Geological disaster forecasting methods, systems, equipment, and media based on big data analysis

Abstract

This application discloses a geological disaster forecasting method, system, equipment, and medium based on big data analysis, relating to the field of geological disasters. The method includes: acquiring geological monitoring data of a target geological area; preprocessing the acquired geological monitoring data to obtain corrected geological monitoring data; performing at least one pre-simulation operation until the pre-simulation indicates that a target geological disaster will occur in the target geological area; wherein, the pre-simulation operation includes: inputting the corrected geological monitoring data into a constructed disaster pre-simulation model, and using the disaster pre-simulation model to predict whether a target geological disaster will occur in the target geological area in a future target period based on the corrected geological monitoring data; if the pre-simulation indicates that a target geological disaster will occur in the target geological area in a future target period, issuing an early warning information. This application can provide timely early warning of possible target geological disasters, enabling target personnel to promptly carry out management and prevention of possible target geological disasters.

Description

Technical Field

[0001] This application relates to the field of geological disaster technology, and in particular to a geological disaster forecasting method, system, equipment and medium based on big data analysis. Background Technology

[0002] Geological disasters refer to catastrophic events related to geological processes caused by natural factors or human activities, such as earthquakes, landslides, mudslides, and ground subsidence. These disasters often pose a serious threat to human life and property and have a destructive impact on the environment.

[0003] To mitigate the losses caused by geological disasters, geological disaster monitoring and early warning technologies have emerged. Traditional geological disaster monitoring and early warning technologies mostly rely on analyzing data collected by monitoring sensors to provide early warnings. However, this approach only detects potential geological disasters when the collected data reaches critical thresholds. Consequently, it cannot promptly send early warnings to residents in the vicinity of potential disaster areas, or to relevant departments involved in disaster prevention and mitigation, via mobile telecommunications services (such as SMS). This hinders timely prevention and mitigation of geological disasters, leading to significant losses for surrounding residents after a disaster occurs. Summary of the Invention

[0004] The purpose of this application is to provide a geological disaster forecasting method, system, and equipment based on big data analysis, in order to solve the problem that traditional geological disaster monitoring and early warning technologies cannot provide timely early warning of geological disasters, and thus cannot prevent and control geological disasters in a timely manner.

[0005] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a geological disaster forecasting method based on big data analysis, including: Obtain geological monitoring data for the target geological area; The acquired geological monitoring data is preprocessed to obtain corrected geological monitoring data; At least one pre-simulation operation shall be performed until the pre-simulation indicates that a target geological hazard will occur in the target geological area; wherein, the pre-simulation operation includes: The corrected geological monitoring data is input into the constructed disaster prediction model, and the disaster prediction model predicts whether the target geological area will experience the target geological disaster in the future target period based on the corrected geological monitoring data. If a geological disaster is predicted to occur in the target geological area during a future target time period, an early warning message will be issued.

[0006] Secondly, this application provides a geological disaster forecasting system based on big data analysis, including: The detection unit is used to acquire geological monitoring data of the target geological area; Processing unit, used for: The real-time acquired geological monitoring data is preprocessed to obtain corrected geological monitoring data; At least one pre-simulation operation shall be performed until the pre-simulation indicates that a target geological hazard will occur in the target geological area; wherein, the pre-simulation operation includes: The corrected geological monitoring data is input into the constructed disaster prediction model, and the disaster prediction model predicts whether the target geological area will experience the target geological disaster in the future target period based on the corrected geological monitoring data. The notification unit is used to issue early warning information when a geological disaster is predicted to occur in the target geological area during a future target period.

[0007] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the geological disaster forecasting method based on big data analysis as described above.

[0008] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the geological disaster prediction method based on big data analysis described above.

[0009] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a geological disaster forecasting method, system, equipment, and medium based on big data analysis. It achieves geological monitoring of a target geological area by acquiring geological monitoring data; preprocesses the acquired geological monitoring data to obtain corrected geological monitoring data, and performs at least one pre-simulation operation until the pre-simulation indicates that a target geological disaster will occur in the target geological area; the pre-simulation operation includes: inputting the corrected geological monitoring data into a constructed disaster pre-simulation model, and using the disaster pre-simulation model to predict whether a target geological disaster will occur in the target geological area in a future target time period based on the corrected geological monitoring data, thus enabling early detection of potential target geological disasters based on geological monitoring data; and issuing an early warning if the pre-simulation indicates that a target geological disaster will occur in the target geological area in a future target time period, enabling timely management and prevention of potential target geological disasters. In summary, the embodiments of this application solve the problem that traditional geological disaster monitoring and early warning technologies cannot provide timely early warnings of geological disasters, preventing surrounding residents from taking timely preventative measures. Furthermore, by preprocessing the geological monitoring data, the accuracy of the geological monitoring data is improved, thereby improving the accuracy of the predicted target geological disasters. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A flowchart illustrating a geological disaster prediction method based on big data analysis, provided as an embodiment of this application; Figure 2 A schematic diagram of the structure of a geological disaster prediction system based on big data analysis provided in an embodiment of this application; Figure 3 for Figure 2 A detailed functional module diagram of the processing unit in a geological disaster forecasting system based on big data analysis. Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] In one exemplary embodiment, such as Figure 1 As shown, a geological disaster forecasting method based on big data analysis is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is described using a server as an example, including the following steps 201 to 204. Wherein: Step 201: Obtain geological monitoring data for the target geological area.

[0015] In this embodiment, the target geological area is the geological area requiring geological disaster forecasting (early warning). The geological monitoring data includes geological factor data, meteorological factor data, and human engineering activity data. The geological factor data includes real-time monitoring data and the physical and mechanical parameters of the soil and rock mass used to construct the model (such as shear strength, elastic modulus, permeability coefficient, and other core mechanical parameters). The real-time monitoring data includes displacement, water level, and seismic data; the seismic data includes seismic wave type (S&P or P-wave) and seismic intensity. The physical and mechanical parameters of the soil and rock mass are obtained through geological exploration and testing.

[0016] Meteorological data includes measured weather data and weather forecasts for future periods. Measured weather data includes measured wind direction, wind speed, air pressure, rainfall, sunshine duration, and solar radiation intensity. Human engineering activity data refers to monitoring videos of human engineering activities. Human engineering activities refer to various construction and development projects undertaken by humans, including but not limited to building engineering (such as the construction of residential, commercial buildings and public facilities (such as schools and hospitals)), transportation engineering (such as the construction of transportation infrastructure such as roads, bridges, tunnels, railways, airports, and ports), water conservancy engineering (such as the construction of dams, irrigation systems, and flood control dikes), energy engineering projects (including power plants (such as coal-fired, nuclear, hydropower, wind power, and solar power), transmission lines, and the laying of oil and gas extraction and transportation pipelines), mining engineering (the extraction of underground mineral resources), and environmental engineering (such as wastewater treatment, solid waste management, and soil remediation).

[0017] In this embodiment, geological disaster forecasting analysis is performed not only through geological factor data, but also through meteorological factor data and human engineering activity factor data, which improves the completeness of geological disaster forecasting, reduces the probability of forecast errors, and improves the accuracy of geological disaster forecasting.

[0018] Step 202: Preprocess the acquired geological monitoring data to obtain corrected geological monitoring data.

[0019] Step 203: Perform a pre-simulation operation at least once until the pre-simulation indicates that the target geological area will experience a target geological hazard in the future within a target time period; the pre-simulation operation includes: The revised geological monitoring data is input into the constructed disaster prediction model, which then predicts whether the target geological disaster will occur in the target geological area in the future target period based on the revised geological monitoring data.

[0020] In this embodiment of the application, the target geological disaster refers to a geological disaster that may occur in the target monitoring area.

[0021] Step 204: If the simulation indicates that a target geological disaster will occur in the target geological area in the future target time period, issue an early warning message.

[0022] In this embodiment, the early warning information includes the target geological disasters that will occur in the future within a target time period. Early warning information can be sent through multiple channels to residents located within the target geological area, as well as relevant departments responsible for geological disaster management and prevention. The specific target personnel are not limited here and can be set according to actual needs.

[0023] By implementing steps 201 to 204 above, geological monitoring of the target geological area is achieved by acquiring geological monitoring data of the target geological area; the acquired geological monitoring data is preprocessed to obtain corrected geological monitoring data, and at least one pre-simulation operation is performed until the pre-simulation indicates that a target geological disaster will occur in the target geological area; wherein, the pre-simulation operation includes: inputting the corrected geological monitoring data into a constructed disaster pre-simulation model, and using the disaster pre-simulation model to predict whether a target geological disaster will occur in the target geological area in a future target period based on the corrected geological monitoring data, thereby enabling early knowledge of possible target geological disasters based on geological monitoring data; if the pre-simulation indicates that a target geological disaster will occur in the target geological area in a future target period, an early warning information is issued, enabling timely management and prevention of possible target geological disasters; in summary, the embodiments of this application solve the problem that traditional geological disaster monitoring and early warning technologies cannot provide timely early warnings of geological disasters, preventing surrounding residents from taking timely preventive measures. In addition, by preprocessing the geological monitoring data, the accuracy of the geological monitoring data is improved, thereby improving the accuracy of the predicted target geological disasters.

[0024] In another exemplary embodiment of this application, the geological monitoring data includes geological factor data. In step 202 above, the geological factor data is preprocessed, including steps a1 to a3, wherein: Step a1 involves cleaning the real-time monitoring data of geological factors to obtain corrected monitoring data.

[0025] In this embodiment, data cleaning includes noise removal, missing value filling, and correction of contradictory information. Simultaneously, to ensure the accuracy and convergence of subsequent machine learning or numerical models, geological factor data with significant differences in dimensions and numerical ranges must be normalized or standardized.

[0026] Outliers can be detected and removed using wavelet transform or isolated forest algorithms to achieve noise removal.

[0027] Missing values ​​are automatically identified and corrected by examining the continuity of the data stream, timestamp sequence, and sensor status. Specifically, Kriging interpolation or Generative Adversarial Networks (GANs) can be used to generate reasonable data to fill in the missing values.

[0028] The system automatically identifies and corrects contradictory information by examining data stream continuity, timestamp sequences, sensor status, physical laws, statistical laws, spatial correlation rules, and expert knowledge-based judgment rules.

[0029] Conflicting information includes contradictions in physical extreme values ​​(such as water content > 1), contradictions in spatial correlation (such as drastic opposite displacement directions of adjacent points), contradictions in multi-source data (such as conflict between GPS and InSAR results), and contradictions in temporal logic (such as drastic changes in water level without replenishment).

[0030] Example of expert knowledge setting: Fault zone strength rule: The strength parameters of the rock mass in the fault zone should be lower than those of the intact surrounding rock; Hydrological unit rule: The water level of the same aquifer should have a continuous hydraulic gradient; Displacement compatibility rule: The displacement vector of the landslide body should generally point towards the free face.

[0031] Step a2: Under a unified spatiotemporal reference, the corrected monitoring data is fused and mapped onto the corresponding grid of the pre-constructed high-precision basic geographic information model (including the initial field of terrain and geological attributes) to obtain the gridded input field, and invalid values ​​in the gridded input field data are detected and removed.

[0032] In this embodiment, invalid values ​​include null values, values ​​outside the physical range (such as negative elevation), and logically incorrect values ​​(such as river area elevation being higher than that of both banks). Invalid values ​​can be detected through range checks, consistency checks, topology rule checks, and visual checks.

[0033] Step a3 involves dimensionality reduction of the gridded input field data to extract combinations of disaster-sensitive factors.

[0034] In this embodiment, step a3 automatically learns the inherent structure of geological factor data, filters out features strongly correlated with disasters (i.e., combinations of disaster-sensitive factors), eliminates redundant parameters, focuses on disaster-sensitive factors, simplifies computational complexity, and improves computational efficiency. Data dimensionality reduction can be performed using autoencoder neural networks or principal component analysis (PCA) to map high-dimensional data to a low-dimensional space (e.g., 10-20 principal components), retaining features strongly correlated with disasters while reducing data volume.

[0035] For example, suppose there are 10 monitoring points, each with 6 indicators, forming 60-dimensional data.

[0036] PCA extracts the first two principal components with a cumulative contribution rate of 85%, such as PC1 (overall slip trend) and PC2 (local deformation pattern), reducing the 60-dimensional data to 2-dimensional feature factors, reducing the amount of computation and focusing on the main risk patterns.

[0037] In another exemplary embodiment of this application, the geological monitoring data includes meteorological factor data. In step 202 above, the meteorological factor data is preprocessed, including steps b1 to b2, wherein: Step b1: Fill in the missing values ​​in the measured weather data and correct the outliers in the measured weather data to obtain the initially corrected measured weather data.

[0038] In this embodiment of the application, to correct outliers in measured weather data, the outliers in the measured weather data can be deleted first, and then the missing values ​​and removed outliers in the measured weather data can be filled by interpolation or mean filling.

[0039] Step b2 involves standardizing and normalizing the preliminarily corrected measured weather data to obtain the final corrected measured weather data.

[0040] In this embodiment of the application, the purpose of standardizing and normalizing the preliminarily corrected measured weather data is to eliminate dimensional differences.

[0041] In another exemplary embodiment of this application, the weather forecast data includes multi-source weather forecast data, and step 202 above, which involves preprocessing the meteorological factor data, further includes: Step b3: Fuse the multi-source weather forecast data for the future target period to obtain the fused weather forecast data.

[0042] In this embodiment, the multi-source weather forecast data includes weather forecast data fed back from automatic weather stations, satellites, radar, etc. The multi-source weather forecast data can be fused using methods such as direct / weighted averaging, optimal interpolation, super-ensemble (machine learning), and Bayesian model averaging. Specifically, the direct / weighted averaging method assigns weights based on the historical errors of each forecast source; the optimal interpolation method considers the statistical characteristics of the errors of each source to perform spatially optimal weighted combination; the super-ensemble (machine learning) method trains a model (such as a neural network) to learn how to optimally combine information from each source; and the Bayesian model averaging performs probabilistic fusion based on historical performance-based probability weights.

[0043] By integrating multi-source weather forecast data from automatic weather stations, satellites, radar, and other sources, the accuracy of weather forecast data can be improved.

[0044] In another exemplary embodiment of this application, the geological monitoring data includes human engineering activity data. In step 202 above, the human engineering activity data is preprocessed, including steps c1 to c2, wherein: Step c1: Perform image preprocessing on the monitoring video of human engineering activities.

[0045] In this embodiment, the purpose of image preprocessing is to improve the clarity of each frame in the video data of human engineering activity monitoring. The specific process of image preprocessing is not limited and can be set according to actual needs. For example, image preprocessing may include resizing, grayscale conversion, and data enhancement.

[0046] In this application, the size adjustment is to adapt the image data obtained from image preprocessing to the input requirements of the three-dimensional geological model.

[0047] Step c2 involves extracting features from the preprocessed image data to obtain feature data of human engineering activities.

[0048] In this application embodiment, the characteristic data of human engineering activities generally include spatial distribution characteristics, dynamic intensity characteristics, and / or temporal variation characteristics. Spatial distribution characteristics include construction type, construction area range (such as excavation boundary coordinates, fill area geometry) and / or heavy machinery operation trajectory (such as vibration source location, movement path). Dynamic intensity characteristics include mechanical vibration parameters (such as pile driver impact force, blasting vibration wave energy) and / or earthwork excavation parameters (such as daily excavation volume, soil pile height). Temporal variation characteristics include activity intensity variation data within the construction cycle (such as day-night construction intensity differences) and temporary support structure status (such as anchor bolt stress, retaining wall displacement). For example, in tunnel excavation engineering, characteristic data may include "daily excavation depth 3 meters", "blasting vibration peak acceleration 0.5g", and "support lag time 2 hours", etc.

[0049] In another exemplary embodiment of this application, in step 203 above, the corrected geological monitoring data includes corrected geological factor data, corrected meteorological factor data, and / or corrected human engineering activity data. The corrected geological factor data includes corrected topographic data and a combination of disaster sensitivity factors. The corrected meteorological factor data includes the final corrected measured weather data and the fused weather forecast data. The corrected human engineering activity data includes human engineering activity characteristic data.

[0050] In another exemplary embodiment of this application, in step 203 above, the corrected geological monitoring data is input into the constructed disaster prediction model. The disaster prediction model predicts whether a target geological disaster will occur in the target geological area in a future target period based on the corrected geological monitoring data, including steps 301 to 303. Wherein: Step 301: Assign the corrected geological monitoring data to the three-dimensional mesh nodes (such as finite element mesh nodes) of the disaster prediction model.

[0051] Step 302: Based on the corrected geological monitoring data, the disaster simulation model simulates the stress distribution and deformation of the geological body in the target monitoring area under the influence of gravity, seepage, earthquake motion and / or human activities in the future target period (such as the decrease of slope safety factor over time), and simulates the dynamic impact of groundwater on rock mass stability (such as accelerated landslides after rainstorms), and outputs key parameters.

[0052] In this embodiment, the disaster prediction model, based on corrected geological monitoring data, predicts the stress distribution and deformation process of the geological body in the target monitoring area under gravity, seepage, seismic motion, and / or human activities in the future target period through the following steps, and simulates the dynamic influence of groundwater on rock mass stability, outputting key parameters: (1) Multiphysics coupling: simultaneous seepage equations ( ), stress balance equation ( ), simulating dynamic processes; in, Here, k is the permeability tensor of the soil mass, representing the ease with which fluid flows in the medium; h is the total head, which is the sum of the position head and pressure head, driving groundwater flow; S is the source-sink term, representing the amount of water flowing into or out of a unit volume of aquifer per unit time (such as rainfall infiltration, pumping, etc.); the seepage equation describes the flow law of groundwater in the pores of the soil mass (Darcy's law) and the conservation of mass, and is used to calculate the spatial distribution and time variation of pore water pressure.

[0053] σ ij Let F be the Cauchy stress tensor, representing the stress state at a point within the soil or rock mass in different directions, and j represent the divergence of the stress tensor; i The stress is the force per unit volume, usually referring to gravity; the stress balance equation describes the condition under which any point in the rock or soil is in static equilibrium (the form of Newton's second law under static conditions), and is used to calculate the stress field and deformation field of the rock or soil under the action of external forces and body forces.

[0054] Coupling relationship: The pore water pressure calculated by the seepage equation will change the effective stress of the soil and rock mass. , The pore water pressure affects the stress distribution in the stress balance equation. Conversely, stress changes can cause variations in the porosity and permeability coefficient k of the soil and rock, influencing the seepage field. Solving these two equations simultaneously can simulate the interaction between groundwater and soil deformation.

[0055] (2) Output the first key parameter directly related to disaster sensitivity factors (such as safety factor) Displacement rate v).

[0056] The key output parameters are directly related to disaster-sensitive factors (such as pore water pressure), emphasizing that changes in the key output parameters are directly driven by disaster-sensitive factors in the input data (such as an increase in pore water pressure), exhibiting a clear physical causal relationship. For example, an increase in pore water pressure leads to a decrease in effective stress, which in turn leads to a decrease in anti-slip force, ultimately affecting the safety factor. The decrease enhances the interpretability of the model.

[0057] Step 303: Check whether the key parameters output by the disaster simulation model exceed the preset threshold. If the key parameters output by the disaster simulation model exceed the preset threshold (such as displacement rate exceeding 10 mm / day, safety factor <1.0), it is considered that the target geological area will experience the target geological disaster in the future target time period; Otherwise, it is assumed that the target geological area will not experience the target geological disaster in the future target time period.

[0058] In another exemplary embodiment of this application, while outputting key parameters in step 302, stress cloud map, displacement vector and crack propagation path of the geological body in the target monitoring area in the future target time period are also output.

[0059] In this embodiment, the stress cloud map visually reveals the stress concentration area (potential failure initiation point), the displacement vector map clearly shows the deformation mode and main slip direction, and the crack propagation path predicts the location and morphology of the potential failure surface. The disaster prediction model (numerical simulation software) outputs results after numerical calculation based on input data and physical equations, and uses visualization technology to render and generate stress cloud maps, displacement vectors, and crack propagation paths of the geological body in the target monitoring area for a future target time period.

[0060] In another exemplary embodiment of this application, the disaster simulation model described above includes a disaster simulation animation model, and step 203 further includes: Step 304: If the key parameters output by the disaster simulation model (such as displacement rate exceeding 10 mm / day, safety factor < 1.0) reach the preset threshold, execute the disaster animation simulation operation. The stress cloud map, displacement vector, and crack propagation path of the geological body in the target monitoring area will be rendered into a dynamic 3D animation in real time during the target period, which will intuitively show the evolution process and impact range of the disaster.

[0061] In another exemplary embodiment of this application, the disaster prediction model includes a geological factor disaster prediction model, a meteorological factor disaster prediction model, and a human engineering activity disaster prediction model.

[0062] In step 301 above, the corrected geological factor data is input into the geological factor disaster prediction model, the corrected meteorological factor data is input into the meteorological factor disaster prediction model, and / or the corrected human engineering activity data is input into the human engineering activity disaster prediction model.

[0063] Step 302 above includes: The geological factor disaster prediction model, based on modified geological factor data, simulates the stress distribution and deformation of the geological body in the target monitoring area under geological conditions only during the future target period, and outputs a second key parameter, including the long-term safety factor. ), geological creep displacement rate (v); The meteorological disaster prediction model, based on modified meteorological data, simulates the stress distribution and deformation of the geological body in the target monitoring area under the influence of only the predicted meteorological event during the future target period, and outputs a third key parameter, including the transient safety factor (Fs_met(t)) and the pore water pressure surge (…). ), precipitation displacement increment (dt); The human activity model, based on modified human engineering activity data, simulates the stress distribution and deformation of the geological body in the target monitoring area under the influence of human engineering activities only during the future target period, and outputs the fourth key parameter after engineering disturbance. The fourth key parameter includes the safety factor (Fs_eng) and the blasting vibration velocity (v_eng).

[0064] Step 303 above includes: If the key parameters output by the geological factor disaster prediction model, meteorological factor disaster prediction model, or human engineering activity disaster prediction model reach the preset threshold, it is considered that the target geological area will experience the target geological disaster in the future target period; otherwise, it is considered that the target geological area will not experience the target geological disaster in the future target period.

[0065] For example, if the slope is currently basically stable ( ≈1.2), if a 200mm torrential rain is forecast in the next 24 hours, the meteorological model simulates the impact of rainstorm infiltration and outputs: at the peak of rainfall, the transient safety factor Fs_met(t) drops to a minimum of 0.95, and the preset threshold is triggered. The warning message reads "<1.0", indicating that "landslides are likely to occur in the target area during heavy rain", and is issued accordingly.

[0066] In this embodiment, by determining whether the key parameters output by the geological factor disaster prediction model, meteorological factor disaster prediction model, or human engineering activity disaster prediction model reach a preset threshold, the cause of the target geological disaster in the future target period can be determined. Specifically, if the key parameters output by the geological factor disaster prediction model reach the preset threshold, the cause of the target geological disaster in the future target period is geological factors; if the key parameters output by the meteorological factor disaster prediction model reach the preset threshold, the cause of the target geological disaster in the future target period is meteorological factors; and if the key parameters output by the human engineering activity disaster prediction model reach the preset threshold, the cause of the target geological disaster in the future target period is human engineering activities, thereby enabling the implementation of corresponding preventative measures.

[0067] In another exemplary embodiment of this application, the disaster prediction model further includes a multi-factor coupled disaster prediction model, wherein each multi-factor coupled disaster prediction model is associated with at least two of the geological factor disaster prediction model, meteorological factor disaster prediction model and human engineering activity disaster prediction model through correlation analysis technology.

[0068] Summary association method: Input / state coupling: Using the output or intermediate state variables of a single-factor model as the input of a multi-factor coupled disaster simulation model; Equation coupling: Directly construct coupled control equations that consider the interaction of multiple factors (such as fluid-structure interaction equations); Data-driven correlation: Analyze the probability distribution of multiple factors jointly causing disasters in historical data and construct a joint probability model.

[0069] Multi-factor coupled disaster prediction models, coupled control equations, and joint probability models are methods for addressing multi-factor disaster problems from different perspectives and levels. Their relationships and differences are as follows: Coupled control equations (physical mechanism model): Core: Based on physical laws (such as mechanics and seepage mechanics), a system of partial differential equations is established to describe the physical mechanism of interaction between various factors (such as stress, water, and earthquake motion).

[0070] Multi-factor coupled disaster prediction model (numerical implementation): The core is the numerical solution of the coupled control equations on a computer (such as using the finite element method or discrete element method). It discretizes the physical equations and simulates the evolution of disasters through iterative calculations.

[0071] Relationship with equations: The pre-model is the "solver" of the coupled control equations.

[0072] Joint probability model (data statistical model): Core principle: Based on a large amount of historical disaster case data, statistical or machine learning methods are used to directly establish a "black box" or "grey box" mapping relationship between various disaster-causing factors (such as rainfall, seismic intensity, and intensity of human activity) and the probability of disaster occurrence. For example, a probabilistic graphical model or a neural network can be constructed.

[0073] Relationship with physical models: Complementary: Physical models start from "first principles" and are suitable for mechanism research and specific scenario simulation; statistical models start from "empirical data" and are suitable for rapid, large-scale probabilistic early warning.

[0074] Validation and Fusion: Simulation results from physical models can serve as a source for generating "synthetic data," which can supplement or validate statistical models. Conversely, combinations of key factors identified by the statistical model can guide the simplification or focused simulation direction of the physical model. In advanced systems, the two can be fused (e.g., physical information neural networks).

[0075] Objective: To quickly assess the combined risk probability when multiple factors coexist, especially applicable to situations where the interactions between factors are complex and difficult to describe with precise physical equations.

[0076] Each multi-factor coupled disaster prediction model simulates the occurrence process of geological disasters under the interaction and mutual influence of two or more disaster-causing factors, capturing nonlinear coupling effects. By combining at least two of the geological factor disaster prediction models, meteorological factor disaster prediction models, and human engineering activity disaster prediction models, synergistic / antagonistic effects can be simulated: capturing the nonlinear superposition effect of "1+1>2" (such as weak interlayer + rainfall), while improving forecast accuracy: more closely reflecting the reality of multiple factors working together, reducing missed and false alarms. Furthermore, analyzing the contribution rate of each factor after coupled early warning can distinguish the dominant factor.

[0077] In another exemplary embodiment of this application, step 302 further includes: Each multi-factor coupled disaster simulation model outputs a fifth key parameter based on at least two of the following: corrected geological factor data, corrected meteorological factor data, and corrected human engineering activity data.

[0078] The key parameters output by each multi-factor coupled disaster simulation model include the coupling safety factor (Fs_coupled), the coupled displacement field, and the plastic zone considering the coupling effect.

[0079] In another exemplary embodiment of this application, step 303 further includes: If the fifth key parameter output by any multi-factor coupled disaster prediction model reaches the preset threshold, it is considered that the target geological area will experience the target geological disaster in the future target period; otherwise, it is considered that the target geological area will not experience the target geological disaster in the future target period.

[0080] For example, the slope of a reservoir bank has weakened due to long-term immersion. =1.1). Input to the multi-factor coupled disaster simulation model: Heavy rainfall forecast + planned rapid water level drop from the reservoir. Simulation: The multi-factor coupled disaster simulation model simultaneously considers rainfall infiltration (increased weight, increased pore pressure) and a sudden drop in reservoir water level (generating outward seepage force, loss of water support). Output: Coupling safety factor. It plummeted to 0.88. Determination: Triggering threshold " The risk level is "<1.0", thus indicating a high risk. However, individual geological disaster prediction models and meteorological disaster prediction models may not have reached this warning level.

[0081] In another exemplary embodiment of this application, in step 302 above, while the geological factor disaster prediction model, meteorological factor disaster prediction model and human engineering activity disaster prediction model output the second key parameter, the third key parameter and the fourth key parameter respectively, they also output the stress cloud map, displacement vector and crack propagation path of the geological body in the target monitoring area in the future target period.

[0082] In another exemplary embodiment of this application, step 304 above further includes: If the second, third, or fourth key parameter output by the geological factor disaster prediction model, meteorological factor disaster prediction model, or human engineering activity disaster prediction model exceeds the preset threshold (e.g., displacement rate exceeds 10 mm / day, safety factor < 1.0), execute the disaster animation prediction operation: The geological factor disaster prediction model, meteorological factor disaster prediction model, or human engineering activity disaster prediction model outputting the second, third, or fourth key parameters that reach the preset threshold will render the stress cloud map, displacement vector, and crack propagation path of the geological body in the target monitoring area in the future target period into a real-time visualized dynamic 3D animation, showing the disaster evolution process and the scope of impact.

[0083] In another exemplary embodiment of this application, step 302 further includes: Each multi-factor coupled disaster simulation model outputs key parameters, as well as stress cloud maps, displacement vectors, and crack propagation paths of the geological body in the target monitoring area during the future target time period.

[0084] Due to different dominant factors, the visualization results output by each model differ in distribution, size, and time series: Geological disaster prediction model: may show long-term stress concentration zones and slow creep displacements related to deep structures; Meteorological disaster prediction model: Stress changes and displacements mainly occur in the shallow layer, and are either synchronous with or delayed by the rainfall process; Human engineering activity disaster prediction model: stress concentration and displacement abrupt change are strictly limited to the engineering disturbance zone and are synchronized with construction activities; Multi-factor coupled disaster simulation model: The result is a complex superposition of the above-mentioned effects, which is closest to the reality.

[0085] In another exemplary embodiment of this application, step 304 above further includes: If any of the fifth key parameters output by the multi-factor coupled disaster simulation model exceeds a preset threshold (e.g., displacement rate exceeds 10 mm / day, safety factor < 1.0), execute the disaster animation simulation operation: The multi-factor coupled disaster simulation model outputting the fifth key parameter exceeding the preset threshold renders the stress cloud map, displacement vector, and crack propagation path of the geological body in the target monitoring area in the future target period into a real-time visualized dynamic 3D animation, showing the disaster evolution process and impact range.

[0086] In this embodiment, the cause of a target geological disaster occurring in a future target time period can be intuitively determined through the disaster evolution process. If both a single-factor model and a multi-factor coupled model trigger the disaster animation pre-simulation process, the cause of the disaster is identified as a synergistic effect of multiple factors, and attribution analysis is performed in the early warning information. For example, if the geological factor disaster pre-simulation model triggers the process alone ( =0.95), the geological-meteorological coupled disaster prediction model is also triggered and the risk is higher ( =0.8). The early warning is based on a geological-meteorological coupled disaster prediction model. The analysis shows that the current extreme rainfall (the dominant factor) is very likely to trigger a landslide given the poor stability of the slope itself (the background factor).

[0087] In another exemplary embodiment of this application, the aforementioned warning information also includes the coordinates of the area in the target geological region where a target geological disaster will occur in the future target time period.

[0088] In another exemplary embodiment of this application, the aforementioned early warning information also includes the probability of a target geological disaster occurring in the target geological area during a future target period and the severity of the target geological disaster.

[0089] In this application embodiment, the probability of causing a target geological disaster includes the probability of a geological disaster caused by human engineering activities, the probability of a geological disaster caused by geological factors, and the probability of a geological disaster caused by meteorological factors. The severity of a target geological disaster includes the severity of a geological disaster caused by human engineering activities, the severity of a geological disaster caused by geological factors, and the severity of a geological disaster caused by meteorological factors. The probability of a geological disaster caused by human engineering activities refers to the probability that a target geological disaster (landslide, collapse, etc.) will occur in the future due to changes in the stability of the geological environment caused by human engineering activities. The probability of a geological disaster caused by geological factors refers to the probability that a target geological disaster will occur in the future due to changes in geological conditions. The probability of a geological disaster caused by meteorological factors refers to the probability that a target geological disaster will occur in the future due to changes in meteorological factors exacerbating the instability of the geological environment. The severity of a geological disaster caused by human engineering activities refers to the maximum potential loss (e.g., burial area, economic loss) that a target geological disaster may cause in the future target period due to human engineering activities. The severity of a geological disaster caused by geological factors refers to the area and intensity of damage that a target geological disaster may affect in the future target period due to changes in geological conditions. The severity of geological disasters caused by meteorological factors refers to the quantifiable losses that may be caused by the chain reaction of a target geological disaster in the future due to changes in meteorological factors (such as debris flows destroying bridges, landslides blocking river channels and forming barrier lakes).

[0090] In another exemplary embodiment of this application, the probability of a geological hazard caused by human engineering activities is calculated according to the following process: Based on Bayesian or logistic regression models that quantify the relationship between data on human engineering activities and the probability of geological disasters caused by human engineering activities, the probability P of geological disasters caused by human engineering activities is calculated using real-time collected data on human engineering activities (characteristic data of human engineering activities), thus obtaining the probability of geological disasters caused by human engineering activities.

[0091] In this embodiment of the application, by collecting a large number of historical cases, the correlation between historical data of human engineering activities and the geological disasters they cause is analyzed. Input feature variables such as excavation slope S, daily excavation volume V, and labels (accident occurred = 1, no = 0) are used to train a Bayesian model or logistic regression model that quantifies the relationship between data of human engineering activities and the probability of geological disasters caused by human engineering activities.

[0092] In another exemplary embodiment of this application, the severity of geological hazards caused by human engineering activities is calculated according to the following process: Using GIS spatial analysis, we can calculate population density and asset value within the threat area and quantify the potential for loss. Severity levels (e.g., mild, moderate, severe) are classified according to the scale of casualties and the threshold of economic losses.

[0093] In this embodiment of the application, the threatened area is the range that may be affected after a geological disaster occurs, including: Directly affected areas: such as the area covered by the landslide and the path of the debris flow; The impact range of secondary disasters includes areas such as the inundation range of landslide dams and the spread range of floods.

[0094] Threat area determination method: determined through the visualization results of the disaster simulation model (crack propagation path, displacement vector range).

[0095] Economic loss is the actual value of the potential loss. When a disaster occurs, the potential loss is converted into economic loss; when it does not occur, the economic loss is zero.

[0096] Loss potential is the expected value of potential losses to people and assets within a threat area. Quantification steps include: The threat area is divided into high, medium, and low risk zones; Compile basic data for each risk area, including population density and asset value (such as buildings and infrastructure); Calculate the potential loss: Potential personnel loss = Σ(population size in risk zone i × population vulnerability coefficient of the zone); Potential economic loss = Σ(asset value in risk zone j × asset vulnerability coefficient in that zone × damage rate).

[0097] Dividing the target area into risk zones based on risk level facilitates the implementation of differentiated monitoring densities, early warning thresholds, emergency resources, and defense measures. Both "basic statistical data" and "loss potential calculations" must be conducted within the defined risk zones. Different risk levels have varying population densities, asset values, and defense capabilities; therefore, separate statistical analysis and calculations are necessary to obtain more accurate and spatially targeted risk assessment results.

[0098] Vulnerability coefficient / damage rate: Derived from historical disaster damage statistics, expert experience, or physical model simulations, it represents the proportion of casualties or asset damage under a disaster of a specific intensity. For example, the population vulnerability coefficient is the probability of casualties based on historical disaster statistics (e.g., 0.1).

[0099] The steps for classifying severity levels based on the scale of casualties and thresholds of economic loss include: The quantitative results of personnel loss and economic loss are combined into a severity level using a matrix method or a weighted scoring method. For example: mild (few casualties and minor losses), moderate (either casualties or losses reach the intermediate level), and severe (either casualties or losses reach the advanced level, or both are intermediate level).

[0100] The severity level is quantified into a numerical value, which is used as the severity value F of geological hazards caused by human engineering activities. 工 For example, the severity of geological hazards caused by human engineering activities is quantified as follows: mild = 0.3, moderate = 0.6, severe = 1.0 (F = 0.0). 工 .

[0101] In another exemplary embodiment of this application, the data sources used to calculate the likelihood of geological factors causing geological disasters mainly include: physical parameters of the soil and rock mass reflecting its basic properties (such as shear strength, elastic modulus, and permeability coefficient); geological structural characteristic data reflecting regional tectonic activity (such as fault slip rate, derived from historical exploration data or inverted from monitoring data (microseismic)); and groundwater monitoring data reflecting real-time dynamic changes in the geological environment. The shear strength of the soil and rock mass (cohesion c, internal friction angle) is also considered. Its inherent and fundamental mechanical properties, such as elastic modulus (E) and Poisson's ratio, are indispensable input parameters for any quantitative stability calculation (such as limit equilibrium method and finite element method).

[0102] The probability of geological disasters caused by geological factors is calculated using the following process: Geological stability index: Physical parameters of the soil and rock mass are obtained through geotechnical mechanics tests. Based on these physical parameters and groundwater monitoring data, the instability probability P is calculated using the limit equilibrium method or numerical model. 地下水诱发 ; Tectonic activity model: Based on fault slip rate and historical earthquake data, predict the probability P of tectonic activity triggering geological hazards. 地震触发 ‎.

[0103] The probability of instability is a calculated result or core component of the likelihood of geological accidents. When using reliability analysis methods (such as Monte Carlo simulation), the physical parameters of the soil and rock mass and groundwater monitoring data are treated as random variables (considering the uncertainty of their experimental values), input into the limit equilibrium model for multiple calculations, and the proportion of times the long-term safety factor Fs < 1 is counted to obtain the probability of instability.

[0104] Fault slip rate reflects the current activity intensity of the fault (mm / year), historical earthquake data: including the magnitude, location, and recurrence cycle of historical earthquakes.

[0105] Based on fault slip rate and historical earthquake data, the probability of tectonic activity triggering geological hazards is predicted, specifically including the following steps 1 to 4. Wherein: Step 1: Establish a regional seismic activity model to determine potential seismic source areas and earthquake recurrence patterns.

[0106] Step 2: Based on the fault slip rate and the moment-slip relationship, estimate the probability of earthquakes of different magnitudes occurring on the fault within a certain period in the future (e.g., 50 years).

[0107] Step 3: Calculate the probability P (earthquake motion = IM) of the target geological area encountering earthquakes of different intensities (such as peak ground acceleration PGA).

[0108] Step 4: Combine the instability probability of the slope in the target area under a specific seismic intensity (obtained through experiments or experience) to comprehensively calculate the triggering probability P (landslide|earthquake occurrence) of geological disasters caused by tectonic (earthquake) activity.

[0109] Steps 1-3 (Seismic Hazard Analysis): The purpose is to calculate the probability that the target area will experience earthquakes of different intensities (e.g., PGA of 0.1g, 0.2g, etc.) in the future, i.e., P (earthquake motion = IM).

[0110] Step 4 (Geological Hazard Vulnerability Analysis): The purpose is to determine the conditional probability of the target slope becoming unstable (landslide) under a given ground motion intensity (IM), i.e., P(landslide|ground motion=IM).

[0111] Comprehensive calculation (total probability formula): The probability P(landslide|earthquake occurrence) of geological disasters caused by tectonic (earthquake) activity is the result of integrating all possible seismic motion intensities: P(landslide|earthquake) = Σ[ P(earthquake motion = IM) × P(landslide|earthquake motion = IM)] Obtaining P(landslide | ground motion = IM): Empirical statistical method: Collect the proportion of landslides in different PGA regions during historical earthquakes and establish empirical relationships.

[0112] Model test method: Through shaking table tests, the instability threshold of different types of slopes under different ground motions is studied.

[0113] Numerical simulation method: Using dynamic finite element / discrete element models, the stability of slopes under random seismic waves is calculated extensively, and the relationship between the probability of instability and the PGA is statistically analyzed.

[0114] Calculation Example: Assume that, based on steps 1-3, the probability of PGA > 0.2g within the next 50 years is 1%, and the probability of PGA being between 0.1g and 0.2g is 5%. Research shows that the probability of a landslide in this area is 80% when PGA > 0.2g and 20% when PGA is between 0.1g and 0.2g. Then: P(landslide|earthquake) ≈ 1% × 80% + 5% × 20% = 0.8% + 1% = 1.8% This 1.8% is the probability of a landslide triggered solely by tectonic (earthquake) factors.

[0115] Geological hazards can be triggered by a variety of geological factors, including long-term creep under gravity, changes in groundwater, and earthquake triggering. In active fault zones, P 地震触发 It may dominate in P_geo.

[0116] P_geo is the union or weighted sum of probabilities from multiple mechanisms, which can be synthesized using Bayesian updates, fuzzy logic, or empirical formulas. For example: P_geo = 1 - [(1 - P 重力蠕变 ) (1 - P 地下水诱发 ) (1 - P 地震触发 This means that as long as the probability of any geological disaster mechanism increases, the total P_geo will increase.

[0117] In another exemplary embodiment of this application, the data sources on which the severity of geological disasters is based mainly include: the extent of geological structural damage (such as landslide volume, collapse area) and the potential impact of disaster chains (such as secondary debris flows, formation of barrier lakes).

[0118] When simulating instability, disaster simulation models calculate the volume of the rock and soil mass (landslide volume) surrounded by the slip surface and the deformation area of ​​the ground surface (collapse area), which is the range of geological structural damage.

[0119] The potential impact of a disaster chain is obtained through chain simulation: The results of the main disaster simulation (such as landslide volume and accumulation zone extent) are used as input; Initiate secondary models; for example, if a landslide blocks a river channel, initiate a hydrodynamic model to simulate the formation of a landslide dam, water level rise, and potential dam failure; if a large amount of loose material is generated, initiate a debris flow model to simulate its downstream flow. The output of the secondary model (such as the flood inundation range and the debris flow impact range) represents the potential impact of the disaster chain.

[0120] For emergency decision-making, the extent of damage directly identifies the core areas requiring urgent evacuation and alert. The potential impact of the disaster chain is used to identify secondary risk areas that need to be prevented in advance. For example, residents in downstream valleys may need to be evacuated in advance due to the risk of a landslide dam, even if they are not directly threatened by the main landslide. This enables proactive and expanded emergency preparedness.

[0121] Methods for calculating the severity of geological disasters caused by geological factors: Disaster scale classification: The damage level is classified according to the initial landslide volume or collapse energy; Chain reaction assessment: Predicting disaster spread paths and secondary disaster superposition effects through numerical simulations (such as DEM, FLO-D); Based on the results of the cascading effect assessment, the impact range of secondary disasters is simulated; Estimate the losses from the initial disaster based on the level of damage caused by the initial disaster; Based on the impact range of secondary disasters, estimate the losses from secondary disasters (the additional losses that secondary disasters may cause). The sum of primary disaster losses and secondary disaster losses is used to obtain the comprehensive loss. The comprehensive loss is then mapped to a preset severity range to obtain the severity of geological disasters caused by geological factors.

[0122] For example, if the estimated loss is less than 1 million, the corresponding severity F of the geological disaster caused by geological factors is... 地 = 0.2 (mild), 1 million ≤ estimated loss < 10 million, corresponding to the severity F of geological disaster caused by geological factors. 地 =0.5 (moderate), estimated loss ≥10 million, corresponding to the severity F of geological disaster caused by geological factors. 地 = 0.8 (severe).

[0123] In this embodiment of the application, primary geological hazards (such as landslides) may trigger a series of secondary hazards (such as landslides blocking rivers to form barrier lakes, dam breaks causing floods; collapse bodies providing material sources for debris flows), and the impact range of secondary hazards may be much greater than that of primary hazards. The purpose of chain reaction assessment is to comprehensively identify all potential threats in order to more accurately delineate the real risk areas and avoid underestimating the severity of the hazards.

[0124] Damage levels are usually classified according to national or industry standards. For example, my country's "Code for Investigation of Landslide Prevention Engineering" classifies landslides by volume as follows: small: <100,000 cubic meters, medium: 100,000 to 1,000,000 cubic meters, large: 1,000,000 to 10,000,000 cubic meters, and extra-large: >10,000,000 cubic meters.

[0125] Collapse energy: Kinetic energy E = 1 / 2 quality speed 2 Mass comes from volume, while velocity can be estimated through model simulation or empirical formulas, and can also be classified according to energy level (such as low energy, medium energy, and high energy).

[0126] Classifying disasters into levels provides an intuitive concept of disaster scale for emergency response (whether it is a small-scale collapse or a large-scale landslide) and links it to loss potential: different levels of damage usually correspond to different ranges of loss potential, which is an important bridge for converting physical quantities into severity weights.

[0127] In this embodiment of the application, the estimation of primary disaster losses based on the initial disaster damage level includes the following steps a to e. Wherein: Step a: Determine the scope of impact: Combine the damage level (such as landslide volume) and topography, and use models (such as DAN, FLO-2D) to simulate the movement and accumulation range of the disaster body.

[0128] Step b, Disaster-bearing body survey: In GIS, overlay the impact range layer with disaster-bearing body layers such as population distribution, buildings, infrastructure, and farmland.

[0129] Step c: Apply vulnerability curves / matrix: For each type of disaster-bearing body (e.g., brick-concrete houses, people, roads), query the corresponding loss rate (between 0 and 1) based on disaster intensity parameters (e.g., landslide accumulation thickness, flow velocity). For example, a 2-meter-thick accumulation may cause complete destruction of brick-concrete houses (loss rate 1.0).

[0130] Step d: Calculate the loss: Individual loss = Number of affected buildings × Average value × Loss rate. For example, housing loss = Number of affected buildings × Average value × Average loss rate.

[0131] Step e, Monetization and Summarization: Monetize personal injury (which can be calculated based on the statistical value of life), property damage, infrastructure repair costs, etc., and then sum them up to obtain the estimated comprehensive loss.

[0132] Based on the impact range of secondary disasters, the estimated losses of secondary disasters are made by taking the impact range of secondary disasters (such as the flooded area of ​​a landslide dammed lake) as a new impact range, repeating the above steps of "estimating initial disaster losses", surveying the disaster-bearing bodies in the range, applying the corresponding vulnerability curves (such as flood depth-loss rate curves), and calculating the additional losses that secondary disasters may cause.

[0133] In another exemplary embodiment of this application, when calculating the probability of geological disasters caused by meteorological factors, the data used, in addition to the real-time measured and forecast meteorological factor data, also includes historical meteorological statistics related to the region (such as historical rainfall data (such as hourly rainfall intensity, cumulative rainfall), freeze-thaw cycle frequency (important for slope stability in cold regions)), climate indicators, and advanced early warning information issued by meteorological departments (such as typhoon paths, orange rainstorm warnings).

[0134] Calculating the likelihood of meteorological factors causing geological disasters specifically includes: By inputting future rainfall forecasts or current monitoring data (I, R) into a pre-established rainfall-disaster probability relationship curve, the probability of geological disasters can be predicted through the rainfall-disaster probability relationship curve. The resulting probability of geological disasters is the likelihood that meteorological factors will cause geological disasters.

[0135] In this embodiment, a rainfall-disaster probability relationship curve is established, that is, a quantitative statistical relationship is established between rainfall parameters (such as hourly rainfall intensity and cumulative rainfall) and the probability of geological disasters. The rainfall-disaster probability relationship curve is an empirical-statistical model and a core tool for quickly assessing the risk of meteorological disasters. For any forecasted or ongoing rainfall event, a corresponding disaster probability estimate can be immediately obtained by looking up a table or calculating.

[0136] The operational procedure for establishing a rainfall-disaster probability relationship curve: Collect data (I = rainfall intensity, R = cumulative rainfall) and corresponding geological disaster occurrence records for each rainfall event over a historical period (occurrence = 1, non-occurrence = 0). Using statistical methods such as logistic regression, the fitted function P = f(I, R) represents the rainfall-hazard probability relationship curve. For example: P_met = 1 / (1 + exp(-(a I + b R - c))).

[0137] When future rainfall forecasts or current monitoring data (I, R) are obtained, they can be directly substituted into the function f(I, R) to calculate the P_met value, which represents the probability of geological disasters caused by meteorological factors.

[0138] In another exemplary embodiment of this application, calculating the probability of meteorological factors causing geological disasters further includes: Multiple indicators (such as soil moisture content, groundwater level, and previous rainfall index) were selected, and the probability of hydrogeological conditions deteriorating under each indicator was calculated. The weight of each indicator is determined by expert scoring or machine learning, and the probability of hydrogeological condition deterioration under multiple indicators is weighted and summed based on the determined weights to obtain the meteorological coupling coefficient. By using the meteorological coupling coefficient as an adjustment factor, the probability of geological disasters predicted by the rainfall-disaster probability relationship curve is corrected, thus obtaining the final probability that meteorological factors will lead to geological disasters.

[0139] In this embodiment, the probability of hydrogeological conditions deteriorating under soil saturation is approximately equal to the current volumetric water content θ / saturated volumetric water content θs. The closer the soil is to saturation, the closer the probability of hydrogeological conditions deteriorating under soil saturation is to 1. The probability of hydrogeological conditions deteriorating under the previous rainfall index is equal to f(API), where API is the previous rainfall index, which reflects the soil moisture content. The higher the API, the greater the probability of hydrogeological conditions deteriorating under the previous rainfall index. The probability of hydrogeological conditions deteriorating under the groundwater level is calculated as (current water level - average annual water level) / (critical disaster water level - average annual water level), standardized to between 0 and 1.

[0140] The meteorological coupling coefficient is an indicator reflecting the sensitivity of current geological and hydrological conditions to rainfall. For example, if the slope soil and rock are close to saturation (high soil moisture content), even a small amount of rainfall will result in a high infiltration rate, which quickly translates into increased pore water pressure, significantly impacting stability. In this case, the meteorological coupling coefficient is high.

[0141] The meteorological coupling coefficient is used to correct the probability estimate of rainfall-hazard curves. A simple rainfall-hazard curve does not take into account the previous state of the geological body. The meteorological coupling coefficient introduces current status information, making the probability assessment more dynamic and accurate.

[0142] Specifically, the meteorological coupling coefficient (denoted as C, ranging from 0 to 1) is usually applied as a multiplier or adjustment factor to the base probability (P_met) obtained from the rainfall-disaster curve.

[0143] If the slope is very dry (C close to 0), even if heavy rainfall is forecast (high P_met), the adjusted meteorological factors will be less likely to cause geological disasters. If the slope is already in a critical saturation state (C close to 1), under the same rainfall, the adjusted meteorological factors will be very likely to cause geological disasters.

[0144] In another exemplary embodiment of this application, when calculating the severity of geological disasters caused by meteorological factors, the data used includes the duration of meteorological events (such as the number of consecutive days of heavy rain) and regional topographic features (such as catchment area and valley slope), wherein the regional topographic features are derived from basic geographic information data.

[0145] Calculating the severity of geological disasters caused by meteorological factors, specifically including: The relationship between rainfall intensity and debris flow flow / velocity was established through regression analysis to predict impact force and burial depth; Hydrological models (such as HEC-RAS) are used to calculate runoff and assess the extent of inundation and the risk of infrastructure damage. Based on the predicted impact force and burial depth, the initial economic losses caused by meteorological factors to geological disasters are determined through vulnerability models and exposure analysis of disaster-bearing bodies. Based on the extent of inundation and the risk of infrastructure damage, determine the potential secondary economic losses caused by secondary flood disasters; The total economic loss is obtained by summing the initial economic loss and the secondary economic loss. Based on the total economic loss, the severity of geological disasters caused by meteorological factors is determined.

[0146] In this embodiment of the application, meteorological factors (such as torrential rain) may induce primary geological disasters (such as landslides and debris flows), which may block river channels and form barrier lakes. The purpose of using a hydrological model (such as HEC-RAS) is as follows: Inputs: upstream water flow (rainfall runoff), landslide dam morphology (from geological disaster simulation); Simulation: The impoundment process of a landslide dammed lake, the water level-reservoir capacity relationship, and the flood evolution after a potential dam failure; Output: Changes in flood inundation area, water depth, and flow velocity over time.

[0147] The flood inundation map output by the hydrological model is a direct basis for assessing the potential casualties and economic losses caused by secondary flood disasters. These losses must be included in the overall severity assessment of geological hazards caused by meteorological factors. Therefore, the hydrological model quantifies the end-of-chain impacts of the disaster and thus comprehensively calculates F... 气 A necessary tool.

[0148] Based on the predicted impact force and burial depth, the economic losses caused by geological disasters due to meteorological factors are determined through vulnerability models and exposure analysis of the affected bodies. Specifically, these losses include: Establish vulnerability curves / matrices: For different types of disaster-bearing bodies (such as brick-concrete residential buildings, reinforced concrete frame buildings, people, and roads), establish the relationship between their loss rate and the intensity of disaster-causing factors through historical disaster damage data, experiments, or expert experience.

[0149] Example: For buildings: a vulnerability curve might indicate that when the impact force of a debris flow reaches X kPa, the loss rate (repair cost / replacement cost) of a brick-and-wood structure house is Y; for people: an empirical matrix might indicate that, without warning, the mortality rate is approximately Z when the burial depth reaches 1 meter; for roads: the probability of traffic disruption is W when the flooding depth reaches 0.5 meters.

[0150] Overlaying disaster-bearing body information with simulation results: In GIS, the intensity field output by the disaster simulation is overlaid with the spatial distribution map of the disaster-bearing bodies.

[0151] For each disaster-bearing unit (such as a building, a road section, or the population of a statistical grid), read the disaster intensity value at that location.

[0152] Based on its type, query the corresponding vulnerability curve / matrix to obtain the loss rate or probability of being affected for that unit. Multiply the loss rate by the value of that unit (building replacement value, road repair cost, demographic value) to obtain the expected loss for that unit.

[0153] The aggregation yields the initial loss value: The expected losses of all affected units are summed to obtain the total direct economic loss and the number of affected people (which can be further converted into expected loss of life).

[0154] The total direct economic loss is already a monetized value, measured in yuan, ten thousand yuan, etc. It is the sum of property damage, infrastructure repair costs, etc. The number of affected people is a physical quantity, measured in persons. In some more comprehensive assessments, casualties are also monetized using methods such as "Statistical Value of Life (VSL)," and then added to the direct economic loss to obtain a total economic value including loss of life, which is the initial loss value. The method for determining secondary economic losses is the same as that for the initial loss value.

[0155] In the context of disaster simulation output, the intensity field refers to a broad concept encompassing all models capable of outputting spatialized disaster intensity distributions. This includes: primary geological disaster simulations (e.g., landslide / debris flow models outputting depositional thickness and velocity vector fields); secondary disaster simulations (e.g., hydrological models outputting flood inundation depth and velocity fields); and relational model outputs (e.g., debris flow flow / velocity distribution predicted through rainfall-debris flow relationships). The intensity field is gridded data describing how the destructive forces (such as force, depth, and velocity) of a disaster are distributed spatially.

[0156] The total economic loss, after being normalized or compared with a preset threshold, can be converted into the severity of geological disasters caused by meteorological factors. : F 气 =Total economic loss / Lmax, F 气 It will be a value between 0 and 1, and L_max is the upper limit value, which is the historical maximum loss value or the theoretically acceptable maximum loss value.

[0157] In another exemplary embodiment of this application, the above-described geological disaster prediction method based on big data analysis further includes: Step 401: Based on the probability and severity of the geological disaster that will occur in the target geological area in the future target period, risk classification is performed to obtain the risk classification results of geological disasters in the target monitoring area.

[0158] In this embodiment, the risk classification result of geological disasters in the target monitoring area is the risk level of the target geological disaster occurring in the target monitoring area. By quantifying the probability and severity of the target geological disaster and classifying the risk, it is possible to achieve precise allocation of resources, dynamic adjustment of early warning, and differentiated implementation of emergency measures, ensuring multi-party collaborative disaster prevention among the government, engineering units, and the public, and ultimately reducing casualties and economic losses.

[0159] In another exemplary embodiment of this application, the aforementioned early warning information also includes the risk classification results of geological hazards in the target monitoring area.

[0160] In another exemplary embodiment of this application, step 401 described above includes steps 501 to 502. Wherein: Step 501: Calculate the risk according to the following formula. : ; in, This indicates the probability of accidents occurring during human engineering activities. This indicates the severity of accidents that occur during human engineering activities. This indicates the likelihood of an accident caused by geological factors. This indicates the severity of the accident caused by geological factors. This indicates the likelihood of an accident caused by meteorological factors. Z represents the severity of accidents caused by meteorological factors, and Z represents the comprehensive susceptibility index of geological disasters.

[0161] In this application embodiment, the probability of an accident caused by human engineering activities represents the possible probability of an accident caused by human engineering activities; the severity of an accident caused by human engineering activities represents the quantitative value of the severity of an accident caused by human engineering activities; the probability of an accident caused by geological factors represents the possible probability of an accident caused by geological factors; the severity of an accident caused by geological factors represents the quantitative value of the severity of an accident caused by geological factors; the probability of an accident caused by meteorological factors represents the possible probability of an accident caused by meteorological factors; and the severity of an accident caused by meteorological factors represents the quantitative value of the severity of an accident caused by meteorological factors.

[0162] The comprehensive geological disaster susceptibility index is calculated using the following formula: = + ; in, The potential geological hazard intensity index reflects the potential risks of geological conditions (such as soil and rock properties and tectonic activity) in the target geological area. This represents the current geological hazard intensity index, reflecting the current risk based on historical hazard data (such as the frequency of landslides and debris flows). , This represents the weighting coefficient (usually 0.6 or 0.4), reflecting... and Differences in contributions. The parameters are obtained by weighted superposition using the Analytic Hierarchy Process (AHP) through a combination of topographic slope, soil shear strength, fault density, and other parameters. By statistically analyzing the density, scale, and activity of historical disaster points within grid units, existing disaster threats can be quantified.

[0163] In this embodiment of the application, Zq is obtained by weighted superposition using the Analytic Hierarchy Process (AHP) by comprehensively considering parameters such as terrain slope, soil shear strength, and fault density. Specifically, this includes the following steps 5011 to 5015: Step 5011, establish the hierarchical structure: Target layer: Potential geological hazard intensity index Zq; Criterion layer: Factors affecting Zq, such as topographic conditions, soil and rock conditions, and structural conditions; Index layer: Specific parameters under each criterion, such as slope under topographic conditions, shear strength under soil and rock conditions, and fault density under tectonic conditions. There may also be other indicators, such as slope height, lithology, joint density, etc.

[0164] Step 5012, construct the judgment matrix: compare each factor in the same level pairwise to determine its relative importance (using the 1-9 scale method, for example, determine the importance of "slope" to "shear strength" in disaster susceptibility) to obtain the judgment matrix.

[0165] The judgment matrix is ​​a square matrix used in the Analytic Hierarchy Process (AHP) to express the relative importance of each pair of factors within the same level. It is filled in by experts or decision-makers after comparing and scoring each pair using the 1-9 scale. a ij = 1: i and j are equally important.

[0166] a ij = 3: i is slightly more important than j.

[0167] a ij = 5: i is significantly more important than j.

[0168] a ij = 7: i is significantly more important than j.

[0169] a ij = 9: i is extremely more important than j.

[0170] 2, 4, 6, 8 are intermediate values, and a ji = 1 / a ij .

[0171] In AHP, approximation algorithms such as the sum-product method or the square root method are typically used to solve the problem manually or through programming.

[0172] Step 5013: For each judgment matrix, calculate its maximum eigenvalue and eigenvector. The eigenvector is the weight of the factor in this layer relative to the factor in the previous layer.

[0173] For example, the weights of the index layer (slope, shear strength, fault density, etc.) relative to the criterion layer (topography, soil, structure) and the weights of the criterion layer relative to the target layer are calculated.

[0174] Step 5014: Normalize the collected raw data of each indicator (slope value, shear strength value, fault density value, etc.) to eliminate the influence of dimensions and make them comparable.

[0175] Step 5015, Linear Weighting: Zq = Σ (Standardized value of index i) The overall weight of indicator i is calculated as follows: (weight of indicator i relative to its criterion). (The criteria are relative to the weights of the target layer).

[0176] The weight of indicator i relative to its criterion: For example, under the "topographic conditions" criterion, the relative weights among the three indicators "slope", "slope height" and "slope aspect".

[0177] The weights of the criteria relative to the target layer: For example, the relative importance weights of the three criteria, "topographic conditions", "rock and soil conditions" and "tectonic conditions", in achieving the overall goal of "potential geological hazard intensity index Zq".

[0178] Overall weighting: The final influence of an indicator (such as "slope") on the overall objective (Zq) needs to be passed through two levels of weighting. For example: The overall weight of slope on Zq = (weight of slope relative to terrain conditions) × (weight of terrain conditions relative to Zq).

[0179] Example: Assume the final calculation yields: Slope weight: 0.3, shear strength weight: 0.4 (inverse index, the smaller the value, the more likely it is to occur), fault density weight: 0.3.

[0180] For a given grid cell, its standardized slope is 0.8, shear strength is 0.2 (low strength, prone to faulting), and fault density is 0.6. Therefore, the Zq of this cell is 0.8. 0.3 + 0.2 0.4 + 0.6 0.3 = 0.24 + 0.08 + 0.18 = 0.50.

[0181] Spatialization: The above calculations are performed on each grid cell within the study area to obtain a Zq index map reflecting the spatial distribution of potential geological hazard intensity. The Zq index map is a grid map, and each grid cell (e.g., 100m × 100m) has a calculated Zq value (e.g., 0.5). When calculating the risk value P of a specific grid cell, the Zq in the formula is taken as the Zq value corresponding to that grid cell on the Zq index map.

[0182] The current geological hazard intensity index Zx reflects the sequelae of disasters that have occurred in the past or the possibility of recurrence.

[0183] The quantitative concept of the current geological hazard intensity index Zx is as follows: within a grid, the more historical hazard points there are, the larger their scale, and the more unstable (highly active) they are at present, the higher their Zx value will be, indicating that the current risk at that location is greater.

[0184] The density, scale, and activity of historical disaster points within the statistical grid cell are used to quantify existing disaster threats and obtain Zx, specifically including: Data preparation: Collect detailed information on historical geological disasters, including the location (coordinates), type, volume (scale), occurrence time, and current activity status (stable, unstable, reactivating) of each disaster point.

[0185] Spatial gridding: Dividing the study area into regular grids of preset size (e.g., 100m×100m).

[0186] Calculate the Zx value for each grid cell (usually a composite index): Disaster point density: Count the number of historical disaster points falling into each grid, and divide the number of historical disaster points by the area of ​​the grid to obtain the density D of the grid; Disaster Scale: Calculate the average or maximum volume of all historical disaster points falling within each grid, as the scale index V; Disaster activity: Based on the survey data, each historical disaster point is assigned an activity level A (e.g., stable=1, unstable=3, resurrection=5), and the average activity A_avg of all historical disaster points in each grid is calculated; Indicator standardization and weighted overlay: Standardize D, V, and A_avg of each grid separately (e.g., min-max normalization). Based on expert experience or AHP, assign corresponding weights Wd, Wv, and Wa to D, V, and A_avg; for example, consider Wa to be the highest weight for A_avg. Calculate the Zx value for each grid cell: Zx = Wd D_norm + Wv V_norm + a A_avg_norm is used to obtain a Zx index map that reflects the spatial distribution of the current status of historical disaster threats.

[0187] Where D is the density of disaster points (number / km²) 2 The original statistical value of ), where V is the average volume of the disaster point (m³). 3 The original statistical values ​​of D, V, and A_avg are the average activity levels of disaster points (e.g., 1, 3, 5). D_norm, V_norm, and A_avg_norm are the standardized (normalized) values ​​of D, V, and A_avg.

[0188] Step 502: Based on the preset risk P and risk classification rules, through risk... The risk classification results of geological hazards in the target monitoring area are obtained.

[0189] In another exemplary embodiment of this application, the rules for dividing risk P and risk classification results include: If P is located in (0, ... a [1] Within the interval, the risk classification result of geological disasters in the target monitoring area is low risk level, and a1 is the first threshold; low risk level indicates that the risk of geological disasters is low; If P is located at ( a 1, a [2] Within the interval, the risk classification result of geological disasters in the target monitoring area is medium risk level, and a2 is the second threshold; medium risk level indicates that the risk of geological disasters is relatively high; If P is located at ( a Within the interval [2, 1], the risk classification result of geological disasters in the target monitoring area is high risk level; high risk level indicates the highest risk of geological disasters.

[0190] In this embodiment of the application, for a 1 and a 2. No specific limitations are set; you can configure it according to your actual needs. For example, set... a 1 = 0.3 a 2 = 0.5.

[0191] In another exemplary embodiment of this application, the above-described geological disaster prediction method based on big data analysis further includes: A risk level layer is overlaid on the three-dimensional topography of the target geological area, and the coordinates of areas with a risk level of high risk are marked.

[0192] In another exemplary embodiment of this application, the above-described geological disaster prediction method based on big data analysis further includes: Store the preprocessed geological monitoring data.

[0193] In another exemplary embodiment of this application, the disaster simulation model is constructed according to steps 601 to 603 below. Wherein: Step 601: Collect historical geological disaster data for the target monitoring area. The historical geological disaster data includes geological monitoring data before the occurrence of historical geological disasters and geological monitoring data at the time of the occurrence of historical geological disasters.

[0194] In this embodiment of the application, historical geological disasters refer to geological disasters that have occurred.

[0195] Step 602: Preprocess the historical geological disaster data to obtain corrected historical geological disaster data.

[0196] For details regarding the preprocessing in this embodiment, please refer to the descriptions in the above embodiments, which will not be repeated here.

[0197] Step 603: A disaster prediction model is constructed using numerical simulation methods (such as the finite element method or the discrete element method), and the constructed disaster prediction model is calibrated and verified by back analysis or parameter calibration based on the corrected historical geological disaster data.

[0198] In this embodiment of the application, the numerical model is responsible for calculating (outputting data such as stress, displacement, and failure). Then, these calculation results can be imported into animation software or professional scientific visualization tools to render and generate intuitive, pre-show animations for display.

[0199] Based on the same inventive concept, this application also provides a geological disaster management and forecasting system based on big data analysis for implementing the geological disaster forecasting method based on big data analysis described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more embodiments of the geological disaster management and forecasting system based on big data analysis provided below can be found in the limitations of the geological disaster forecasting method based on big data analysis described above, and will not be repeated here.

[0200] In one exemplary embodiment, such as Figure 2 As shown, a geological disaster management and forecasting system 70 based on big data analysis is provided, including: Detection unit 701 is used to acquire geological monitoring data of the target geological area; Processing unit 702 is used for: The acquired geological monitoring data is preprocessed to obtain corrected geological monitoring data; At least one pre-simulation operation must be performed until the pre-simulation indicates that the target geological hazard will occur in the target geological area; the pre-simulation operation includes: The corrected geological monitoring data is input into the constructed disaster prediction model, and the disaster prediction model predicts whether the target geological area will experience the target geological disaster in the future target period based on the corrected geological monitoring data. Notification unit 703 is used to issue early warning information when it is predicted that a target geological disaster will occur in the target geological area in the future target period.

[0201] In this application embodiment, for the disaster prediction model, and the disaster prediction model constructed by inputting the modified geological monitoring data, and the relevant introduction of predicting whether the target geological disaster will occur in the target geological area in the future target period based on the modified geological monitoring data, please refer to the description of the above method embodiment, which will not be repeated here.

[0202] In another exemplary embodiment of this application, the geological monitoring data includes geological factor data, climate factor data, and human engineering activity data.

[0203] For details regarding geological factor data, climate factor data, and human engineering activity data in this application embodiment, please refer to the description of the above method embodiment.

[0204] Accordingly, the detection unit 701 includes: Geological disaster detection equipment is used to acquire real-time monitoring data of geological factors; Climate monitoring equipment is used to acquire data on climate factors; Human engineering activity monitoring equipment, used to acquire data on human engineering activities.

[0205] In another exemplary embodiment of this application, the geological factor data includes surface displacement, subsurface displacement, water level, and seismic data.

[0206] Accordingly, geological hazard detection equipment includes: Surface displacement monitoring equipment is used to acquire surface displacement data. Underground displacement monitoring equipment is used to acquire underground displacement data. Hydrological monitoring equipment is used to acquire water level data; Earthquake monitoring instruments are used to acquire earthquake data.

[0207] In this embodiment of the application, the seismic data includes seismic wave type and seismic intensity.

[0208] In another exemplary embodiment of this application, the surface displacement monitoring device adopts any one of the following: GNSS receiver, total station, laser displacement meter, crack gauge, inclinometer, crack meter, and hydrostatic level.

[0209] In another exemplary embodiment of this application, the meteorological factor data includes weather forecast data and measured weather data, and the measured weather data includes wind direction, wind speed, air pressure, rainfall, sunshine duration and solar radiation intensity.

[0210] Accordingly, climate monitoring equipment includes: Weather forecast monitoring equipment is used to acquire weather forecast data in real time; Wind direction monitoring equipment is used to obtain wind direction in real time; Wind speed monitoring equipment is used to obtain wind speed in real time; Barometric pressure monitoring equipment is used to obtain barometric pressure in real time. Rainfall monitoring equipment is used to obtain rainfall data in real time. Sunshine monitoring equipment is used to obtain sunshine duration in real time; Radiation monitoring equipment is used to obtain real-time solar radiation intensity.

[0211] In another exemplary embodiment of this application, the meteorological factor data also includes multi-source weather forecast data. Accordingly, the weather forecast monitoring equipment includes multi-source weather forecast monitoring equipment, such as automatic weather stations, satellite weather forecast monitoring equipment, and radar weather forecast monitoring equipment.

[0212] In another exemplary embodiment of this application, the human engineering activity data includes human engineering activity monitoring video. Accordingly, the human engineering activity monitoring device includes: Video surveillance systems are used to acquire video footage of human engineering activities.

[0213] In another exemplary embodiment of this application, such as Figure 3 As shown, the processing unit 702 includes: The preprocessing module 7021 is used to preprocess the acquired geological monitoring data to obtain corrected geological monitoring data; The virtual module 7022 is used to perform at least one pre-simulation operation until the pre-simulation determines that a target geological disaster will occur in the target geological area.

[0214] In this embodiment of the application, the acquisition of geological monitoring data is preprocessed to obtain corrected geological monitoring data. For details, please refer to the description of the above method embodiments, which will not be repeated here.

[0215] In another exemplary embodiment of this application, the virtual module 7022 is further configured to: After performing at least one pre-simulation operation until the target geological area is found to be prone to a target geological disaster, if the target geological area is found to be prone to a target geological disaster, the probability and severity of the target geological disaster will be output.

[0216] In another exemplary embodiment of this application, such as Figure 3 As shown, the processing unit 702 further includes: The judgment module 7023 is used to classify the risk based on the probability and severity of the target geological disaster, and obtain the risk classification result of the geological disaster in the target monitoring area.

[0217] In another exemplary embodiment of this application, the determination module 7023 is further configured to: Calculate the risk according to the following formula. : ; in, This indicates the probability of accidents occurring during human engineering activities. This indicates the severity of accidents that occur during human engineering activities. This indicates the likelihood of an accident caused by geological factors. This indicates the severity of the accident caused by geological factors. This indicates the likelihood of an accident caused by meteorological factors. Z represents the severity of accidents caused by meteorological factors, and Z represents the comprehensive susceptibility index of geological disasters. Based on risk The risk classification results of geological hazards in the target monitoring area are obtained.

[0218] In this application embodiment, for details regarding the probability of accidents occurring during human engineering activities, the severity of accidents occurring during human engineering activities, the probability of accidents occurring due to geological factors, the severity of accidents occurring due to geological factors, the probability of accidents occurring due to meteorological factors, the severity of accidents occurring due to meteorological factors, and the comprehensive susceptibility index of geological disasters, please refer to the descriptions in the above method embodiments, which will not be repeated here.

[0219] In another exemplary embodiment of this application, the determination module 7023 is further configured to: If P is located in (0, ... a [1] Within the interval, the risk classification result of geological disasters in the target monitoring area is low risk level, and a1 is the first threshold; low risk level indicates that the risk of geological disasters is low; If P is located at ( a 1, a [2] Within the interval, the risk classification result of geological disasters in the target monitoring area is medium risk level, and a2 is the second threshold; medium risk level indicates that the risk of geological disasters is relatively high; If P is located at ( a Within the interval [2, 1], the risk classification result of geological disasters in the target monitoring area is high-risk; high-risk level indicates the highest risk of geological disasters.

[0220] In this embodiment of the application, fora 1 and a 2. No specific limitations are set; you can configure it according to your actual needs. For example, set... a 1 = 0.3 a 2 = 0.5.

[0221] In another exemplary embodiment of this application, the virtual module 7022 is further configured to: Build a disaster simulation model by following these steps: Collect historical geological disaster data for the target monitoring area. The historical geological disaster data includes geological monitoring data before the occurrence of historical geological disasters and geological monitoring data at the time of the occurrence of historical geological disasters. Historical geological disaster data is preprocessed to obtain corrected historical geological disaster data; Numerical simulation methods (such as finite element method and discrete element method) are used to construct disaster prediction models. Based on corrected historical geological disaster data, the constructed disaster prediction models are calibrated and verified through back analysis or parameter calibration.

[0222] In another exemplary embodiment of this application, such as Figure 2 As shown, the geological disaster management and forecasting system 70 based on big data analysis also includes: Database 704 is used to store the preprocessed geological monitoring data output by processing unit 702.

[0223] In another exemplary embodiment of this application, database 704 employs a cloud-based database.

[0224] In another exemplary embodiment of this application, such as Figure 2 As shown, the geological disaster management and forecasting system 70 based on big data analysis also includes: The first transmission unit 705 is used to transmit the geological monitoring data output by the detection unit 701 to the processing unit 702.

[0225] In this embodiment of the application, the input terminal of the first transmission unit 705 is communicatively connected to the output terminal of the detection unit 701, and the output terminal of the first transmission unit 705 is communicatively connected to the input terminal of the processing unit 702.

[0226] In another exemplary embodiment of this application, such as Figure 2 As shown, the geological disaster management and forecasting system 70 based on big data analysis also includes: The second transmission unit 706 is used to transmit the current geological disaster data output by the processing unit 702 to the notification unit 703.

[0227] In this embodiment of the application, the input terminal of the second transmission unit 706 is communicatively connected to the output terminal of the processing unit 702, and the output terminal of the second transmission unit 706 is communicatively connected to the input terminal of the notification unit 703.

[0228] In another exemplary embodiment of this application, the first transmission unit 705 and the second transmission unit 706 employ a network server.

[0229] In another exemplary embodiment of this application, the processing unit 702 employs server 104.

[0230] In another exemplary embodiment of this application, the notification unit 703 employs at least one of a speaker and a terminal 102.

[0231] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores geological monitoring data and geological hazard data. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a geological hazard prediction method based on big data analysis.

[0232] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0233] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0234] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0235] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0236] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0237] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0238] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0239] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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, they should be considered to be within the scope of this specification.

[0240] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A geological disaster forecasting method based on big data analysis, characterized in that, The geological disaster forecasting method based on big data analysis includes: Obtain geological monitoring data for the target geological area; The acquired geological monitoring data is preprocessed to obtain corrected geological monitoring data; At least one pre-simulation operation shall be performed until the pre-simulation indicates that a target geological hazard will occur in the target geological area; wherein, the pre-simulation operation includes: The revised geological monitoring data is input into the constructed disaster prediction model, which then predicts whether the target geological disaster will occur in the target geological area in the future target period based on the revised geological monitoring data. If a geological disaster is predicted to occur in the target geological area during a future target time period, an early warning message will be issued.

2. The geological disaster forecasting method based on big data analysis according to claim 1, characterized in that, The geological monitoring data includes geological factor data, meteorological factor data, and human engineering activity data; The geological disaster forecasting method based on big data analysis also includes: If the pre-simulation indicates that a target geological disaster will occur in the target geological area, the probability and severity of the target geological disaster will be output. The probability of the target geological disaster includes the probability of accidents caused by human engineering activities, the probability of accidents caused by geological factors, and the probability of accidents caused by meteorological factors. The severity of the target geological disaster includes the severity of accidents caused by human engineering activities, the severity of accidents caused by geological factors, and the severity of accidents caused by meteorological factors. The geological hazard data also includes the probability and severity of the target geological hazard occurring.

3. The geological disaster forecasting method based on big data analysis according to claim 2, characterized in that, Also includes: Based on the probability and severity of the target geological disaster, a risk classification is performed to obtain the risk classification results of the geological disaster in the target monitoring area; The geological disaster data also includes the risk classification results of geological disasters in the target monitoring area.

4. The geological disaster forecasting method based on big data analysis according to claim 3, characterized in that, The risk classification based on the probability and severity of the target geological disaster, to obtain the geological disaster risk classification result for the target monitoring area, includes: Calculate the risk according to the following formula. : ； in, This indicates the probability of accidents occurring during human engineering activities. This indicates the severity of accidents that occur during human engineering activities. This indicates the likelihood of an accident caused by geological factors. This indicates the severity of the accident caused by geological factors. This indicates the likelihood of an accident caused by meteorological factors. Z represents the severity of accidents caused by meteorological factors, and Z represents the comprehensive susceptibility index of geological disasters. Based on risk The risk classification results of geological hazards in the target monitoring area are obtained.

5. The geological disaster forecasting method based on big data analysis according to claim 4, characterized in that, According to risk The risk classification results of geological hazards in the target monitoring area are obtained, including: If P is located in (0, ... a If the geological hazard risk classification result of the target monitoring area is within the interval [1], then the risk level is low. a 1 represents the first threshold; If P is located at ( a 1, a 2] Within the specified interval, the geological hazard risk classification result for the target monitoring area is medium risk. a 2 is the second threshold; If P is located at ( a If the geological hazard risk classification result of the target monitoring area is within the interval [2, 1], then the geological hazard risk classification result of the target monitoring area is high risk level.

6. The geological disaster forecasting method based on big data analysis according to claim 4, characterized in that, The comprehensive geological hazard susceptibility index is calculated according to the following formula: = + ； in, Indicates the potential geological hazard intensity index. This indicates the current geological hazard intensity index; , Represents the weighting coefficient, reflecting and Differences in contributions.

7. The geological disaster forecasting method based on big data analysis according to claim 1, characterized in that, The process of inputting the corrected geological monitoring data into the constructed disaster prediction model, and using the disaster prediction model to predict whether a target geological disaster will occur in the target geological area in a future target period based on the corrected geological monitoring data, specifically includes: The corrected geological monitoring data is assigned to the three-dimensional grid nodes of the disaster prediction model; The disaster simulation model is based on modified geological monitoring data. It simulates the stress distribution and deformation of the geological body in the target monitoring area under the influence of gravity, seepage, earthquake motion and / or human activities in the future target period, and simulates the dynamic impact of groundwater on rock mass stability, outputting key parameters. Check whether the key parameters output by the disaster simulation model exceed the preset threshold: If the key parameters output by the disaster prediction model exceed the preset threshold, it is assumed that the target geological area will experience the target geological disaster in the future target time period. Otherwise, it is assumed that the target geological area will not experience the target geological disaster in the future target time period.

8. A geological disaster forecasting system based on big data analysis, characterized in that, The geological disaster forecasting system based on big data analysis includes: The detection unit is used to acquire geological monitoring data of the target geological area; Processing unit, used for: The real-time acquired geological monitoring data is preprocessed to obtain corrected geological monitoring data; At least one pre-simulation operation shall be performed until the pre-simulation indicates that a target geological hazard will occur in the target geological area; wherein, the pre-simulation operation includes: The revised geological monitoring data is input into the constructed disaster prediction model, which then predicts whether the target geological disaster will occur in the target geological area in the future target period based on the revised geological monitoring data. The notification unit is used to issue early warning information when a geological disaster is predicted to occur in the target geological area during a future target period.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the geological disaster forecasting method based on big data analysis as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the geological disaster prediction method based on big data analysis as described in any one of claims 1-7.

Images