88 results about "Natural hazard" patented technology

The invention discloses an urban inland inundation risk multi-level prediction method and device based on time-space diagram learning, a storage medium and a computer program product, and relates to the technical field of natural disaster risk prediction, and the method comprises the steps: collecting multi-modal urban hydrological data; performing hierarchical time modeling on the multi-modal urban hydrological data, and extracting a time embedding vector; constructing a heterogeneous graph based on the time embedding vector, and performing spatial feature aggregation calculation on the heterogeneous graph to obtain spatial embedding representation; and performing multi-level prediction according to the spatial embedding representation to obtain a multi-granularity waterlogging risk index. Through multi-modal data acquisition and preprocessing, layered time modeling, heterogeneous graph construction, spatial feature aggregation calculation and multi-level prediction, multi-modal urban hydrological data are effectively fused, and comprehensive and accurate urban inland inundation risk prediction is realized.

The invention relates to a BIM-based building structure intelligent management method. The method comprises the following steps: step 1, dynamic monitoring factor mining and multi-modal sensing: collecting meteorological hydrology, geological conditions, microenvironment and natural disaster historical data; 2, neural operator-driven virtual-real fusion dynamic analysis: converting the format of the collected data, mapping the data into a BIM model, integrating a virtual-real synchronization engine, and pre-training a Fourier neural operator through real-time displacement and strain data reverse calibration model boundary conditions in a construction stage, and establishing a direct mapping relationship between an external load and a structural response; 3, multi-modal data collaboration and hierarchical permission interaction: forming a multi-dimensional feature matrix through association of timestamps and space coordinates; 4, multi-modal graph convolutional network damage identification: learning dynamic relevance between nodes through an adaptive graph convolutional layer, and extracting damage sensitive features; and step 5, dynamic life early warning of digital twinning drive: predicting the remaining service period of the structure.

The embodiment of the invention provides a power grid natural disaster early warning method and system based on artificial intelligence, and belongs to the technical field of artificial intelligence. The method comprises the following steps of: firstly, constructing a multi-source power grid and disaster information evolution association set which comprises power grid equipment operation dynamic information and natural disaster occurrence dynamic information of a corresponding region and is associated in real time through a region identifier and annotated with an information evolution trend, and then carrying out sequential risk conduction analysis on the multi-source power grid and disaster information evolution association set; then calling a pre-trained power grid disaster intelligent hierarchical prediction model to carry out multi-stage feedback prediction processing on the result, generating a power grid natural disaster hierarchical prediction result, executing early warning strategy dynamic iteration optimization according to the result, generating early warning strategy dynamic iteration adjustment parameters, and carrying out early warning strategy dynamic iteration adjustment. And finally, generating a power grid natural disaster accurate early-warning instruction based on the early-warning strategy dynamic iteration adjustment parameters, and sending the power grid natural disaster accurate early-warning instruction to a power grid monitoring center terminal. The power grid natural disaster early-warning method can comprehensively, accurately and dynamically perform early warning on the power grid natural disaster.

The invention relates to the technical field of optical signal processing and analysis, and particularly provides a submarine optical fiber vibration event detection method based on credibility weighting, and the method comprises the steps: obtaining an optical pulse echo matrix according to a received signal, and carrying out the decomposition of the signal through a low-rank and sparse decomposition model. A credibility coefficient is introduced to measure the reliability of the data, an optimization model is established, and the credibility coefficient is set to be a tide credibility coefficient or a natural disaster credibility coefficient according to the marine environment. Solving the optimization model to obtain a sparse item estimation matrix corresponding to the sparse item; and detecting a natural disaster event and a vibration event in the received signal according to the sparse term estimation matrix. According to the method, reliability evaluation of the signals can be dynamically adjusted according to different marine environments, noise and real vibration events are effectively distinguished, high robustness is kept in the complex and multi-interference marine environment, the method is not prone to being affected by interference sources, and event detection precision and accuracy are good.

The invention discloses a risk management and control method, system and equipment based on a water conservancy project, and relates to the technical field of risk management and control. A sensor data drift abnormal index is calculated by obtaining multi-source monitoring data of a target water conservancy project after an earthquake, and an engineering structure change abnormal index is calculated based on a preset structure health model and epicenter parameters; and calculating an earthquake propagation influence index by combining the seismic source distance, calculating an earthquake influence coefficient by combining the three factors, and judging whether an original risk early warning evaluation system needs to be updated according to the earthquake influence coefficient. According to the method, after a natural disaster event (such as a slight earthquake) is encountered, whether the risk early warning evaluation system needs to be updated or not can be judged according to the actual disaster degree; it is ensured that early warning is not delayed or invalid, and the possibility of causing larger potential safety hazards is reduced; meanwhile, whether the risk early warning evaluation system is updated or not can be reasonably judged according to the influence of the actual earthquake on the construction of the hydraulic engineering project, and unnecessary waste of computing resources is reduced.

The invention discloses a satellite precipitation product improvement method, device and system based on a dual-time scale minimum error, and the method comprises the steps: carrying out the improvement of a first satellite precipitation product, a second satellite precipitation product and an interpolation precipitation product according to the obtained first satellite precipitation product, second satellite precipitation product and interpolation precipitation product; sequentially adopting a dual tool variable algorithm, a least square fusion algorithm and an inverse error weighting algorithm to respectively calculate first fusion rainfall products under a daily scale and a monthly scale; and fusing the first fused precipitation products under the daily scale and the monthly scale to generate a second fused precipitation product. Ground observation data are not needed, daily scale and monthly scale rainfall errors can be minimized at the same time, the thought of data fusion is incorporated, the performance of satellite rainfall products is enhanced, and detailed data support can be provided for natural disaster monitoring and early warning and regional water resource management.

The invention relates to a sensor assembly (100) for continuous monitoring in the region of natural hazards and / or for detecting natural hazard events, in particular gravimetric mass movements, preferably avalanches, landslides and rockslides, the assembly comprising: - a housing (1) for contacting, in particular partially and / or entirely, a mass to be monitored for the occurrence of a gravimetric mass movement, in particular a snow mass, soil mass or rock mass, which housing can be cast off by means of an aircraft; - at least one position sensor which is designed to determine the position of the housing, in particular in the form of geographic coordinates; - at least one acceleration / orientation sensor (Acc), in particular for detecting orientation, translational movement, acceleration and / or vibrations; - at least one temperature sensor; - a wireless communication device and / or a data memory; - an energy supply device; and - a control unit, - wherein the at least one position sensor, the at least one acceleration / orientation sensor (Acc), the at least one temperature sensor, the control unit, the communication device and the power supply device are each disposed in and / or on the housing, - wherein the control unit is connected to the position sensor, to the acceleration / orientation sensor (Acc) and to the temperature sensor and is designed to control the sensors in order to measure sensor measurement values, and - wherein the control unit is connected to the communication device and / or to the data memory and to the energy supply device and is designed to transmit sensor measurement values determined by the sensors by means of the communication device and / or to store said sensor measurement values in the data memory.

The invention provides an overhead transmission line fault loss assessment method based on spherical fuzzy and entropy weight methods, and relates to the technical field of transmission line fault loss assessment. Comprising the following steps: S1, acquiring natural disaster data of a to-be-evaluated line area; s2, dividing the fault loss into a plurality of evaluation dimensions, and constructing a hierarchical fault loss evaluation system by taking natural disasters as influence factors; s3, based on the disaster data, calculating fault loss cost of each evaluation dimension under different natural disasters; s4, using a fuzzy set theory to quantify the influence degree of the natural disaster on each evaluation dimension, and calculating a subjective weight; s5, calculating the objective weight of each evaluation dimension by adopting an entropy weight method; and S6, determining a comprehensive weight through the subjective and objective weight fusion model, and evaluating the fault loss of the overhead transmission line based on the comprehensive weight. In the whole life cycle cost calculation, fault loss evaluation is added, accurate quantification of multi-dimensional loss is realized, and the reliability of the line is improved in the later line design process.

The invention relates to the technical field of intelligent driving, and discloses an intelligent risk avoiding method and system for a vehicle, and the method comprises the steps: obtaining disaster early warning information from a natural disaster early warning platform; acquiring sensing data of the surrounding environment of the vehicle; according to the disaster early warning information and the sensing data, determining that the vehicle has a risk avoiding demand; and controlling the vehicle to execute risk avoiding operation according to the type of the disaster. In combination with disaster early warning information of the natural disaster early warning platform and sensing data of the vehicle to the surrounding environment, risk avoiding control is carried out on the vehicle, and the driving safety is improved.

The invention discloses an elasticity evaluation and improvement method for a power transmission system under the action of debris flow, is used for solving the technical problem of single fault rate consideration factor in the prior art, and belongs to the technical field of evaluation of anti-disaster capability and recovery capability of the power transmission system under extreme natural disasters (such as debris flow). The method comprises the following specific steps: step 1, constructing a debris flow model considering an impact force attenuation effect; s11, describing the influence area of the debris flow by using the three parameters; s12, carrying out stress analysis on the tower impacted by the debris flow; 2, constructing a fault model considering the accumulated impact process of debris flow on the power transmission element; s21, calculating a tower failure probability; s22, calculating the failure probability of the power transmission line; s23, calculating the overall failure probability of the power transmission line; 3, sampling at a constant time interval delta t based on the solved failure probability of the power transmission line within the debris flow duration delta T; 4, drawing an expected power supply power curve of the power transmission system; and 5, calculating an elastic index.

The present invention discloses a method and system for the site selection of disaster relief material reserve warehouses based on earthquakes of magnitude 7 or above, relating to the technical field of the site selection of disaster relief material reserve warehouses, including: determining an intensity area where the earthquake magnitude is 7 or above and the earthquake intensity is 8 or above according to the earthquake literature materials and earthquake intensity attenuation models of the research area; using remote sensing building recognition technology and eight-connected mode to process the intensity area to obtain the disaster-affected area; determining the farthest safe flight distance when helicopters transport disaster relief materials and determining the candidate areas for the site selection of disaster relief material reserve warehouses within the research area; screening the candidate areas for the site selection according to the constraint relationship between the farthest safe flight distance and the disaster-affected area, as well as natural disasters, terrain indices, traffic conditions, urban planning, etc. of the research area, so as to determine the optimal site selection area. The address area selected by the present invention can achieve the purpose of practical feasibility.

The invention belongs to the field of ionosphere detection, artificial intelligence and deep learning, and particularly relates to a lightning electromagnetic pulse-based large-area ionosphere D layer height measurement method, system and equipment. The problems that a traditional detection method is high in cost, limited in coverage range and prone to being interfered by natural disasters and strong electromagnetic pulses are solved. The method comprises the following steps: acquiring lightning event information based on an electromagnetic pulse detection station; calculating a ground wave and sky wave arrival time difference based on a pre-trained deep learning model, and calculating to obtain the ionosphere D layer height of a single position; performing gridding processing on the obtained ionosphere D layer height data at different positions according to certain time and spatial resolution, performing interpolation null value filling on the gridded data, and performing smoothing processing; and outputting a calculation result of the D layer height of the large-area ionized layer. According to the invention, wide-area low-cost ionosphere detection is realized by adopting the natural lightning radiation source, and the measurement precision is improved by optimizing the model for different lightning types.

The invention discloses a mountain disaster susceptibility prediction method and device considering a composite terrain distance, and relates to the field of natural disaster risk assessment, and the method comprises the steps: obtaining a mountain disaster sample and a multi-source environment factor in a research area range; terrain factors are screened from the multi-source environment factors, and the weight of each terrain factor is calculated based on an entropy weight method; on the basis of the Euclidean space distance, a weighted terrain difference distance is introduced, and a composite terrain distance is constructed through fusion; sorting and screening the mountain disaster sample data according to the composite terrain distance, and determining a local reference sample set and a geographic weighted weight corresponding to each local reference sample; adopting a Bayesian optimization method to perform joint optimization on the bandwidth parameter of the geographical weighted weight, the weight parameter of the composite terrain distance and the structure parameter of the random forest model to obtain an optimal parameter combination; and under the constraint of the optimal parameter combination, carrying out point-by-point prediction on the research area. According to the invention, the precision of mountain disaster susceptibility prediction can be improved.

The present application relates to the technical field of power meteorological prediction. In order to solve the problem that the current power grid natural disaster warning system has deficiencies in feature extraction and fails to make full use of the knowledge in related fields to enhance the prediction ability of the model, a power grid natural disaster warning method, a computer device, a computer-readable storage medium, and a computer program product based on a knowledge graph are disclosed. The method includes step S1 of constructing a power meteorological knowledge graph based on meteorological data, geographic information, and power equipment data corresponding to the occurrence of power grid natural disasters; step S2 of enhancing the features of the meteorological data according to the semantic features in the knowledge graph to form fused features; step S3 of constructing a disaster weather prediction model based on a graph convolutional neural network and a self-attention mechanism according to the fused features; and step S4 of warning of power grid natural disasters under target weather according to the disaster weather prediction model. Using this method can improve the accuracy of power grid natural disaster prediction.

The present invention discloses a construction external failure fault diagnosis method, device, equipment and storage medium; the method comprises: acquiring fault information, identifying the fault information through a fault identification model to obtain identification result information, and determining the fault type based on the identification result information and first environmental monitoring information; when the fault type belongs to a non-lightning fault, determining whether the fault type belongs to a metallic fault by performing feature analysis on the fault waveform information in the fault information, and determining whether the fault type belongs to a non-natural disaster fault through second environmental monitoring information; when the fault type belongs to a metallic fault and a non-natural disaster fault, determining the probability of the fault being a construction external failure fault based on the fault information and the second environmental monitoring information according to a preset construction external failure fault correlation factor analysis rule, so as to accurately analyze and determine the probability of the occurrence of a construction external failure fault, and improve the accuracy and reliability of line fault diagnosis.

The utility model provides a windproof anti-seismic photovoltaic tracking support system, which relates to the field of photovoltaic supports, and comprises a plurality of groups of synchronously adjusted photovoltaic support mechanisms and two groups of symmetrically arranged connecting rod damping mechanisms, the plurality of groups of photovoltaic support mechanisms are arranged side by side, and each group of connecting rod damping mechanism is connected in series with two ends of the photovoltaic support mechanism. The gravity center design of the photovoltaic tracking support system is optimized, the multiple sets of photovoltaic support mechanisms arranged side by side form an integral structure through the connecting rod damping mechanism, it is ensured that under the action of strong wind, the multiple sets of photovoltaic support mechanisms can jointly resist the extreme environment and keep balance, and the overturning risk is reduced; according to the utility model, the influence of natural disasters such as strong wind and earthquake on the photovoltaic tracking support system is effectively reduced while the power generation efficiency is improved, and the popularization value is high.

The application discloses a kind of based on ST-GAN's multi-source data dynamic risk early warning method and system, the method includes the following steps: S1. constructing the spatiotemporal generation confrontation network suitable for spatiotemporal data characteristics, utilize the network to generate extreme precipitation data, realize spatiotemporal unbalanced data reduction, obtain the equalization data of final output;S2. the equalization data of final output with high-dimensional space variable is carried out multi-source data fusion, and nonlinear spatiotemporal information conversion equation, and by local linearization obtains linear approximation model, to predict future time series;S3. under the present situation that extreme rainfall data amount is relatively insufficient, neural network based on dual learning theory accurately learns the parameter of nonlinear spatiotemporal conversion, estimates extreme weather event.The application is through spatiotemporal generation confrontation network (ST-GAN) and multi-source data fusion engine, significantly improves the precision and efficiency of natural disaster warning.

A method for displaying an augmented reality (AR) representation of physical effects and property damage resulting from a parametric earthquake event is provided. The method includes scanning scenes with a common boundary and identifying the common boundary and objects in the first and second scenes. The method also includes creating an AR boundary for the common boundary and AR objects for the objects. The method further includes receiving a seismic characteristic. The method additionally includes, while scanning the first scene, displaying an aperture effect on the AR boundary and at least one seismic effect on the AR objects in the second scene based on the at least one received seismic characteristic, wherein the aperture effect allows viewing the seismic effect on AR objects in the second scene on the user device while the second scene is not visible to the one or more sensors of the user device.

The invention discloses an intelligent management system mainly used for reducing the damage risk of a photovoltaic power station caused by extreme weather, and the system comprises an equipment installation backboard, one side of the equipment installation backboard is fixedly provided with a photovoltaic inverter, and the other side of the photovoltaic inverter is provided with an equipment installation frame in a penetrating manner. And equipment mounting bolts are mounted in the equipment mounting racks in a threaded manner. According to the intelligent management system mainly used for reducing the damage risk of the photovoltaic power station caused by the extreme weather, in the daily use process, the photovoltaic inverter is fixedly installed on one side of the installation backboard through cooperation of the equipment installation frame and the equipment installation bolts; and then the photovoltaic intelligent management system is utilized to perform intelligent management on the photovoltaic power station, so that the damage to the photovoltaic power station caused by natural disasters can be detected in real time, and when the natural disasters encounter, the management can automatically judge a response mode according to the current state, and the whole set of system can be controlled to take corresponding response measures.

The following relates generally to calculating estimated power outages during a natural disaster event. In this regard, some embodiments create a synthetic network of a power infrastructure of a geographic area by: determining a location of a power substation; determining a location of a customer; determining a location of a power line linking the power substation to the customer; and determining if the power line is overhead or underground. Some embodiments then use the created synthetic network to simulate an event to calculate the estimated power outages during the event.

This invention relates to a method for constructing a disaster-bearing entity based on a structure-function-behavior model, comprising the following steps: S1 entity structural semantic modeling; S2 entity functional semantic modeling; S3 behavioral semantic modeling; S4 constructing a semantic fusion module and building urban natural disaster risk factor data and knowledge data, and using the risk factor data and knowledge data to fuse the semantics formed in the models built in steps S1-S3 to realize the construction of the disaster-bearing entity. The method also includes a data and mechanism-driven twin model update method, specifically including: updating the twin model using a change detection model, and update management, wherein the change detection model includes the detection of structural and semantic changes. The method also includes a disaster risk prediction method, which uses a hybrid approach consisting of decision trees and node data formed by urban natural disaster risk factor data and knowledge data. This achieves data analysis and prediction of urban entity disaster situations with visualization of the sum of structure-function-behavior factors.

This invention discloses a parallelized method for surface water dynamic simulation based on a pipeline network. First, it constructs a water flow path from the water source to the watershed outlet, simplifying the three-dimensional surface into a one-dimensional pipeline network. Then, it improves the classic Manning formula using topographic parameters to estimate the flow velocity of each water flow path. Combined with a preset travel time, it simulates the surface runoff. Finally, it uses parallel computing to parallelize the velocity estimation function and the surface runoff simulation function, resulting in a high-precision and high-efficiency parallelized method for surface water dynamic simulation. This invention not only considers the influence of topographic parameters on runoff velocity in surface water dynamic simulation but also applies the parallel technology CUDA to the field of surface water dynamic simulation. While ensuring the accuracy of the surface water dynamic simulation algorithm, it improves the real-time response efficiency of surface water dynamic simulation, laying the foundation for scientific land planning, water resource management, and natural disaster response strategies.

The invention provides a multi-detection-end natural disaster analysis system based on a large model, and the system comprises a data collection module, a scene judgment module, a scene environment simulation module, a scene noise module, a denoising module, a standard disaster module, and a scene impact analysis module. The scene environment simulation module establishes a scene simulation environment, the scene judgment module is used for judging an application scene to which current input data belongs, the scene noise module is used for determining the influence of basic environment noise on data, the denoising module is used for denoising disaster data results, and the standard disaster module establishes a disaster model without scene factors. The scene influence analysis module is used for establishing a feature influence database of different scenes on disaster detection data. The method fully considers the mutual influence between the two dimensions of the environment and the disaster type, and is higher in scene pertinence and more accurate in data compared with the prior art. And a data basis is provided for mutual synthesis of multiple disasters and classification of disasters with similar influence.

1. The name of the design product: lighting column assembly. 2. The use of the design product: used as a lighting column assembly for safety and hazard identification in various work areas (such as infrastructure, road construction and maintenance, aviation, port, railway, mining, traffic congestion area, and public access area); or used to guide pedestrian traffic during work area, public activity, during natural disaster, and help identify the location of various landmarks. 3. The design points of the design product: in shape. 4. The picture or photo that best shows the design points: perspective view 1. 5. The rear view and right view are omitted as the design points have been shown in other views. 6. Other changes in the state of the case need to be explained: the product with color is shown in the figure.

This invention discloses a method and system for comprehensive risk assessment of natural disasters in coal mines under complex geological conditions, belonging to the field of coal mine disaster risk assessment technology. The method involves collecting relevant coal mine data; determining comprehensive risk assessment indicators for natural disasters in coal mines under complex geological conditions, and calculating the comprehensive indicator weights using a combined weighting method; constructing an improved transfer learning-BP neural network model, optimizing scoring parameters by combining relevant coal mine data under complex geological conditions, and determining the scoring values ​​for the comprehensive risk assessment of natural disasters in coal mines under complex geological conditions; and conducting a comprehensive risk assessment of natural disasters in coal mines under complex geological conditions based on the comprehensive indicator weights and the scoring values ​​to obtain the assessment results. This invention achieves comprehensive, accurate, and real-time assessment of natural disasters in coal mines under complex geological conditions, effectively preventing and reducing the risks of underground natural disasters under complex geological conditions, and ensuring safe production and the safety of life and property of coal mine workers.

The utility model relates to the technical field of track engineering, in particular to an adjustable track fastener which comprises a left-right adjusting component set arranged between a track fastener body and a track bed, the left-right adjusting component set is used for adjusting the left-right position of the track fastener body and the left-right position of a track piece, and the left-right adjusting component set comprises an embedded sleeve piece. A base plate piece is arranged at the top of the embedded sleeve piece, and a distance adjusting block is arranged in the base plate piece. The rail fastener system has a larger adjusting range and flexibility, so that challenges caused by factors such as topographic changes and natural disasters can be better dealt with, the construction cost and the maintenance cost are reduced, and the practicability and the adaptability of a railway system are improved.

The present application discloses an intelligent railway operation and maintenance system, method, equipment, medium and product, which relate to the field of railway operation and maintenance. The system includes an air-ground-vehicle integrated transmission module for collecting and transmitting sensor data of key train components; a drone cluster module for collecting and transmitting environmental image data within a set range of the railway line; a digital twin big model training module for constructing a digital twin big model based on the sensor data and environmental image data, determining an integrated big model, and using the integrated big model to predict train fault types and natural disasters, and generate emergency guidance plans; the integrated big model is a neural network model. The present application improves the real-time detection and is more suitable for future ultra-high-speed scenarios.

A method and system for monitoring low-temperature rain, snow, and freezing in complex terrain areas comprises the following steps: Step 1: Collecting satellite observation data of the monitoring area for large-scale monitoring of cold air movement, snow cover, and surface temperature, and identifying lower atmospheric inversion layer data; Step 2: Obtaining atmospheric vertical structure data through a ground-based remote sensing vertical observation system; Step 3: Collecting image data of the monitoring area through UAV-based remote sensing; Step 4: Deploying ground data acquisition units to collect data including rainfall and snowfall, snow thickness, temperature, road icing, and power line icing; Step 5: Collecting historical disaster data for the monitoring area; Step 6: Cleaning, fusing, and extracting features from various data using machine learning algorithms; Step 7: Constructing an assessment model reflecting low-temperature rain, snow, and freezing disasters based on the extracted feature data, and generating prediction results for low-temperature rain, snow, and freezing natural disasters in the monitoring area.

The invention provides a non-coal mine natural disaster-causing risk assessment method and system, and the method comprises the steps: carrying out the division according to the factors of non-coal mine natural disaster-causing disasters and an assessment target, and forming a disaster-causing risk assessment system which comprises a target layer, a criterion layer and an index layer; calculating the index weight of each layer of index through an analytic hierarchy process; adopting a fuzzy analytic hierarchy process to obtain a fuzzy evaluation matrix of each layer of indexes; obtaining a comprehensive evaluation vector according to the index weight of each layer of index and the fuzzy evaluation matrix; according to the method, an occurrence possibility grade value and a consequence severity grade value are calculated according to evaluation vectors, a final risk grade is determined, influence weights of all indexes on disaster-causing risks are calculated by introducing fuzzy mathematics, contributions of different indexes to the disaster-causing risks are optimized, and the risk assessment accuracy is improved. According to the method, the defect of lack of difference in calculation of the weight of each disaster-causing index of the non-coal mine in a traditional method is overcome, the evaluation result can better conform to the actual situation, and the method has a feasible guiding effect.

The application discloses a mountain torrent disaster early warning method and system, and the method comprises the following steps: acquiring cloud image data; processing the cloud image data based on a pre-trained deep learning model to obtain a predicted rainfall sequence; constructing a hydrological and hydrodynamic model and performing scenario simulation to demarcate a disaster threshold rainfall; and combining the predicted rainfall sequence and the disaster threshold rainfall to perform early warning. The system comprises an image acquisition module, a rainfall prediction module, a threshold determination module and an early warning module. By using the application, mountain torrent disaster early warning can be performed in advance, important disaster relief and risk avoidance time can be obtained for mountain torrent disasters, and the prediction accuracy is well guaranteed due to the combination of the rainfall prediction module and the atmospheric pressure observation and the cloud rainfall estimation method based on deep learning. The application can be widely applied in the field of natural disaster early warning.