Underground settlement risk dynamic prediction method and system based on time sequence remote sensing

Abstract

The invention discloses an underground settlement risk dynamic prediction method and system based on time sequence remote sensing, and relates to the technical field of prediction data processing, and the method comprises the steps: obtaining a target ground region and a sub-region, and collecting an initial deformation sequence of the sub-region; obtaining a stability ratio of the sub-regions, and obtaining a difference index of adjacent sub-regions; obtaining an average stability ratio, if a difference index between the ith sub-region and the jth sub-region is lower than a product of the average stability ratio and a first preset threshold value, fusing the ith sub-region and the jth sub-region to form a consistency prediction region, and fusing to form a target deformation sequence of the consistency prediction region; and obtaining a risk index of the consistency prediction region according to the target deformation sequence of the consistency prediction region, and obtaining a risk prediction result according to the risk index of the consistency prediction region, the average stability ratio and a second preset threshold. The method has the advantages of self-adaptability, collaborative association and sensitive prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of predictive data processing, in particular to a dynamic prediction method and system for underground subsidence risk based on time-series remote sensing. BACKGROUND

[0002] The problem of land subsidence induced by overexploitation of groundwater, large-scale engineering construction or geological activities is increasingly prominent, and its potential risks include major disasters such as rupture of underground pipelines, instability of building structures and ground collapse.

[0003] Currently, the prediction of ground subsidence risk mainly relies on the surface height change data collected by time-series remote sensing technology. However, on the one hand, existing methods can usually only analyze the deformation trend of a single monitoring area independently, ignoring the fact that subsidence often has correlation and synergy in spatial distribution. This isolated analysis mode cannot effectively capture the linked deformation of adjacent areas due to the continuity of geological structure or load transmission effect, which may lead to missing of key risk information or misjudgment of abnormal areas. On the other hand, existing risk assessment models have deficiencies in adaptability. The deformation sequence itself is often accompanied by fluctuations caused by measurement noise, environmental interference (such as cloud cover) or seasonal changes. Due to differences in geological conditions, the stability of data in different areas is also different (e.g., large fluctuations in soft soil areas and small fluctuations in rock foundation areas). If a fixed threshold is used to uniformly judge subsidence risk or divide areas, the area with low stability is easy to trigger frequent false alarms due to noise interference, while the area with high stability may mask slow but continuous accelerated subsidence signals, which seriously affects the accuracy and reliability of prediction. In addition, the existing technology often relies on cumulative subsidence or a single rate value to quantify risk, which is difficult to sensitively identify potential dangerous acceleration trends in the deformation process (such as sudden change from uniform subsidence to accelerated subsidence), and the early warning of critical risk is delayed. In summary, there is an urgent need for a dynamic prediction method that can effectively integrate spatio-temporal correlation information, adaptively process data volatility, and sensitively capture signs of subsidence acceleration, to improve the accuracy and timeliness of early warning of regional subsidence risk in complex geological environments. SUMMARY

[0004] In view of the technical problems described in the background art, the present application provides a dynamic prediction method and system for underground subsidence risk based on time-series remote sensing.

[0005] The application discloses a kind of underground subsidence risk dynamic prediction methods based on time sequence remote sensing, method includes: obtaining target ground area and dividing it into multiple sub-regions, collect the ground surface remote sensing subsidence of each sub-region in multiple detection periods before current time and form initial deformation sequence;According to the initial deformation sequence of each sub-region, the stable proportion of each sub-region is obtained, and the difference index of each adjacent sub-region is obtained according to the initial deformation sequence of each adjacent sub-region;The average stable proportion between the i th sub-region and the j th sub-region is obtained, if the difference index between the i th sub-region and the j th sub-region is lower than the product of average stable proportion and first preset threshold, then the i th sub-region and the j th sub-region are fused and form a consistent prediction region, the target deformation sequence of the consistent prediction region is fused by the two initial deformation sequences corresponding to the consistent prediction region;According to the target deformation sequence of each consistent prediction region, the risk index of each consistent prediction region is obtained, and the risk prediction result is obtained according to the risk index of each consistent prediction region, average stable proportion and second preset threshold.

[0006] Optionally, according to the target deformation sequence of each consistent prediction region, the risk index of each consistent prediction region includes: the ground surface remote sensing subsidence of the next adjacent data in the target deformation sequence of each consistent prediction region is subtracted from the previous ground surface remote sensing subsidence, and the subsidence difference value is obtained;All subsidence difference values in the target deformation sequence of each consistent prediction region are added, and the difference total is obtained;The risk index of each consistent prediction region is obtained by dividing the difference total of the target deformation sequence of each consistent prediction region by the average value of the ground surface remote sensing subsidence of the target deformation sequence.

[0007] Optionally, according to the risk index of each consistent prediction region, average stable proportion and second preset threshold, the risk prediction result includes: the average stable proportion is multiplied by the second preset threshold to obtain a modified threshold;If the risk index of the k th consistent prediction region exceeds the modified threshold, the risk prediction result that the k th consistent prediction region exists subsidence anomaly is generated, otherwise the risk prediction result that the k th consistent prediction region normally subsides is generated.

[0008] Optionally, according to the initial deformation sequence of each sub-region, the stable proportion of each sub-region includes: the absolute value of the difference value of adjacent ground surface remote sensing subsidence in the initial deformation sequence of each sub-region is obtained, and the absolute difference value is obtained;The proportion of the absolute difference value exceeding the difference threshold value in the initial deformation sequence of each sub-region is obtained, and is used as the stable proportion of each sub-region.

[0009] Optionally, the acquiring the difference index of each adjacent sub-region according to the initial deformation sequence of each adjacent sub-region comprises: obtaining the difference between the ground surface remote sensing subsidence amount of the initial deformation sequence of the i-th sub-region and the initial deformation sequence of the j-th sub-region at the n-th detection period and taking the absolute value, and obtaining the absolute difference amount at the n-th detection period; and acquiring the difference index of the i-th sub-region and the j-th sub-region according to the absolute difference amount of each detection period in the initial deformation sequence of the i-th sub-region and the j-th sub-region.

[0010] Optionally, the fusing the two initial deformation sequences corresponding to the consistency prediction region and forming the target deformation sequence of the consistency prediction region comprises: taking the average value of the sum of the ground surface remote sensing subsidence amount of the two initial deformation sequences corresponding to the consistency prediction region at the n-th detection period as the data of the n-th detection period in the target deformation sequence of the consistency prediction region.

[0011] Also provided is a dynamic prediction system for underground subsidence risk based on time-series remote sensing, comprising: an acquisition module configured to acquire a target ground surface region and divide the target ground surface region into a plurality of sub-regions, and collect the ground surface remote sensing subsidence amount of each sub-region in a plurality of detection periods before the current time and form an initial deformation sequence; a first data processing module configured to acquire the stable proportion of each sub-region according to the initial deformation sequence of each sub-region, and acquire the difference index of each adjacent sub-region according to the initial deformation sequence of each adjacent sub-region; a second data processing module configured to acquire the average stable proportion between the i-th sub-region and the j-th sub-region, and if the difference index between the i-th sub-region and the j-th sub-region is lower than the product of the average stable proportion and a first preset threshold, fuse the i-th sub-region and the j-th sub-region and form a consistency prediction region, and fuse the two initial deformation sequences corresponding to the consistency prediction region and form the target deformation sequence of the consistency prediction region; and a risk prediction module configured to acquire the risk index of each consistency prediction region according to the target deformation sequence of each consistency prediction region, and acquire a risk prediction result according to the risk index of each consistency prediction region, the average stable proportion and a second preset threshold.

[0012] Optionally, the risk prediction module is further configured to: subtract the ground surface remote sensing subsidence amount of the previous one from the ground surface remote sensing subsidence amount of the next one of adjacent data in the target deformation sequence of each consistency prediction region, and obtain a subsidence amount difference value; add all the subsidence amount difference values in the target deformation sequence of each consistency prediction region, and obtain a difference value sum; and divide the difference value sum of the target deformation sequence of each consistency prediction region by the average value of the ground surface remote sensing subsidence amount of the target deformation sequence, and obtain the risk index of each consistency prediction region.

[0013] Optionally, the risk prediction module is further configured to: multiply the average stability ratio by a second preset threshold to obtain a modified threshold; if the risk index of the kth consistency prediction region exceeds the modified threshold, generate a risk prediction result that the kth consistency prediction region has a risk of subsidence anomaly, and otherwise generate a risk prediction result that the kth consistency prediction region has normal subsidence.

[0014] Optionally, the first data processing module is further configured to: take the absolute value of the difference of the remote sensing subsidence amount of adjacent ground surfaces in the initial deformation sequence of each sub-region, and obtain an absolute difference value; and obtain the proportion of the absolute difference value that exceeds a difference threshold in the initial deformation sequence of each sub-region as the stability ratio of each sub-region.

[0015] The beneficial effects of the present application are embodied in:

[0016] In the entire underground subsidence risk dynamic prediction method based on time-series remote sensing, first, the consistency prediction region is dynamically divided based on the difference index and the stability ratio, breaking through the limitations of traditional fixed grid isolated analysis, avoiding information fragmentation in geological continuous areas, retaining signal strength in abnormal independent areas, and improving spatial risk coverage; further, in the regional fusion stage, the lower the average stability ratio, the more stringent the difference index threshold (such as the need for strict cooperation in the backfill area to merge), to avoid pollution of analysis data by low reliability areas; further, in the risk determination stage, the modified threshold is automatically scaled with the average stability ratio (the threshold compression triggers sensitive early warning in soft soil areas, and the threshold relaxation resists interference false alarms in bedrock areas), realizing tolerance adaptation driven by geological characteristics; further, the risk index is calculated through the nonlinear coupling of the difference sum and the sequence mean (such as the index jump caused by the rapid expansion of the industrial area cavity), which is more sensitive to the dangerous inflection point of uniform speed acceleration (even if the total amount of subsidence does not exceed the threshold), providing an intervention window of several periods ahead for the development of underground diseases. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual proportions.

[0018] Figure 1 Part of the process schematic diagram of S1 to S4 in the underground subsidence risk dynamic prediction method based on time-series remote sensing of the present application; Figure 2 Part of the process schematic diagram of S4 in the underground subsidence risk dynamic prediction method based on time-series remote sensing of the present application; Figure 3A schematic diagram of steps of the underground subsidence risk dynamic prediction method based on time-series remote sensing of the present application; Figure 4 A schematic diagram of part of the steps of S2 in the underground subsidence risk dynamic prediction method based on time-series remote sensing of the present application; Figure 5 Another schematic diagram of part of the steps of S2 in the underground subsidence risk dynamic prediction method based on time-series remote sensing of the present application; Figure 6 A schematic diagram of part of the steps of S4 in the underground subsidence risk dynamic prediction method based on time-series remote sensing of the present application; Figure 7 Another schematic diagram of part of the steps of S4 in the underground subsidence risk dynamic prediction method based on time-series remote sensing of the present application. DETAILED DESCRIPTION

[0019] In order to make the objects, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some but not all of the embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0021] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", etc. are only used for differentiation in description, and cannot be understood as indicating or implying relative importance.

[0022] As shown in Figures 1 to 3 , a dynamic prediction method for underground subsidence risk based on time-series remote sensing is provided, in one embodiment, the method comprises:

[0023] S1, obtaining a target ground area and dividing it into multiple sub-areas, collecting the ground surface remote sensing subsidence amount of each sub-area in multiple detection periods before the current time and forming an initial deformation sequence;

[0024] S2, obtaining the stable proportion of each sub-area according to the initial deformation sequence of each sub-area, and obtaining the difference index of each adjacent sub-area according to the initial deformation sequence of each adjacent sub-area;

[0025] S3, obtaining an average stability ratio between the i-th sub-region and the j-th sub-region, and if a difference index between the i-th sub-region and the j-th sub-region is lower than a product of the average stability ratio and a first preset threshold, merging the i-th sub-region and the j-th sub-region to form a consistent prediction region, and merging two initial deformation sequences corresponding to the consistent prediction region to form a target deformation sequence of the consistent prediction region;

[0026] S4, obtaining a risk index of each consistent prediction region according to the target deformation sequence of each consistent prediction region, and obtaining a risk prediction result according to the risk index of each consistent prediction region, the average stability ratio, and a second preset threshold.

[0027] In this embodiment, it should be noted that in S1, the key to obtaining the target ground region and dividing it into multiple sub-regions lies in adapting to the spatial heterogeneity of geological structure. For example, when monitoring the subsidence risk of a new urban area, the groundwater level change gradient, soil compressibility distribution and building load difference need to be considered comprehensively: if the entire new area is regarded as a single region, local accelerated subsidence points (such as soft soil layers along subway tunnels) may be hidden; therefore, the sub-region division needs to take into account the geological exploration results and infrastructure layout, such as taking a 500m x 500m grid as a basic unit, further encrypting the grid to 100m x 100m around important facilities (bridge pile foundation, underground pipe gallery), to ensure that high-risk areas are finely monitored, and this division can be a simple geometric cutting.

[0028] Subsequently, assuming that the interferometric synthetic aperture radar (InSAR) technology of satellite is used, the elevation data is obtained every 12 days, and each detection period is 12 days interval; for a sub-region of a filled soil area, the average height change amount of each period (such as the surface decreases by 1.2mm from period 1 to period 2, decreases by 0.8mm from period 2 to period 3...) in the continuous 24 months (about 60 periods) is extracted, and these height change amounts arranged in time sequence constitute the "initial deformation sequence".

[0029] Further, the essence of the initial deformation sequence is to reflect the dynamic fingerprint of the geological response of the sub-region, and the construction process implies an adaptability process for the subsidence mechanism. For example, in a coastal soft soil area, due to the hysteresis effect of pore water pressure dissipation, periodic fluctuations (accelerated subsidence in rainy season, slow in dry season) may appear in the sequence, which needs to be distinguished from the real geological risk, therefore, the surface remote sensing subsidence amount in the sequence is not the original observation value, but the relative change amount after calibration by the ground control point.

[0030] For example, in a clay sub-region of an industrial park, the sequence from January 2023 to January 2024 shows that the settlement amount stabilized at 0.5-0.7 mm per month in the first 6 months (factory uniform loading period), and then increased to 1.5-2.0 mm per month in the last 6 months (groundwater pipe rupture induced aquifer compression). This sequence containing an acceleration inflection point can better reveal potential risks than simply accumulating the settlement amount.

[0031] For example, in a highway sub-region, if the InSAR sequence shows that the settlement amount exceeds the threshold value (such as >3 mm / cycle) for three consecutive cycles, the tiltmeter data of this section should be checked synchronously to determine whether the horizontal displacement is abnormal (which may cause slope instability), thereby confirming the engineering significance of remote sensing data. It is worth emphasizing that the sequence at the sub-region scale overcomes the drawbacks of regional averaging: a residential sub-region rebounds 0.3 mm / cycle due to the closure of deep water wells, while an adjacent commercial area settles 1.1 mm / cycle due to new foundation pit excavation. If they are combined for analysis, it will lead to a risk misjudgment. Independent sequences can capture the stability signal of the rebound area and the acceleration signal of the excavation area, providing data basis for the stability evaluation of S2 and the regional fusion of S3, ultimately achieving the goal of dynamic prediction and identifying the acceleration settlement inflection point before surface cracks are visible.

[0032] In S2, the stability ratio is obtained from the time dimension fluctuation analysis of the initial deformation sequence of each sub-region: for any sub-region, the calculation result of the absolute value of the settlement amount difference between adjacent detection cycles in its sequence (S21) directly reflects the intensity of short-term deformation shock. For example, in a coastal backfill area, if the groundwater level is frequently fluctuated by tides, it may cause the settlement amount difference between adjacent cycles in the sequence to exceed the preset difference threshold value (such as 1.0 mm) multiple times. In this case, the proportion of difference values exceeding the threshold value is very high, and the stability ratio tends to 0. Conversely, in rock foundation distribution areas, due to the low compressibility of soil, the settlement amount difference between adjacent cycles is very small (such as the absolute value of the difference is generally less than 0.2 mm), the proportion of exceeding the threshold value is close to 0, and the stability ratio tends to 1. This parameter not only identifies the reliability of the region's own data (weak noise interference in high proportion areas), but also deeply reflects the inherent properties of geological media: the fluctuation caused by the lag of pore water pressure dissipation in soft soil areas is quantified as a low stability ratio, and the mechanical rigidity of rock foundation areas is manifested as a high stability ratio.

[0033] Further, the difference index focuses on the spatial dimension, revealing the spatial correlation of geological response by comparing the settlement behaviors of adjacent sub-regions at the same time sequence node (S23). A typical case is two sub-regions across the same aquifer: if the aquifer pumping triggers uniform settlement, the difference in settlement amount between the two regions at each cycle is minimal, and the proportion of statistical absolute difference exceeding the preset difference amount (such as the allowable fluctuation tolerance of the region) is extremely low, and the difference index tends to 0; but if there is a hidden fault under a certain sub-region, causing its settlement rate to be significantly higher than that of the adjacent region, the absolute difference amount at each cycle continues to exceed the tolerance, the proportion of statistical exceedance rises, and the difference index significantly increases. The essence of the difference index is to measure the degree of failure of spatial cooperative deformation - an index of 0 indicates that the load transfer or geological continuity is not damaged, and an increasing index indicates local abnormal disturbance or boundary effects.

[0034] Further, the calculation of the stability ratio uses the form of super-threshold proportion, which has the physical meaning of distinguishing between normal fluctuations and abnormal jumps: in the seasonal frozen soil area, the settlement difference between adjacent cycles during the melting period may periodically exceed the threshold value due to frost heaving effects, but if the super-threshold point proportion does not exceed 60% (the difference threshold is set according to the characteristics of the region), it is still considered as controllable fluctuation, and the stability ratio remains 0.4; while in a certain mining area, the single-cycle settlement difference increases by 3 times the threshold value due to the collapse of the mined-out area, and the subsequent high level continues, the super-threshold proportion reaches 90%, and the stability ratio drops to below 0.1. This mechanism avoids the defect that a fixed difference standard cannot adapt to the characteristics of the region, providing an adaptive basis for the fusion judgment of S3.

[0035] Further, the acquisition of the difference index relies on the cross-cycle difference accumulation statistics, emphasizing the verification of spatio-temporal consistency: taking adjacent regions across a subway tunnel as an example, the difference index of the two regions before the tunnel construction is stable at 0.1 (indicating uniformity of the stratum); during the shield tunneling period, the settlement of the sub-region directly above the tunnel accelerates, while the deformation of the lateral region lags behind, causing the absolute difference amount to exceed the preset value for 5 consecutive periods (such as the difference amount increasing from the normal 1 mm to 4 mm), and the proportion of the super-standard period increases significantly, causing the difference index to rise to 0.8. This parameter breaks through the existing static neighborhood analysis method, coupling time evolution and spatial comparison, and provides a quantitative basis for identifying local disconnection caused by engineering activities.

[0036] In S3, the prediction unit with geological consistency is dynamically divided based on the quantified stability ratio and difference index in S2. The prerequisite for regional fusion is that the settlement behaviors of adjacent sub-regions have spatial synergy, that is, the difference index needs to be lower than the product threshold composed of the average stability ratio of the two regions and the first preset threshold. When the first preset threshold is taken, the first preset threshold is between 0.4 and 0.6 through historical limited engineering experiments, the first preset threshold is 0.5 (default value), which is suitable for most homogeneous soil layers (such as alluvial plains); the first preset threshold is 0.4 (strict mode), which is used in high-risk areas (such as along the subway line) to prevent backfill soil disturbance area from polluting the bedrock data; the first preset threshold is 0.6 (lenient mode), which is suitable for homogeneous soft soil area and enhances the noise suppression capability.

[0037] For example, two sub-regions (sub-region A stability ratio 0.9, sub-region B stability ratio 0.8) on a certain alluvial plain, the average stability ratio is 0.85; if they are synchronous due to sharing aquifer, the difference index is only 0.1 (such as the deviation of the settlement amount of each cycle is less than the tolerance), and the first preset threshold is 0.5, the product threshold is 0.425, and the difference index is 0.1, which is significantly lower than 0.425, so the fusion mechanism is triggered.

[0038] Further, the consistent prediction region generated after fusion essentially represents a geomechanical response unit, and its boundary is no longer constrained by the initial geometric grid, but is naturally defined by load transmission continuity or rock-soil layer homogeneity. The sequence fusion uses cycle-level mean calculation: the settlement amounts of two original sequences at the same time node are averaged (such as in cycle 1, the settlement of region A is 1.0 mm, and the settlement of region B is 1.2 mm, and the fusion sequence is recorded as 1.1 mm), which suppresses local noise in the spatial dimension. However, it should be noted that if there is strong isolated disturbance in a certain region (such as sub-region B local position suddenly accelerated settlement due to pipeline leakage), the difference index rises, resulting in that the fusion condition is not met, so independent analysis is retained to avoid signal dilution.

[0039] Further, the lower the average stability ratio (indicating that at least one region data fluctuates greatly), the smaller the product threshold, and the more stringent the fusion condition. For example, a backfill soil sub-region in a certain mining area (stability ratio 0.3) and a neighboring bedrock sub-region (stability ratio 0.9), the average stability ratio is only 0.6; at this time, even if the difference index of the two is 0.25 (due to the fluctuation of backfill soil settlement, part of the cycle exceeds the standard), the product threshold is 0.3 when the first preset threshold is 0.5. Although the difference index 0.25 is lower than 0.3, it is close to the critical value, and the abnormal fluctuation of the backfill soil in the target deformation sequence after fusion will be partially covered by the bedrock data. Therefore, low reliability regions must be fused with highly synergistic regions (such as backfill soil area can only be merged with the same type of soil region and the difference is extremely small).

[0040] For example, in the case of a fault, the sub-regions on both sides of the fault have a difference index of 0.7 due to the fault movement of the rock mass (such as the north side continuously subsiding and the south side stable), even if the stable proportion of both is relatively high (0.8), the product threshold of the average stable proportion 0.8 and the preset threshold 0.5 is still far lower than 0.7, so the fusion is rejected. This ensures that the function of the fault as a geological danger boundary is not blurred, and the independent early warning ability of the high-risk area is maintained.

[0041] In S4, the calculation of the risk index reveals the trend mutation strength inherent in the sequence: the continuous accumulation of the difference between adjacent subsidence values in the target deformation sequence (the sum of the differences) essentially captures the cumulative fluctuation energy, and dividing by the average value of the sequence eliminates the disturbance of the absolute subsidence scale.

[0042] For example, the target sequence of a certain industrial area after fusion shows uniform subsidence in the early stage (the adjacent difference is stable at 0.5 mm), and the difference increases sharply to 2.0 mm in the later stage due to the expansion of the underground cavity, and the sum of the differences increases significantly; while the average value of the sequence only increases from 0.6 mm to 0.9 mm due to the influence of the early stage, at this time the risk index still jumps due to the sharp increase in the numerator, and is more sensitive than the existing rate index in identifying the acceleration turning point. This design effectively distinguishes between dangerous acceleration (sharp increase in the sum of the differences) and low-risk uniform subsidence (stable sum of the differences), and can provide early warning even if the total subsidence does not reach the warning value.

[0043] Further, the risk determination introduces the average stable proportion to dynamically modify the second threshold value: for low stable proportion areas (such as a mine backfill soil fusion area with a stable proportion of 0.3), the modified threshold value is compressed to 30% of the original threshold value (assuming the second threshold value is 0.5→0.15 after modification), and a small acceleration (risk index 0.18) triggers an alarm; while the high stable proportion area (such as the bedrock fusion area with a stable proportion of 0.9) has a modified threshold value of 0.45, and an alarm is triggered only when the risk index exceeds 0.45, avoiding false alarms caused by slight shaking of the rock layer.

[0044] Further, the dynamic scaling of the modified threshold value is essentially an adaptation of the risk tolerance of the reliability of the fusion area: the sequence in the low stable proportion area itself has large fluctuations (such as the soft soil area with high sum of differences due to rainfall disturbance), and by reducing the modified threshold value, the monitoring sensitivity is forced to be improved, which is equivalent to amplifying the dangerous signal in the noise background.

[0045] For example, the risk index of a certain reclamation area fusion target sequence is 0.25, if the area stability ratio is only 0.4 (affected by tides), the correction threshold is adjusted to 0.2 (0.4x0.5), then 0.25>0.2 triggers an early warning, and the possible dam leakage hidden danger can be responded in time; on the contrary, the same risk index appears in a clay fusion area with a stability ratio of 0.8 (correction threshold of 0.4), it is determined to be normal settlement. For the strong interference area with a stability ratio of 0.2, the correction threshold is reduced to 0.1, and the risk index needs to exceed 0.1 to report an alarm, so as to avoid frequent false alarms due to normal fluctuations; for the area with a stability ratio of 0.9, the correction threshold is increased to 0.45, and the risk index needs to break through a higher threshold to report an alarm, so as to prevent missing the deep landslide risk of slow acceleration. The final output result is a binary decision - settlement anomaly or normal settlement, and its physical meaning is to lock the fusion area whose accelerated energy exceeds the geological reliable boundary (correction threshold), so as to provide priority basis for engineering disposal.

[0046] In summary, in the whole underground settlement risk dynamic prediction method based on time sequence remote sensing, the consistency prediction area is dynamically divided based on the difference index and the stability ratio, breaking through the isolated analysis limitation of the traditional fixed grid, avoiding the information fragmentation of the geological continuous area, retaining the signal strength of the abnormal independent area, and improving the spatial risk coverage rate; further, in the area fusion stage, the lower the average stability ratio, the more stringent the difference index threshold (such as the backfill area which needs to be strictly coordinated to merge), so as to avoid pollution of the analysis data in the low reliability area; further, in the risk judgment stage, the correction threshold is automatically scaled with the average stability ratio (the threshold compression triggers sensitive early warning in soft soil area, and the threshold relaxation resists interference false alarm in bedrock area), so as to realize the tolerance adaptation driven by geological characteristics; further, the risk index is calculated through the nonlinear coupling of the difference sum and the sequence mean (such as the index jump caused by the sudden increase of the difference value due to the cavity expansion in industrial area), which is more sensitive to identify the dangerous inflection point of uniform speed to acceleration (even if the total settlement amount does not exceed the threshold), so as to provide an intervention window of several periods in advance for the development of underground diseases.

[0047] As shown in Figure 6 In one embodiment, the risk index of each consistency prediction area is obtained according to the target deformation sequence of each consistency prediction area in S4, which includes:

[0048] S41, subtracting the surface remote sensing settlement of the previous one from the surface remote sensing settlement of the next one of the adjacent data in the target deformation sequence of each consistency prediction area, and obtaining the settlement difference value;

[0049] S42, adding all the settlement difference values in the target deformation sequence of each consistency prediction area, and obtaining the difference sum;

[0050] S43, divide the difference sum of the target deformation sequence of each consistency prediction area by the surface remote sensing settlement average of the target deformation sequence, and obtain the risk index of each consistency prediction area.

[0051] In the present embodiment, it should be noted that in S41, the change gradient of adjacent time points is extracted from the target deformation sequence of the consistency prediction area. Specifically, for the target deformation sequence of each consistency prediction area (such as the industrial area fusion sequence), the settlement amount of the next detection period is calculated in time sequence, and the difference value of the previous period is calculated.

[0052] For example, the settlement amount of a certain area sequence in period 1 to period 5 is [1.2mm, 1.5mm, 1.0mm, 2.3mm, 3.1mm], then the difference value of period 2 minus period 1 is +0.3mm, the difference value of period 3 minus period 2 is -0.5mm, the difference value of period 4 minus period 3 is +1.3mm, and so on. These difference values represent the short-time settlement acceleration, and the positive value represents accelerated settlement, and the negative value represents rebound or deceleration, and the physical meaning is to capture the change intensity of deformation kinetic energy in adjacent periods.

[0053] In S42, the settlement amount difference sequence generated in S41 is accumulated. Continuing the previous example, the sum of all difference values [+0.3mm, -0.5mm, +1.3mm, +0.8mm] is 0.3-0.5+1.3+0.8=1.9mm.

[0054] For example, when a certain area uniformly settles, the difference sum tends to be zero (such as the sequence [1.0, 1.1, 1.2, 1.3] has a difference sum of only 0.3), and when there is a cavity collapse, the total sum will significantly increase due to the sharp increase in period difference (such as from +0.2mm to +1.5mm).

[0055] In S43, taking the difference sum of 1.9mm as an example, the mean value of the target sequence [1.2, 1.5, 1.0, 2.3, 3.1] is about 1.82mm, and the risk index = 1.9 / 1.82≈1.04. Among them, the numerator (difference sum) represents the cumulative fluctuation intensity, and the denominator (sequence mean) eliminates the interference of absolute settlement amount (such as the acceleration characteristics of a 5mm area and a 50mm area can be compared); the uniform settlement sequence has a low difference sum and a moderate mean, so the index tends to 0 (ideal state), and the sequence containing the acceleration inflection point (such as the sudden increase in difference value after the collapse of the area) makes the numerator increase significantly more than the denominator. For example, the mean value of a certain area is 1.5mm, the difference sum is 1.0 (the index is 0.67), the mean value increases to 2.0mm after the collapse, and the difference sum increases to 4.0 (the index is 2.0), the increase is significantly higher than the linear rate calculation.

[0056] For example, the settlement amount of a certain area sequence in period 1 to period 5 is [1.2mm, 1.5mm, 1.0mm, 2.3mm, 3.1mm], then the difference value of period 2 minus period 1 is +0.3mm, the difference value of period 3 minus period 2 is -0.5mm, the difference value of period 4 minus period 3 is +1.3mm, and so on. These difference values represent the short-time settlement acceleration, and the positive value represents accelerated settlement, and the negative value represents rebound or deceleration, and the physical meaning is to capture the change intensity of deformation kinetic energy in adjacent periods. Figure 7In one embodiment, as shown, obtaining the risk prediction result according to the risk index, the average stability ratio and the second preset threshold of each consistency prediction region in S4 includes:

[0057] S44, multiplying the average stability ratio by the second preset threshold to obtain a modified threshold;

[0058] S45, if the risk index of the kth consistency prediction region exceeds the modified threshold, generating a risk prediction result that the kth consistency prediction region has a subsidence anomaly, otherwise generating a risk prediction result that the kth consistency prediction region has normal subsidence.

[0059] In this embodiment, it should be noted that in S44, the second preset threshold is modified based on the average stability ratio of the consistency prediction region (calculated during fusion in S3). When the second preset threshold is taken, based on historical limited times of accelerated subsidence disaster data, the accelerated inflection point of 95% of the risk index is counted, for example, when the risk index is 0.6, the number of historical limited times of accelerated subsidence disasters increases, and the second preset threshold is 0.6*0.95=0.57. The second preset threshold is calibrated by limited historical accelerated events (such as 3 periods of data before collapse), which is generally between 0.4 and 0.6.

[0060] For example, the stability ratio of the backfill soil fusion region is 0.3, and the second threshold is 0.5, then the modified threshold = 0.3*0.5 = 0.15; when the stability ratio of the bedrock region is 0.9, the modified threshold is 0.45. Among them, the data of low stability ratio region (such as 0.3) has large fluctuation, and the alarm sensitivity is improved by compressing the threshold (0.15) to force small acceleration to trigger alarm; the noise interference of high stability ratio region (such as 0.9) is weak, so the threshold is relaxed (0.45) to avoid false alarm of rock mass normal creep. The modification process is essentially to set different risk tolerance boundaries for different geological units - soft soil area needs to tolerate high frequency fluctuation, so the alarm threshold is lowered; the natural fluctuation of rock mass area is small, so the alarm requirement is improved.

[0061] In S45, the risk index of a reclamation area fusion region is 0.25, if its stability ratio is 0.4 and the modified threshold is 0.2 (0.4*0.5), then 0.25>0.2 triggers the "subsidence anomaly" alarm; the same risk index 0.25 appears in the clay area (stability ratio 0.8, modified threshold 0.4) and is determined as "normal subsidence".

[0062] Among them, the modified threshold of the low reliability region (stability ratio 0.2) is reduced to 0.1 to avoid false alarm of normal high frequency fluctuation (the index needs to be greater than 0.1 to trigger alarm); the threshold of the high reliability region (stability ratio 0.9) is increased to 0.45 to prevent false alarm of deep slow landslide (which needs to be triggered by severe acceleration).

[0063] For example, Figure 4As shown, in one embodiment, obtaining the stable proportion of each sub-region based on the initial deformation sequence of each sub-region in S2 includes:

[0064] S21. Calculate the absolute value of the difference between the remotely sensed subsidence amounts of adjacent surfaces in the initial deformation sequence of each sub-region, and obtain the absolute difference.

[0065] S22. Obtain the percentage of absolute differences exceeding the difference threshold in the initial deformation sequence of each sub-region, and use it as the stable proportion of each sub-region.

[0066] In this embodiment, it should be noted that in S21, the short-term fluctuation intensity of the deformation sequence is captured by calculating the absolute value of the difference in settlement between adjacent detection cycles. Taking a certain frozen soil sub-region as an example, its initial deformation sequence shows a sudden increase in settlement during the snowmelt period (e.g., cycle 2 is 1.8 mm more than cycle 1) and a slight rebound during the freezing period (cycle 3 is 0.3 mm less than cycle 2). Then, the absolute value sequence of the difference between adjacent cycles is [|1.8|,|-0.3|,...]=[1.8,0.3,...]. This operation removes the influence of the sign and purely quantifies the deformation oscillation amplitude during the cycle: high values ​​(e.g., 1.8 mm) indicate a strong geological response (frozen soil thawing and compression), and low values ​​(e.g., 0.3 mm) reflect a stable state.

[0067] In S22, the proportion exceeding a preset difference threshold is used as the stability proportion. The difference threshold can be referenced by the 80th percentile of the absolute value of adjacent differences in the previous 10% of detection periods (e.g., if the historical difference sequence for a certain area is [0.3, 0.7, 0.2, 0.6, 1.1] mm, the 80th percentile ≈ 0.85 mm). Example: In a sub-region of soft soil, due to seasonal fluctuations in groundwater level, the difference sequence shows 0.3 mm, 1.1 mm, and 0.9 mm. Calculating the threshold according to the 80th percentile rule, it is 0.95 mm. Therefore, the proportion exceeding the threshold is only 33% (1.1 mm in period 2 exceeds the standard), and the stability proportion = 1 - 33% = 0.67.

[0068] For example, a sub-regional sequence in a mining area contains 20 adjacent periodic differences. 15 of these differences exceed the soft soil threshold of 1.0 mm due to disturbance in the goaf, representing 75% of the differences. Therefore, the stability ratio is 1 - 75% = 0.25. This ratio reflects the region's resistance to disturbance—in the permafrost region, when the periodic exceedance ratio is 40% (due to freeze-thaw cycles), the stability ratio remains at 0.6, unlike the 0.1 ratio in the subsidence region where 90% exceed the threshold.

[0069] like Figure 5 As shown, in one embodiment, obtaining the difference index of each adjacent sub-region based on the initial deformation sequence of each adjacent sub-region in S2 includes:

[0070] S23, obtaining the difference between the initial deformation sequence of the ith sub-region and the initial deformation sequence of the jth sub-region at the nth detection period and taking the absolute value, and obtaining the absolute difference amount at the nth detection period;

[0071] S24, obtaining the difference index of the ith sub-region and the jth sub-region according to the absolute difference amount of each detection period in the initial deformation sequence of the ith sub-region and the jth sub-region.

[0072] In this embodiment, it should be noted that in S23, the synchronous deformation behavior is compared cycle by cycle for adjacent sub-regions (such as A and B regions sharing an aquifer). In a homogeneous soil layer without disturbance, the settlement of A region in period 1 is 1.0 mm, and the settlement of B region is 1.05 mm, and the absolute difference amount is |1.0-1.05|=0.05 mm; when fault activity causes B region to suddenly drop to 1.8 mm in period 5 (A region is still 1.0 mm), the difference amount jumps to 0.8 mm.

[0073] This operation essentially detects the instantaneous breaking point of spatial coordination - low difference amount (<0.1 mm) represents uniform load transfer, and high difference amount (>0.5 mm) warns of local abnormalities (such as fault dislocation or pipeline leakage), and the output is the difference amount in the period.

[0074] In S24, the proportion of each period absolute difference amount generated by S23 exceeding the preset difference amount (such as 0.5 mm in the tectonic zone and 0.2 mm in the soil layer) is calculated. For example, the two sub-regions across the subway tunnel have only 1 difference amount exceeding 0.5 mm (proportion 10%, difference index 0.1) in the 10 periods before construction; and the shield passes through the period for 6 consecutive periods exceeding the threshold (proportion 60%, difference index 0.6). This index combines time and space dimensions: index 0.1 represents 10% of the periods that fail to coordinate (which can be ignored), and 0.6 represents 60% of the periods that are disconnected (which need to be isolated and analyzed).

[0075] It should also be noted that the difference index of the ith sub-region and the jth sub-region is obtained according to the absolute difference amount of each detection period in the initial deformation sequence of the ith sub-region and the jth sub-region in S24, which is represented as:

[0076] ; wherein,

[0077] is the difference index of the ith sub-region and the jth sub-region, is the number of detection periods in the initial deformation sequence, is the absolute difference amount of the nth detection period in the initial deformation sequence of the ith sub-region and the jth sub-region, is the standard difference amount of the same detection period in the initial deformation sequence of the adjacent sub-regions.

[0078] It is also necessary to explain that the whole expression, The absolute difference amount is binarized by comparing with the standard difference amount . Specifically, when (the difference is within the tolerance range), 0 or -1 is output, and after the function processing, 0 is output; when (the difference exceeds the tolerance range), +1 is output, and after the function processing, 1 is output. Among them, when taking the value, if the geological scene is a homogeneous aquifer / alluvial plain, the standard difference amount is 10% of the average value of the settlement amount (such as an average of 1.0 mm→0.1 mm), because the spatial difference of settlement caused by hydraulic gradient change is ≤10%; if the geological scene is a fault-affected zone, the standard difference amount of the rock mass area is 0.3 mm, and the standard difference amount of the soil area is 0.5 mm, which is based on the limit value of rock mass displacement / shear deformation critical value of soil; if the geological scene is the disturbance area of underground engineering (along the tunnel), the standard difference amount is 15% of the average value of the settlement amount (such as an average of 2.0 mm→0.3 mm), which is due to the upper limit of the differential settlement gradient caused by excavation unloading; if the historical difference amount of the two areas is stable at a certain value (such as 0.15 mm) for a long time during the undisturbed period, then .

[0079] Further, the existing method is difficult to distinguish between slight over-standard and significant over-standard, and through the above binarization, it is forced to focus on whether the synergy is destroyed, avoiding interference in judgment due to small deviations (such as normal fluctuation of 1.1 mm vs. tolerance of 1.0 mm in soft soil area). For boundary sensitive areas such as faults and tunnels, once the difference exceeds the geological allowable value (such as 0.8 mm caused by rock mass displacement vs. 0.3 mm of rock mass tolerance), it is immediately marked as 1, providing clear basis for spatial isolation.

[0080] Further, all the number of synergy failures in all detection periods is accumulated, and all M periods are traversed, if the difference exceeds the standard in any period, 1 is added, otherwise 0 is added. If the difference exceeds the standard for consecutive multiple periods (such as 5 / 6 periods exceeding the standard during shield crossing period), the accumulated value will increase significantly (0.83), while transient interference (such as equipment error) only exceeds the standard for a single period (1 / 6≈0.17). Short-term fluctuations (such as single-period data anomaly caused by cloud cover) are automatically filtered out as they do not form a sustained failure signal (accounting for <20%).

[0081] Further, Convert the cumulative failure times to the proportion in the interval [0, 1]. The difference index of the homogeneous aquifer tends to 0 (failure proportion <5%), triggering fusion to form a large area analysis; the fault zone index tends to 1 (failure proportion >80%), forcing spatial isolation to reserve independent risk units. Provide quantitative basis for fusion: when Less than the average stable proportion multiplied by the threshold value (such as 0.1<0.8x0.5=0.4), it is determined that the geological continuity is not damaged, and fusion is allowed; otherwise (such as 0.7>0.4) fusion is refused.

[0082] In one embodiment, the fusion of the two initial deformation sequences corresponding to the consistent prediction area in S3 and the formation of the target deformation sequence of the consistent prediction area include:

[0083] The average of the sum of the surface remote sensing subsidence in the n th detection period in the two initial deformation sequences corresponding to the consistent prediction area is taken as the data in the n th detection period in the target deformation sequence of the consistent prediction area.

[0084] In this embodiment, it should be noted that for two sub-regions determined to have geological consistency (such as A region and B region sharing aquifer), the target deformation sequence is generated by aligning the data points cycle by cycle and taking the average.

[0085] Specifically, for any detection period n, the surface remote sensing subsidence of the two sub-regions in that period is added and divided by 2 as the n th period subsidence of the fused consistent prediction area. For example, when the subsidence of A region is 1.5 mm and the subsidence of B region is 1.7 mm in period 3, the fused sequence is recorded as 1.6 mm ((1.5+1.7) / 2) in that period. This fusion mechanism realizes double optimization in the spatial dimension - it suppresses local measurement noise (such as 0.3 mm abnormal jump caused by single-point instrument error in A region, which is diluted by normal data in B region) through the average algorithm, while retaining the common characteristics of the trend deformation (such as the 0.2 mm acceleration signal appearing synchronously in the two regions due to aquifer compression, which is completely retained).

[0086] A time-series remote sensing-based underground subsidence risk dynamic prediction system is also provided, which comprises:

[0087] The acquisition module is configured to acquire a target ground area and divide it into multiple sub-regions, collect the surface remote sensing subsidence of each sub-region in multiple detection periods before the current time, and form an initial deformation sequence;

[0088] The first data processing module is configured to obtain the stable proportion of each sub-region according to the initial deformation sequence of each sub-region, and obtain the difference index of each adjacent sub-region according to the initial deformation sequence of each adjacent sub-region;

[0089] The second data processing module is configured to obtain an average stability ratio between the i-th sub-region and the j-th sub-region, and if a difference index between the i-th sub-region and the j-th sub-region is lower than a product of the average stability ratio and a first preset threshold, fuse the i-th sub-region and the j-th sub-region to form a consistent prediction region, and fuse two initial deformation sequences corresponding to the consistent prediction region to form a target deformation sequence of the consistent prediction region. The risk prediction module is configured to obtain a risk index of each consistent prediction region according to the target deformation sequence of each consistent prediction region, and obtain a risk prediction result according to the risk index of each consistent prediction region, the average stability ratio, and a second preset threshold.

[0090] In an embodiment, the risk prediction module is further configured to: subtract a previous ground surface remote sensing subsidence amount from a next ground surface remote sensing subsidence amount of adjacent data in the target deformation sequence of each consistent prediction region, and obtain a subsidence amount difference value; add all the subsidence amount difference values in the target deformation sequence of each consistent prediction region, and obtain a difference value sum; and divide the difference value sum of the target deformation sequence of each consistent prediction region by an average value of the ground surface remote sensing subsidence amount of the target deformation sequence, and obtain the risk index of each consistent prediction region.

[0091] In an embodiment, the risk prediction module is further configured to: multiply the average stability ratio by the second preset threshold to obtain a modified threshold; if the risk index of the k-th consistent prediction region exceeds the modified threshold, generate a risk prediction result that the k-th consistent prediction region has a subsidence anomaly, and otherwise generate a risk prediction result that the k-th consistent prediction region has normal subsidence.

[0092] In an embodiment, the first data processing module is further configured to: obtain an absolute difference value by taking an absolute value of a difference value of adjacent ground surface remote sensing subsidence amounts in the initial deformation sequence of each sub-region; and obtain a proportion of the absolute difference value that exceeds a difference value threshold in the initial deformation sequence of each sub-region as a stability ratio of each sub-region.

[0093] In the embodiment, it should be noted that, as for the above-mentioned underground subsidence risk dynamic prediction system based on time-series remote sensing, the specific manner of performing operations has been described in detail in the embodiments of the underground subsidence risk dynamic prediction method based on time-series remote sensing, and will not be described in detail here.

[0094] The preferred embodiments of the present disclosure are described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the specific details in the above-described embodiments. Within the technical concept range of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all belong to the protection range of the present disclosure.

[0095] It should be further noted that various specific technical features described in the above specific embodiments can be combined in any suitable manner, as long as there is no contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0096] In addition, various different embodiments of the present disclosure can also be combined in any manner, as long as it does not deviate from the idea of the present disclosure, it should also be considered as disclosed by the present disclosure.

[0097] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, and not to limit it; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and description of the present application.

Claims

1. A method for dynamic prediction of underground subsidence risk based on time-series remote sensing, characterized in that, include: The target ground area is acquired and divided into multiple sub-regions. The surface remote sensing subsidence of each sub-region is collected in multiple detection cycles before the current time and an initial deformation sequence is formed. The stable proportion of each sub-region is obtained based on the initial deformation sequence of each sub-region, and the difference index of each adjacent sub-region is obtained based on the initial deformation sequence of each adjacent sub-region. Obtain the average stability ratio between the i-th sub-region and the j-th sub-region. If the difference index between the i-th sub-region and the j-th sub-region is lower than the product of the average stability ratio and the first preset threshold, then merge the i-th sub-region and the j-th sub-region to form a consistent prediction region. Merge the two initial deformation sequences corresponding to the consistent prediction region to form the target deformation sequence of the consistent prediction region. Risk indicators for each consistent prediction region are obtained based on the target deformation sequence of each consistent prediction region, and risk prediction results are obtained based on the risk indicators, average stability ratio, and second preset threshold of each consistent prediction region.

2. The method for dynamic prediction of underground subsidence risk based on time-series remote sensing according to claim 1, characterized in that, The process of obtaining risk indicators for each consistent prediction region based on the target deformation sequence of each consistent prediction region includes: Subtract the previous land surface remote sensing subsidence from the next land surface remote sensing subsidence in the target deformation sequence of each consistent prediction area, and obtain the subsidence difference. Add up all the settlement differences in the target deformation sequence of each consistent prediction region and get the sum of the differences; The risk index for each consistent prediction area is obtained by dividing the sum of the differences in the target deformation sequences of each consistent prediction area by the average value of the surface remote sensing subsidence of the target deformation sequence.

3. The method for dynamic prediction of underground subsidence risk based on time-series remote sensing according to claim 1, characterized in that, The step of obtaining risk prediction results based on risk indicators, average stability ratios, and a second preset threshold for each consistent prediction region includes: Multiply the average stability ratio by the second preset threshold to obtain the correction threshold; If the risk index of the kth consistent prediction area exceeds the correction threshold, a risk prediction result of abnormal settlement is generated for the kth consistent prediction area; otherwise, a risk prediction result of normal settlement is generated for the kth consistent prediction area.

4. The method for dynamic prediction of underground subsidence risk based on time-series remote sensing according to claim 1, characterized in that, The step of obtaining the stable proportion of each sub-region based on the initial deformation sequence of each sub-region includes: The absolute value of the difference in remotely sensed subsidence between adjacent surfaces in the initial deformation sequence of each sub-region is calculated, and the absolute difference is obtained. Obtain the percentage of absolute differences exceeding the difference threshold in the initial deformation sequence of each sub-region, and use this percentage as the stable proportion of each sub-region.

5. The method for dynamic prediction of underground subsidence risk based on time-series remote sensing according to claim 1, characterized in that, The step of obtaining the difference index of each adjacent sub-region based on the initial deformation sequence of each adjacent sub-region includes: Obtain the difference between the initial deformation sequence of the i-th subregion and the initial deformation sequence of the j-th subregion in the n-th detection period, calculate the absolute value, and obtain the absolute difference in the n-th detection period. The difference index between the i-th and j-th sub-regions is obtained based on the absolute difference between the i-th and j-th sub-regions in the initial deformation sequence at each detection cycle.

6. The method for dynamic prediction of underground subsidence risk based on time-series remote sensing according to claim 1, characterized in that, The step of fusing the two initial deformation sequences corresponding to the consistent prediction region to form the target deformation sequence of the consistent prediction region includes: The average of the sum of the surface remote sensing subsidence amounts in the nth detection period of the two initial deformation sequences corresponding to the consistent prediction area is used as the data in the nth detection period of the target deformation sequence of the consistent prediction area.

7. A dynamic prediction system for underground subsidence risk based on time-series remote sensing, characterized in that, The system includes: The acquisition module is used to acquire the target ground area and divide it into multiple sub-regions, collect the surface remote sensing subsidence of each sub-region within multiple detection cycles before the current time, and form an initial deformation sequence. The first data processing module is used to obtain the stable proportion of each sub-region based on the initial deformation sequence of each sub-region, and to obtain the difference index of each adjacent sub-region based on the initial deformation sequence of each adjacent sub-region. The second data processing module is used to obtain the average stability ratio between the i-th sub-region and the j-th sub-region. If the difference index between the i-th sub-region and the j-th sub-region is lower than the product of the average stability ratio and the first preset threshold, the i-th sub-region and the j-th sub-region are merged to form a consistent prediction region. The two initial deformation sequences corresponding to the consistent prediction region are merged to form the target deformation sequence of the consistent prediction region. The risk prediction module is used to obtain the risk indicators of each consistent prediction region based on the target deformation sequence of each consistent prediction region, and to obtain the risk prediction results based on the risk indicators, average stability ratio and second preset threshold of each consistent prediction region.

8. The dynamic prediction system for underground subsidence risk based on time-series remote sensing according to claim 7, characterized in that, The risk prediction module is also used for: Subtract the previous land surface remote sensing subsidence from the next land surface remote sensing subsidence in the target deformation sequence of each consistent prediction area, and obtain the subsidence difference. Add up all the settlement differences in the target deformation sequence of each consistent prediction region and get the sum of the differences; The risk index for each consistent prediction area is obtained by dividing the sum of the differences in the target deformation sequences of each consistent prediction area by the average value of the surface remote sensing subsidence of the target deformation sequence.

9. The dynamic prediction system for underground subsidence risk based on time-series remote sensing according to claim 7, characterized in that, The risk prediction module is also used for: Multiply the average stability ratio by the second preset threshold to obtain the correction threshold; If the risk index of the kth consistent prediction area exceeds the correction threshold, a risk prediction result of abnormal settlement is generated for the kth consistent prediction area; otherwise, a risk prediction result of normal settlement is generated for the kth consistent prediction area.

10. The dynamic prediction system for underground subsidence risk based on time-series remote sensing according to claim 7, characterized in that, The first data processing module is also used for: The absolute value of the difference in remotely sensed subsidence between adjacent surfaces in the initial deformation sequence of each sub-region is calculated, and the absolute difference is obtained. Obtain the percentage of absolute differences exceeding the difference threshold in the initial deformation sequence of each sub-region, and use this percentage as the stable proportion of each sub-region.

Images