Geological disaster early warning method and related device

Abstract

The invention discloses a geological disaster early warning method, a geological disaster early warning system, geological disaster early warning equipment and a computer readable storage medium, and the method comprises the steps: carrying out the feature data collection of a target region, and carrying out the data collection through a ground deformation monitoring network and underground water dynamic monitoring equipment, performing feature parameter extraction based on the plateau environment basic data set, the mining feature data set, the plateau environment correction coefficient, the ground deformation monitoring data set and the underground water dynamic data set to obtain a fusion feature vector; adopting a machine learning algorithm, constructing a plateau rock-soil body constitutive model, and fusing the feature vectors to construct and train a settlement prediction model to obtain the settlement prediction model; reasoning the fusion feature vector based on a settlement prediction model to obtain a ground settlement prediction value and spatial distribution thereof; and generating early warning information based on the land subsidence predicted value and the spatial distribution thereof, and issuing the early warning information. And the early warning accuracy of geological disasters is improved.

Description

Technical Field

[0001] This invention belongs to the field of geological disaster early warning technology, and more specifically, relates to a geological disaster early warning method, a geological disaster early warning system, a geological disaster early warning device, and a computer-readable storage medium. Background Technology

[0002] As a clean energy source, geothermal resources have significant applications in heating and hot spring development in plateau regions. With economic development and increasing energy demand in plateau areas, the scale of geothermal extraction is continuously expanding. However, large-scale extraction of geothermal water leads to a drop in groundwater levels and a decrease in pore water pressure, which in turn triggers geological disasters such as land subsidence and ground fissures, seriously threatening the safety of infrastructure and the stability of the ecological environment in plateau regions.

[0003] The relevant technologies, primarily developed for plains or low-altitude areas, suffer from the following problems: They fail to fully consider the impact of unique environmental factors such as low air pressure, large temperature differences, and strong radiation on the mechanical properties of soil and rock, resulting in poor applicability of early warning models in plateau regions; they neglect the impact of unique geological phenomena such as freeze-thaw cycles and seasonal permafrost on ground stability, leading to low early warning accuracy; existing monitoring methods are not optimized for the harsh plateau environment, making data collection difficult and unreliable; and they lack a risk assessment system that considers plateau environmental corrections, resulting in unreasonable early warning thresholds and a high risk of false alarms or missed alarms. These factors contribute to the low accuracy of geological disaster early warning technologies applied to plateau regions.

[0004] Therefore, how to develop a geological disaster early warning method suitable for underground hot water extraction in plateau environments and improve the accuracy of early warning for geological disasters caused by underground hot water extraction in plateau areas is a key issue of concern to those skilled in the art. Summary of the Invention

[0005] The purpose of this application is to provide a geological disaster early warning method, a geological disaster early warning system, a geological disaster early warning device, and a computer-readable storage medium to improve the accuracy of early warning for geological disasters caused by underground hot water extraction in plateau areas.

[0006] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a geological disaster early warning method, comprising: Feature data was collected from the target area to obtain a basic dataset of the plateau environment, a dataset of mining features, and plateau environment correction coefficients. Data was collected through a ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain ground deformation monitoring datasets and groundwater dynamic datasets; Feature parameters are extracted based on the plateau environment basic dataset, the mining feature dataset, the plateau environment correction coefficient, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain a fused feature vector; A settlement prediction model was constructed and trained using machine learning algorithms, a constitutive model of plateau rock and soil, and the fused feature vectors, resulting in a settlement prediction model. Based on the settlement prediction model, the fused feature vector is inferred to obtain the predicted ground settlement value and its spatial distribution. Early warning information is generated and released based on the predicted ground subsidence values ​​and their spatial distribution.

[0007] Optionally, feature data is collected from the target area to obtain a basic dataset of the plateau environment, a mining feature dataset, and plateau environment correction coefficients, including: Based on the geographic coordinate range of the target area, obtain elevation data, terrain slope, aspect data, atmospheric pressure, oxygen content, and diurnal temperature range, and use them as the basic dataset for the plateau environment. Data extraction was performed based on the acquired historical data of underground hot water extraction to obtain an extraction characteristic dataset; The plateau environment correction coefficient is calculated based on the plateau environment basic dataset.

[0008] Optionally, feature parameters are extracted based on the plateau environment basic dataset, the mining feature dataset, the plateau environment correction coefficient, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain a fused feature vector, including: Based on the plateau environment correction coefficient, the ground deformation monitoring dataset is corrected to obtain deformation data; A mining deformation correlation model is constructed based on the mining feature dataset and the deformation data; Feature parameters are extracted from the plateau environment dataset, the mining feature dataset, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain deformation rate, deformation acceleration, and basic feature parameters; The mining deformation correlation model, the deformation rate, the deformation acceleration, and the basic feature parameters are used as the fused feature vector.

[0009] Optionally, a settlement prediction model is constructed and trained using machine learning algorithms, a constructed constitutive model of plateau rock and soil, and the fused feature vectors, resulting in a settlement prediction model, including: Constitutive equations were constructed based on the physical and mechanical parameters of the soil and rock mass and the plateau environment correction coefficients, resulting in a parameter set for the constitutive model of the plateau soil and rock mass. The settlement prediction model is constructed and trained based on machine learning algorithms, the parameter set of the constitutive model of the plateau rock and soil, and the fused feature vector, and the settlement prediction model is obtained.

[0010] This application also provides a geological disaster early warning system, including: The regional data acquisition module is used to collect feature data of the target area to obtain the plateau environment basic dataset, mining feature dataset, and plateau environment correction coefficient; The ground deformation data acquisition module is used to acquire data through the ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain ground deformation monitoring datasets and groundwater dynamic datasets. The feature parameter extraction module is used to extract feature parameters based on the plateau environment basic dataset, the mining feature dataset, the plateau environment correction coefficient, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain a fused feature vector; The model building module is used to construct and train a settlement prediction model by employing machine learning algorithms, a constructed constitutive model of plateau rock and soil, and the fused feature vectors, thereby obtaining a settlement prediction model. The model inference module is used to infer the fused feature vector based on the settlement prediction model to obtain the predicted ground settlement value and its spatial distribution. The early warning information generation module is used to generate early warning information based on the predicted ground subsidence value and its spatial distribution, and to release the early warning information.

[0011] Optionally, the regional data acquisition module is specifically used to acquire elevation data, terrain slope, aspect data, atmospheric pressure, oxygen content, and diurnal temperature range based on the geographical coordinate range of the target area, and use these as the basic dataset of the plateau environment; extract data based on the acquired historical data of underground hot water extraction to obtain an extraction feature dataset; and calculate the plateau environment correction coefficient based on the basic dataset of the plateau environment.

[0012] Optionally, the feature parameter extraction module is specifically used to correct the ground deformation monitoring dataset based on the plateau environment correction coefficient to obtain deformation data; construct a mining deformation correlation model based on the mining feature dataset and the deformation data; extract feature parameters from the plateau environment basic dataset, the mining feature dataset, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain deformation rate, deformation acceleration, and basic feature parameters; and use the mining deformation correlation model, the deformation rate, the deformation acceleration, and the basic feature parameters as the fused feature vector.

[0013] Optionally, the model building module is specifically used to construct constitutive equations based on the physical and mechanical parameters of the soil and rock mass and the plateau environment correction coefficient to obtain a constitutive model parameter set for the plateau soil and rock mass; and to construct and train a settlement prediction model based on machine learning algorithms, the constitutive model parameter set for the plateau soil and rock mass, and the fused feature vector to obtain the settlement prediction model.

[0014] This application also provides a geological disaster early warning device, including: Memory, used to store computer programs; A processor is used to execute the computer program to implement the steps of the geological disaster early warning method as described above.

[0015] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the geological disaster early warning method described above.

[0016] This application provides a geological disaster early warning method, comprising: collecting feature data of a target area to obtain a plateau environmental basic dataset, a mining feature dataset, and a plateau environmental correction coefficient; collecting data through a ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain a ground deformation monitoring dataset and a groundwater dynamic dataset; extracting feature parameters based on the plateau environmental basic dataset, the mining feature dataset, the plateau environmental correction coefficient, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain a fused feature vector; constructing and training a settlement prediction model using a machine learning algorithm, a constructed plateau rock and soil constitutive model, and the fused feature vector to obtain a settlement prediction model; inferring from the fused feature vector based on the settlement prediction model to obtain ground settlement prediction values ​​and their spatial distribution; generating early warning information based on the ground settlement prediction values ​​and their spatial distribution, and issuing the early warning information.

[0017] It has the following beneficial effects: By comprehensively collecting feature data from the target area, the impact of the unique plateau environment on the development of geological disasters was fully considered. A plateau environment correction coefficient was introduced to correct the monitoring data, improving its accuracy and reliability. A three-dimensional monitoring system for ground deformation and groundwater dynamics was established, enabling monitoring of the geological disaster development process. Multi-source data fusion technology was used to extract comprehensive feature vectors, and a prediction model was constructed by combining machine learning algorithms and a plateau soil and rock constitutive model. This ensured the physical rationality of the predictions while enhancing the model's learning ability and prediction accuracy. The overall solution is fully adapted to the characteristics of the plateau environment, providing scientific, accurate, and timely geological disaster early warning services for geothermal water extraction in plateau areas, effectively ensuring the safety and sustainability of geothermal resource development. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a geological disaster early warning method provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of a geological disaster early warning system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the geological disaster early warning device provided in the embodiments of this application. Detailed Implementation

[0020] The purpose of this application is to provide a geological disaster early warning method, a geological disaster early warning system, a geological disaster early warning device, and a computer-readable storage medium to improve the accuracy of early warning for geological disasters caused by underground hot water extraction in plateau areas.

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0022] The following embodiment illustrates a geological disaster early warning method provided in this application.

[0023] Please refer to Figure 1 , Figure 1 This is a flowchart of a geological disaster early warning method provided in an embodiment of this application.

[0024] In this embodiment, the method may include: S101, collect feature data of the target area to obtain the plateau environment basic dataset, mining feature dataset, and plateau environment correction coefficient; This step aims to collect characteristic data of the target area to obtain a basic dataset of the plateau environment, a dataset of mining characteristics, and plateau environmental correction coefficients. In other words, it comprehensively acquires the basic data and environmental parameters affecting the development of geological disasters in plateau regions. Furthermore, considering the unique environmental characteristics of plateau regions, such as high altitude, low air pressure, low oxygen content, large temperature differences, and strong ultraviolet radiation, these factors significantly affect the physical and mechanical properties of soil and rock masses and the migration patterns of groundwater; therefore, it is essential to collect these characteristic data first.

[0025] Optionally, this step may include: Step 1: Based on the geographic coordinate range of the target area, obtain elevation data, terrain slope, aspect data, atmospheric pressure, oxygen content, and diurnal temperature range, and use them as the basic dataset for the plateau environment. Step 2: Extract data based on the acquired historical data of underground hot water extraction to obtain an extraction feature dataset; Step 3: Calculate the plateau environment correction coefficient based on the plateau environment basic dataset.

[0026] Optionally, this step may also include: acquiring elevation data, terrain slope and aspect information of the target area through remote sensing satellites, and collecting atmospheric parameters such as atmospheric pressure, oxygen concentration and diurnal temperature range using a network of meteorological stations to form a basic dataset of plateau environment; Simultaneously, the location coordinates and mining history of each hot water well are obtained from the geothermal extraction management department. Data such as the time series of extraction volume, water temperature change series, and water level change series are extracted to calculate the cumulative extraction volume and the trend of extraction rate, forming an extraction characteristic dataset. Based on the collected plateau environmental parameters, a plateau environmental correction coefficient calculation model was established, which takes into account factors such as the influence of atmospheric pressure on the effective stress of soil and rock, the thermal expansion and contraction effect caused by the diurnal temperature range, and the impact of low oxygen environment on the performance of monitoring equipment. The plateau environmental correction coefficient applicable to the target area was obtained by comprehensive calculation through mathematical formulas.

[0027] It is evident that by implementing this step, basic data reflecting the unique environment and mining activities of the plateau can be obtained, providing necessary input parameters for subsequent monitoring data correction and model construction, and ensuring that the early warning method can adapt to the special characteristics of the plateau environment.

[0028] S102, data is collected through the ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain ground deformation monitoring dataset and groundwater dynamic dataset; Building upon S101, this step aims to collect data through a surface deformation monitoring network and groundwater dynamic monitoring equipment to obtain surface deformation monitoring datasets and groundwater dynamic datasets. Specifically, this step establishes a multi-source, collaborative, three-dimensional monitoring system to acquire direct data characterizing the development of geological hazards in real time. Land subsidence is the most direct manifestation of geological hazards caused by geothermal water extraction, and changes in groundwater level and pressure are the root causes of land subsidence; therefore, it is necessary to monitor both surface deformation and groundwater dynamic changes simultaneously.

[0029] Optionally, this step may include: In terms of ground deformation monitoring, a monitoring network is constructed using a variety of technologies, including GNSS continuously operating reference stations, precise leveling, and InSAR satellite remote sensing. GNSS stations are deployed in the mining area and its influence range according to the grid principle to record three-dimensional displacement changes in real time. Leveling benchmarks are set up along the main settlement direction and elevation measurements are carried out regularly. InSAR technology provides large-scale areal deformation information. The three technologies verify and complement each other to form a complete ground deformation monitoring dataset.

[0030] In terms of groundwater dynamic monitoring, monitoring wells are reasonably deployed around hot water extraction wells, and high-precision water level gauges, water pressure sensors, temperature sensors and other equipment are installed to monitor changes in groundwater level, pore water pressure and water temperature in real time; at the same time, stress and strain sensors are installed at key locations to monitor changes in the stress state of the rock and soil mass and form a groundwater dynamic dataset.

[0031] It is evident that by establishing such an integrated air-space-ground monitoring network, we can comprehensively grasp the development process and evolution trend of geological disasters, providing real-time and reliable data support for accurate early warning.

[0032] S103, based on the plateau environment basic dataset, mining characteristic dataset, plateau environment correction coefficient, ground deformation monitoring dataset, and groundwater dynamic dataset, feature parameters are extracted to obtain a fused feature vector; Building upon S102, this step aims to extract feature parameters from the plateau environment basic dataset, mining characteristic dataset, plateau environment correction coefficient, ground deformation monitoring dataset, and groundwater dynamic dataset to obtain a fused feature vector. This step fuses and extracts features from multi-source heterogeneous data to generate a feature vector that comprehensively reflects the development status of geological hazards. Raw monitoring data often contains noise and redundant information, and the spatiotemporal resolution of different data sources is inconsistent, requiring data preprocessing and feature extraction before it can be used for model training.

[0033] Optionally, this step may include: Step 1: Correct the ground deformation monitoring dataset based on the plateau environment correction coefficient to obtain deformation data; Step 2: Construct a mining deformation correlation model based on the mining feature dataset and deformation data; Step 3: Extract feature parameters from the plateau environment basic dataset, mining feature dataset, ground deformation monitoring dataset, and groundwater dynamic dataset to obtain deformation rate, deformation acceleration, and basic feature parameters; Step 4: Use the mining deformation correlation model, deformation rate, deformation acceleration, and basic feature parameters as the fusion feature vector.

[0034] Optionally, this step may also include: First, the ground deformation monitoring data is corrected for the plateau environment. A plateau environment correction coefficient is used to eliminate the influence of factors such as low air pressure and large temperature differences on deformation measurements, thus obtaining the true ground subsidence. Then, time series analysis is employed to extract characteristic parameters such as subsidence rate, subsidence acceleration, and subsidence range from the corrected deformation data. For groundwater dynamic data, derived characteristics such as water level drop rate, water pressure gradient, and stress concentration factor are calculated.

[0035] By spatiotemporally matching mining characteristic data with monitoring data, the correlation between mining volume and deformation is established, and comprehensive features such as mining intensity index and influence radius are extracted. Principal component analysis and correlation analysis are used to reduce the dimensionality and optimize the extracted features, eliminating redundant features and retaining the feature parameters most sensitive to the development of geological hazards, ultimately forming a fused feature vector that includes deformation features, hydrological features, mining features, and environmental features.

[0036] This step of data fusion and feature extraction can transform multi-source monitoring data into structured feature representations, improving the information density and usability of the data and providing high-quality input for machine learning models.

[0037] S104. A settlement prediction model was constructed and trained using machine learning algorithms, a constitutive model of plateau rock and soil, and fused feature vectors to obtain the settlement prediction model. Building upon S103, this step aims to construct and train a settlement prediction model using machine learning algorithms, a constructed constitutive model of plateau soil and rock, and fused feature vectors. This step establishes a hybrid prediction model that integrates physical mechanisms and data-driven approaches to achieve accurate prediction of ground subsidence. Purely physical models are insufficient to fully describe complex geological processes, while purely data-driven models lack physical constraints; therefore, a physical-data fusion modeling strategy is adopted.

[0038] Optionally, this step may include: Step 1: Based on the physical and mechanical parameters of the soil and rock mass and the plateau environment correction coefficient, the constitutive equation is constructed to obtain the parameter set of the constitutive model of the plateau soil and rock mass; Step 2: Based on machine learning algorithms, the parameter set of the constitutive model of plateau rock and soil, and the fused feature vector, a settlement prediction model is constructed and trained to obtain the settlement prediction model.

[0039] Optionally, this step may also include: First, based on the theory of rock and soil mechanics, considering the special mechanical behavior of rock and soil in the low-pressure environment of the plateau, a modified constitutive model of plateau rock and soil is established. This model includes physical mechanisms such as stress-strain relationship, consolidation theory, and thermo-mechanical coupling effect, and introduces plateau environment correction parameters.

[0040] Then, suitable machine learning algorithms for spatiotemporal prediction are selected, such as Long Short-Term Memory (LSTM), Random Forest (RF), Support Vector Machine (SVM), etc., or ensemble learning methods are used to combine multiple algorithms. The output of the plateau soil and rock constitutive model is used as the prior knowledge and constraints of the machine learning model, and feature vectors are fused as input, with historical ground subsidence data as labels, for model training.

[0041] During training, methods such as cross-validation and grid search are used to optimize model hyperparameters, and regularization techniques are used to prevent overfitting.

[0042] By using physical models to provide mechanistic constraints and machine learning models to capture complex nonlinear relationships, the two complement each other to form a subsidence prediction model with good generalization ability and prediction accuracy. Implementing this step enables the establishment of an intelligent prediction model adapted to the characteristics of the plateau environment and integrating multi-source information, providing technical support for early warning of geological disasters.

[0043] S105, Based on the settlement prediction model, the fused feature vector is inferred to obtain the predicted ground settlement value and its spatial distribution; Building upon S104, this step aims to infer the fused feature vectors based on the settlement prediction model to obtain the predicted ground settlement value and its spatial distribution. This step utilizes the trained prediction model to perform inference calculations on real-time monitoring data to predict the development trend of ground settlement over a certain period in the future.

[0044] Optionally, this step may include: The real-time collected monitoring data undergoes the same preprocessing and feature extraction process as the model training to generate a real-time fused feature vector. This feature vector is then input into the settlement prediction model, which, based on the learned geological hazard evolution patterns and the current state, predicts the ground settlement at different time scales (e.g., 24 hours, 7 days, 30 days).

[0045] The prediction process not only provides settlement values ​​for individual points, but also generates a spatial distribution map of settlement prediction for the entire target area by combining spatial interpolation and geostatistical analysis methods with the spatial distribution of monitoring points and the spatial heterogeneity of geological conditions. This distribution map can visually display the location, extent, and depth of settlement funnels and identify dangerous areas with large settlement gradients.

[0046] Simultaneously, the model also outputs confidence intervals for the prediction results, quantifying the uncertainty of the prediction. For the locations of critical infrastructure, it provides refined predictions, offering settlement prediction information with higher spatiotemporal resolution.

[0047] Through the reasoning and calculation in this step, we can grasp the development trend and spatial distribution characteristics of ground subsidence in advance, providing a scientific basis for formulating targeted disaster prevention and mitigation measures, and realizing the transformation from passive response to proactive prevention.

[0048] S106 generates and issues early warning information based on predicted ground subsidence values ​​and their spatial distribution.

[0049] Building upon S105, this step aims to generate and issue early warning information based on predicted ground subsidence values ​​and their spatial distribution. This step transforms the prediction results into actionable early warning information and disseminates it promptly through appropriate channels, ultimately realizing the implementation of the early warning function.

[0050] Optionally, this step may include: Based on predicted ground subsidence values ​​and their spatial distribution, combined with the deformation resistance of buildings and infrastructure in plateau areas, and referring to relevant national and local technical standards, a tiered early warning system is established. This system typically includes four warning levels: blue, yellow, orange, and red, corresponding to different subsidence rates and risk levels. The generation of early warning information includes not only the warning level but also the affected area, the expected duration of impact, potential hazards, and recommended countermeasures.

[0051] Considering the unique communication conditions in plateau regions, a multi-channel information dissemination strategy is adopted, including issuing official warnings through the government emergency management platform, sending warning information to relevant responsible persons and affected residents via mobile phone text messages and APP push notifications, setting up warning display screens in key areas to display the warning status in real time, and expanding the warning coverage through traditional media such as radio and television.

[0052] Customized warning content is generated for different warning targets, such as providing detailed technical analysis reports to government departments, providing mining control suggestions to enterprises, and providing easy-to-understand risk avoidance guidelines to the public.

[0053] By establishing a sound mechanism for generating and disseminating early warning information, it is possible to ensure that early warning information is delivered to relevant parties in a timely, accurate, and effective manner, thereby minimizing losses caused by geological disasters and safeguarding the lives and property of people in plateau regions as well as the sustainable use of geothermal resources.

[0054] In summary, this embodiment comprehensively collects feature data from the target area, fully considering the impact of the unique plateau environment on the development of geological disasters. A plateau environment correction coefficient is introduced to correct the monitoring data, improving its accuracy and reliability. By establishing a collaborative monitoring system for ground deformation and groundwater dynamics, three-dimensional monitoring of the geological disaster development process is achieved. Multi-source data fusion technology is used to extract comprehensive feature vectors, and a prediction model is constructed by combining machine learning algorithms and a plateau soil and rock constitutive model. This ensures the physical rationality of the predictions while enhancing the model's learning ability and prediction accuracy. The overall solution is fully adapted to the characteristics of the plateau environment, providing scientific, accurate, and timely geological disaster early warning services for geothermal water extraction in plateau areas, effectively ensuring the safety and sustainability of geothermal resource development.

[0055] The following describes a geological disaster early warning system provided by an embodiment of this application. The geological disaster early warning system and geological disaster early warning method described below can be referred to in correspondence with each other.

[0056] Please refer to Figure 2 , Figure 2 This is a schematic diagram of the structure of a geological disaster early warning system provided in an embodiment of this application.

[0057] In this embodiment, the system may include: The regional data acquisition module 100 is used to collect feature data of the target area to obtain the plateau environment basic dataset, mining feature dataset, and plateau environment correction coefficient. The ground deformation data acquisition module 200 is used to acquire data through the ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain ground deformation monitoring dataset and groundwater dynamic dataset. The feature parameter extraction module 300 is used to extract feature parameters based on the plateau environment basic dataset, mining feature dataset, plateau environment correction coefficient, ground deformation monitoring dataset, and groundwater dynamic dataset to obtain a fused feature vector. The model building module 400 is used to build and train a settlement prediction model by using machine learning algorithms, constructing a constitutive model of plateau rock and soil, and fusing feature vectors to obtain a settlement prediction model. The model inference module 500 is used to infer the fused feature vector based on the settlement prediction model to obtain the predicted ground settlement value and its spatial distribution. The early warning information generation module 600 is used to generate and issue early warning information based on the predicted ground subsidence value and its spatial distribution.

[0058] Optionally, the regional data acquisition module is specifically used to acquire elevation data, terrain slope, aspect data, atmospheric pressure, oxygen content, and diurnal temperature range based on the geographic coordinate range of the target area, and to use this as the basic dataset for the plateau environment; to extract data based on the acquired historical data of underground hot water extraction to obtain the extraction characteristic dataset; and to calculate the plateau environment correction coefficient based on the basic dataset for the plateau environment.

[0059] Optionally, a feature parameter extraction module is used to correct the ground deformation monitoring dataset based on the plateau environment correction coefficient to obtain deformation data; construct a mining deformation correlation model based on the mining feature dataset and deformation data; extract feature parameters from the plateau environment basic dataset, mining feature dataset, ground deformation monitoring dataset, and groundwater dynamic dataset to obtain deformation rate, deformation acceleration, and basic feature parameters; and use the mining deformation correlation model, deformation rate, deformation acceleration, and basic feature parameters as a fusion feature vector.

[0060] Optionally, the model building module is specifically used to construct constitutive equations based on the physical and mechanical parameters of the soil and rock mass and the plateau environment correction coefficient, thereby obtaining a constitutive model parameter set for the plateau soil and rock mass; and to construct and train a settlement prediction model based on machine learning algorithms, the constitutive model parameter set for the plateau soil and rock mass, and fused feature vectors, thereby obtaining a settlement prediction model.

[0061] This application also provides geological disaster early warning equipment; please refer to it. Figure 3 , Figure 3 This is a schematic diagram of the geological disaster early warning device provided in the embodiments of this application. The geological disaster early warning device may include: Memory, used to store computer programs; A processor, used to execute computer programs, can implement the steps of any of the geological disaster early warning methods described above.

[0062] like Figure 3 The diagram shows the structural composition of a geological disaster early warning device. The device may include a processor 10, a memory 11, a communication interface 12, and a communication bus 13. The processor 10, memory 11, and communication interface 12 all communicate with each other via the communication bus 13.

[0063] In this embodiment, the processor 10 may be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field-programmable gate array, or other programmable logic devices.

[0064] The processor 10 can call the program stored in the memory 11. Specifically, the processor 10 can execute the operations in the embodiment of the abnormal IP identification method.

[0065] The memory 11 is used to store one or more programs. The programs may include program code, which includes computer operation instructions. In this embodiment, the memory 11 stores at least a program for implementing the following functions: Feature data was collected from the target area to obtain a basic dataset of the plateau environment, a dataset of mining features, and plateau environment correction coefficients. Data was collected through a ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain ground deformation monitoring datasets and groundwater dynamic datasets; Feature parameters were extracted from the plateau environment basic dataset, mining feature dataset, plateau environment correction coefficient, ground deformation monitoring dataset, and groundwater dynamic dataset to obtain a fused feature vector. A settlement prediction model was constructed and trained using machine learning algorithms, a constitutive model of plateau rock and soil, and fused feature vectors. Based on the settlement prediction model, the fused feature vector is inferred to obtain the predicted ground settlement value and its spatial distribution. Early warning information is generated and released based on the predicted values ​​of ground subsidence and their spatial distribution.

[0066] In one possible implementation, the memory 11 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function; and the data storage area may store data created during use.

[0067] In addition, memory 11 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device or other volatile solid-state storage device.

[0068] Communication interface 12 can be an interface for the communication module, used to connect with other devices or systems.

[0069] Of course, it should be noted that, Figure 3 The structure shown does not constitute a limitation on the geological disaster early warning device in the embodiments of this application. In practical applications, the geological disaster early warning device may include more than Figure 3 More or fewer components as shown, or combinations of certain components.

[0070] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement the steps of any of the geological disaster early warning methods described above.

[0071] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0072] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0073] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0074] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0075] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0076] The foregoing has provided a detailed description of a geological disaster early warning method, a geological disaster early warning system, a geological disaster early warning device, and a computer-readable storage medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A geological disaster early warning method, characterized in that, include: Feature data was collected from the target area to obtain a basic dataset of the plateau environment, a dataset of mining features, and plateau environment correction coefficients. Data was collected through a ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain ground deformation monitoring datasets and groundwater dynamic datasets; Feature parameters are extracted based on the plateau environment basic dataset, the mining feature dataset, the plateau environment correction coefficient, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain a fused feature vector; A settlement prediction model was constructed and trained using machine learning algorithms, a constitutive model of plateau rock and soil, and the fused feature vectors, resulting in a settlement prediction model. Based on the settlement prediction model, the fused feature vector is inferred to obtain the predicted ground settlement value and its spatial distribution. Early warning information is generated and released based on the predicted ground subsidence values ​​and their spatial distribution.

2. The geological disaster early warning method according to claim 1, characterized in that, Feature data was collected from the target area to obtain a basic dataset of the plateau environment, a mining feature dataset, and plateau environment correction coefficients, including: Based on the geographic coordinate range of the target area, obtain elevation data, terrain slope, aspect data, atmospheric pressure, oxygen content, and diurnal temperature range, and use them as the basic dataset for the plateau environment. Data extraction was performed based on the acquired historical data of underground hot water extraction to obtain an extraction characteristic dataset; The plateau environment correction coefficient is calculated based on the plateau environment basic dataset.

3. The geological disaster early warning method according to claim 1, characterized in that, Based on the aforementioned plateau environment basic dataset, mining feature dataset, plateau environment correction coefficient, ground deformation monitoring dataset, and groundwater dynamic dataset, feature parameters are extracted to obtain a fused feature vector, including: Based on the plateau environment correction coefficient, the ground deformation monitoring dataset is corrected to obtain deformation data; A mining deformation correlation model is constructed based on the mining feature dataset and the deformation data; Feature parameters are extracted from the plateau environment dataset, the mining feature dataset, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain deformation rate, deformation acceleration, and basic feature parameters; The mining deformation correlation model, the deformation rate, the deformation acceleration, and the basic feature parameters are used as the fused feature vector.

4. The geological disaster early warning method according to claim 1, characterized in that, A settlement prediction model is constructed and trained using machine learning algorithms, a constitutive model of plateau soil and rock mass, and the fused feature vectors, resulting in a settlement prediction model including: Constitutive equations were constructed based on the physical and mechanical parameters of the soil and rock mass and the plateau environment correction coefficients, resulting in a parameter set for the constitutive model of the plateau soil and rock mass. The settlement prediction model is constructed and trained based on machine learning algorithms, the parameter set of the constitutive model of the plateau rock and soil, and the fused feature vector, and the settlement prediction model is obtained.

5. A geological disaster early warning system, characterized in that, include: The regional data acquisition module is used to collect feature data of the target area to obtain the plateau environment basic dataset, mining feature dataset, and plateau environment correction coefficient; The ground deformation data acquisition module is used to acquire data through the ground deformation monitoring network and groundwater dynamic monitoring equipment to obtain ground deformation monitoring datasets and groundwater dynamic datasets. The feature parameter extraction module is used to extract feature parameters based on the plateau environment basic dataset, the mining feature dataset, the plateau environment correction coefficient, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain a fused feature vector; The model building module is used to construct and train a settlement prediction model by employing machine learning algorithms, a constructed constitutive model of plateau rock and soil, and the fused feature vectors, thereby obtaining a settlement prediction model. The model inference module is used to infer the fused feature vector based on the settlement prediction model to obtain the predicted ground settlement value and its spatial distribution. The early warning information generation module is used to generate early warning information based on the predicted ground subsidence value and its spatial distribution, and to release the early warning information.

6. The geological disaster early warning system according to claim 5, characterized in that, The regional data acquisition module is specifically used to acquire elevation data, terrain slope, aspect data, atmospheric pressure, oxygen content, and diurnal temperature range based on the geographical coordinate range of the target area, and use these as the basic dataset for the plateau environment; extract data based on the acquired historical data of underground hot water extraction to obtain an extraction feature dataset; and calculate the plateau environment correction coefficient based on the basic dataset for the plateau environment.

7. The geological disaster early warning system according to claim 6, characterized in that, The feature parameter extraction module is specifically used to correct the ground deformation monitoring dataset based on the plateau environment correction coefficient to obtain deformation data; to construct a mining deformation correlation model based on the mining feature dataset and the deformation data; and to extract feature parameters from the plateau environment basic dataset, the mining feature dataset, the ground deformation monitoring dataset, and the groundwater dynamic dataset to obtain deformation rate, deformation acceleration, and basic feature parameters. The mining deformation correlation model, the deformation rate, the deformation acceleration, and the basic feature parameters are used as the fused feature vector.

8. The geological disaster early warning system according to claim 7, characterized in that, The model construction module is specifically used to construct constitutive equations based on the physical and mechanical parameters of the soil and rock mass and the plateau environment correction coefficient, thereby obtaining a constitutive model parameter set for the plateau soil and rock mass; and to construct and train a settlement prediction model based on machine learning algorithms, the constitutive model parameter set for the plateau soil and rock mass, and the fused feature vector, thereby obtaining the settlement prediction model.

9. A geological disaster early warning device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the geological disaster early warning method as described in any one of claims 1 to 4 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the geological disaster early warning method as described in any one of claims 1 to 4.

Images