Cloud-based geological disaster analysis method and system

Abstract

The invention relates to the technical field of data analysis, in particular to a geological disaster analysis method and system based on cloud, and the method comprises the following steps: obtaining and coding a rainfall intensity, gradient, wind speed and humidity data source of a geological area, recognizing a disaster factor mapping structure to generate a coding group, and quantifying a gradient difference superposition humidity screening high-frequency unit. And performing connectivity analysis to extract an aggregated risk area boundary, extracting a wind speed and humidity ratio to screen out an overrun factor, outputting a high-coupling factor combination, and identifying a trend change to output a risk grading identification group. According to the method, multi-source environment data are synchronously collected and indexed according to time and space, the matching precision of disaster factors and space grids is enhanced, the high-frequency disturbance recognition stability is improved through multi-layer fusion analysis, risk boundaries are extracted through communication modeling to improve the positioning accuracy, and the coupling factor analysis capability is improved through factor ratio fusion; trend matching is combined with a response gradient to improve risk grading perspectiveness, and the overall process enhances early warning efficiency and expands data analysis depth and breadth.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a cloud-based geological disaster analysis method and system. BACKGROUND

[0002] The technical field of data analysis includes the use of various mathematical models and statistical methods to organize and analyze data to discover potential laws and support decision-making processes. The core content of this technical field includes data collection, data cleaning, data modeling, data visualization, and result interpretation. Through the combination of computer science, mathematics, statistics, and other disciplines, data analysis technology can be widely applied in meteorology, transportation, geology, and other industries. In the field of geology, data analysis is particularly useful for identifying geological structures, analyzing disaster warning, and studying strata evolution. By integrating large amounts of raw data and real-time monitoring information, the cognitive ability of natural phenomena and potential risks can be improved.

[0003] Among them, the geological disaster analysis method refers to the risk assessment, cause identification and evolution trend judgment of geological disasters such as landslides, collapses and mudslides. By combining remote sensing image information, original disaster records, geological and geomorphic data, and environmental parameters such as precipitation, the method can identify and warn areas prone to geological disasters through time series analysis, spatial distribution comparison, and multi-factor cross comparison. It covers geological unit division criteria, environmental evolution data analysis standards, and rainfall and terrain slope coupling relationship analysis methods. Trend prediction is achieved through data fusion analysis methods, including remote sensing image change recognition, time series-based terrain index evolution analysis, original disaster database pattern matching, and surface hydrological factor correlation analysis.

[0004] The existing technology has the problem of slow response to dynamic factors, lacks the ability to perceive the continuity of real-time data and spatial pattern changes, and the disaster cause identification process is easily affected by short-term fluctuations, lacking systematic dynamic verification means. In the spatial clustering process, fixed thresholds and static region division rules are often used, which cannot reflect the evolution characteristics of environmental factor aggregation trends, thereby affecting the integrity of risk boundary identification. In the factor screening and trend judgment process, the processing method is biased towards single variable correlation analysis, which cannot reflect the cross-coupling characteristics between factors, and is prone to misjudgment caused by variable imbalance. The original data calling is mainly based on periodic pattern matching, and the trend change response is insufficient, resulting in a lack of dynamic adaptation ability in snapshot comparison. For example, during the concentrated rainfall period, the coupling response between regional slope change and humidity rise cannot be timely captured by traditional models, directly affecting the timely prediction of disaster risks, and easily forming a problem of warning delay or blind area, limiting the practical effect of traditional schemes in multi-dimensional information interweaving and fast response scenarios, and cannot meet the risk assessment needs in complex geological dynamic background. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art, and the cloud-based geological disaster analysis method and system are proposed.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme: a cloud-based geological disaster analysis method, comprising the following steps: S1: Obtain the rainfall intensity, slope, wind speed and soil moisture data sources in the monitored geological area, collect the real-time distribution of the data in the cloud platform, encode each type of data according to the time stamp and latitude and longitude, identify the mapping structure between the disaster factors and the grid area, and generate a disaster factor grid encoding group; S2: Call the slope and rainfall intensity layers in the disaster factor grid encoding group, select the continuously changing units in the same time window, perform gradient difference quantization, superimpose the humidity sequence, analyze the frequency distribution of the fluctuation value, select the units with fluctuation frequency exceeding the set threshold, and output a high-frequency fluctuation grid list; S3: Based on the high-frequency fluctuation grid list, perform connectivity analysis on the adjacency relationship and spatial density, select the grid cluster that meets the connectivity rate threshold, identify the corresponding spatial boundary identifier, and output the aggregated risk area boundary value; S4: Call the limited area of the aggregated risk area boundary value, extract the wind speed and soil humidity sequence, calculate the corresponding change rate, and compare it with the slope data weighted deviation, exclude the factors with deviation exceeding the limit, and retain the factor combination with high cross-coupling degree, and output a geological disaster area reconstruction factor set.

[0007] As a further scheme of the present application, the disaster factor grid encoding group includes spatial positioning number, time synchronization label, disaster factor type identifier, the high-frequency fluctuation grid list includes abnormal intensity value, fluctuation period parameter, frequency threshold label, the aggregated risk area boundary value includes boundary coordinate range, aggregated grid number, spatial aggregation identifier, and the geological disaster area reconstruction factor set includes factor combination weight, coupling strength coefficient and change rate index.

[0008] As a further scheme of the present application, the acquisition step of the disaster factor grid encoding group is specifically: S111: Obtain the rainfall intensity, slope, wind speed and soil moisture data sources in the monitored geological area, unify the data time and spatial resolution according to the latitude and longitude field and time stamp field of each record, and generate four types of environmental factor original data table; S112: Based on the four types of environmental factor original data table, extract the time sequence under the same latitude and longitude in each type of data, use a sliding window to segment by time, map the sequence start and end time and coordinates to a unique encoding, and generate a single-factor grid mapping encoding value group; S113: Based on the single-class factor grid mapping coding value group, the grid area is determined according to the latitude and longitude field, the factor value combination in the same area is merged, the spatial coverage proportion of the factor combination under the different time segments is compared, the coupling relationship characteristics of the regional factor combination are extracted, and a disaster factor grid coding group is generated.

[0009] As a further scheme of the present application, the high-frequency fluctuation grid list acquisition step is specifically: S211: The disaster factor grid coding group is called, the slope layer and the precipitation intensity layer are extracted, the continuously changing grid in the same time window is selected, the difference between the unit slope change rate and the precipitation intensity change rate is identified, the time sequence gradient difference distribution is identified, and a slope coupling difference graph is generated; S212: Based on the slope coupling difference graph, the humidity sequence layer is superimposed, the fluctuation frequency of the unit humidity value in the period is counted, and the coupling difference value is combined to generate a humidity disturbance frequency graph; S213: According to the humidity disturbance frequency graph, the grid unit with a frequency value greater than the humidity fluctuation frequency threshold value is screened, the corresponding precipitation change amplitude, slope change cumulative amount and disturbance duration are combined, the composite disturbance index value of the high-sensitivity unit is calculated, the high-sensitivity unit with concentrated fluctuation is identified, and a cloud high-sensitivity sliding area grid list is generated.

[0010] As a further scheme of the present application, the acquisition step of the aggregation risk area boundary value is specifically: S311: Based on the high-frequency fluctuation grid list, the boundary coincidence value and the corresponding coordinates of the grid and the adjacent grid are extracted, the geometric connection state between the grids is judged, and the grid adjacency connection degree value is obtained; S312: The grid adjacency connection degree value is called, the spatial density is calculated, the Euclidean distance between the grid centers, the number of overlapping grids per unit area and the adjacency connection degree deviation are identified, the spatial aggregation situation is judged, the spatial aggregation strength ratio is calculated, the connectivity threshold is set, the grid group with aggregation degree exceeding the threshold is screened, and the grid connectivity rate screening set is obtained. S313: According to the grid connectivity rate screening set, the edge grid contour coordinates and sequence are extracted, the boundary of the closed connected region is identified, and the aggregation risk area boundary value is output.

[0011] As a further scheme of the present application, the acquisition step of the geological disaster area reconstruction factor set is specifically: S411: The aggregation risk area boundary value is called, the wind speed sequence and the soil humidity sequence of the corresponding position are extracted, the ratio sequence of the wind speed and the humidity is identified, the ratio difference between adjacent time points is calculated, whether the difference value exceeds the wind and humidity fluctuation threshold value is judged, the time period exceeding the limit is screened out, and a wind and humidity ratio constraint matrix is established. S412: According to the rheumatism ratio constraint matrix, the slope data in the region is called to calculate the rheumatism slope offset coupling degree, the high coupling points in the region are screened by comparing the set offset threshold value, the spatial intersection point set is extracted and the data binding coordinates under the cloud platform are located, and the geological disaster area reconstruction factor set is output.

[0012] As a further scheme of the present application, the method further comprises a step S5: S5: Based on the time sequence change trend of the geological disaster area reconstruction factor set, the accumulated precipitation and terrain snapshot group in the region are extracted as matching reference, the consistency section between the trend and the original snapshot is judged, the response gradient is identified according to the trend change direction, and the geological disaster risk trend grading identification group is output. The geological disaster risk trend grading identification group comprises trend level code, response direction parameter and consistency verification label.

[0013] As a further scheme of the present application, the acquisition step of the geological disaster risk trend grading identification group is specifically: S511: Based on the time sequence change trend of the geological disaster area reconstruction factor set, the factor value snapshot of the continuous time node is identified, the terrain snapshot corresponding to the accumulated precipitation time is extracted, the coupling strength of the spatial distribution of the accumulated precipitation and the terrain factor group is analyzed, and the coupling strength trend sequence is obtained. S512: According to the coupling strength trend sequence, the trend change difference of adjacent time period is extracted, the consistency with the fluctuation direction of the grid cell value in the terrain snapshot is analyzed, the region with consistent change direction in space is identified, the continuous consistent section is marked, and the proportion in the whole region is counted, the spatial consistency section distribution coefficient is obtained. S513: The spatial consistency section distribution coefficient is called to extract the region section with continuous consistent direction, the response amplitude difference in the trend positive and negative directions is determined, the disaster risk level interval is divided, and the geological disaster risk trend grading identification group is output.

[0014] The cloud-based geological disaster analysis system is used to execute the above-mentioned cloud-based geological disaster analysis method, and the system comprises: The mapping relationship analysis module obtains the precipitation intensity, slope, wind speed and soil moisture data source in the monitored geological region, extracts the time stamp and latitude and longitude of each item, identifies the time and space combination code across categories, analyzes the mapping relationship between the disaster factor and the monitoring grid, and generates the geological factor space index group; The disturbance identification module extracts the change amount difference of the code sequence of slope and precipitation intensity in the same time window based on the geological factor space index group, superimposes the humidity sequence to calculate the fluctuation frequency, screens the code grid with frequency exceeding the set value, and outputs the high disturbance response grid set. The aggregation identification module analyzes the adjacency relationship and dense distribution between the grids based on the high disturbance response grid set, labels the continuous connected units and extracts the boundary range, identifies the potential aggregation disaster area spatial profile, and generates a risk aggregation area boundary box group; The factor reconstruction module extracts the regional wind speed and soil moisture change rate, performs deviation comparison with the slope change value based on the risk aggregation area boundary box group, removes the deviation out-of-limit combination, retains the coupling combination in the high linkage region, and establishes a geological disaster area reconstruction factor set; The trend classification module matches the regional cumulative rainfall graph and the slope snapshot graph based on the sequence variation trend of the geological disaster area reconstruction factor set, extracts the consistent section of variation, analyzes the direction of wind speed and humidity sequence fluctuation, divides the response level, and outputs a geological disaster risk trend classification identification group.

[0015] Compared with the prior art, the advantages and positive effects of the present application are: In the present application, by synchronously collecting and uniformly indexing multi-source environmental data based on time and geographical position in real time, the recognition accuracy of disaster factors and spatial grid relationship is strengthened, and the fine-grained control ability of dynamic change characteristics is improved. Multi-level data fusion analysis is performed on the gradient change unit in a specific time window, combined with the fluctuation frequency filtering rule, the stable recognition performance of the high-frequency disturbance area is enhanced. In the spatial relationship level, the connectivity and density of the fluctuation area are modeled, the aggregation identification of the geological abnormal area boundary is realized, and the accuracy of the regional spatial direction is improved. In the local area factor analysis, the wind speed and humidity variation rate are combined with the slope weighted deviation comparison mechanism to avoid single factor error interference, effectively screen high-coupling disaster factor combinations, and enhance the analytical ability of the internal relationship between disaster factors. Combined with the original snapshot and the cumulative rainfall and other dynamic evolution sequences, a trend expression consistency matching mechanism is constructed, and the response gradient analysis method is embedded in the trend identification, so that the risk level judgment has higher timeliness and foresight. Through the construction of a multi-factor spatio-temporal fusion and change trend driven decision-making process, the response efficiency of disaster warning is improved, and the depth and breadth of information utilization are widened in the data analysis dimension, which promotes the evolution of geological disaster risk identification to high precision and high dynamic adaptation. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The workflow schematic diagram of the present application is shown in the figure; Figure 2 The acquisition flowchart of the disaster factor grid code group in the present application is shown in the figure; Figure 3 The acquisition flowchart of the high-frequency fluctuation grid list in the present application is shown in the figure; Figure 4 The acquisition flowchart of the aggregation risk area boundary value in the present application is shown in the figure; Figure 5 The flow chart for obtaining the geological disaster region reconstruction factor set in the present application is shown in Figure 1. Figure 6 The flow chart for obtaining the geological disaster risk trend classification group in the present application is shown in Figure 2. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0018] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, which are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited. EMBODIMENT

[0019] Please refer to Figure 1 The present application provides a technical solution: a cloud-based geological disaster analysis method, comprising the following steps: S1: Obtain the rainfall intensity, slope, wind speed and soil moisture data sources in the monitored geological region, collect the real-time distribution of the data in the cloud platform, perform time and space index coding on each type of data according to the time stamp and latitude and longitude, identify the mapping structure between the disaster factors and the grid region, and generate a disaster factor grid coding group; S2: Call the slope and rainfall intensity layers in the disaster factor grid coding group, select the continuously changing units within the same time window, perform gradient difference quantization, superimpose the humidity sequence, analyze the frequency distribution of the fluctuation value, filter out the units with fluctuation frequency exceeding the set threshold, and output a high-frequency fluctuation grid list; S3: Based on the high-frequency fluctuation grid list, perform connectivity analysis on the adjacency relationship and spatial density, select the grid clusters that meet the connectivity rate threshold, identify the corresponding spatial boundary identifier, and output the aggregated risk area boundary value; S4: Call the limited area of the aggregated risk area boundary value, extract the wind speed and soil moisture sequence, calculate the corresponding change rate, and compare it with the weighted deviation of the slope data, filter out the factors with deviation exceeding the limit, and retain the factor combination with high cross-coupling degree, and output the geological disaster region reconstruction factor set; S5: Based on the time series trend of the geological disaster area reconstruction factor set, the accumulated precipitation and terrain snapshot group in the area are extracted as the matching reference, the consistency section between the trend and the original snapshot is judged, the response gradient is identified according to the trend change direction, and the geological disaster risk trend classification identification group is output.

[0020] The disaster factor grid encoding group includes spatial positioning number, time synchronization label, disaster factor type identification, high-frequency fluctuation grid list includes abnormal intensity value, fluctuation period parameter, frequency threshold label, aggregated risk area boundary value includes boundary coordinate range, aggregated grid number, spatial aggregation identification, geological disaster area reconstruction factor set includes factor combination weight, coupling strength coefficient, change rate index, and geological disaster risk trend classification identification group includes trend level code, response direction parameter and consistency verification label.

[0021] Please refer to Figure 2 The acquisition steps of the disaster factor grid encoding group are as follows: S111: Obtain the precipitation intensity, slope, wind speed and soil moisture data sources in the monitored geological area, unify the data time and spatial resolution according to the latitude and longitude field and time stamp field of each record, and generate four types of environmental factor original data table; For each record in the data source, the latitude and longitude and time stamp field are extracted, and based on this, the spatial and time adjustment is carried out to ensure that all data have consistent resolution. For example, for precipitation intensity data, if there is a time difference between different time stamps, it needs to be adjusted by interpolation or averaging method; for spatial resolution, the latitude and longitude of all records need to be aligned, and appropriate grid division method is used to uniformly process the precipitation intensity, slope, wind speed and soil moisture values in each grid, to ensure that the data have consistency in geographical space and meet certain time interval requirements. The process can ensure that the environmental factor table after data processing can fully and accurately reflect the changes of the environment in the monitoring area. Taking a certain monitoring area as an example, assuming that the collection time points of precipitation intensity, slope, wind speed and soil moisture are different in the time interval from January 1st to January 10th, the time stamp and latitude and longitude fields in the data source will show the inconsistency of the data, and the blank time period needs to be filled by interpolation calculation to form a data set with uniform time resolution, and a four-type environmental factor original data table is generated, which contains all the environmental information at different times and spatial positions as the basis data for subsequent analysis.

[0022] S112: Based on the four-type environmental factor original data table, the time series under the same latitude and longitude in each type of data is extracted, the sliding window is used for time segmentation, the sequence start and end time and coordinates are mapped to unique encoding, and the single-factor grid mapping encoding value group is generated; Based on the latitude and longitude information of each monitoring point, the time series data of four types of environmental factors corresponding to the location are extracted, and the time series is divided by time period using the sliding window technique. The size of the sliding window can be set as needed, for example, set to 3 hours per window. Within each window, a unique time code is generated based on the data in that time period, combined with the latitude and longitude information, to generate a grid mapping code value for each factor in that time window. For example, if the precipitation intensity, slope, wind speed, and soil moisture data of a certain location between 0:00 on January 1 and 3:00 on January 1 are known, a unique code is generated based on the timestamp of the window and the latitude and longitude of the location. This code can be used as an identifier for subsequent analysis to help distinguish between environmental factor data at different time periods and different locations. The core of the process is the time segmentation of the sliding window and how to generate a unique grid code through the dual mapping of time and space to ensure accurate correspondence of each factor data in time and space.

[0023] S113: Based on the single-factor grid mapping code value group, the grid area is determined according to the latitude and longitude field, the factor value combination in the same area is merged, and the spatial coverage ratio of the factor combination in the differentiated time segments is compared to extract the coupling relationship characteristics of the regional factor combination, and generate a disaster factor grid code group; According to the single-factor grid mapping code value group, different grid areas are determined within the region according to the latitude and longitude field, for example, the monitoring area is divided into multiple 1 square kilometer grids to ensure that each grid contains data from multiple monitoring points. The factor value combination in each grid is merged, and the environmental factor data in each time segment within the same grid area is merged. According to the factor combination in different time segments, the spatial coverage ratio is calculated to ensure the accuracy of the coverage calculation, and then the spatial coverage ratio of the factor combination in different time segments is compared to extract its coupling relationship characteristics. Assuming that in a certain area, the precipitation intensity and soil moisture show a high coupling relationship in a certain time period, this point is identified and the corresponding coupling relationship code value is generated to reflect the interaction between precipitation and soil moisture in that area. The coupling relationship characteristics and grid code value will constitute a disaster factor grid code group for subsequent disaster analysis.

[0024] Please refer to Figure 3 The steps for obtaining the high-frequency fluctuation grid list are as follows: S211: Call the disaster factor grid code group, extract the slope layer and precipitation intensity layer, select continuous change grids within the same time window, identify the difference between the unit slope change rate and the precipitation intensity change rate, identify the time series gradient difference distribution, and generate a slope precipitation coupling difference map; By monitoring continuously changing grids within the same time window, slope and precipitation intensity data for each grid cell are acquired. This data is updated and processed in real time through a Geographic Information System (GIS) platform. First, the slope change rate and precipitation intensity change rate of each grid are calculated. By comparing the differences between these two rates, potential landslide hazard areas can be preliminarily identified. For example, in a typical mountainous geological disaster monitoring project, real-time data shows a sudden increase in the slope change rate and a significant increase in the precipitation intensity change rate of a certain grid. At this time, the grid is marked as a high-risk area. Through further data analysis, such as using time series analysis to predict future trends, the data is comprehensively considered to update the slope-gradient coupling difference map. This map can intuitively show the relationship between topography and precipitation changes, providing a basis for subsequent disaster early warning. This process not only improves the practicality of the data but also increases the accuracy of predictions.

[0025] S212: Based on the slope coupling difference map, the humidity sequence layer is superimposed to perform periodic fluctuation frequency statistics on the unit humidity value, and combined with the coupling difference value merging analysis, a humidity disturbance frequency map is generated. By overlaying a humidity sequence layer and statistically analyzing the frequency of humidity fluctuations within a period, the process first utilizes data collected by humidity sensors located in different geographical locations. Each sensor records data at set time intervals (e.g., hourly). Through statistical analysis of the data, the frequency of humidity changes in each grid cell within a specific time window is calculated, revealing patterns and trends in humidity fluctuations. For example, in a study conducted in a coastal area, analysis of data from several consecutive days revealed abnormally high frequency of humidity fluctuations at night in certain areas, which is related to topography and local climate conditions. This analysis helps identify areas requiring focused monitoring. By combining the frequency data with the aforementioned coupling difference values, it is possible to more accurately predict which areas are prone to geological disasters due to the combined effects of humidity and precipitation. Ultimately, a humidity disturbance frequency map is generated, showing the frequency of humidity fluctuations in different regions, providing important input for geological disaster early warning systems.

[0026] S213: Based on the humidity disturbance frequency map, select grid cells with frequency values ​​greater than the humidity fluctuation frequency threshold. Combine this with the corresponding precipitation variation amplitude, cumulative slope variation, and disturbance duration, using the following formula:

[0027] Calculate the composite perturbation index value of the high-sensitivity cells, identify the high-sensitivity cells with concentrated fluctuations, and generate a list of high-sensitivity sliding zone grids in the cloud. in, This represents the composite perturbation index value of the highly sensitive unit. Let represent the variation in precipitation intensity for the i-th grid cell. Let be the humidity disturbance value of the i-th grid cell. Let be the cumulative slope change value of the i-th grid cell. Let be the humidity fluctuation frequency of the i-th grid cell, and n be the total number of grid cells that meet the screening criteria. Grid cells with fluctuation frequencies exceeding the humidity fluctuation frequency threshold were selected. Specifically, the humidity fluctuation frequency threshold was set based on the raw statistical distribution of humidity changes in the study area over the past 5 years. Anomaly frequency thresholds were determined using quantiles; here, it was set to >8 fluctuations per day, meaning an hourly humidity change exceeding ±3% was considered a valid disturbance, and cells accumulating more than 8 fluctuations were deemed abnormal. Subsequently, precipitation intensity change data for the past 72 hours were extracted from the corresponding cells, and the hourly increase was calculated using time-series differences. The maximum increase was then taken as the precipitation change amplitude value for that cell. Simultaneously, the cumulative increase in slope layer values ​​within this period is calculated, that is, the sum of the absolute values ​​of the differences between the hourly slope grid data and the previous hour, which is used as the cumulative slope change value. For humidity disturbance values The peak-to-valley humidity difference over 24 hours was used as the evaluation index. The parameter dimensions were standardized through normalization. (Unit: mm / h) (unit:%), (Unit: degrees) Normalize to [0, 1] respectively, as follows: ; Where X is the original value. These are normalized values. The maximum and minimum values ​​are referenced from the original maximum and minimum monitoring data within the region, as detailed below: The value range is [0, 60] mm / h. The value range is [0, 45]%. The value range is [0, 20] degrees, and the fluctuation frequency is... The unit is "times / day", and the maximum value is set to 20. Substitute the measured parameters of the following example mesh cells into the formula for calculation. Table 1 shows the original data for each parameter: Table 1 Examples of disturbance parameter values ​​for different grid cells

[0028] As shown in Table 1, the three grid cells that meet the frequency threshold condition are selected above, and normalized before being substituted into the formula: A1: ; ; A2: ; A3: ; Substitute the above normalized parameters into the formula: ; The calculations are as follows: ; ; ; The denominator is ; Final calculation formula value: ; The results show that the composite disturbance index value in the analyzed area is 0.0214. Comparing this value with the regional empirical benchmark value of 0.018 confirms that it is higher than the benchmark, indicating that the unit has a high degree of disturbance aggregation and should be included in the cloud-based high-sensitivity sliding zone grid list as a key early warning target. By introducing the square root of the sum of squares of slope change and humidity disturbance, the ability to express the nonlinear change amplitude under the action of heavy precipitation is enhanced. The composite disturbance index can integrate the changing trends of multi-source data and realize cross-dimensional feature response integration, thereby improving the spatial resolution and parameter sensitivity of geological sliding potential identification.

[0029] Please see Figure 4 The specific steps for obtaining the boundary values ​​of the aggregated risk zone are as follows: S311: Based on the high-frequency fluctuating grid list, extract the boundary overlap value and corresponding coordinates of the grid and its adjacent grids, determine the geometric connection status between grids, and obtain the grid adjacency connection value; Based on the grid list, the boundary overlap and relative position coordinate set between each grid and its adjacent grids are extracted in detail to determine the contact relationship between grids. The core of this process is to accurately describe the spatial layout interaction of each grid. In the embodiment, it is assumed that a monitoring area is divided into several grids, each grid representing a monitoring point. The data between monitoring points are used to calculate the spatial contact frequency and position deviation, thereby assessing the monitoring coverage and existing monitoring blind spots in the area. The calculation process requires comparing and analyzing the boundary data and position coordinates of each grid to determine the geometric connection status of each grid and converting the data into numerical data that can be used for further analysis. For example, the boundary overlap of two grids is evaluated by the length of the boundary line and the number of intersections, and the position coordinate set is determined by calculating the relative position of the grid center point to obtain the grid adjacency connectivity value, which is used for the next step of spatial density assessment.

[0030] S312: Calculates spatial density by calling the grid adjacency connectivity value, identifying the Euclidean distance between grid centers, the number of overlapping grids per unit area, and the adjacency connectivity deviation, and determining the spatial clustering situation using the following formula: ; Calculate the spatial clustering intensity ratio, set a connectivity threshold, and filter grid groups with clustering intensity exceeding the threshold to obtain the grid connectivity filter set; in, Indicates the ratio of spatial aggregation intensity. The position difference along the X-axis of the k-th grid. The position difference along the Y-axis of the k-th grid. Let Euclidean distance be the center point of the k-th grid. Let k be the adjacency connectivity of the k-th grid. The mean connectivity of the k-th grid. Let be the number of overlaps in the k-th grid. Let Euclidean mean distance be the neighborhood of the k-th grid. Let k be the spatial density of the k-th grid. The density mean Total number of grid cells; To establish a spatial clustering assessment mechanism, it is necessary to first comprehensively quantify the structural relationships between each grid and its neighboring grids, and then collect the coordinate data of the center point of each grid unit through the geographic data interface integrated into the cloud platform. Calculate the Euclidean distance between the k-th grid and its neighboring grids by combining the coordinate data of the adjacent grids. Using formula , The coordinates are the center coordinates of the adjacent grid. In the example, if... , ,but ; Obtain the coordinate differences in the X and Y directions between this grid and its adjacent grids. Based on the data in the example above , serving as a key input for spatial location differences; Calculate the grid adjacency connectivity This represents the number of overlapping areas between meshes at the boundary, which can be obtained by counting collinear points on the boundary. Let's assume that mesh k in the detection area shares 8 edge segments with its surrounding meshes. =8, and the adjacency connectivity of the 10 grids in this region is calculated, and the numerical set is as follows. Calculate the average adjacency connectivity. =7.3, then, the number of overlapping grids per unit area was counted. This can be achieved by dividing the sub-region containing the grid into 1km² units and counting the number of overlapping grids within that area. Let's say the current area is... =4, further collect the center-to-center distance of several grids in the neighborhood, and calculate the average distance as... If the sampling distance value is ,but Furthermore, grid spatial density The effective number of particles within a grid cell is calculated by dividing its area. Let the area of ​​the grid be... If the number of valid points is 300, then The average value is obtained after collecting the spatial density of multiple grids. =0.027; To ensure consistency in the dimensions of parameters in the formula, the units of each parameter need to be standardized by using a standard normalization method. ,in The mean, Standard deviation, with For example, let's define the data collection area. The set is Calculated , ,but The values ​​of each participating item are normalized in this way and then transferred to the standard scale. Then substitute into the formula: The current R=0.651. If the connectivity threshold is set to 0.6, this value is in the over-threshold interval and is included in the grid connectivity screening set. By combining the spatial location difference, Euclidean distance, adjacency degree and density deviation, a multi-dimensional cross-judgment mechanism is formed to improve the sensitivity of geological disaster cluster identification. The result shows that the k-th grid has clustering characteristics and should be used as the input basis for subsequent boundary identification.

[0031] S313: Based on the mesh connectivity set, extract the edge mesh contour coordinates and sequences, identify the boundaries of closed connected regions, and output the aggregated risk zone boundary values; In the process of extracting the contour coordinate point set of the edge grid and the sequence number of the adjacent grid, a boundary identification framework is constructed using the circumscribed boundary determination method to determine the edge position of the closed connected region. The key point of this process is how to accurately identify and delineate the boundary of the risk area. In the embodiment, it is assumed that an urban planning department uses technology to identify high-risk flood areas in the city. By collecting the geographical and environmental data of each grid and calculating the connectivity of each grid, it determines which grid combinations form potential flood risk areas. Then, the specific boundary coordinate points of the area are extracted for the specific implementation of urban planning and disaster prevention and mitigation work. Through the data, the planning department can more accurately plan the city's drainage system and flood control measures to ensure the safety and sustainable development of the city.

[0032] Please see Figure 5 The specific steps for obtaining the set of reconstructed factors for geological disaster areas are as follows: S411: Call the boundary value of the aggregated risk area, extract the wind speed sequence and soil moisture sequence at the corresponding location, identify the ratio sequence of wind speed and humidity, calculate the ratio difference between adjacent time points, determine whether the difference exceeds the wind and dampness fluctuation threshold, screen out the time period exceeding the limit, and establish a wind and dampness ratio constraint matrix. Wind speed and soil moisture sequences at corresponding locations are extracted to provide foundational data for geological hazard analysis on a cloud platform. In practical applications, such as in a specific mountainous area, data collected by remote sensors is received via a cloud computing platform, transmitting wind speed and soil moisture data from various monitoring points. After refinement, software is configured to perform a preliminary data quality assessment immediately upon receiving the data, eliminating abnormal data introduced by equipment malfunctions or other issues. For example, if the data from a monitoring station suddenly differs significantly from that of neighboring stations, this data is marked as suspicious and excluded from subsequent analysis. Based on the aggregated risk zone boundary information in a Geographic Information System (GIS), valid data points within the corresponding boundaries are filtered to ensure spatial consistency of the analyzed data, and a wind-moisture ratio constraint matrix is ​​obtained.

[0033] S412: Based on the wind-humidity ratio constraint matrix, retrieve the slope data within the region using the following formula: ; Calculate the wind-damp slope offset coupling degree, compare the high coupling points in the filter area with the set offset threshold, extract the spatial intersection point set and locate its data binding coordinates under the cloud platform, and output the set of geological disaster area reconstruction factors. Where B represents the aeolian slope offset coupling degree, and v represents the wind speed to humidity ratio. The value represents the average ratio of wind speed to humidity, and h represents the slope value. L represents the average slope, L represents the variation in the ratio of wind speed to humidity, and N represents the variation in the slope value. The set of ratio differences between adjacent time points is calculated to determine whether the difference exceeds the wind-humidity fluctuation threshold, filtering out time periods exceeding the limit, and providing accurate input for the next step of cloud base geological disaster analysis. For example, in a preset flood warning area, the monitoring station records data every 10 minutes, and the calculated wind speed to humidity ratio (v) is the ratio of real-time wind speed (e.g., 5 m / s) to soil moisture percentage (e.g., 30%). Assuming that the wind speed is 5 m / s and the soil moisture is 30% at 10:00 AM, and the wind speed is 6 m / s and the soil moisture is 35% at 10:00 AM, then the wind speed to humidity ratios at these two time points are 0.1667 and 0.1714, respectively; based on this, the difference between these two ratios can be calculated. The data is then compared to a predefined rheumatism fluctuation threshold (assumed to be 0.005). Since the difference does not exceed the threshold, the data for that time period is retained. The rheumatism ratio is then normalized to eliminate the influence of dimensions and improve the model's generality. For example, normalization can be achieved by dividing the rheumatism ratio by the maximum rheumatism ratio at that monitoring station (assumed to be 0.2), resulting in a normalized rheumatism ratio of... ; In addition, the normalized value of the slope (h) needs to be calculated. Assuming the slope at this location is 15 degrees and the maximum slope in the area is 30 degrees, the normalized slope value is h = 15 / 30 = 0.5. Now, the wind-damp slope offset coupling degree B can be calculated using the previously defined formula; Where v = 0.875 and h = 0.5, assume and The values ​​represent the raw mean, and L=0.05 and N=0.1 represent the standard deviation. Substitute into the formula to calculate: ; The calculation results show that the current wind-humidity deviation coupling strength is low, indicating that the changes in wind speed and humidity at this point in time have little impact on the slope. This data can be used in geological hazard analysis models to further monitor and predict potential geological hazard risks.

[0034] Please see Figure 6 The specific steps for obtaining the geological disaster risk trend classification indicator group are as follows: S511: Based on the temporal variation trend of the set of reconstructed factors in geological disaster areas, identify factor numerical snapshots at continuous time nodes, extract topographic snapshots corresponding to the cumulative precipitation time, analyze the coupling strength between cumulative precipitation and the spatial distribution of topographic factor groups, and obtain the coupling strength trend sequence. First, time-series data of the reconstructed factor set for the geological disaster area is acquired and analyzed to identify factor value snapshots at consecutive time points. To achieve this goal, geological disaster-related models and raw data are used to ensure accurate recording of factor values ​​at each time point, such as temperature, precipitation, soil moisture, and geological factors. Topographic snapshots corresponding to the cumulative precipitation time need to be extracted. This process can be achieved by monitoring real-time precipitation data and matching it with topographic data to obtain accurate topographic snapshots. For example, if the cumulative precipitation exceeds a certain threshold (e.g., 50 mm), the corresponding time point can be extracted along with the topographic data to analyze the coupling strength between cumulative precipitation and the spatial distribution of topographic factor groups. The analysis steps include calculating the spatial correlation between precipitation and different topographic factors (e.g., slope, soil type), and using correlation coefficients or regression analysis to assess the coupling strength. This process can be implemented using algorithmic models such as regression analysis to calculate the correlation coefficient between each pair of factors, further revealing the trend of coupling strength changes. These changes at consecutive time points provide data support for subsequent analysis, ultimately yielding a coupling strength trend sequence through analysis.

[0035] S512: Based on the coupling strength trend sequence, extract the trend change difference between adjacent time periods, analyze the consistency with the direction of fluctuation of raster cell values ​​in the terrain snapshot, identify areas with consistent spatial change directions, mark continuous consistent segments, and calculate their proportion in the whole area to obtain the spatial consistency segment distribution coefficient. The trend difference between adjacent time periods is extracted from the coupling strength trend sequence. The change is calculated by comparing the coupling strength values ​​of adjacent time periods. For example, if the coupling strength value is 0.65 in one time period and 0.75 in the next, the difference is 0.10. The consistency of this difference with the direction of fluctuation in raster cell values ​​in the terrain snapshot is analyzed. This analysis requires comparing the trend of coupling strength changes between adjacent time periods with the direction of change in raster cell values ​​(such as terrain slope, geological structure, etc.) to determine if consistency exists. For instance, if the slope changes from 5 degrees to 7 degrees and the coupling strength changes from 0.6 to 0.7 in a certain time period, if the directions of change are consistent, it indicates that the change direction in the region is consistent. The next step is to identify areas with consistent spatial change directions, which can be achieved through spatial analysis methods such as spatial clustering. The clustered areas represent terrain and coupling strength changes with consistent directions within that region. Each region is labeled as a "continuous consistent segment," and spatial consistency is assessed by calculating the proportion of consistent segments in the entire region. By comparing the distribution of consistent segments, the spatial consistency segment distribution coefficient is calculated and output as a quantitative indicator of the geological hazard risk in the region.

[0036] S513: Call the spatial consistency segment distribution coefficient, extract the continuous direction consistent area segments, measure the response amplitude difference in the positive and negative trends, divide the disaster risk level intervals, and output the geological disaster risk trend classification label group. The previously obtained spatial consistency segment distribution coefficient is used to identify areas with consistent continuous directions. The selection of these areas is based on spatial consistency analysis, ensuring that the changing trends of topographic factors and coupling strength remain consistent within the selected areas. The response amplitude difference is measured, analyzing the changes in coupling strength within the area under positive and negative trends. For example, assuming a positive trend where the coupling strength of a certain area increases from 0.7 to 0.85, and a negative trend where it decreases from 0.75 to 0.6, the response amplitude difference is calculated as 0.85-0.7 (positive) and 0.75-0.6 (negative). The difference between these two amplitude values ​​can serve as an important parameter for disaster risk assessment. Based on the response amplitude difference, disaster risk level intervals are defined. For instance, based on the magnitude of the amplitude difference, three level intervals are defined: an amplitude difference greater than 0.15 indicates a high-risk area, 0.1-0.15 indicates a medium-risk area, and less than 0.1 indicates a low-risk area. By outputting the calculation results, a geological disaster risk trend classification identifier group is generated, which reflects the disaster risk level of different regions at different time periods, providing decision support for disaster prevention and mitigation work.

[0037] The cloud-based geological hazard analysis system is used to execute the aforementioned cloud-based geological hazard analysis method. The system includes: The mapping relationship analysis module acquires data sources of precipitation intensity, slope, wind speed, and soil moisture within the monitored geological area, extracts the timestamp and latitude and longitude of each item, identifies cross-category spatiotemporal combination codes, analyzes the mapping relationship between disaster factors and monitoring grids, and generates a spatial index group of geological factors. The disturbance identification module is based on the spatial index group of geological factors. It extracts the difference in change of the encoded sequences of slope and precipitation intensity within the same time window, calculates the fluctuation frequency by superimposing the humidity sequence, filters the encoded grids with frequencies exceeding the set value, and outputs a set of grids with high disturbance response. The aggregation identification module is based on a high-perturbation response grid set. It analyzes the adjacency relationship and dense distribution between grids, marks continuous connected units and extracts the boundary range, identifies the spatial outline of potential aggregated disaster areas, and generates risk cluster boundary box groups. The factor reconstruction module is based on the risk cluster area boundary box group, extracts the wind speed and soil moisture change rate in the area, compares the deviation with the slope change value, removes combinations with excessive deviation, retains the coupled combinations in the high linkage area, and establishes a set of geological disaster area reconstruction factors. The trend classification module is based on the sequence variation trend of the geological disaster area reconstruction factor set, matches the regional cumulative precipitation map and slope snapshot map, extracts the consistent variation segments, performs directional analysis on the wind speed and humidity sequence fluctuations, classifies the response level, and outputs the geological disaster risk trend classification label group.

[0038] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A cloud-based geological hazard analysis method, characterized in that, Includes the following steps: S1: Acquire data sources of precipitation intensity, slope, wind speed, and soil moisture within the monitored geological area, collect the real-time distribution of the data in the cloud platform, perform spatiotemporal indexing and encoding on each type of data based on timestamps and latitude and longitude, identify the mapping structure between disaster factors and grid areas, and generate disaster factor grid coding groups. S2: Call the slope and precipitation intensity layers in the disaster factor grid coding group, select continuously changing units within the same time window, perform gradient difference quantization, overlay the humidity sequence, analyze the frequency distribution of fluctuation values, filter units whose fluctuation frequency exceeds the set threshold, and output a list of high-frequency fluctuation grids. S3: Based on the high-frequency fluctuation grid list, perform connectivity analysis on adjacency and spatial density, select grid clusters that meet the connectivity threshold, identify the corresponding spatial boundary markers, and output the aggregated risk zone boundary values; S4: Call the defined area of ​​the aggregated risk zone boundary value, extract the wind speed and soil moisture sequences, calculate the corresponding change ratio, and compare the weighted deviation with the slope data. Screen out factors with excessive deviation, retain factor combinations with high cross-coupling degree, and output the set of geological disaster area reconstruction factors.

2. The cloud-based geological disaster analysis method according to claim 1, characterized in that, The disaster factor grid coding group includes spatial positioning number, time synchronization label, and disaster factor type identifier; the high-frequency fluctuation grid list includes abnormal intensity value, fluctuation period parameter, and frequency threshold label; the aggregated risk zone boundary value includes boundary coordinate range, aggregated grid number, and spatial cluster identifier; and the geological disaster area reconstruction factor set includes factor combination weight, coupling strength coefficient, and change ratio index.

3. The cloud-based geological disaster analysis method according to claim 1, characterized in that, The specific steps for obtaining the disaster factor grid coding group are as follows: S111: Obtain data sources of precipitation intensity, slope, wind speed, and soil moisture within the monitored geological area. Based on the latitude and longitude fields and timestamp fields of each record, unify the data time and spatial resolution to generate raw data tables for four types of environmental factors. S112: Based on the original data tables of the four types of environmental factors, extract the time series under the same latitude and longitude in each type of data, use a sliding window to segment by time, map the start and end time of the series to the coordinates as a unique code, and generate a single-type factor grid mapping code value group. S113: Based on the single-type factor grid mapping code value group, the grid area is delineated according to the latitude and longitude fields, the factor value combinations in the same area are merged, the spatial coverage ratio of factor combinations under different time segments is compared, the coupling relationship features of regional factor combinations are extracted, and disaster factor grid coding group is generated.

4. The cloud-based geological disaster analysis method according to claim 3, characterized in that, The specific steps for obtaining the high-frequency fluctuation grid list are as follows: S211: Call the disaster factor grid coding group, extract the slope layer and precipitation intensity layer, select continuously changing grids within the same time window, identify the difference between the unit slope change rate and the precipitation intensity change rate, identify the temporal gradient difference distribution, and generate a slope coupling difference map. S212: Based on the slope coupling difference map, a humidity sequence layer is superimposed, and the frequency of fluctuations in the unit humidity value within the period is statistically analyzed. Combined with the coupling difference value merging analysis, a humidity disturbance frequency map is generated. S213: Based on the humidity disturbance frequency map, filter grid cells with frequency values ​​greater than the humidity fluctuation frequency threshold, combine the corresponding precipitation change amplitude, slope change accumulation and disturbance duration, calculate the composite disturbance index value of the high-sensitivity cells, identify the high-sensitivity cells with concentrated fluctuations, and generate a list of cloud-based high-sensitivity sliding zone grids.

5. The cloud-based geological disaster analysis method according to claim 4, characterized in that, The specific steps for obtaining the boundary value of the aggregated risk zone are as follows: S311: Based on the high-frequency fluctuation grid list, extract the boundary overlap value and corresponding coordinates of the grid and adjacent grids, determine the geometric connection status between grids, and obtain the grid adjacency connection value; S312: Call the grid adjacency connectivity value, calculate the spatial density, identify the Euclidean distance between grid centers, the number of overlapping grids per unit area and the adjacency connectivity deviation, determine the spatial clustering situation, calculate the spatial clustering intensity ratio, set the connectivity threshold, filter grid groups with clustering exceeding the threshold, and obtain the grid connectivity filter set. S313: Based on the mesh connectivity filter set, extract the edge mesh contour coordinates and sequence, identify the boundaries of closed connected regions, and output the boundary values ​​of aggregated risk areas.

6. The cloud-based geological disaster analysis method according to claim 5, characterized in that, The specific steps for obtaining the set of geological disaster area reconstruction factors are as follows: S411: Call the aggregated risk zone boundary value, extract the wind speed sequence and soil moisture sequence at the corresponding location, identify the ratio sequence of wind speed and humidity, calculate the ratio difference between adjacent time points, determine whether the difference exceeds the rheumatism fluctuation threshold, screen out the time period exceeding the limit, and establish a rheumatism ratio constraint matrix. S412: Based on the wind-humidity ratio constraint matrix, call the slope data in the region, calculate the wind-humidity slope offset coupling degree, compare with the set offset threshold to filter high coupling points in the region, extract the spatial intersection point set and locate its data binding coordinates under the cloud platform, and output the set of geological disaster area reconstruction factors.

7. The cloud-based geological disaster analysis method according to claim 1, characterized in that, The method also includes step S5: S5: Based on the temporal change trend of the set of geological disaster area reconstruction factors, extract the cumulative precipitation and topographic snapshot map group in the area as the matching benchmark, determine the consistency segment between the trend and the original snapshot, identify the response gradient according to the trend change direction, and output the geological disaster risk trend classification label group. The geological disaster risk trend classification and identification group includes trend level code, response direction parameter, and consistency verification label.

8. The cloud-based geological disaster analysis method according to claim 7, characterized in that, The specific steps for obtaining the geological disaster risk trend classification identifier group are as follows: S511: Based on the temporal variation trend of the set of reconstructed factors in the geological disaster area, identify the factor numerical snapshots at consecutive time nodes, extract the topographic snapshots corresponding to the cumulative precipitation time, analyze the coupling strength between the cumulative precipitation and the spatial distribution of the topographic factor group, and obtain the coupling strength trend sequence. S512: Based on the coupling strength trend sequence, extract the trend change difference between adjacent time periods, analyze the consistency with the direction of fluctuation of grid cell values ​​in the terrain snapshot, identify areas with consistent spatial change directions, mark continuous consistent segments, and calculate their proportion in the whole area to obtain the spatial consistency segment distribution coefficient. S513: Call the spatial consistency segment distribution coefficient, extract the continuous direction consistent area segments, measure the response amplitude difference in the positive and negative trends, divide the disaster risk level intervals, and output the geological disaster risk trend classification label group.

9. A cloud-based geological disaster analysis system, characterized in that, The system is used to implement the cloud-based geological hazard analysis method according to any one of claims 1-8, and the system comprises: The mapping relationship analysis module acquires data sources of precipitation intensity, slope, wind speed, and soil moisture within the monitored geological area, extracts the timestamp and latitude and longitude of each item, identifies cross-category spatiotemporal combination codes, analyzes the mapping relationship between disaster factors and monitoring grids, and generates a spatial index group of geological factors. Based on the geological factor spatial index group, the disturbance identification module extracts the difference in change of the coded sequences of slope and precipitation intensity within the same time window, calculates the fluctuation frequency by superimposing the humidity sequence, filters the coded grids with frequencies exceeding the set value, and outputs a set of high disturbance response grids. Based on the high-disturbance response grid set, the aggregation identification module analyzes the adjacency relationship and dense distribution between grids, marks continuous connected units and extracts boundary ranges, identifies the spatial outline of potential aggregated disaster areas, and generates risk cluster area boundary box groups. The factor reconstruction module extracts the rate of change of wind speed and soil moisture within the risk cluster area based on the boundary box group, compares the deviation with the slope change value, eliminates combinations with excessive deviation, retains coupled combinations in the high linkage area, and establishes a set of geological disaster area reconstruction factors. The trend classification module, based on the sequence variation trend of the set of geological disaster area reconstruction factors, matches the regional cumulative precipitation map and slope snapshot map, extracts the consistent variation segments, performs directional analysis on the fluctuation of wind speed and humidity sequences, classifies the response level, and outputs the geological disaster risk trend classification identifier group.

Images