Geological disaster intelligent monitoring method and system

Abstract

The present application relates to the technical field of data processing, and more particularly to a geological disaster intelligent monitoring method and system. The method comprises: acquiring subsidence data of a geological disaster monitoring area, and pore water pressure data and high-frequency vibration data synchronized with the subsidence data; for each subsidence data point in the subsidence data, determining a water-soil coupling factor of each subsidence data point; determining the data reliability of each subsidence data point; acquiring multiple groups of optimized coarse-grained sequences; based on the optimized coarse-grained sequences, using a multi-scale entropy algorithm to realize precursor identification of geological disasters. The present application establishes a reliability mechanism by evaluating the water-soil coupling factor and combining the high-frequency vibration data, and performs weighted coarse-graining on the subsidence data, thereby suppressing noise interference and improving the accuracy and reliability of geological disaster precursor identification.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to an intelligent monitoring method and system for geological disasters. Background Technology

[0002] With the acceleration of urbanization and the large-scale development of underground space, ground subsidence has become a frequent and serious geological hazard. Ground subsidence poses a serious threat to infrastructure and may lead to significant casualties and economic losses. Therefore, high-precision and timely intelligent monitoring and early warning of potential subsidence areas has become an urgent need in the field of urban public safety.

[0003] Currently, with the geological disaster professional monitoring and early warning cloud platform as the core, combined with an integrated monitoring network of intelligent sensors such as GNSS receivers, pore water pressure gauges, and vibration sensors, it is possible to collect multi-source heterogeneous data of geological bodies in real time. Although ground collapse is sudden, it is accompanied by weak precursor signals before the disaster occurs. The multiscale entropy (MSE) algorithm can evaluate the system complexity by analyzing the entropy values ​​of time series at different scales, and is a powerful tool for capturing the precursors of critical points.

[0004] However, when using the multi-scale entropy algorithm to analyze settlement data, existing technologies often encounter problems. The core coarsening process of this algorithm typically involves simply averaging the original time series within a scale window. In the complex field environment of geological monitoring, the data not only contains weak precursor signals but also a large amount of transient, strong noise caused by external factors such as construction sites and heavy vehicles. During the standard multi-scale entropy algorithm's simple mean coarsening process, this strong noise significantly interferes with the mean calculation, leading to artifacts in the coarsened sequence and distorting the calculated multi-scale entropy curve. The system may misinterpret this as a drastic change in complexity, resulting in false alarms and ultimately affecting the accuracy of intelligent geological disaster monitoring. Summary of the Invention

[0005] To address the problem that standard multi-scale entropy algorithms are susceptible to noise interference during the coarsening process, leading to a decrease in the accuracy of identifying precursors to ground subsidence and thus increasing the likelihood of false alarms, thereby affecting the accuracy of intelligent geological disaster monitoring, this invention provides an intelligent geological disaster monitoring method and system.

[0006] In a first aspect, the present invention provides an intelligent monitoring method for geological disasters, employing the following technical solution:

[0007] A method for intelligent monitoring of geological hazards includes: acquiring settlement data of a geological hazard monitoring area, as well as pore water pressure data and high-frequency vibration data synchronized with the settlement data; determining the soil-water coupling factor for each settlement data point based on the settlement data value and corresponding pore water pressure data value of each settlement data point within its window, and the mean settlement data value and corresponding pore water pressure data value within the window to which each settlement data point belongs; determining the data reliability of each settlement data point based on the soil-water coupling factor, the high-frequency vibration data value corresponding to each settlement data point, and the maximum value among the high-frequency vibration data values ​​corresponding to each settlement data point within its window; weighting and coarsening the settlement data within the data segment to which each settlement data point belongs at different scales based on the data reliability to obtain multiple sets of optimized coarse-grained sequences; using a multi-scale entropy algorithm to obtain the sample entropy of each set of optimized coarse-grained sequences; generating a multi-scale entropy curve based on the sample entropy; and realizing the identification of precursors of geological hazards based on the multi-scale entropy curve.

[0008] The beneficial effects are as follows: By analyzing the coupling relationship between settlement data and pore water pressure data, an evaluation model for the soil-water coupling factor was constructed, which can effectively identify the true deformation characteristics of geological bodies; by combining high-frequency vibration data, a comprehensive evaluation mechanism for data credibility was constructed, realizing intelligent identification and weight adjustment of noise data; by weighting settlement data with data credibility, the interference of strong noise on the coarsening process was effectively suppressed, avoiding the artifact problem caused by traditional mean coarsening, ensuring that high-quality data points receive higher weights in the coarsening process, and improving the accuracy of the coarsening sequence; by analyzing the optimized coarsening sequence through the multi-scale entropy algorithm, a high-precision multi-scale entropy curve was generated, which can accurately capture the subtle changes in the complexity of geological bodies, effectively solving the problem of false alarms easily generated by the standard MSE algorithm in complex noise environments, improving the accuracy and reliability of geological disaster precursor signals, and providing more reliable intelligent monitoring technology support for urban public safety.

[0009] Furthermore, the acquisition of settlement data in the geological disaster monitoring area, as well as pore water pressure data and high-frequency vibration data synchronized with the settlement data, includes: using a GNSS receiver to collect three-dimensional coordinate data on the surface of the geological disaster monitoring area and extracting the vertical displacement component as settlement data; using a pore water pressure gauge to collect pore water pressure data at the stratigraphic monitoring points in the geological disaster monitoring area; and using a micro-core vibration sensor to collect high-frequency vibration data at the vibration monitoring points in the geological disaster monitoring area.

[0010] Furthermore, it also includes: preprocessing the settlement data, pore water pressure data, and high-frequency vibration data. The preprocessing includes: digitizing the settlement data, pore water pressure data, and high-frequency vibration data; aligning the digitized settlement data, pore water pressure data, and high-frequency vibration data with timestamps; and normalizing the timestamp-aligned settlement data, pore water pressure data, and high-frequency vibration data.

[0011] Furthermore, the water-soil coupling factor satisfies: In the formula, For the first Water-soil coupling factor for each settlement data point For the first The length of the window to which each settlement data point belongs. and The first Within the window of the settlement data point, the first... The settlement data values ​​of each settlement data point and the corresponding pore water pressure data values. and The first The average settlement data and the average pore water pressure data within the window to which each settlement data point belongs. For hyperparameters, It is the minimum-maximum normalization function. It is the absolute value symbol.

[0012] The beneficial effects are as follows: By calculating the covariance between settlement data and pore water pressure data, an evaluation model for the soil-water coupling factor was constructed, which can effectively reflect the correlation between geological settlement and pore water pressure changes. The larger the covariance value, the stronger the soil-water coupling relationship and the more obvious the geological significance of the data. The absolute value operation ensures a positive measurement of the coupling strength, the introduction of hyperparameters avoids calculation anomalies when the value is zero, and the normalization process eliminates the influence of differences in the data magnitude of different monitoring points, making the soil-water coupling factor have good comparability. Through the calculation of the soil-water coupling factor, a geological and physical basis is provided for subsequent data reliability assessment, and the ability to distinguish between real precursor signals and noise data is improved.

[0013] Furthermore, the data credibility satisfies: In the formula, For the first The reliability of data from individual settlement data points For the first Water-soil coupling factor for each settlement data point For the first The vibration data point value corresponding to each settlement data point For the first The maximum value among the high-frequency vibration data values ​​corresponding to each settlement data point within the window to which the settlement data point belongs. It is the minimum-maximum normalization function. It is the absolute value symbol.

[0014] The beneficial effects are as follows: By normalizing the product of the soil-water coupling factor and the relative difference of vibration data, a comprehensive evaluation model for data credibility is constructed, realizing the combination of geological and physical characteristics and environmental noise characteristics; the relative difference of vibration data reflects the degree of difference between the current data point and the maximum vibration value within the window. When the difference is small, it indicates that it may be affected by strong vibration noise, while when the difference is large, it indicates that the data is relatively accurate; the product of the soil-water coupling factor and the vibration difference ensures that high credibility is obtained only when the data simultaneously has strong soil-water coupling characteristics and relatively small vibration noise, effectively suppressing the influence of strong noise data and improving the accuracy of credibility assessment.

[0015] Furthermore, the coarse-grained sequence is obtained as follows:

[0016] Settlement data at scale The following weighted coarsening is performed to obtain the optimized value of each coarse-grained data point in the coarse-grained sequence, wherein the optimized value satisfies: In the formula, In order to scale Next Optimization values ​​for coarse-grained data points For coarse-grained data point indexing, For the first The reliability of data from individual settlement data points For the first Settlement data values ​​for each settlement data point.

[0017] The beneficial effects are as follows: By using a weighted average based on data credibility, an improvement over traditional mean coarsening is achieved, ensuring that high-credibility data points receive higher weights during the coarsening process and effectively suppressing the interference of low-quality data on the coarsening results; when the data credibility of settlement data points is high, their settlement value contributes more to the coarsening results, and when the credibility is low, their contribution decreases accordingly, achieving intelligent suppression of noise data and protection of true signals; the optimized coarsening mechanism improves the accuracy of multi-scale entropy analysis, effectively solving the problem of false alarms easily generated by traditional methods in complex noise environments, and providing a reliable data processing foundation for the accurate identification of geological disaster precursor signals.

[0018] Furthermore, the number of scales is 10.

[0019] Furthermore, the method for identifying precursors of geological disasters includes: recording the average entropy of all samples on the multi-scale entropy curve as the average entropy value; in response to the average entropy value obtained at the current moment being lower than the average of the average entropy values ​​obtained at each moment in the past set time period by a set proportion, identifying the current geological disaster monitoring area as having precursors of collapse, issuing an early warning, and completing intelligent monitoring of geological disasters.

[0020] The beneficial effects are as follows: By calculating the mean entropy of all samples on the multi-scale entropy curve as the average entropy value, a comprehensive assessment of the complexity of geological bodies is achieved, effectively reflecting the overall complexity variation characteristics of the geological system; by comparing the current average entropy value with a set proportion of the historical average entropy value, an intelligent judgment mechanism for precursor identification is constructed. When the average entropy value decreases, it indicates an abnormal change in the complexity of the geological body, which may be a signal of a collapse precursor; the calculation of the historical data mean over a set period ensures the stability of the judgment benchmark and avoids the impact of short-term fluctuations; the introduction of the set proportion provides flexible adjustment of the early warning threshold, which can be optimized and adjusted according to the characteristics of different monitoring areas; the precursor identification mechanism fully utilizes the advantages of multi-scale entropy analysis, combined with dynamic threshold judgment, effectively improving the accuracy and timeliness of geological disaster early warning, and providing support for the prevention and emergency response of geological disasters through intelligent judgment and early warning prompts.

[0021] Furthermore, the previously set duration was 24 hours, and the set ratio was 50%.

[0022] Secondly, the present invention provides an intelligent monitoring system for geological disasters, which adopts the following technical solution:

[0023] A geological disaster intelligent monitoring system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the aforementioned geological disaster intelligent monitoring method is implemented.

[0024] By adopting the above technical solution, a computer program for the above-mentioned intelligent monitoring method of geological disasters is generated and stored in a memory so that it can be loaded and executed by a processor. A terminal device can then be made based on the memory and processor for convenient use.

[0025] The present invention has the following technical effects:

[0026] (1) Breaking through the limitations of traditional multi-scale entropy simple mean coarse-graining which is easily affected by strong noise interference, the data credibility achieves accurate identification and weight suppression of noise data. The data credibility integrates the water-soil coupling factor reflecting the physical relationship between settlement and pore water pressure, and high-frequency vibration data reflecting the intensity of external interference. When weighting coarse-graining, low weights are given to noise data with low credibility, reducing its interference on the coarse-grained sequence, avoiding the artifact sequence and complexity misjudgment caused by noise in traditional methods, and reducing false alarms that misidentify noise as a precursor to disaster.

[0027] (2) Abandoning the limitations of traditional methods that rely solely on settlement data, settlement data is integrated with pore water pressure data reflecting the internal hydrological state of the geology and high-frequency vibration data reflecting external disturbances. Among them, the soil-water coupling factor assesses the geological rationality of the data through the correlation between settlement and pore water pressure, while the high-frequency vibration data assesses the data's resistance to interference. The combination of the two makes the data credibility no longer dependent on a single dimension, providing a more comprehensive basis for weighted coarse-graining and improving the quality of the coarse-grained sequence.

[0028] (3) The sample entropy calculated based on the optimized coarse-grained sequence can truly reflect the complexity changes of the settlement data. Traditional methods cause the entropy curve to be distorted due to noise interference, which may misjudge noise artifacts as sudden changes in complexity and generate false alarms. This invention filters noise and retains precursors through weighted coarse-grained filtering, and the generated multi-scale entropy curve can accurately reflect the true complexity of the geological state, thereby providing a more accurate basis for the identification of geological disaster precursors. Attached Figure Description

[0029] Figure 1 This is a flowchart of a method for intelligent monitoring of geological disasters according to an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram comparing the multi-scale entropy curves of standard MSE and weighted MSE in an intelligent geological disaster monitoring method according to an embodiment of the present invention.

[0031] Figure 3 This is a schematic diagram comparing the average entropy values ​​of the standard MSE and the weighted MSE with the early warning threshold in an intelligent geological disaster monitoring method according to an embodiment of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] This invention discloses an intelligent monitoring method for geological disasters, referring to... Figure 1 This includes steps S001-S005:

[0034] S001: Acquire settlement data of the geological disaster monitoring area, as well as pore water pressure data and high-frequency vibration data synchronized with the settlement data.

[0035] This embodiment aggregates data from smart sensors deployed in geological disaster monitoring areas, i.e., areas prone to ground subsidence, such as around urban foundation pits, within a cloud platform. This includes: collecting three-dimensional surface coordinate data using a high-precision GNSS receiver and extracting the vertical displacement component as settlement data. For example, the acquisition frequency is... Auxiliary data 1: Pore water pressure data of key strata are collected synchronously using pore water pressure gauges at stratum monitoring points; Auxiliary data 2: Vibration monitoring data of ground-based or shallow-surface fixed micro-core vibration sensors are collected synchronously at vibration monitoring points to sense high-frequency vibration data of the geological body at the monitoring point in real time; The above data are uploaded to the cloud platform in real time, and then the data are digitized using analog-to-digital converters to obtain a digital representation, and immediately timestamped and normalized to facilitate subsequent analysis.

[0036] S002: For each settlement data point in the settlement data, determine the water-soil coupling factor for each settlement data point.

[0037] It should be noted that the core of this step of the analysis lies in identifying the real settlement driven by internal geological factors, such as changes in groundwater. The precursor to ground collapse is often the disruption of the geological body's equilibrium. At this time, the settlement deformation will show a high sensitivity to changes in pore water pressure, that is, the water-soil coupling is enhanced. Therefore, if a settlement data point and its corresponding pore water pressure data at the same time show a strong correlation within a short time window, such as rising water pressure and accelerated settlement, then the settlement data point is more likely to reflect the real geological response, and it should be assigned a larger water-soil coupling factor.

[0038] Based on the settlement data value and corresponding pore water pressure data value of each settlement data point within the window to which each settlement data point belongs, as well as the average settlement data and corresponding pore water pressure data within the window to which each settlement data point belongs, the water-soil coupling factor of each settlement data point is determined.

[0039] The implementers can set the length of the window to which each settlement data point belongs according to the specific implementation situation. For example, a time window containing 60 data points with the settlement data point as the final data point.

[0040] Specifically, the water-soil coupling factor satisfies:

[0041] ;

[0042] In the formula, For the first Water-soil coupling factor for each settlement data point For the first The length of the window to which each settlement data point belongs. and The first Within the window of the settlement data point, the first... The settlement data values ​​of each settlement data point and the corresponding pore water pressure data values. and The first The average settlement data and the average pore water pressure data within the window to which each settlement data point belongs. For hyperparameters, It is the minimum-maximum normalization function. It is the absolute value symbol.

[0043] Implementers can set hyperparameters according to the specific implementation situation, for example, The hyperparameter exists to prevent the calculated value from being 0.

[0044] in, Indicates the first The absolute value of the covariance between the settlement data and the pore water pressure data within the window for each settlement data point. The larger this value, the stronger the correlation between settlement changes and water pressure changes during the same period. The higher the reliability of a settlement data point as a true geological response rather than external noise, the better. The larger it is, the greater it becomes; conversely, the smaller it is, the greater it becomes.

[0045] S003: Determine the reliability of the data at each settlement data point.

[0046] It should be noted that the soil-water coupling factor for each settlement data point is used to determine the authenticity of the response based on the correlation between settlement data and pore water pressure data. However, the soil-water coupling factor cannot distinguish between a true geological response and external noise. This is because high-intensity external disturbances can cause noise in the settlement data, which may also be transmitted through the saturated soil medium, resulting in a sudden pressure change in the pore water pressure data at the same time. This leads to the calculation of an inflated soil-water coupling factor, and the incorrect labeling of this common vibration event caused by external noise as a true geological response. In the scenario corresponding to this scheme, the generation of noise data is often accompanied by significant vibrations. Therefore, high-frequency vibration data simultaneously recorded at each settlement data point can be monitored to further filter out noise data. Thus, this step analyzes the energy characteristics of the high-frequency vibration data and, combined with the soil-water coupling factor for each settlement data point, calculates the data reliability of each settlement data point. The larger the high-frequency vibration data value simultaneously recorded at a settlement data point, the greater the likelihood that the settlement data point is caused by noise pollution, and the lower its data reliability.

[0047] The reliability of each settlement data point is determined based on the water-soil coupling factor, the high-frequency vibration data value corresponding to each settlement data point, and the maximum value among the high-frequency vibration data values ​​corresponding to each settlement data point within the window to which each settlement data point belongs.

[0048] Specifically, the data credibility satisfies the following:

[0049] ;

[0050] In the formula, For the first The reliability of data from individual settlement data points For the first Water-soil coupling factor for each settlement data point For the first The vibration data point value corresponding to each settlement data point For the first The maximum value among the high-frequency vibration data values ​​corresponding to each settlement data point within the window to which the settlement data point belongs. It is the minimum-maximum normalization function. It is the absolute value symbol.

[0051] in, The larger the value, the higher the correlation between the settlement data and the pore water pressure data. The greater the likelihood that a settlement data point represents a genuine geological response rather than external noise, the better. The larger it is, the greater it will be; conversely, the smaller it is, the greater it will be. The smaller the value, the better. The larger the vibration data value at each settlement data point at the same time, the more it indicates that the... The greater the likelihood that a settlement data point is caused by noise pollution, the better. The smaller it will be; and vice versa.

[0052] S004: Obtain multiple sets of optimized coarse-grained sequences.

[0053] It should be noted that in this step, the coarsening process of the settlement data at multiple scales will be weighted and optimized based on the data reliability of each settlement data point.

[0054] Based on the data reliability, the settlement data within the data segment to which each settlement data point belongs is weighted and coarsened at different scales to obtain multiple sets of optimized coarsened sequences.

[0055] The implementers can set the length of the data segment to which each settlement data point belongs according to the specific implementation situation. For example, a data segment containing 1000 data points with the settlement data point as the final data point; the first 999 data points are only used as auxiliary reference data points and are not used for the analysis of this invention.

[0056] Specifically, the number of scales is 10, that is, scales It changes from 1 to 10.

[0057] Specifically, the coarse-grained sequence is obtained as follows:

[0058] Settlement data at scale The following weighted coarsening is performed to obtain the optimized value of each coarse-grained data point in the coarse-grained sequence, wherein the optimized value satisfies:

[0059] ;

[0060] In the formula, In order to scale Next Optimization values ​​for coarse-grained data points For coarse-grained data point indexing, For the first The reliability of data from individual settlement data points For the first Settlement data values ​​for each settlement data point.

[0061] The relationship between the optimized numerical values ​​is a weighted average calculation formula, which is based on the scale. The higher the confidence level of the settlement data points, that is, the settlement data that is related to water pressure and not to vibration, the greater the contribution weight to the coarse-grained results; conversely, the lower the confidence level of the data points, the closer their weight is to 0, and they are automatically filtered out in the coarse-grained calculation. This ensures that the coarse-grained sequence can reflect the real geological settlement trend with high fidelity, and provides a reliable data basis for the subsequent calculation of sample entropy values.

[0062] S005: Based on the optimized coarse-grained sequence, the multi-scale entropy algorithm is used to identify precursors of geological disasters.

[0063] Using a multi-scale entropy algorithm, the sample entropy of each optimized coarse-grained sequence is obtained. Based on the sample entropy, a multi-scale entropy curve is generated. (See [link]). Figure 2 ,from Figure 2 As can be seen, the standard MSE has a lower sample entropy value than the weighted MSE at multiple scales, which can easily lead to false alarms due to the low value. Based on the multi-scale entropy curve, the precursors of geological disasters can be identified.

[0064] Specifically, the process of identifying precursors to geological disasters includes:

[0065] The mean value of the entropy of all samples on the multi-scale entropy curve is denoted as the average entropy value.

[0066] If the average entropy value obtained at the current moment is lower than the average of the average entropy values ​​obtained at each moment in the past set time period by a set proportion, it is determined that there are signs of collapse in the current geological disaster monitoring area, and an early warning is issued. The relevant responsible persons are automatically notified through the cloud platform, and it is recommended to immediately activate the emergency plan, such as sealing off the collapse hazard area, evacuating personnel, and stopping construction in the surrounding area, thus completing the intelligent monitoring of geological disasters.

[0067] Implementers can set the previously set duration and percentage according to the specific implementation situation. For example, the set duration is 24 hours and the set percentage is 50%.

[0068] like Figure 3 As shown, the average entropy value of the standard MSE is lower than the preset dynamic warning threshold, thus triggering a false alarm. The weighted MSE can avoid false alarms.

[0069] This invention also discloses an intelligent geological disaster monitoring system, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement an intelligent geological disaster monitoring method according to the present invention.

[0070] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0071] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A geological disaster intelligent monitoring method, characterized in that, The method comprises the following steps: obtaining subsidence data of a geological disaster monitoring area, and pore water pressure data and high-frequency vibration data synchronized with the subsidence data; determining a water-soil coupling factor of each subsidence data point according to the subsidence data value of each subsidence data point in the window to which the subsidence data point belongs and the corresponding pore water pressure data value, and the mean value of the subsidence data in the window to which the subsidence data point belongs and the corresponding pore water pressure data mean value; determining the data reliability of each subsidence data point according to the water-soil coupling factor, the corresponding high-frequency vibration data value of each subsidence data point, and the maximum value of the corresponding high-frequency vibration data value of each subsidence data point in the window to which the subsidence data point belongs; ； In the formula, is the data reliability of the first settlement data point, is the water-soil coupling factor of the first settlement data point, is the value of the vibration data point corresponding to the first settlement data point, is the maximum value in the high-frequency vibration data values corresponding to each settlement data point in the window to which the first settlement data point belongs, is the minimum-maximum normalization function, is the absolute value symbol; performing weighted coarse-graining on the subsidence data in different scales in the data segment to which each subsidence data point belongs according to the data reliability, to obtain a plurality of groups of optimized coarse-grained sequences; using a multi-scale entropy algorithm to obtain the sample entropy of each group of optimized coarse-grained sequences, generating a multi-scale entropy curve according to the sample entropy, and realizing the precursor identification of geological disasters based on the multi-scale entropy curve. 2.The geological disaster intelligent monitoring method according to claim 1, characterized in that, The method comprises the following steps: using a GNSS receiver to collect three-dimensional coordinate data on the ground surface of the geological disaster monitoring area, and extracting the vertical displacement component as subsidence data; using a pore water pressure gauge to collect pore water pressure data at the layer monitoring point of the geological disaster monitoring area; using a micro-chip vibration sensor to collect high-frequency vibration data at the vibration monitoring point of the geological disaster monitoring area. 3.The geological disaster intelligent monitoring method according to claim 1, characterized in that, The method further comprises the following steps: preprocessing the subsidence data, pore water pressure data and high-frequency vibration data, which comprises the following steps: digitizing the subsidence data, pore water pressure data and high-frequency vibration data; aligning the time stamps of the digitized subsidence data, pore water pressure data and high-frequency vibration data; normalizing the subsidence data, pore water pressure data and high-frequency vibration data after time stamp alignment. 4.The geological disaster intelligent monitoring method according to claim 1, characterized in that, The water-soil coupling factor satisfies the following condition: ； In the formula, For the first Water-soil coupling factor for each settlement data point For the first The length of the window to which each settlement data point belongs. and The first Within the window of the settlement data point, the first... The settlement data values ​​of each settlement data point and the corresponding pore water pressure data values. and The first The average settlement data and the average pore water pressure data within the window to which each settlement data point belongs. For hyperparameters, It is the minimum-maximum normalization function. It is the absolute value symbol. 5.The geological disaster intelligent monitoring method according to claim 1, characterized in that, The coarse-grained sequence is obtained in the following manner: The data points of the sequence are weighted and coarsened in scale down to obtain an optimized numerical value for each coarsened data point in the coarsened sequence, the optimized numerical value satisfying: ； wherein is an optimized numerical value for the th coarse-grained data point, th coarse-grained data point, is an index of the coarse-grained data point, is a data confidence for the th settling data point, th settling data point, is a settling data value for the 6.The geological disaster intelligent monitoring method according to claim 1, characterized in that, The number of scales is 10. 7.The geological disaster intelligent monitoring method according to claim 1, characterized in that, The method for realizing the precursor identification of geological disasters comprises the following steps: taking the mean value of all sample entropies on the multi-scale entropy curve as the average entropy value; in response to the average entropy value obtained at the current time being lower than the set proportion of the mean value of the average entropy values obtained at each time within a set time period in the past, determining that a collapse precursor has occurred in the current geological disaster monitoring area, and issuing a warning prompt to complete intelligent monitoring of geological disasters. 8.The geological disaster intelligent monitoring method of claim 7, wherein, The set time period is 24 hours, and the set proportion is 50%.

9. A geological disaster intelligent monitoring system, characterized in that, The method comprises the following steps: a processor and a memory, the memory storing computer program instructions which, when executed by the processor, realize the method for intelligent monitoring of geological disasters according to any one of claims 1-8.

Images