A method and system for debris flow geological disaster monitoring and emergency response

Abstract

Embodiments of the present invention relate to the technical field of debris flow monitoring, and specifically disclose a method and system for debris flow geological disaster monitoring and emergency response. Embodiments of the present invention determine a debris flow monitoring area, divide it into multiple monitoring sub-areas; conduct multi-source sensing monitoring; perform data preprocessing on multiple sensing monitoring data; perform data synchronization and fusion on multiple standard monitoring data; when there is a debris flow hazard, perform hazard grading, generate area grading information, and starting from the key risk sub-areas, conduct UAV emergency warning control. It can divide the debris flow monitoring area into multiple monitoring sub-areas, conduct multi-source sensing monitoring of different sub-areas, and perform data preprocessing, synchronization and fusion, and hazard grading, generate area grading information, and then conduct corresponding UAV emergency warning control to achieve different hazard gradings and emergency responses, thereby solving the defects of insufficient data collection, low warning accuracy, and poor warning timeliness.

Description

Technical Field

[0001] The present invention belongs to the technical field of debris flow monitoring, and particularly relates to a method and system for monitoring and emergency response to debris flow geological disasters. Background Art

[0002] Debris flow is a natural disaster phenomenon in which substances such as water, sediment, and stones are mixed together and rush downhill at a high speed along terrains such as mountainsides and river valleys.

[0003] Debris flows usually occur under triggering conditions such as heavy rainfall, snowmelt, or earthquakes, and can quickly carry a large amount of sediment and gravel, posing a serious threat to human life and property.

[0004] In the prior art, since debris flows usually occur in areas with complex terrains, for the monitoring and early warning of debris flow geological disasters, there are often defects such as insufficient data collection, low early warning accuracy, and poor early warning timeliness, and different risk levels and emergency responses cannot be carried out. Summary of the Invention

[0005] The purpose of the embodiments of the present invention is to provide a method and system for monitoring and emergency response to debris flow geological disasters, aiming to solve the problems raised in the background art.

[0006] To achieve the above purpose, the embodiments of the present invention provide the following technical solutions:

[0007] A method for monitoring and emergency response to debris flow geological disasters, the method specifically includes the following steps:

[0008] Determine the debris flow monitoring area, obtain the geological and topographical data of the debris flow monitoring area, divide the debris flow monitoring area into multiple monitoring sub-areas, and select key risk sub-areas;

[0009] Perform multi-source sensing monitoring on multiple monitoring sub-areas to obtain multiple sensing monitoring data;

[0010] Perform data preprocessing on multiple sensing monitoring data to obtain multiple standard monitoring data;

[0011] Perform data synchronization and fusion on multiple standard monitoring data to generate fusion monitoring data;

[0012] According to the fusion monitoring data, when there is a debris flow hazard, perform risk grading on multiple monitoring sub-areas to generate area grading information, and starting from the key risk sub-areas, perform UAV emergency warning control.

[0013] A debris flow geological disaster monitoring and emergency response system, the system includes a monitoring area division unit, a multi-source sensing monitoring unit, a monitoring data preprocessing unit, a data synchronization and fusion unit, and an emergency warning control unit, where:

[0014] The monitoring area division unit is used to determine the debris flow monitoring area, obtain the geological and topographical data of the debris flow monitoring area, divide the debris flow monitoring area into multiple monitoring sub-areas, and select key risk sub-areas;

[0015] The multi-source sensing monitoring unit is used to perform multi-source sensing monitoring on multiple said monitoring sub-areas and obtain multiple sensing monitoring data;

[0016] The monitoring data preprocessing unit is used to preprocess the multiple sensing monitoring data to obtain multiple standard monitoring data;

[0017] The data synchronization and fusion unit is used to synchronize and fuse the multiple standard monitoring data to generate fused monitoring data;

[0018] The emergency warning control unit is used to, according to the fused monitoring data, when there is a debris flow danger, perform danger grading on multiple monitoring sub-areas, generate area grading information, and starting from the key risk sub-areas, perform unmanned aerial vehicle (UAV) emergency warning control.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] In the embodiment of the present invention, by determining the debris flow monitoring area, dividing multiple monitoring sub-areas; performing multi-source sensing monitoring; preprocessing the multiple sensing monitoring data; synchronizing and fusing the multiple standard monitoring data; when there is a debris flow danger, performing danger grading, generating area grading information, and starting from the key risk sub-areas, performing UAV emergency warning control. It can divide the debris flow monitoring area into multiple monitoring sub-areas, perform multi-source sensing monitoring on different sub-areas, and perform data preprocessing, synchronization fusion, and danger grading, generate area grading information, and then perform corresponding UAV emergency warning control to achieve different danger gradings and emergency responses, thereby solving the defects of insufficient data collection, low warning accuracy, and poor warning timeliness. Description of the Drawings

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0022] Figure 1 Shows the flowchart of the method provided by the embodiment of the present invention.

[0023] Figure 2 The flowchart of dividing multiple monitoring sub - regions in the method provided by the embodiment of the present invention is shown.

[0024] Figure 3 The flowchart of performing multi - source sensing monitoring in the method provided by the embodiment of the present invention is shown.

[0025] Figure 4 The flowchart of obtaining multiple standard monitoring data in the method provided by the embodiment of the present invention is shown.

[0026] Figure 5 The flowchart of performing data synchronization and fusion in the method provided by the embodiment of the present invention is shown.

[0027] Figure 6 The flowchart of UAV emergency early - warning control in the method provided by the embodiment of the present invention is shown.

[0028] Figure 7 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0029] Figure 8 The structural block diagram of the monitoring area division unit in the system provided by the embodiment of the present invention is shown.

[0030] Figure 9 The structural block diagram of the multi - source sensing monitoring unit in the system provided by the embodiment of the present invention is shown.

[0031] Figure 10 The structural block diagram of the emergency early - warning control unit in the system provided by the embodiment of the present invention is shown. Detailed implementation manners

[0032] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] It can be understood that in the prior art, since debris flows usually occur in areas with complex terrains, for the monitoring and early - warning of debris - flow geological disasters, there are often defects such as insufficient data collection, low early - warning accuracy and poor early - warning timeliness, and different risk levels and emergency responses cannot be carried out.

[0034] To solve the above problems, in the embodiments of the present invention, a debris flow monitoring area is determined, geological and topographical data of the debris flow monitoring area is obtained, the debris flow monitoring area is divided into multiple monitoring sub-areas, and key risk sub-areas are selected; multi-source sensing monitoring is performed on the multiple monitoring sub-areas to obtain multiple sensing monitoring data; data preprocessing is performed on the multiple sensing monitoring data to obtain multiple standard monitoring data; data synchronization and fusion are performed on the multiple standard monitoring data to generate fused monitoring data; according to the fused monitoring data, when there is a debris flow hazard, risk grading is performed on the multiple monitoring sub-areas to generate area grading information, and starting from the key risk sub-areas, UAV emergency warning control is carried out. It is possible to divide the debris flow monitoring area into multiple monitoring sub-areas, perform multi-source sensing monitoring of different sub-areas, and perform data preprocessing, synchronization fusion and risk grading to generate area grading information, and then perform corresponding UAV emergency warning control to achieve different risk gradings and emergency responses, thereby solving the defects of insufficient data collection, low warning accuracy and poor warning timeliness.

[0035] Figure 1 The flowchart of the method provided by the embodiments of the present invention is shown.

[0036] Specifically, a method for debris flow geological disaster monitoring and emergency response, the method specifically includes the following steps:

[0037] Step S101, determine a debris flow monitoring area, obtain geological and topographical data of the debris flow monitoring area, divide the debris flow monitoring area into multiple monitoring sub-areas, and select key risk sub-areas.

[0038] In the embodiments of the present invention, by determining a debris flow monitoring area with the need for debris flow geological disaster monitoring and emergency response, obtaining the geological and topographical data of the debris flow monitoring area, through terrain analysis of the geological and topographical data, the debris flow monitoring area is divided into multiple monitoring sub-areas, and then from the geological and topographical data, multiple geological and topographical sub-area data corresponding to the multiple monitoring sub-areas are matched, risk initialization analysis is performed on the multiple geological and topographical sub-area data, and one key risk sub-area is selected from the multiple monitoring sub-areas.

[0039] It can be understood that the key risk sub-area can be a monitoring sub-area determined by analyzing the historical records corresponding to the multiple monitoring sub-areas and having the most frequent occurrence of debris flow geological disasters.

[0040] Specifically, Figure 2 The flowchart of dividing multiple monitoring sub-areas in the method provided by the embodiments of the present invention is shown.

[0041] Among them, in the preferred embodiment provided by the present invention, the steps of determining the debris flow monitoring area, obtaining the geological and topographical data of the debris flow monitoring area, dividing the debris flow monitoring area into multiple monitoring sub-areas, and selecting key risk sub-areas specifically include the following steps:

[0042] Step S1011, determine the debris flow monitoring area;

[0043] Step S1012, obtain the geological and topographical data of the debris flow monitoring area;

[0044] Step S1013, divide the debris flow monitoring area into multiple monitoring sub-areas according to the geological and topographical data;

[0045] Step S1014, based on the geological and topographical data, conduct an initial risk analysis on multiple monitoring sub-areas and select key risk sub-areas.

[0046] Among them, in the preferred embodiment provided by the present invention, the steps of conducting an initial risk analysis on multiple monitoring sub-areas based on the geological and topographical data and selecting key risk sub-areas specifically include the following steps:

[0047] According to the geological and topographical data, extract the terrain feature triple data of each monitoring sub-area. The terrain feature triple data includes the slope angle, rock and soil density, and coverage index of the center point;

[0048] Input the terrain feature triple data of each monitoring sub-area into the principal component analysis model, calculate the contribution rate of the first principal component variance, and use the contribution rate of the first principal component variance as the comprehensive terrain index to obtain the principal component scores of each monitoring sub-area;

[0049] Set the membership function threshold according to the regional geological mechanics parameters and obtain the historical disaster critical value. When the principal component scores of each monitoring sub-area exceed the historical disaster critical value, use the S-type function to calculate the membership degree of the principal component scores of each monitoring sub-area to obtain the geological risk membership degree of each monitoring sub-area;

[0050] Conduct an analysis of the terrain surface curvature gradient for each monitoring sub-area to obtain the terrain mutation feature index;

[0051] Obtain the historical risk time record and the corresponding terrain data of each monitoring sub-area;

[0052] Train the neural network model with the historical risk time record and the corresponding terrain data of each monitoring sub-area. Through the correlation analysis between the disaster occurrence frequency and the terrain parameters, iteratively optimize the weight coefficients of each parameter to obtain the dynamic weight of the terrain features;

[0053] The geological risk membership degree is weighted with the dynamic weights of the corresponding terrain features to obtain the weighted sum of the geological risk membership degrees of each monitoring sub-region;

[0054] The terrain mutation feature index is amplified and fused with the weighted sum of the geological risk membership degrees of each monitoring sub-region to obtain the comprehensive risk value of each monitoring sub-region;

[0055] The comprehensive risk values of each monitoring sub-region are sorted in descending order. The top 20% are classified as the first-level risk area, the middle 30% as the second-level risk area, and the rest as the third-level risk area, so as to obtain the regional risk distribution map with classification labels.

[0056] In the above solution, the present invention captures both macroscopic terrain features and microscopic geomorphic mutation information through the collaboration of geological parameter principal component analysis and terrain surface curvature gradient analysis. Compared with traditional single slope analysis, potential slip surfaces can be identified. The recognition accuracy of the gully area by the dual non-linear processing of fuzzy membership degree and gradient index amplification is higher than that of the linear model.

[0057] Furthermore, the debris flow geological disaster monitoring and emergency response method further includes the following steps:

[0058] Step S102, perform multi-source sensing monitoring on multiple said monitoring sub-regions to obtain multiple sensing monitoring data.

[0059] In the embodiment of the present invention, soil moisture sensing monitoring is performed on multiple monitoring sub-regions to obtain soil monitoring data, rainfall sensing monitoring is performed on multiple monitoring sub-regions to obtain rainfall monitoring data, displacement sensing monitoring of the surface and / or rock strata is performed on multiple monitoring sub-regions to obtain displacement monitoring data. At the same time, vibration sensing monitoring is performed on multiple monitoring sub-regions to obtain vibration monitoring data. The soil monitoring data, rainfall monitoring data, displacement monitoring data, and vibration monitoring data together constitute multiple sensing monitoring data.

[0060] Specifically, Figure 3 shows the flowchart of multi-source sensing monitoring in the method provided by the embodiment of the present invention.

[0061] Among them, in the preferred embodiment provided by the present invention, the performing multi-source sensing monitoring on multiple said monitoring sub-regions to obtain multiple sensing monitoring data specifically includes the following steps:

[0062] Step S1021, perform soil moisture sensing monitoring on multiple said monitoring sub-regions to obtain soil monitoring data;

[0063] Step S1022, perform rainfall sensing monitoring on multiple said monitoring sub-regions to obtain rainfall monitoring data;

[0064] Step S1023, perform displacement sensing monitoring on multiple said monitoring sub-regions to obtain displacement monitoring data;

[0065] Step S1024, perform vibration sensing monitoring on multiple said monitoring sub-regions to obtain vibration monitoring data.

[0066] Furthermore, the debris flow geological disaster monitoring and emergency response method further includes the following steps:

[0067] Step S103, perform data preprocessing on multiple said sensing monitoring data to obtain multiple standard monitoring data.

[0068] In the embodiment of the present invention, by identifying and removing noise, redundancy, and anomalies in multiple sensing monitoring data, multiple effective monitoring data are obtained, and then according to a preset data standard, multiple effective monitoring data are standardized to obtain multiple standard monitoring data.

[0069] It can be understood that abnormal data can be data with sensor failures or excessive errors, and abnormal data is identified through machine learning.

[0070] Specifically, Figure 4 shows a flowchart of obtaining multiple standard monitoring data in the method provided by the embodiment of the present invention.

[0071] Among them, in the preferred embodiment provided by the present invention, the performing data preprocessing on multiple said sensing monitoring data to obtain multiple standard monitoring data specifically includes the following steps:

[0072] Step S1031, identify and remove noise, redundancy, and anomalies in multiple said sensing monitoring data to obtain multiple effective monitoring data;

[0073] Step S1032, standardize multiple said effective monitoring data to obtain multiple standard monitoring data.

[0074] Among them, in the preferred embodiment provided by the present invention, the standardizing multiple said effective monitoring data to obtain multiple standard monitoring data specifically includes the following steps:

[0075] Obtain the sensor numbers corresponding to the effective monitoring data, and according to the sensor numbers, obtain the corresponding sensor factory accuracy identification codes;

[0076] Analyze the sensor factory accuracy identification codes to obtain the reference mean and standard deviation of each type of sensor;

[0077] Obtain the sensing monitoring data sequence in the effective monitoring data within the previous T minutes before the current moment;

[0078] Sort the same type of data in the sensing monitoring data sequence by time to obtain a sliding window data sequence, and take the value at the middle position in the sorting to obtain the median value of the sliding window for each type of sensor.

[0079] Calculate the interquartile range based on the values at the 25% and 75% positions in the sliding window data sequence.

[0080] Generate an adaptive adjustment factor based on the number of data in the sliding window data sequence and the interquartile range.

[0081] Calculate the difference between the current data timestamp and the data reception time in the sliding window data sequence, and substitute the difference into the S-shaped function for non-linear conversion to obtain the aging strengthening coefficient.

[0082] Standardize each type of current sensing monitoring data in the effective monitoring data using the reference mean and standard deviation of the sensor to generate several initial standardized results.

[0083] Calculate the absolute deviation between the current sensing monitoring data in the effective monitoring data and the median value of the sliding window, and non-linearly map the absolute deviation using the hyperbolic tangent function according to the adaptive adjustment factor to obtain the robust adjustment amount.

[0084] Fuse several initial standardized results with the generated robust adjustment amount to obtain a fusion result, and then strengthen the aging of the fusion result using the aging strengthening coefficient to obtain the final multiple standard monitoring data.

[0085] In the above solution, the present invention parses the factory encoding of the sensor, converts the hardware parameters into mathematical features using the binary bit parsing algorithm, and automatically matches the accuracy level and range parameters to achieve the self-adaptation of the sensor characteristics. To ensure the advantage of aging optimization, the aging coefficient adopts an S-shaped attenuation function, and the weight of the newly collected data is 3.2 times that of the data 10 minutes ago, thereby realizing the strengthening of data freshness.

[0086] Furthermore, the debris flow geological disaster monitoring and emergency response method further includes the following steps:

[0087] Step S104, perform data synchronization fusion on the multiple standard monitoring data to generate fusion monitoring data.

[0088] In the embodiment of the present invention, by obtaining the sampling period and / or timestamp of multiple standard monitoring data, and then using Kalman filtering to align the multiple sampling periods and / or multiple timestamps to generate multiple aligned time points to ensure the synchronization of data time, and then perform data fusion on the multiple standard monitoring data according to the multiple aligned time points to generate fusion monitoring data.

[0089] Specifically, Figure 5The flowchart of data synchronization and fusion in the method provided by the embodiment of the present invention is shown.

[0090] Among them, in the preferred embodiment provided by the present invention, the data synchronization and fusion of the multiple standard monitoring data to generate the fused monitoring data specifically includes the following steps:

[0091] Step S1041, obtain the sampling period and / or timestamp of the multiple standard monitoring data;

[0092] Step S1042, use Kalman filtering to align the multiple sampling periods and / or the multiple timestamps to generate multiple aligned time points;

[0093] Step S1043, perform data fusion on the multiple standard monitoring data according to the multiple aligned time points to generate the fused monitoring data.

[0094] Among them, in the preferred embodiment provided by the present invention, the performing data fusion on the multiple standard monitoring data according to the multiple aligned time points to generate the fused monitoring data specifically includes the following steps:

[0095] Obtain the occurrence frequency of each sensor data value in the standard monitoring data to obtain the data frequency distribution;

[0096] Calculate the first information entropy value according to the data frequency distribution and convert the first information entropy value into the second weight coefficient;

[0097] Normalize the second weight coefficient to obtain the dynamic weight coefficient of each sensor;

[0098] Obtain the timestamp of each sensor data value in the standard monitoring data and calculate the time difference between the current moment and the timestamp of the sensor data value;

[0099] Obtain the historical warning times, perform logarithmic smoothing processing on the time difference according to the historical warning times to obtain the smoothed time difference;

[0100] Compare each sensor data value in the standard monitoring data with the reference range of the corresponding sensor, and apply exponential decay suppression to the data deviating from the reference range according to the comparison result and the smoothed time difference to obtain the spatio-temporal double-domain adjusted data value of each sensor;

[0101] Obtain the surface curvature change rate of the monitored sub-region, calculate the ratio of the surface curvature of the current monitored sub-region to the maximum surface curvature value in all monitored sub-regions, and non-linearly amplify the ratio to generate the terrain feature amplification coefficient;

[0102] Obtain the comprehensive data distribution of all sensors in the monitored sub-region, and calculate the second information entropy value of the entire monitored sub-region according to the comprehensive data distribution of all sensors in the monitored sub-region;

[0103] Map the second information entropy value to a smooth interval of 0-1 using a hyperbolic function to generate a regional conflict adjustment coefficient;

[0104] Weight the spatio-temporal double-domain adjusted data values of each sensor using the dynamic weight coefficient of each sensor, and then amplify the weighted result using the terrain feature amplification coefficient to obtain the data processing result of each sensor. Apply the regional conflict adjustment coefficient to the data processing result of each sensor, and then perform cumulative synthesis to obtain the fused monitoring data.

[0105] In the above solution, the present invention calculates the information entropy value of sensor data in real time and dynamically adjusts the weights of each sensor. This mechanism can automatically identify faulty sensors. When the data of a certain sensor is abnormal, its weight automatically drops below the reference value. And by combining the time decay factor with the spatial reference comparison, the weight decay of historical stale data can suppress the number of errors. And introduce the terrain surface curvature gradient parameter to deeply fuse the terrain features, effectively improving the monitoring sensitivity in steep slope areas (≥25°).

[0106] Further, the debris flow geological disaster monitoring and emergency response method further includes the following steps:

[0107] Step S105, according to the fused monitoring data, when there is a debris flow danger, perform risk grading on multiple monitored sub-regions to generate regional grading information, and starting from the key risk sub-region, perform unmanned aerial vehicle (UAV) emergency warning control.

[0108] In the embodiment of the present invention, according to the preset danger threshold data, perform a comparative analysis of the danger of the fused monitoring data to determine whether there is a debris flow danger. In the case of determining that there is a debris flow danger, perform risk grading on multiple monitored sub-regions, record the regional grading information, and then starting from the key risk sub-region, plan a flight warning route according to the regional grading information, and then perform flight control for UAV emergency warning according to the flight warning route.

[0109] It can be understood that after the selection of the key risk sub-region is completed, it is necessary to control the UAV to fly to the key risk sub-region in advance for emergency warning standby.

[0110] Specifically, Figure 6 The flowchart of the UAV emergency warning control in the method provided by the embodiment of the present invention is shown.

[0111] Among them, in the preferred embodiment provided by the present invention, when there is a debris flow hazard according to the fusion monitoring data, the following steps are specifically included for hazard grading of multiple monitoring sub-regions, generating regional grading information, and starting from the key risk sub-regions for UAV emergency warning control:

[0112] Step S1051: Conduct hazard analysis on the fusion monitoring data to determine whether there is a debris flow hazard;

[0113] Step S1052: When there is a debris flow hazard, conduct hazard grading on multiple monitoring sub-regions to generate regional grading information;

[0114] Step S1053: Starting from the key risk sub-region, plan a flight warning route according to the regional grading information;

[0115] Step S1054: Conduct UAV emergency warning control according to the flight warning route.

[0116] Among them, in the preferred embodiment provided by the present invention, the following steps are specifically included for conducting hazard grading on multiple monitoring sub-regions to generate regional grading information when there is a debris flow hazard:

[0117] Extract the total number of warning events that occurred in the same monitoring sub-region within 3 months from the current time;

[0118] Obtain the second information entropy value and the historical maximum entropy value during the current detection period, and normalize the second information entropy value during the current detection period using the historical maximum entropy value to obtain a normalized entropy value;

[0119] Perform weighted superposition of the normalized entropy value and the total number of warning events to obtain a dynamic entropy weight coefficient;

[0120] Extract soil moisture time series data and rainfall monitoring data from the fusion monitoring data, and input the rainfall monitoring data into an S-type function converter to output a soil risk index of 0-1;

[0121] Calculate the percentage value of the slope of the current monitoring sub-region to the maximum slope among all monitoring sub-regions to obtain a slope value;

[0122] Set corresponding slope weights according to the soil moisture time series data, and perform weighting on the slope value and the slope weight to obtain an enhanced slope value;

[0123] Fuse the enhanced slope value and the soil risk index to obtain a terrain coupling weight;

[0124] Calculate the covariance relationship value of the data within the past hour from the time series data of soil moisture and the rainfall monitoring data, and standardize the variance relationship value to obtain the standardized covariance relationship value;

[0125] Introduce a time decay factor to decay the standardized covariance relationship value, and use the hyperbolic tangent function to constrain the decay result within the interval [-1, 1] to obtain the covariance weight coefficient;

[0126] Establish a mapping table of surface curvature - decay rate, and substitute the surface curvature change rate of the monitoring sub - region into the mapping table of surface curvature - decay rate to obtain the surface curvature adjustment coefficient;

[0127] Calculate the time difference between the current moment and the nearest warning, and use the surface curvature adjustment coefficient to adjust the time between the current moment and the nearest warning to obtain the terrain - corrected time difference;

[0128] Perform an asymmetric Gaussian decay calculation based on the terrain - corrected time difference to obtain the primary decay coefficient;

[0129] Apply a linear decay compensation to the flat - area part corresponding to the primary decay coefficient according to the surface curvature change rate of the monitoring sub - region to obtain the spatio - temporal decay coefficient;

[0130] Fuse the dynamic entropy weight coefficient and the terrain - coupling weight to construct a hierarchical decision sub - item;

[0131] Obtain the terrain surface curvature reference value, and fuse the ratio of the surface curvature change rate of the monitoring sub - region to the terrain surface curvature reference value with the covariance weight coefficient to construct a hierarchical decision denominator item;

[0132] Perform a division operation on the hierarchical decision sub - item and the hierarchical decision denominator item, and apply the spatio - temporal decay coefficient to generate a final hierarchical value from 0 to 10 to obtain the regional hierarchical information.

[0133] In the above solution, the present invention forms an adaptive risk perception ability by calculating the information entropy value of the monitoring data in real time (a quantization index of data chaos) and combining the historical occurrence frequency of disaster events. The classification stability under abnormal data conditions can be effectively improved compared with the traditional fixed-weight mode; for the surface curvature-soil moisture dual-factor enhancement model, when the slope > 25° and the soil water content > 28%, the risk weight index doubling mechanism is automatically triggered to improve the disaster identification and response speed in steep slope areas. And by adopting the method of collaborative analysis of multi-source data, the covariance time series analysis of soil moisture and rainfall is carried out to realize the dynamic quantitative evaluation of the parameter correlation strength, which can improve the hidden landslide points in rainstorm scenarios and reduce the false alarm rate. And a spatio-temporal attenuation model is designed to fuse the terrain surface curvature compensation mechanism, which implements linear attenuation for flat areas (surface curvature < 0.05) and retains 60% data validity for complex terrains (surface curvature > 0.1), effectively improving the anti-interference robustness of the model.

[0134] Furthermore, Figure 7 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0135] Among them, in another preferred embodiment provided by the present invention, a debris flow geological disaster monitoring and emergency response system includes:

[0136] A monitoring area division unit 101, configured to determine a debris flow monitoring area, obtain geological and topographical data of the debris flow monitoring area, divide the debris flow monitoring area into multiple monitoring sub-areas, and select a key risk sub-area.

[0137] In the embodiment of the present invention, the monitoring area division unit 101 determines a debris flow monitoring area with the need for debris flow geological disaster monitoring and emergency response, obtains the geological and topographical data of the debris flow monitoring area, divides the debris flow monitoring area into multiple monitoring sub-areas through terrain analysis of the geological and topographical data, then matches the multiple geological and topographical sub-area data corresponding to the multiple monitoring sub-areas from the geological and topographical data, conducts risk initialization analysis on the multiple geological and topographical sub-area data, and selects a key risk sub-area from the multiple monitoring sub-areas.

[0138] Specifically, Figure 8 The structural block diagram of the monitoring area division unit 101 in the system provided by the embodiment of the present invention is shown.

[0139] Among them, in the preferred embodiment provided by the present invention, the monitoring area division unit 101 specifically includes:

[0140] An area determination module 1011, configured to determine a debris flow monitoring area;

[0141] A data acquisition module 1012, configured to acquire geological and topographical data of the debris flow monitoring area;

[0142] A region division module 1013, configured to divide the debris flow monitoring area into multiple monitoring sub-regions according to the geological and topographical data;

[0143] A risk initialization analysis module 1014, configured to perform risk initialization analysis on multiple monitoring sub-regions based on the geological and topographical data, and select key risk sub-regions.

[0144] Furthermore, the debris flow geological disaster monitoring and emergency response system further includes:

[0145] A multi-source sensing monitoring unit 102, configured to perform multi-source sensing monitoring on multiple monitoring sub-regions to obtain multiple sensing monitoring data.

[0146] In an embodiment of the present invention, the multi-source sensing monitoring unit 102 obtains soil monitoring data by performing soil moisture sensing monitoring on multiple monitoring sub-regions, obtains rainfall monitoring data by performing rainfall sensing monitoring on multiple monitoring sub-regions, obtains displacement monitoring data by performing displacement sensing monitoring on the surface and / or rock strata of multiple monitoring sub-regions, and at the same time, obtains vibration monitoring data by performing vibration sensing monitoring on multiple monitoring sub-regions. The soil monitoring data, rainfall monitoring data, displacement monitoring data, and vibration monitoring data together constitute multiple sensing monitoring data.

[0147] Specifically, Figure 9 FIG. shows a structural block diagram of the multi-source sensing monitoring unit 102 in the system provided by the embodiment of the present invention.

[0148] Among them, in a preferred embodiment provided by the present invention, the multi-source sensing monitoring unit 102 specifically includes:

[0149] A soil moisture sensor 1021, configured to perform soil moisture sensing monitoring on multiple monitoring sub-regions to obtain soil monitoring data;

[0150] A rainfall sensor 1022, configured to perform rainfall sensing monitoring on multiple monitoring sub-regions to obtain rainfall monitoring data;

[0151] A displacement sensor 1023, configured to perform displacement sensing monitoring on multiple monitoring sub-regions to obtain displacement monitoring data;

[0152] A vibration sensor 1024, configured to perform vibration sensing monitoring on multiple monitoring sub-regions to obtain vibration monitoring data.

[0153] Furthermore, the debris flow geological disaster monitoring and emergency response system further includes:

[0154] The monitoring data preprocessing unit 103 is configured to perform data preprocessing on the multiple sensing monitoring data to obtain multiple standard monitoring data.

[0155] In an embodiment of the present invention, the monitoring data preprocessing unit 103 identifies and removes noise, redundancy, and anomalies from the multiple sensing monitoring data to obtain multiple effective monitoring data, and then performs standardization processing on the multiple effective monitoring data according to a preset data standard to obtain multiple standard monitoring data.

[0156] The data synchronization and fusion unit 104 is configured to perform data synchronization and fusion on the multiple standard monitoring data to generate fused monitoring data.

[0157] In an embodiment of the present invention, the data synchronization and fusion unit 104 obtains the sampling periods and / or timestamps of the multiple standard monitoring data, and then uses Kalman filtering to align the multiple sampling periods and / or multiple timestamps to generate multiple aligned time points to ensure data time synchronization. Furthermore, according to the multiple aligned time points, the data synchronization and fusion unit 104 performs data fusion on the multiple standard monitoring data to generate fused monitoring data.

[0158] The emergency warning and control unit 105 is configured to, according to the fused monitoring data, when there is a debris flow hazard, perform hazard grading on multiple monitoring sub-regions to generate regional grading information, and starting from the key risk sub-regions, perform unmanned aerial vehicle (UAV) emergency warning and control.

[0159] In an embodiment of the present invention, the emergency warning and control unit 105 performs comparative analysis of hazards on the fused monitoring data according to preset hazard threshold data to determine whether there is a debris flow hazard. In the case of determining that there is a debris flow hazard, the emergency warning and control unit 105 performs hazard grading on multiple monitoring sub-regions, records the regional grading information, and then starting from the key risk sub-regions, plans a flight warning route according to the regional grading information. Furthermore, according to the flight warning route, the emergency warning and control unit 105 performs flight control for UAV emergency warning.

[0160] Specifically, Figure 10 FIG. shows the structural block diagram of the emergency warning and control unit 105 in the system provided by the embodiment of the present invention.

[0161] Among them, in a preferred embodiment provided by the present invention, the emergency warning and control unit 105 specifically includes:

[0162] The hazard judgment module 1051 is configured to perform hazard analysis on the fused monitoring data to determine whether there is a debris flow hazard;

[0163] The hazard grading module 1052 is configured to, when there is a debris flow hazard, perform hazard grading on multiple monitoring sub-regions to generate regional grading information;

[0164] A route planning module 1053, configured to plan a flight warning route starting from the key risk sub-region according to the region classification information;

[0165] A warning control module 1054, configured to perform emergency warning control of the UAV according to the flight warning route.

[0166] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0167] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0168] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0169] The above-described embodiments merely represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the appended claims.

[0170] The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A debris flow geological disaster monitoring and emergency response method, characterized in that: The method specifically comprises the following steps: Determine a debris flow monitoring area, obtain geological and topographic data of the debris flow monitoring area, divide the debris flow monitoring area into multiple monitoring sub-areas, and select key risk sub-areas; Performing multi-source sensor monitoring on the plurality of monitoring sub-areas to obtain a plurality of sensor monitoring data; Performing data preprocessing on the plurality of sensor monitoring data to obtain a plurality of standard monitoring data; Synchronously fusing a plurality of the standard monitoring data to generate fused monitoring data; According to the fused monitoring data, when there is a risk of debris flow, multiple monitoring sub-areas are classified into risk categories, regional classification information is generated, and emergency warning control of drones is performed based on key risk sub-areas; The determining of the debris flow monitoring area, obtaining geological and topographic data of the debris flow monitoring area, dividing the debris flow monitoring area into a plurality of monitoring sub-areas, and selecting key risk sub-areas specifically comprises the following steps: Determine debris flow monitoring areas; Acquiring geological and topographic data of the debris flow monitoring area; According to the geological and topographic data, the debris flow monitoring area is divided into a plurality of monitoring sub-areas; Based on the geological and topographic data, initial risk analysis is performed on the plurality of monitoring sub-areas, and key risk sub-areas are selected; The step of performing risk initialization analysis on a plurality of monitoring sub-areas based on the geological and topographic data and selecting key risk sub-areas specifically includes the following steps: According to the geological and topographic data, the terrain characteristic triplet data of each monitoring sub-area is extracted, and the terrain characteristic triplet data includes the slope angle, rock and soil density and vegetation coverage index of the center point; The terrain feature triplet data of each monitoring sub-area is input into the principal component analysis model, the variance contribution rate of the first principal component is calculated, and the variance contribution rate of the first principal component is used as the comprehensive terrain index to obtain the principal component score of each monitoring sub-area; The membership function threshold is set according to the regional geomechanical parameters, and the historical disaster critical value is obtained. When the principal component score of each monitoring sub-region exceeds the historical disaster critical value, the membership degree of the principal component score of each monitoring sub-region is calculated using the S-type function to obtain the geological risk membership degree of each monitoring sub-region; The terrain surface curvature gradient of each monitoring sub-area is analyzed to obtain the terrain mutation characteristic index; Obtain the historical risk time records of each monitoring sub-area and the terrain data at the corresponding time; The neural network model is trained with the historical risk time records of each monitoring sub-area and the terrain data at the corresponding time. Through the correlation analysis between the frequency of disasters and terrain parameters, the weight coefficients of each parameter are iteratively optimized to obtain the dynamic weight of the terrain characteristics. The geological risk membership is weighted with the dynamic weight of the corresponding terrain features to obtain the weighted sum of the geological risk membership of each monitoring sub-area; The terrain mutation characteristic index is amplified and integrated with the weighted sum of the geological risk membership of each monitoring sub-area to obtain the comprehensive risk value of each monitoring sub-area; The comprehensive risk values ​​of each monitoring sub-area are sorted in descending order, and the first 20% are classified as level 1 risk areas, the middle 30% as level 2 risk areas, and the rest as level 3 risk areas, so as to obtain a regional risk distribution map with graded labels.

2. The debris flow geological disaster monitoring and emergency response method according to claim 1 is characterized in that: The performing multi-source sensor monitoring on the plurality of monitoring sub-areas and obtaining a plurality of sensor monitoring data specifically comprises the following steps: Perform soil moisture sensing monitoring on the plurality of monitoring sub-areas to obtain soil monitoring data; Performing rainfall sensing monitoring on the plurality of monitoring sub-areas to obtain rainfall monitoring data; Performing displacement sensing monitoring on the plurality of monitoring sub-areas to obtain displacement monitoring data; Vibration sensing monitoring is performed on the plurality of monitoring sub-areas to obtain vibration monitoring data.

3. The debris flow geological disaster monitoring and emergency response method according to claim 2 is characterized in that: The data preprocessing of the plurality of sensor monitoring data to obtain a plurality of standard monitoring data specifically comprises the following steps: Identify and remove noise, redundancy and abnormal data from the plurality of sensor monitoring data to obtain a plurality of valid monitoring data; The plurality of effective monitoring data are standardized to obtain a plurality of standard monitoring data.

4. The debris flow geological disaster monitoring and emergency response method according to claim 3 is characterized in that: The standardization of the plurality of valid monitoring data to obtain a plurality of standard monitoring data specifically comprises the following steps: Obtain the sensor number corresponding to the valid monitoring data, and obtain the corresponding sensor factory accuracy identification code based on the sensor number; Parse the factory accuracy identification code of the sensor to obtain the benchmark mean and standard deviation of each type of sensor; Obtain the sensor monitoring data sequence from the valid monitoring data within T minutes before the current moment; The same type of data in the sensor monitoring data sequence is sorted by time to obtain a sliding window data sequence, and the middle position value in the sorting is taken to obtain the sliding window median value of each type of sensor; The interquartile range is calculated based on the values ​​at the 25% and 75% positions in the sliding window data sequence; Generate an adaptive adjustment factor according to the number and interquartile range of data in the sliding window data sequence; Calculate the difference between the current data timestamp and the data receiving time in the sliding window data sequence, and substitute the difference into the S-type function for nonlinear transformation to obtain the time-effectiveness enhancement coefficient; Standardize each type of current sensor monitoring data in the effective monitoring data using the sensor's baseline mean and standard deviation to generate a number of initial standardized results; Calculate the absolute deviation between the current sensor monitoring data in the effective monitoring data and the median value of the sliding window, and according to the adaptive adjustment factor, use the hyperbolic tangent function to nonlinearly map the absolute deviation to obtain the generated anti-error adjustment amount; A number of initial standardized results are fused with the generated robustness adjustment amount to obtain a fused result, which is then time-enhanced using a time-enhancement coefficient to obtain the final multiple standard monitoring data.

5. The debris flow geological disaster monitoring and emergency response method according to claim 4 is characterized in that: The synchronous fusion of the plurality of standard monitoring data to generate fused monitoring data specifically comprises the following steps: Obtaining a sampling period and / or timestamp of a plurality of the standard monitoring data; Using Kalman filtering, aligning the plurality of sampling periods and / or the plurality of time stamps to generate a plurality of aligned time points; According to the plurality of aligned time points, data fusion is performed on the plurality of standard monitoring data to generate fused monitoring data.

6. The debris flow geological disaster monitoring and emergency response method according to claim 5 is characterized in that: The step of fusing the plurality of standard monitoring data according to the plurality of aligned time points to generate fused monitoring data specifically comprises the following steps: Obtain the frequency of occurrence of each sensor data value in the standard monitoring data to obtain the data frequency distribution; Calculate a first information entropy value according to the data frequency distribution, and convert the first information entropy value into a second weight coefficient; Normalizing the second weight coefficient to obtain a dynamic weight coefficient of each sensor; Obtain the timestamp of each sensor data value in the standard monitoring data, and calculate the time difference between the current moment and the timestamp of the sensor data value; Obtain the number of historical warnings, perform logarithmic smoothing on the time difference according to the number of historical warnings, and obtain the smoothed time difference; The data value of each sensor in the standard monitoring data is compared with the reference range of the corresponding sensor. According to the comparison result and the smoothed time difference, exponential attenuation suppression is applied to the data deviating from the reference range to obtain the data value adjusted in the time and space dual domains of each sensor. Obtain the surface curvature change rate of the monitoring sub-area, calculate the ratio of the surface curvature of the current monitoring sub-area to the maximum surface curvature value of all monitoring sub-areas, and amplify the ratio nonlinearly to generate the terrain feature amplification coefficient; Obtaining the comprehensive data distribution of all sensors in the monitoring sub-area, and calculating the second information entropy value of the entire monitoring sub-area according to the comprehensive data distribution of all sensors in the monitoring sub-area; The second information entropy value is mapped to a smooth interval of 0-1 using a hyperbolic function to generate a regional conflict adjustment coefficient; The data values ​​adjusted in the time and space dual domains of each sensor are weighted by the dynamic weight coefficient of each sensor, and then the terrain feature amplification coefficient is used to amplify the weighted result to obtain the data processing result of each sensor. The regional conflict adjustment coefficient is added to the data processing result of each sensor, and then accumulated and synthesized to obtain the fused monitoring data.

7. The debris flow geological disaster monitoring and emergency response method according to claim 6 is characterized in that: The method of performing risk classification on multiple monitoring sub-areas according to the fused monitoring data when there is a risk of debris flow, generating regional classification information, and performing emergency warning control of drones based on key risk sub-areas specifically includes the following steps: Performing risk analysis on the fused monitoring data to determine whether there is a risk of debris flow; When there is a risk of debris flow, the plurality of monitoring sub-areas are classified into risk categories to generate regional classification information; Taking the key risk sub-areas as the starting point and according to the regional classification information, plan the flight warning route; Carry out emergency warning control of the drone according to the flight warning route.

8. The debris flow geological disaster monitoring and emergency response method according to claim 7 is characterized in that: When there is a risk of debris flow, the plurality of monitoring sub-areas are classified into risk categories, and the generation of regional classification information specifically comprises the following steps: Extract the total number of warning events that occurred in the same monitoring sub-area within 3 months from the current time; Obtaining a second information entropy value in a current detection cycle and a historical maximum entropy value, and normalizing the second information entropy value in the current detection cycle using the historical maximum entropy value to obtain a normalized entropy value; The normalized entropy value and the total number of warning events are weighted and superimposed to obtain the dynamic entropy weight coefficient; Extract soil moisture time series data and rainfall monitoring data from the fused monitoring data, input the rainfall monitoring data into the S-type function converter, and output a soil risk index of 0-1; Calculate the percentage of the slope of the current monitoring sub-area and the maximum slope of all monitoring sub-areas to obtain the slope value; The corresponding slope weight is set according to the soil moisture time series data, and the slope value and the slope weight are weighted to obtain the enhanced slope value; The enhanced slope value is fused with the soil risk index to obtain the terrain coupling weight; Calculate the covariance relationship value of the soil moisture time series data and the rainfall monitoring data extracted from the data in the past hour, and standardize the variance relationship value to obtain the standardized covariance relationship value; The standardized covariance relationship value is introduced into the time decay factor for decay, and the decay result is constrained to the interval [-1,1] through the hyperbolic tangent function to obtain the covariance weight coefficient; Establish a surface curvature-attenuation rate mapping table, substitute the surface curvature change rate of the monitored sub-area into the surface curvature-attenuation rate mapping table, and obtain the surface curvature adjustment coefficient; Calculate the time difference between the current moment and the most recent warning, and use the surface curvature adjustment coefficient to adjust the time between the current moment and the most recent warning to obtain the terrain-corrected time difference; Asymmetric Gaussian attenuation calculation is performed based on the terrain-corrected time difference to obtain the primary attenuation coefficient; According to the surface curvature change rate of the monitoring sub-area, linear attenuation compensation is applied to the corresponding flat area part of the primary attenuation coefficient to obtain the spatiotemporal attenuation coefficient; The dynamic entropy weight coefficient is integrated with the terrain coupling weight to construct the molecular term of hierarchical decision-making; Obtain the topographic surface curvature reference value, fuse the ratio of the surface curvature change rate of the monitored sub-area to the topographic surface curvature reference value with the covariance weight coefficient, and construct the denominator term of the hierarchical decision; The numerator of the grading decision is divided by the denominator of the grading decision, and the spatiotemporal attenuation coefficient is applied to generate a final grading value of 0-10 to obtain the regional grading information.

9. A debris flow geological disaster monitoring and emergency response system, the system is applied to the debris flow geological disaster monitoring and emergency response method according to any one of claims 1 to 8, characterized in that: The system includes a monitoring area division unit, a multi-source sensor monitoring unit, a monitoring data preprocessing unit, a data synchronization fusion unit and an emergency warning control unit, wherein: A monitoring area division unit is used to determine a debris flow monitoring area, obtain geological and topographic data of the debris flow monitoring area, divide the debris flow monitoring area into a plurality of monitoring sub-areas, and select key risk sub-areas; A multi-source sensor monitoring unit, used to perform multi-source sensor monitoring on the plurality of monitoring sub-areas to obtain a plurality of sensor monitoring data; A monitoring data preprocessing unit, used for performing data preprocessing on the plurality of sensor monitoring data to obtain a plurality of standard monitoring data; A data synchronization fusion unit, used for performing data synchronization fusion on a plurality of the standard monitoring data to generate fused monitoring data; The emergency warning control unit is used to classify the dangers of multiple monitoring sub-areas according to the fused monitoring data when there is a risk of debris flow, generate regional classification information, and perform emergency warning control of drones based on key risk sub-areas.

Images