-validation of the same risk event through three independent dimensions: signal combination breadth, measured-benchmark deviation, and precursor correlation characteristics. This overcomes the problem of false alarms triggered by sensor disturbances or random fluctuations due to single indicators, significantly reducing the false alarm rate on-site. Multi-dimensional cross-validation is achieved, greatly reducing the false alarm rate. Furthermore, by using three correlation analysis indicators—water level response lag time, stress-water level linkage change rate, and displacement-seepage ratio—precursor signals such as structural anomalies and seepage channel formation can be captured before the segment displacement reaches the benchmark value. This enables the identification of precursor characteristics of tunnel segment displacement risks and provides early warning before displacement exceeds limits. Moreover, this application abandons the traditional fixed threshold approach and establishes a displacement benchmark value lookup table based on rainfall condition classification. The benchmark value adaptively changes with geological conditions and rainfall conditions. Different sections and different rainfall scenarios use matching early warning reference standards, significantly improving the accuracy and targeting of early warnings. This allows the early warning threshold to dynamically adapt and accurately match different conditions.

[0008] In one feasible implementation, a monitoring network is deployed in the target subway tunnel to monitor the displacement of tunnel segments under rainfall conditions, resulting in a multi-source monitoring index dataset. Specifically, this involves dividing the target subway tunnel into several monitoring sections based on a predetermined longitudinal spacing, and deploying multi-source monitoring equipment in each monitoring section. The multi-source monitoring equipment includes rainfall monitoring equipment, water level monitoring equipment, water pressure monitoring equipment, segment displacement monitoring equipment, segment stress monitoring equipment, and seepage monitoring equipment. Through this multi-source monitoring equipment, data is monitored on the target subway tunnel under rainfall conditions, and the multi-source monitoring index dataset is synchronously collected based on a unified time label. The multi-source monitoring index dataset includes at least the following multi-source monitoring indicators: real-time rainfall intensity, cumulative rainfall, groundwater level depth, water level change rate, pore water pressure, segment vertical displacement, segment horizontal displacement, segment rotation angle, segment stress, and joint seepage rate.

[0009] In one feasible implementation, based on the multi-source monitoring index dataset, a correlation analysis based on the precursor characteristics of tunnel segment displacement is performed on the multi-source monitoring index data to obtain a multi-dimensional correlation index dataset. Specifically, this includes: calculating the relevant time lag coefficient between the start time of a rainfall event and the groundwater level response time using the multi-source monitoring index dataset; obtaining the linkage characteristics between water level and rainfall based on the real-time calculated water level change rate; correlating the stress data of tunnel segments and the groundwater level depth data at various times under rainfall conditions to obtain a stress-water level linkage curve; and calculating the linkage rate of change between water level change and stress change based on the stress-water level linkage curve. The linkage change rate is defined as the stress-water level linkage characteristic; wherein, the stress-water level linkage characteristic represents the sensitivity of the segment structure to water level changes; through the multi-source monitoring index dataset, the linkage analysis between the segment displacement data and the joint seepage rate of the same monitoring section is performed to calculate the linkage ratio between the segment displacement change and the seepage rate change; and the linkage ratio is defined as the displacement-seepage linkage characteristic; wherein, the displacement-seepage linkage characteristic represents the probability of seepage channels formed by segment joints or back wall cavities; the water level-rainfall linkage characteristic, the stress-water level linkage characteristic, and the displacement-seepage linkage characteristic are combined and processed by index items to obtain the multi-dimensional correlation index dataset.

[0010] In one feasible implementation, risk threshold triggering conditions are determined for both the multi-source monitoring index dataset and the multi-dimensional correlation index dataset to obtain a primary risk signal representing the displacement risk of tunnel segments. Specifically, this includes: classifying the real-time rainfall intensity in the multi-source monitoring index dataset according to a preset rainfall intensity threshold to obtain the primary rainfall intensity risk signal; wherein the primary rainfall intensity risk signal includes a first-class primary rainfall intensity risk signal and a second-class primary rainfall intensity risk signal; classifying the water level change rate in the multi-source monitoring index dataset according to a preset water level change threshold to obtain the primary water level change risk signal; wherein the primary water level change risk signal includes a first-class primary water level change risk signal and a second-class primary water level change risk signal; and classifying the segment stress in the multi-source monitoring index dataset according to a preset stress threshold to obtain the primary segment stress risk signal; wherein the primary segment stress risk signal includes a first-class primary stress risk signal and a second-class primary stress risk signal. The system generates primary segment stress risk signals and secondary primary segment stress risk signals. It then classifies the vertical displacement of segments in the multi-source monitoring index dataset by applying a preset vertical displacement threshold to obtain the primary vertical displacement risk signals. These primary vertical displacement risk signals include: primary type I primary vertical displacement risk signals and secondary type II primary vertical displacement risk signals. Finally, it classifies the joint seepage rate in the multi-source monitoring index dataset by applying a preset seepage rate threshold to obtain the primary seepage rate risk signals. These primary seepage rate risk signals include: primary type I primary seepage rate risk signals and secondary type II primary seepage rate risk signals. The primary risk signals represent risk signals under different risk states. Finally, it jointly stores the primary rainfall intensity risk signal, the primary water level change risk signal, the primary segment stress risk signal, the primary vertical displacement risk signal, and the primary seepage rate risk signal to obtain the primary risk signal.

[0011] In one feasible implementation, a preliminary comprehensive risk level is determined based on different combination types of the primary risk signals. Specifically, this includes: performing an existence type analysis on the primary risk signals of type I and type II contained in the primary risk signals at the same time to generate a current risk signal combination type; obtaining an original risk level based on the current risk signal combination type; wherein the original risk level includes: normal level, attention level, warning level, and alert level; adjusting the original risk level upwards or downwards based on the current rainfall stage conditions to obtain the preliminary comprehensive risk level; wherein the current rainfall stage conditions include: the initial rainfall stage, the continuous rainfall stage, and the rainfall dissipation stage.

[0012] In one feasible implementation, based on the deviation data of tunnel segments and through the multi-dimensional correlation index dataset, the preliminary comprehensive risk level is corrected using a risk level correction process based on the baseline deviation and multi-dimensional correlation index data to obtain the final warning level. Specifically, this includes: using a multi-source monitoring index dataset to determine the rainfall condition type based on the rainfall intensity and duration at the current moment; querying the segment displacement baseline value corresponding to the current rainfall condition; wherein the segment displacement baseline value is obtained from a segment vertical displacement baseline value lookup table, and the segment vertical displacement baseline value is... The displacement reference value lookup table is established based on physical model test data and historical field measurement data; the measured vertical displacement of the tunnel segment at the current moment in the multi-source monitoring index dataset is compared and calculated with the tunnel segment displacement reference value to obtain the current displacement deviation rate of the tunnel segment; the current displacement deviation rate of the tunnel segment is associated and stored with the corresponding tunnel segment location information to obtain the deviation data; based on the comparison relationship between the displacement deviation rate in the deviation data and the preset displacement deviation threshold, a reference value deviation correction strategy is obtained; wherein, the reference value deviation correction strategy is used to adjust the preliminary comprehensive wind... The risk level is initially adjusted upwards or downwards. Based on the linkage feature types present in the multi-dimensional correlation index data, a second upward adjustment is performed on the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy. This includes: if the linkage feature type is water level and rainfall linkage, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy is increased by one level; if the linkage feature type is stress and water level linkage, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy is directly determined as an alarm level; if the linkage feature type is displacement and seepage linkage, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy is increased by one level. After the second upward adjustment of the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy, the final warning level and the corresponding final warning information are obtained. The final warning information includes: the triggered primary risk signal, the corresponding current tunnel segment displacement deviation rate and segment displacement benchmark value, the linkage feature type present in the triggered multi-dimensional correlation index data, and the tunnel segment location information of the current tunnel segment.

[0013] In one feasible implementation, the method for establishing the segment vertical displacement reference value lookup table includes: obtaining segment displacement test data under different geological conditions and different combinations of rainfall conditions based on extreme rainfall simulation physical model tests; correcting and calibrating the segment vertical displacement test data based on historical rainfall events of the target subway tunnel and corresponding measured segment vertical displacement data; and archiving the corrected and calibrated segment vertical displacement test data according to working condition type and geological conditions to generate the segment vertical displacement reference value lookup table.

[0014] In one feasible implementation, based on the final warning level, corresponding emergency response recommendations are generated, specifically including: receiving the final warning level and the corresponding final warning information; if the final warning level is a concern level, then generating concern-level emergency response recommendations based on the final warning information; if the final warning level is a warning level, then generating warning-level emergency response recommendations based on the final warning information; if the final warning level is an alarm level, then generating alarm-level emergency response recommendations based on the final warning information.

[0015] Secondly, embodiments of this application also provide a graded early warning device for subway tunnel segment displacement, the device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to execute a graded early warning method for subway tunnel segment displacement as described in any of the above embodiments.

[0016] Thirdly, embodiments of this application also provide a non-volatile computer storage medium, which is a non-volatile computer-readable storage medium storing at least one program. Each program includes instructions, which, when executed by a terminal, cause the terminal to execute a graded early warning method for subway tunnel segment displacement as described in any of the above embodiments.

[0017] This application provides a graded early warning method, device, and medium for subway tunnel segment displacement. Compared with the prior art, the embodiments of this application have the following beneficial technical effects: 1. This application achieves cross-validation of the same risk event through three independent dimensions: signal combination breadth, measured-benchmark deviation, and precursor correlation characteristics. Only when multiple dimensions simultaneously point to the same conclusion will the warning level be upgraded step by step. A single indicator cannot trigger a false alarm due to sensor disturbances or accidental fluctuations, and the false alarm rate on site is significantly reduced.

[0018] 2. This application uses three correlation analysis indicators—water level response lag time, stress-water level linkage change rate, and displacement-seepage ratio—to capture early warning signals such as abnormal structural response and seepage channel formation before the segment displacement reaches the benchmark value.

[0019] 3. This application abandons the traditional fixed threshold approach and establishes a displacement benchmark lookup table based on rainfall condition classification. The benchmark values ​​are dynamically updated through continuous cumulative averaging of measured data. The benchmark values ​​adapt to geological conditions and rainfall conditions, and matching early warning reference standards are used for different sections and rainfall scenarios, significantly improving the accuracy and relevance of early warnings.

[0020] 4. Complete data for each early warning and response in this application are recorded, archived, and used for subsequent baseline updates and response optimization, so that the performance of the early warning system continues to improve as the operating time extends. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart illustrating a graded early warning method for subway tunnel segment displacement provided in this application embodiment; Figure 2 This is a schematic diagram of a graded early warning device for the displacement of subway tunnel segments provided in an embodiment of this application. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0023] This application provides a graded early warning method for the displacement of subway tunnel segments, such as... Figure 1 As shown, the graded early warning method for subway tunnel segment displacement specifically includes steps S101-S106: S101. By deploying a monitoring network in the target subway tunnel, data monitoring and processing of the displacement status of the subway tunnel segments under the influence of precipitation are carried out to obtain a multi-source monitoring index dataset.

[0024] Specifically, based on the predetermined longitudinal spacing of the target subway tunnel, several monitoring sections are first defined. Multi-source monitoring equipment is then deployed in each monitoring section. This multi-source monitoring equipment includes: rainfall monitoring equipment, water level monitoring equipment, water pressure monitoring equipment, segment displacement monitoring equipment, segment stress monitoring equipment, and seepage monitoring equipment.

[0025] In one embodiment, monitoring sections are divided along the longitudinal direction of the tunnel at certain intervals, and multi-source monitoring equipment is deployed at each monitoring section, including: rainfall monitoring equipment: tipping bucket rain gauges are deployed at corresponding positions on the ground along the tunnel to collect rainfall intensity data in real time; water level monitoring equipment: water level observation wells are deployed on both sides of the tunnel, and pressure water level gauges are installed in the wells to collect groundwater depth data in real time; water pressure monitoring equipment: pore water pressure gauges are buried at different depths on the outside of the tunnel segments to collect pore water pressure data in real time; displacement monitoring equipment: laser displacement sensors and tilt sensors are installed on the inner sidewalls of the tunnel segments to collect vertical displacement, horizontal displacement, and rotation angle data of the segments in real time; stress monitoring equipment: fiber optic grating strain sensors are attached to the inner sidewalls of the tunnel segments to collect strain data of the segments in real time, and the stress value of the segments is calculated based on the material constitutive relationship; seepage monitoring equipment: seepage sensors are deployed at the joints of the tunnel segments to collect seepage rate data of the segment joints in real time.

[0026] Furthermore, by using multi-source monitoring equipment, data monitoring is conducted on the target subway tunnel under rainfall conditions, and multi-source monitoring indicator datasets are collected synchronously based on a unified time label.

[0027] The multi-source monitoring index data in the multi-source monitoring index dataset includes at least the following: real-time rainfall intensity, cumulative rainfall, groundwater level depth, water level change rate, pore water pressure, segment vertical displacement, segment horizontal displacement, segment rotation angle, segment stress, and joint seepage rate.

[0028] In one embodiment, the aforementioned multi-source monitoring data are synchronously collected and stored using a unified time stamp (accurate to the minute) to form a multi-source monitoring indicator dataset. Then, the multi-source monitoring indicator data is packaged and initially screened in 5-minute data packets. The initial screening rules may include: removing obvious outliers exceeding the sensor's measurement range, and averaging the data from similar sensors at the same monitoring section to eliminate random noise.

[0029] S102. Based on the multi-source monitoring index dataset, perform correlation analysis on the multi-source monitoring index data based on the precursor characteristics of tunnel segment displacement to obtain a multi-dimensional correlation index dataset.

[0030] Specifically, by first using a multi-source monitoring index dataset, the time lag coefficient between the start time of a rainfall event and the groundwater level response time is calculated, and based on the real-time calculated rate of water level change, the linkage characteristics between water level and rainfall are obtained.

[0031] In one embodiment, it is necessary to establish a linkage characteristic between water level and rainfall; that is, it is necessary to calculate the time lag coefficient τ between rainfall events and water level responses in real time. First, after the start of a continuous rainfall event, the rainfall start time t0 is recorded; groundwater level data is continuously monitored, and the start time t1 when the water level begins to rise significantly is recorded; then the time lag coefficient τ = t1 - t0 is calculated. Furthermore, depending on different geological conditions, the value of τ can have different physical meanings: when τ ≤ 2h, it indicates that the strata in this section have high permeability, rainfall infiltrates rapidly, and the risk of the tunnel segments being affected by buoyancy increases quickly; when 2h < τ ≤ 8h, it indicates that the strata in this section have a certain buffering capacity, and there is a time lag in the buoyancy effect on the tunnel segments; when τ > 8h, it indicates that the strata in this section have low permeability or good drainage conditions, and the tunnel segments are less affected by rainfall. Meanwhile, the rate of water level change, Vw(t) = [H(t) - H(t - Δt)] / Δt, is calculated in real time. When Vw(t) is continuously positive and its absolute value continues to increase, it indicates that the groundwater level is still rising and the risk of the tunnel lining segments floating is correspondingly increased.

[0032] Furthermore, based on the correlation between the stress data of the tunnel segments and the groundwater level depth data at various times under rainfall conditions, a stress-water level linkage curve was obtained for the same monitoring section. Based on this curve, the linkage rate between water level change and stress change was calculated; this linkage rate was defined as the stress-water level linkage characteristic. The stress-water level linkage characteristic represents the sensitivity of the tunnel segment structure to water level changes.

[0033] In one embodiment, a stress-water level linkage characteristic is established. Linkage analysis can be performed on the segment stress data S(t) and water level data H(t) of the same monitoring section: the segment stress value S(t) at each moment is correlated with the water level depth H(t) at the same moment to form a stress-water level linkage curve; the linkage change rate k(t) between the water level change ΔH(t) and the stress change ΔS(t) is calculated as k(t) = ΔS(t) / ΔH(t). When the value of k(t) suddenly increases (exceeding the mean of k values ​​under the same historical conditions + 2 times the standard deviation), it indicates that the segment structure's sensitivity to water level changes is abnormally high, which may indicate that microcracks have formed in the segment or joints have opened, and is a precursor signal that the segment displacement is about to accelerate.

[0034] Furthermore, it is necessary to conduct a linkage analysis between the segment displacement data and the joint seepage rate at the same monitoring section using a multi-source monitoring index dataset, and calculate the linkage ratio between the change in segment displacement and the change in seepage rate; this linkage ratio will then be defined as the displacement-seepage linkage characteristic. The displacement-seepage linkage characteristic represents the probability of a seepage channel formed by a segment joint or a cavity behind the wall.

[0035] In one embodiment, it is also necessary to establish a linkage feature between displacement and seepage. That is, a linkage analysis is performed on the segment displacement data D(t) and the joint seepage rate data Q(t) of the same monitoring section: the ratio r(t) of the segment displacement change ΔD(t) to the seepage rate change ΔQ(t) is calculated as ΔD(t) / ΔQ(t); when r(t) drops sharply (i.e. the seepage rate increases significantly while the displacement change is not obvious), it indicates that a seepage channel has been formed at the segment joint or behind the wall cavity. Although the current displacement is not significant, the existence of the seepage channel will accelerate the subsequent floating of segments. This phenomenon is regarded as an early warning signal of segment displacement risk.

[0036] Furthermore, the linkage characteristics between water level and rainfall, stress and water level, and displacement and seepage are combined and processed into index items to obtain a multidimensional correlation index dataset.

[0037] S103. The trigger conditions for risk thresholds are determined in the multi-source monitoring index dataset to obtain primary risk signals representing tunnel segment displacement risk. These primary risk signals include: primary rainfall intensity risk signal, primary water level change risk signal, primary segment stress risk signal, primary vertical displacement risk signal, and primary seepage rate risk signal.

[0038] Specifically, it is necessary to first classify the real-time rainfall intensity in the multi-source monitoring index dataset by judging and classifying it according to a preset rainfall intensity threshold to obtain the primary rainfall intensity risk signal. Among them, the primary rainfall intensity risk signal includes Class I primary rainfall intensity risk signal and Class II primary rainfall intensity risk signal.

[0039] In one embodiment, when the real-time rainfall intensity I(t) is ≥16 mm / h and lasts for more than 2 hours, it is recorded as a primary rainfall intensity risk signal R1 (corresponding to a rainstorm condition, a type I primary rainfall intensity risk signal); when the real-time rainfall intensity I(t) is ≥32 mm / h and lasts for more than 0.5 hours, it is recorded as a primary rainfall intensity risk signal R1+ (corresponding to a heavy rainstorm condition, a type II primary rainfall intensity risk signal).

[0040] Furthermore, the water level change rates in the multi-source monitoring indicator dataset are classified according to preset water level change thresholds to obtain primary water level change risk signals. These primary water level change risk signals include: Class I primary water level change risk signals and Class II primary water level change risk signals.

[0041] In one embodiment, when the water level change rate Vw(t) ≥ 0.3 m / h and the upward trend continues for more than 3 hours, it is recorded as the primary water level change risk signal R2 (Class I primary water level change risk signal); when the water level depth H(t) rises above the tunnel arch, it is recorded as the primary water level change risk signal R2+ (Class II primary water level change risk signal).

[0042] Furthermore, the segment stress in the multi-source monitoring index dataset is classified according to a preset stress threshold to obtain primary segment stress risk signals. These primary segment stress risk signals include: Class I primary segment stress risk signals and Class II primary segment stress risk signals.

[0043] In one embodiment, when the segment stress S(t) increases by more than 30% compared to the normal state (the average stress value over 24 consecutive hours on a sunny day), it is recorded as a primary segment stress risk signal R3 (a type I primary segment stress risk signal); when the segment stress S(t) reaches 65% of the segment design strength, it is recorded as a primary segment stress risk signal R3+ (a type II primary segment stress risk signal).

[0044] Furthermore, the vertical displacement of the tunnel segments in the multi-source monitoring index dataset is classified according to a preset vertical displacement threshold to obtain primary vertical displacement risk signals. These primary vertical displacement risk signals include: Class I primary vertical displacement risk signals and Class II primary vertical displacement risk signals.

[0045] In one embodiment, when the vertical displacement Dv(t) of the segment is first detected to exceed twice the baseline value at 0:00 on the same day (i.e., the initial displacement value when there is no rainfall on the same day), it is recorded as a primary vertical displacement risk signal R4 (a type I primary vertical displacement risk signal); when the difference between adjacent time periods of the vertical displacement Dv(t) of the segment is positive for three consecutive sampling periods (15 minutes) and increases successively, it is recorded as a primary vertical displacement risk signal R4+ (representing continuous acceleration of displacement, a type II primary vertical displacement risk signal).

[0046] Furthermore, the joint seepage rate in the multi-source monitoring index dataset is classified according to a preset seepage rate threshold to obtain the primary seepage rate risk signal. The primary seepage rate risk signal includes: a first-class primary seepage rate risk signal and a second-class primary seepage rate risk signal.

[0047] In one embodiment, when the joint seepage rate Q(t) increases by more than 3 times compared to the normal state (the average seepage rate over 24 consecutive hours on a sunny day), it is recorded as a primary seepage rate risk signal R5 (a type I primary seepage rate risk signal); when the joint seepage rate Q(t) increases by more than 8 times compared to the normal state, it is recorded as a primary seepage rate risk signal R5+ (a type II primary seepage rate risk signal).

[0048] Furthermore, by jointly storing the primary rainfall intensity risk signal, primary water level change risk signal, primary segment stress risk signal, primary vertical displacement risk signal, and primary seepage rate risk signal, the primary risk signal can be obtained. Among them, the first-class primary risk signal and the second-class primary risk signal represent the risk signals under different risk states.

[0049] As a feasible implementation method, at the end of each data acquisition cycle (5 minutes), the above triggering conditions are iterated, and all triggering signals (primary risk signals) that meet the conditions at the current moment are summarized into several items in the primary risk signal set Signal_Primary={R1, R1+, R2, R2+, R3, R3+, R4, R4+, R5, R5+}. To avoid false triggering caused by instantaneous sensor disturbances, each primary risk signal is reconfirmed. That is, when a triggering condition is met for the first time, the signal is marked as "pending confirmation"; if the triggering condition is still met in the next acquisition cycle (5 minutes later), "pending confirmation" is changed to "confirmed"; if the triggering condition is no longer met in the next acquisition cycle, it is determined to be a disturbance signal and cleared; only "confirmed" primary risk signals enter the subsequent judgment process; finally, the primary risk signals after validity confirmation are obtained.

[0050] S104. Based on different combinations of primary risk signals, a preliminary comprehensive risk level is determined.

[0051] Specifically, the existence type analysis of the first type of primary risk signal and the second type of primary risk signal contained in the primary risk signal is performed at the same time to generate the current risk signal combination type.

[0052] Furthermore, based on the current combination of risk signals, the original risk levels are obtained. These original risk levels include: Normal Level, Attention Level, Warning Level, and Alert Level.

[0053] In one embodiment, when the current risk signal combination type is no signal or only one item from R1 or R2, the original risk level is level 0 (normal level), the risk is caused by a single abnormal factor, and does not constitute a substantial threat to segment displacement.

[0054] The current risk signal combination type is simultaneous confirmation of R1+R2, the original risk level is level three (attention level), the risk is caused by the simultaneous response of rainfall and hydrology, and rainfall infiltration has already affected the groundwater system; The current risk signal combination type is simultaneous confirmation of R1+R2+R3, the original risk level is level three (attention level), the risk is caused by the structural stress has begun to respond, and monitoring needs to be strengthened; The current risk signal combination type is simultaneous confirmation of R1+R2+R3+(any one of R4 or R5), the original risk level is level two (early warning level), the risk cause is that displacement or seepage has appeared, and the risk of segment displacement is relatively high. The current risk signal combination type is simultaneous confirmation of R1+R2+R3+R4+R5, the original risk level is level two (early warning level), the risk is caused by the full response of five types of indicators, and the risk continues to rise; The current risk signal combination type is either R3+ or R4+, the original risk level is level two (early warning level), the risk is caused by stress exceeding the limit or continuous acceleration of displacement, and it has entered the early warning state; The current risk signal combination type is confirmed R2+ and simultaneously confirmed either R3 or R4. The original risk level is Level II (early warning level). The risk is caused by the water level exceeding the tunnel arch and the structure responding. The current risk signal combination type is confirmed R5+, the original risk level is level two (early warning level), the risk is caused by a sharp increase in seepage, indicating that the joints of the pipe segments are opening or seepage channels behind the wall are forming; The current risk signal combination type is simultaneous confirmation of R3+ and R4+, the original risk level is level one (alarm level), the risk is caused by stress exceeding the limit and displacement continuing to accelerate, and it must be dealt with immediately; The current risk signal combination type is the simultaneous confirmation of R1+R2+R3+R4+R5, which includes either R3+ or R4+. The original risk level is Level 1 (alarm level). The risk is caused by a full indicator response and the existence of an over-limit item. The current risk signal combination type is the confirmation of any three or more strong signals with "+" (R1+, R2+, R3+, R4+, R5+), the original risk level is level one (alarm level), the risk is caused by the simultaneous triggering of multiple strong signals, and the risk is extremely high; Finally, based on the above-mentioned signal combination type, the set of primary risk signals at the current moment is mapped to a preliminary comprehensive risk level L_pre, with a value range of {level 0, level 3, level 2, level 1}.

[0055] Furthermore, based on the current rainfall stage conditions, the original risk level is adjusted upwards or downwards to obtain a preliminary comprehensive risk level. The current rainfall stage conditions include: the initial rainfall stage, the rainfall duration stage, and the rainfall dissipation stage.

[0056] As a feasible implementation method, the timeliness of the preliminary judgment result needs to be marked, so the preliminary comprehensive risk level L_pre needs to be slightly adjusted; that is, while outputting the preliminary risk level, the current rainfall stage is marked: 1) Initial rainfall (0 to 3 hours after the start of rainfall): At this time, the hydrological response has not yet fully manifested. If R3 or R4 signals have appeared, the preliminary level will be automatically upgraded by one level; 2) Rainfall duration (more than 3 hours after the start of rainfall): Normal judgment is made according to the judgment table (4.1); 3) Rainfall dissipation period (rainfall intensity drops to below 2 mm / h and water level begins to fall): If the preliminary level is level II or level I and has not been downgraded for more than 2 hours, the original level will be maintained (considering the lag effect) until the water level falls significantly below the safety line.

[0057] S105. Based on the deviation data of tunnel segments and through the multidimensional correlation index dataset, the preliminary comprehensive risk level is corrected based on the baseline deviation and the multidimensional correlation index data to obtain the final warning level.

[0058] Specifically, by using a multi-source monitoring index dataset, the rainfall intensity and duration at the current moment are used to determine the rainfall condition type, thus obtaining the current rainfall condition.

[0059] In one embodiment, in the real-time determination of rainfall condition type, the current rainfall condition type can be determined based on the current rainfall intensity I(t) and the duration of continuous rainfall T(t): W1 continuous light rain type, I<2.5mm / h and T≥24h; W2 moderate rain type, 2.5≤I<8mm / h and 8≤T<24h; W3 heavy rain type, 8≤I<16mm / h and 4≤T<12h.

[0060] Furthermore, the reference values ​​for segment displacement corresponding to the current rainfall conditions are retrieved. These reference values ​​are obtained from a lookup table of vertical segment displacement reference values, which is established based on physical model test data and historical field measurement data.

[0061] As a feasible implementation method, in establishing the segment vertical displacement reference value lookup table: simulation tests based on extreme rainfall simulation physical models can be used to obtain segment displacement test data under different geological conditions and rainfall combinations. Then, based on historical rainfall events of the target subway tunnel and corresponding measured segment vertical displacement data, the segment vertical displacement test data is corrected and calibrated. Afterwards, the corrected and calibrated segment vertical displacement test data is archived according to working condition type and geological conditions to generate the segment vertical displacement reference value lookup table.

[0062] In one embodiment, extreme rainfall simulation physical model tests (transparent fluid-structure interaction model test device) can be used to obtain segment displacement test data under different geological conditions and rainfall combinations. The test data is then corrected and calibrated by combining historical rainfall events in the target tunnel with corresponding measured segment displacement data. The calibrated displacement data is archived according to the working condition type and geological conditions to form the aforementioned segment vertical displacement reference value lookup table. This segment vertical displacement reference value lookup table includes, for example, a vertical displacement reference value D_base of 3 for W1 continuous light rain type; a vertical displacement reference value D_base of 8 for W2 moderate rain type; a vertical displacement reference value D_base of 16 for W3 heavy rain type; and a vertical displacement reference value D_base of 30 for W4 torrential rain type.

[0063] As a feasible implementation method, in the dynamic updating of the benchmark value, after each rainfall event ends, the actual working condition type, geological conditions and measured segment displacement data of this event can be compiled into a new record; then, a simple arithmetic average can be performed on all existing historical records under the working condition type and geological conditions; finally, the updated average value is used to refresh the corresponding value in the segment vertical displacement benchmark value lookup table.

[0064] Furthermore, the measured vertical displacement of the tunnel segment at the current moment is compared and calculated with the reference value of the segment displacement in the multi-source monitoring index dataset to obtain the displacement deviation rate of the current tunnel segment. The displacement deviation rate of the current tunnel segment is then associated with and stored with the corresponding tunnel segment location information to obtain the deviation data.

[0065] In one embodiment, the measured vertical displacement of the segment at the current moment, Dv(t), can be compared with the baseline value D_base corresponding to the working condition to calculate the displacement deviation rate ε: ε=Dv(t) / D_base.

[0066] Furthermore, based on the comparison between the displacement deviation rate and the preset displacement deviation threshold in the deviation data, a benchmark deviation correction strategy is obtained. This benchmark deviation correction strategy is used to initially adjust the preliminary comprehensive risk level upwards or downwards.

[0067] In one embodiment, the baseline deviation correction strategy includes: when the displacement deviation rate ε < 0.6, it indicates that the current measured displacement is significantly lower than the baseline value, and the segment displacement risk is low. If the preliminary comprehensive risk level L_pre is level three or level two, L_pre is downgraded by one level (e.g., level two to level three, level three to level 0), but not lower than level 0; when 0.6 ≤ ε < 0.9, it indicates that the current measured displacement is close to the baseline value, and L_pre remains unchanged; when 0.9 ≤ ε < 1.1, it indicates that the current measured displacement has reached the baseline value, and the segment displacement risk is increased. If L_pre is level 0 or level three, L_pre is upgraded by one level (e.g., level 0 to level three, level three to level two); when ε ≥ 1.1, it indicates that the current measured displacement has exceeded the baseline value by 10%, and the segment displacement risk is extremely high. Regardless of the preliminary comprehensive risk level L_pre, the final warning level is directly set to level one (alarm level).

[0068] Furthermore, based on the linkage characteristic types present in the multidimensional correlation index data, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy is subject to a second upward adjustment control. This includes: if the linkage characteristic type is a linkage between water level and rainfall, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy is increased by one level. If the linkage characteristic type is a linkage between stress and water level, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy is directly determined as an alarm level. If the linkage characteristic type is a linkage between displacement and seepage, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy is increased by one level.

[0069] In one embodiment, the linkage characteristics present in the multidimensional correlation index data are used to achieve a secondary upward adjustment of the preliminary comprehensive risk level. This includes: if a C1 signal (a signal of linkage between water level and rainfall, with a water level response lag τ≤2h and a continuously rising water level) has been identified, it indicates that the rainfall infiltration path is unobstructed, and the timeliness of the buoyancy effect on the pipe segments is enhanced. In this case, the final warning level is increased by one level based on the baseline deviation correction strategy (but not exceeding one level); if a C2 signal (a signal of linkage between stress and water level, with an abnormal change in the stress-water level linkage rate k(t)) has been identified, it indicates that the pipe segment structure has shown an abnormal response, which is a precursor signal of pipe segment displacement. In this case, regardless of the ε value, the final warning level is directly set to at least level two (if the baseline deviation correction strategy has already determined it to be level one, then level one is maintained); if a C3 signal (a signal of linkage between displacement and seepage, with a sudden drop in the displacement-seepage ratio r(t) exceeding 40%) has been identified, it indicates that a seepage channel has been formed, which is a precursor signal that the pipe segments are about to accelerate upward movement. In this case, the final warning level will be raised by one level based on the baseline deviation correction strategy (but not exceeding one level).

[0070] Furthermore, after a second upward adjustment of the preliminary comprehensive risk level, which has been corrected using the benchmark deviation correction strategy, the final warning level and corresponding final warning information are obtained. The final warning information includes: the triggered primary risk signal, the corresponding current tunnel segment displacement deviation rate and segment displacement benchmark value, the type of linkage characteristics present in the triggered multi-dimensional correlation index data, and the current tunnel segment location information.

[0071] In one embodiment, the final warning level L_final is determined by combining the benchmark deviation correction strategy (first correction rule) and the linkage feature type correction strategy (second correction rule) existing in the multi-dimensional correlation indicator data. The determination process is as follows: First, according to the first correction rule, L_pre is corrected with the ε value to obtain the preliminary corrected warning level L_temp; then, according to the second correction rule, the trigger status of C1, C2, and C3 is checked, and L_temp is corrected a second time to obtain the final warning level L_final; the value range of L_final is {level 0, level 3, level 2, level 1}.

[0072] In one embodiment, the risk states that can directly trigger a Level 1 alarm (alarm level) are: 1) When the segment stress S(t) exceeds 75% of the segment design strength, a Level 1 alarm is directly triggered (regardless of the initial comprehensive risk level L_pre and displacement deviation rate ε); 2) When the segment vertical displacement Dv(t) exceeds 1.5 times the baseline value D_base, a Level 1 alarm is directly triggered (regardless of L_pre); 3) When the segment joint seepage rate Q(t) increases by more than 15 times the initial value within 30 minutes, a Level 1 alarm is directly triggered (regardless of L_pre).

[0073] S106. Based on the final warning level, generate corresponding emergency response recommendations.

[0074] Specifically, the system first receives the final warning level and the corresponding final warning information. If the final warning level is a "Concern" level, emergency response recommendations for that level are generated based on the final warning information. If the final warning level is a "Warning" level, emergency response recommendations for that level are generated based on the final warning information. If the final warning level is an "Alarm" level, emergency response recommendations for that level are generated based on the final warning information.

[0075] In one embodiment, the recommended emergency response measures for Level 1 Concern are as follows: ① Increase the data collection frequency from the usual 10 minutes to 5 minutes; ② Notify maintenance personnel to conduct on-site manual inspections of the warning section to check for visible abnormalities in the segment joints and track bed structure; ③ Report the warning information to the subway operation control center for record-keeping; ④ Track subsequent weather forecasts and water level changes, updating the assessment results every 2 hours. Recommended emergency response measures for Level 2 Warning are as follows: ① Activate the Level 2 emergency response plan; ② Based on the Level 3 measures, arrange a professional detection team with ground-penetrating radar to detect cavities behind the tunnel wall in the warning section; ③ If cavities are detected, immediately mark the location and size of the cavities and initiate the grouting filling process; ④ Check the drainage function of the circumferential drainage pipes in the warning section; if drainage is found to be poor, use high-pressure water jet cleaning or re-lay the drainage pipes; ⑤ Deploy temporary displacement monitoring targets on the tunnel inner wall and conduct manual verification measurements using a total station. Recommended emergency response measures for alarm level: ① Activate the Level 1 emergency response plan and notify relevant departments such as subway operations, emergency management, and fire rescue; ② Immediately organize a professional repair team to rush to the scene; ③ If the segment uplift continues to increase, implement emergency dewatering measures outside the tunnel (open surrounding emergency dewatering wells) to quickly lower the groundwater level outside the tunnel; ④ Simultaneously implement grouting reinforcement behind the tunnel wall, using fast-setting grouting materials to seal the voids behind the wall; ⑤ If the segments have cracked or have obvious water seepage, use epoxy resin to seal the cracks and reinforce the structure; ⑥ Based on the segment displacement and deformation, activate the train speed limit operation plan, and implement section suspension if necessary.

[0076] As a feasible implementation method, based on the obtained final warning level, final warning information and emergency response recommendations, by optimizing the grouting process to fill the voids behind the tunnel segment wall, optimizing the spacing of the circumferential drainage pipes, and improving the permeability of the geotextile, the seepage field around the tunnel can be effectively optimized and the pore water pressure reduced.

[0077] In addition, this application also provides a graded early warning device for the displacement of subway tunnel segments, such as... Figure 2 As shown, the graded early warning device 200 for subway tunnel segment displacement specifically includes: At least one processor 201; and a memory 202 communicatively connected to the at least one processor 201; wherein the memory 202 stores instructions executable by the at least one processor 201 to enable the at least one processor 201 to execute: By deploying a monitoring network in the target subway tunnel, data on the displacement status of the tunnel segments under the influence of precipitation are monitored and processed to obtain a multi-source monitoring index dataset. Based on the multi-source monitoring index dataset, a correlation analysis based on the precursor characteristics of tunnel segment displacement is performed on the multi-source monitoring index data to obtain a multi-dimensional correlation index dataset. The trigger conditions for risk thresholds are determined on the multi-source monitoring index dataset to obtain primary risk signals that represent the displacement risk of tunnel segments. The primary risk signals include: primary rainfall intensity risk signal, primary water level change risk signal, primary segment stress risk signal, primary vertical displacement risk signal, and primary seepage rate risk signal. Based on different combinations of primary risk signals, a preliminary comprehensive risk level is determined; Based on the deviation data of tunnel segments and through a multidimensional correlation index dataset, the preliminary comprehensive risk level is corrected based on the baseline deviation and multidimensional correlation index data to obtain the final warning level. Based on the final warning level, corresponding emergency response recommendations are generated.

[0078] This application achieves cross-validation of the same risk event through three independent dimensions: signal combination breadth, measured-benchmark deviation, and precursor correlation characteristics. This overcomes the problem of false alarms triggered by sensor disturbances or random fluctuations due to a single indicator, significantly reducing the false alarm rate on-site. It achieves multi-dimensional cross-validation, greatly reducing the false alarm rate. Furthermore, by using three correlation analysis indicators—water level response lag time, stress-water level linkage change rate, and displacement-seepage ratio—it can detect precursor signals such as abnormal structural responses and seepage channel formation before the segment displacement reaches the benchmark value. This enables the identification of risk precursor characteristics of tunnel segment displacement and provides early warning before displacement exceeds limits. Moreover, this application abandons the traditional fixed threshold approach and establishes a displacement benchmark value lookup table based on rainfall condition classification. The benchmark value adaptively changes with geological conditions and rainfall conditions. Different sections and different rainfall scenarios use matching early warning reference standards, greatly improving the accuracy and targeting of early warnings. This allows the early warning threshold to dynamically adapt and accurately match different conditions.

[0079] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0080] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0081] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0082] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0083] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0084] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0085] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0086] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0087] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0088] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0089] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of this specification.

Claims

1. A graded early warning method for displacement of subway tunnel segments, characterized in that, The method includes: By deploying a monitoring network in the target subway tunnel, data on the displacement status of the tunnel segments under the influence of precipitation are monitored and processed to obtain a multi-source monitoring index dataset. Based on the multi-source monitoring index dataset, a correlation analysis based on the precursor characteristics of tunnel segment displacement is performed on the multi-source monitoring index data to obtain a multi-dimensional correlation index dataset. The trigger conditions for risk thresholds are determined on the multi-source monitoring index dataset to obtain primary risk signals representing tunnel segment displacement risks; wherein, the primary risk signals include: primary rainfall intensity risk signal, primary water level change risk signal, primary segment stress risk signal, primary vertical displacement risk signal, and primary seepage rate risk signal; Based on the different combinations of the aforementioned primary risk signals, a preliminary comprehensive risk level is determined; Based on the deviation data of the tunnel segments and through the multidimensional correlation index dataset, the preliminary comprehensive risk level is corrected based on the baseline deviation and the multidimensional correlation index data to obtain the final warning level. Based on the final warning level, corresponding emergency response recommendations are generated.

2. The graded early warning method for subway tunnel segment displacement according to claim 1, characterized in that, By deploying a monitoring network in the target subway tunnel, data monitoring and processing of the tunnel segment displacement status under the influence of precipitation are performed to obtain a multi-source monitoring index dataset, specifically including: Based on the predetermined longitudinal spacing of the target subway tunnel, several monitoring sections are divided; and multi-source monitoring equipment is deployed in each of the monitoring sections; wherein, the multi-source monitoring equipment includes: rainfall monitoring equipment, water level monitoring equipment, water pressure monitoring equipment, segment displacement monitoring equipment, segment stress monitoring equipment, and seepage monitoring equipment; The multi-source monitoring equipment is used to monitor the target subway tunnel under rainfall conditions, and the multi-source monitoring index dataset is collected synchronously based on a unified time label. The multi-source monitoring index data in the multi-source monitoring index dataset includes at least: real-time rainfall intensity, cumulative rainfall, groundwater level depth, water level change rate, pore water pressure, segment vertical displacement, segment horizontal displacement, segment rotation angle, segment stress, and joint seepage rate.

3. The graded early warning method for subway tunnel segment displacement according to claim 1, characterized in that, Based on the multi-source monitoring index dataset, a correlation analysis based on the precursor characteristics of tunnel segment displacement is performed on the multi-source monitoring index data to obtain a multi-dimensional correlation index dataset, specifically including: Using the multi-source monitoring index dataset, the time lag coefficient between the start time of the rainfall event and the groundwater level response time is calculated, and the linkage characteristics between water level and rainfall are obtained based on the real-time calculated water level change rate. Based on the correlation between the stress data of the tunnel segments and the groundwater level depth data of the same monitoring section at various times under rainfall conditions, a stress-water level linkage curve is obtained; and based on the stress-water level linkage curve, the linkage change rate between the water level change and the stress change is calculated; the linkage change rate is determined as the stress-water level linkage characteristic; wherein, the stress-water level linkage characteristic represents the sensitivity of the tunnel segment structure to water level changes; Using the multi-source monitoring index dataset, a linkage analysis is performed on the segment displacement data and joint seepage rate of the same monitoring section to calculate the linkage ratio between the segment displacement change and the seepage rate change; and the linkage ratio is determined as the displacement-seepage linkage feature; wherein, the displacement-seepage linkage feature represents the probability of seepage channels formed by segment joints or back wall cavities. The multidimensional correlation index dataset is obtained by combining the water level-rainfall linkage feature, the stress-water level linkage feature, and the displacement-seepage linkage feature with index items.

4. A graded early warning method for subway tunnel segment displacement according to claim 1, characterized in that, Both the multi-source monitoring index dataset and the multi-dimensional correlation index dataset are subjected to risk threshold trigger condition judgment to obtain a primary risk signal representing the displacement risk of tunnel segments, specifically including: The real-time rainfall intensity in the multi-source monitoring index dataset is classified and processed by a preset rainfall intensity threshold to obtain the primary rainfall intensity risk signal; wherein, the primary rainfall intensity risk signal includes a first-class primary rainfall intensity risk signal and a second-class primary rainfall intensity risk signal; The water level change rate in the multi-source monitoring index dataset is judged and divided according to a preset water level change threshold to obtain the primary water level change risk signal; wherein, the primary water level change risk signal includes: a first-class primary water level change risk signal and a second-class primary water level change risk signal. The segment stress in the multi-source monitoring index dataset is classified according to a preset stress threshold to obtain the primary segment stress risk signal; wherein, the primary segment stress risk signal includes: a first-class primary segment stress risk signal and a second-class primary segment stress risk signal. The vertical displacement of the tunnel segments in the multi-source monitoring index dataset is judged and divided according to a preset vertical displacement threshold to obtain the primary vertical displacement risk signal; wherein, the primary vertical displacement risk signal includes: a first-class primary vertical displacement risk signal and a second-class primary vertical displacement risk signal; The joint seepage rate in the multi-source monitoring index dataset is judged and divided according to a preset seepage rate threshold to obtain the primary seepage rate risk signal; wherein, the primary seepage rate risk signal includes: a first-class primary seepage rate risk signal and a second-class primary seepage rate risk signal. Among them, the first type of primary risk signal and the second type of primary risk signal represent risk signals under different risk states; The primary risk signal is obtained by jointly storing the primary rainfall intensity risk signal, the primary water level change risk signal, the primary segment stress risk signal, the primary vertical displacement risk signal, and the primary seepage rate risk signal.

5. A graded early warning method for subway tunnel segment displacement according to claim 1, characterized in that, Based on different combinations of the aforementioned primary risk signals, a preliminary comprehensive risk level is determined, specifically including: The existence type analysis of the first-class primary risk signal and the second-class primary risk signal contained in the primary risk signal is performed at the same time to generate the current risk signal combination type. Based on the current combination of risk signals, an original risk level is obtained; wherein, the original risk level includes: normal level, attention level, warning level, and alarm level; The initial comprehensive risk level is obtained by adjusting the original risk level up or down based on the current rainfall stage conditions; wherein, the current rainfall stage conditions include: the initial rainfall stage, the rainfall duration stage, and the rainfall dissipation stage.

6. A graded early warning method for subway tunnel segment displacement according to claim 1, characterized in that, Based on the deviation data of tunnel segments and through the multidimensional correlation index dataset, the preliminary comprehensive risk level is corrected based on the baseline deviation and the multidimensional correlation index data to obtain the final warning level, specifically including: By using a multi-source monitoring index dataset, the rainfall intensity and duration at the current moment are used to determine the rainfall condition type, thus obtaining the current rainfall condition. Query the segment displacement reference value corresponding to the current rainfall condition; wherein the segment displacement reference value is obtained from the segment vertical displacement reference value lookup table, and the segment vertical displacement reference value lookup table is established based on physical model test data and field measured historical data; The measured vertical displacement of the tunnel segment at the current moment in the multi-source monitoring index dataset is compared and calculated with the displacement reference value of the tunnel segment to obtain the displacement deviation rate of the current tunnel segment; and the displacement deviation rate of the current tunnel segment is associated with and stored with the corresponding tunnel segment position information to obtain the deviation data. Based on the comparison between the displacement deviation rate and the preset displacement deviation threshold in the deviation data, a benchmark deviation correction strategy is obtained; wherein, the benchmark deviation correction strategy is used to perform a preliminary upward or downward adjustment of the risk level of the preliminary comprehensive risk level; Based on the linkage characteristics present in the multidimensional correlation index data, a second upward adjustment is made to the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy, including: If the linkage feature type is a water level and rainfall linkage feature, the preliminary comprehensive risk level after being corrected by the benchmark value deviation correction strategy will be upgraded by one level. If the linkage feature type is stress and water level linkage feature, then the preliminary comprehensive risk level after being corrected by the benchmark value deviation correction strategy will be directly judged as the alarm level. If the linkage feature type is a displacement and seepage linkage feature, the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy will be upgraded by one level. After a second upward adjustment of the preliminary comprehensive risk level after correction by the benchmark deviation correction strategy, the final warning level and the corresponding final warning information are obtained; wherein, the final warning information includes: the triggered primary risk signal, the corresponding current tunnel segment displacement deviation rate and segment displacement benchmark value, the linkage feature type existing in the triggered multi-dimensional correlation index data, and the tunnel segment location information of the current tunnel segment.

7. A graded early warning method for subway tunnel segment displacement according to claim 6, characterized in that, The method for establishing the reference value lookup table for the vertical displacement of the tunnel segments includes: Based on the simulation test of the physical model of extreme rainfall, test data of tunnel segment displacement under different geological conditions and different combinations of rainfall conditions were obtained. Based on historical rainfall events and corresponding measured vertical displacement data of the tunnel segments, the test data of vertical displacement of the tunnel segments are corrected and calibrated. The corrected and calibrated vertical displacement test data of the tunnel segments are archived according to the working condition type and geological conditions, and a reference table of vertical displacement values ​​of the tunnel segments is generated.

8. A graded early warning method for subway tunnel segment displacement according to claim 1, characterized in that, Based on the final warning level, corresponding emergency response recommendations are generated, including: Receive the final warning level and the corresponding final warning information; If the final warning level is a concern level, then based on the final warning information, a concern-level emergency response recommendation will be generated. If the final warning level is a warning level, then based on the final warning information, emergency response recommendations for the warning level are generated; If the final warning level is an alarm level, then based on the final warning information, alarm-level emergency response recommendations will be generated.

9. A graded early warning device for displacement of subway tunnel segments, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, enabling the at least one processor to execute a graded early warning method for subway tunnel segment displacement according to any one of claims 1-8.

10. A non-volatile computer storage medium, characterized in that, The storage medium is a non-volatile computer-readable storage medium that stores at least one program, each program including instructions that, when executed by a terminal, cause the terminal to perform a graded early warning method for subway tunnel segment displacement according to any one of claims 1-8.

Images