CN-122024430-A - Geological disaster real-time monitoring and early warning system and method based on multi-source data fusion

CN122024430ACN 122024430 ACN122024430 ACN 122024430ACN-122024430-A

Abstract

The invention discloses a geological disaster real-time monitoring and early warning system based on multi-source data fusion, which relates to the technical field of geological disaster monitoring and comprises a field monitoring layer, an edge computing layer, a cloud platform layer and an application layer; the geological disaster real-time monitoring and early warning system based on multi-source data fusion comprises a multi-source data depth fusion method, wherein satellite remote sensing, a ground sensor, an unmanned plane LiDAR and ground water level data are fused through an improved D-S evidence theory for the first time, the problem of monitoring one side by a single data source is solved, the calculation accuracy of a comprehensive disaster risk value is improved by more than 30%, an edge-cloud cooperative early warning mechanism is used for realizing localized early warning (response time is less than or equal to 10 seconds) at an edge end, and cloud realizes multi-dimensional fusion and dynamic threshold adjustment, so that the real-time performance and accuracy are considered, and the early warning hysteresis problem when a network in a remote mountain area is interrupted is solved.

Inventors

YANG XUQIANG
WANG HAI
HUO YUNLONG
JIANG WENMING
WANG SIJIE

Assignees

宏大爆破工程集团有限责任公司

Dates

Publication Date: 20260512
Application Date: 20260114

Claims (6)

1. The geological disaster real-time monitoring and early warning system based on multi-source data fusion is characterized by comprising a field monitoring layer, an edge calculating layer, a cloud platform layer and an application layer.
2. The geological disaster real-time monitoring and early warning system based on multi-source data fusion as set forth in claim 1, wherein said on-site monitoring layer comprises: The multi-parameter sensor group comprises a GNSS displacement sensor, a soil moisture content sensor, an underground water level sensor, an inclination angle sensor and a tipping bucket type rain gauge, wherein all the sensors have IP68 waterproof grades and are suitable for an extreme environment of-30 ℃ to 70 ℃; The unmanned aerial vehicle inspection unit is a multi-rotor unmanned aerial vehicle carrying a laser radar (LiDAR) and a high-definition camera, presets 9:00 automatic take-off every day, performs terrain scanning and image acquisition on a monitoring area, and supplements a coverage blind area of a ground sensor; the solar power supply module consists of a 200W monocrystalline silicon solar panel and a 12V/100Ah lithium battery, supports continuous power supply on overcast days for 72 hours, and ensures continuous operation of equipment.
3. The geological disaster real-time monitoring and early warning system based on multi-source data fusion as set forth in claim 1, wherein said edge calculation layer comprises: The hardware carrier is characterized in that an industrial edge gateway is adopted, an ARM Cortex-A53 processor is carried, and 4G/5G/Beidou short message multimode communication is supported; Core functions: The data preprocessing, namely eliminating abnormal values of a sensor through a Kalman filtering algorithm, and performing time synchronization on multi-source data; the localization early warning is carried out, namely a lightweight data fusion model is built in based on a random forest algorithm, a displacement risk index is calculated in real time by combining a GNSS displacement rate and an accumulated displacement, a hydrological risk index is calculated by combining a rainfall, a soil moisture content and a groundwater level, and when a weighted value of the displacement risk index and the hydrological risk index exceeds a preset threshold value, the local acousto-optic early warning is triggered immediately; And data distribution transmission, namely transmitting the preprocessed effective data to a cloud platform through a 5G network, simultaneously storing the effective data to a local 128GB SSD, automatically caching the data if the network is interrupted, and recovering the data for later retransmission.
4. The geological disaster real-time monitoring and early warning system based on multi-source data fusion as set forth in claim 1, wherein the cloud platform layer comprises: The data storage and management, namely, adopting a distributed database (HBase) to store historical monitoring data, and supporting quick retrieval according to monitoring points, time and data types; The multisource data fusion algorithm is based on an improved D-S evidence theory, carries out multidimensional fusion by combining the displacement risk index and the hydrological risk index uploaded by an edge calculation layer and terrain gradient data acquired by an unmanned plane LiDAR and ground vegetation coverage data of satellite remote sensing, and outputs a comprehensive disaster risk value, wherein the comprehensive disaster risk value is 0-100, and the higher the score is, the higher the risk is; The dynamic threshold model is used for automatically adjusting the early warning threshold according to the geological type of the monitored area and optimizing model parameters in real time by combining with the historical disaster cases; the visual platform displays a monitoring point distribution map, a real-time data curve, an unmanned aerial vehicle aerial image and a comprehensive disaster risk thermodynamic diagram through a Web end and supports one-key export of a monitoring report.
5. The geological disaster real-time monitoring and early warning system based on multi-source data fusion as set forth in claim 1, wherein the application layer comprises: The method comprises the steps of grading early warning pushing, namely automatically pushing early warning information to users at different levels when the risk value of the comprehensive disaster exceeds a threshold value, wherein blue early warning is used for pushing to a village emergency office, yellow early warning is used for pushing to a county emergency administration, orange/red early warning is used for synchronously pushing to a provincial emergency hall and a local village committee, and the pushing mode comprises short messages, APP notification and micro-message public number reminding; the emergency command function is used for supporting marking of the danger avoiding route and the temporary placement point position on the platform, associating the contact modes of nearby rescue teams and providing visual support for disaster prevention decision-making; The remote management of the equipment comprises the steps of remotely controlling the field equipment through a cloud platform, namely adjusting the sampling frequency of a sensor, restarting an edge gateway, triggering the unmanned aerial vehicle to take off emergently, and reducing the times of manual field maintenance.
6. The method for real-time monitoring and early warning of geological disasters based on multi-source data fusion according to any one of claims 1 to 5, which is characterized by comprising the following steps: Step S1, data acquisition The multi-parameter sensor group of the field monitoring layer acquires displacement, soil moisture content, ground water level and rainfall data in real time, the unmanned plane automatically takes off and acquires terrain and image data 9:00 a day, and satellite remote sensing data are updated 1 time a day; Step S2, edge end pretreatment and local early warning The edge gateway receives the sensor data, eliminates abnormal values through Kalman filtering, calculates a displacement risk index and a hydrological risk index, triggers local acousto-optic early warning if the threshold is exceeded, and caches the data; step S3, data transmission The edge gateway uploads the preprocessed effective data to the cloud platform through the 5G network, and the effective data is automatically cached when the network is interrupted and is restored for later retransmission; Step S4, cloud multi-source data fusion The cloud platform fuses the edge data, the unmanned plane LiDAR data and the satellite remote sensing data, and calculates the comprehensive disaster risk value through improving the D-S evidence theory; Step S5, dynamic threshold judgment and grading early warning The cloud platform calls a corresponding early warning threshold according to the geological type of the monitoring area, and if the comprehensive disaster risk value exceeds the threshold, early warning information is pushed to a corresponding user according to the grade; step S6, data visualization and real-time emergency support The user views the monitoring data and the risk thermodynamic diagram through the Web/APP, the platform provides the function of associating the risk avoidance route with rescue resources, and the field device is remotely managed.

Description

Geological disaster real-time monitoring and early warning system and method based on multi-source data fusion Technical Field The invention relates to the technical field of geological disaster monitoring, in particular to a geological disaster real-time monitoring and early warning system and method based on multi-source data fusion. Background Geological disasters (landslide, collapse and debris flow) are one of main natural disasters threatening the life and property safety of residents in mountainous areas, and according to data of emergency management parts, economic losses caused by the geological disasters in China exceed 50 hundred million yuan, and disaster events occur in the China for more than 1 ten thousand years. Currently, the main stream monitoring technologies in the industry can be divided into three categories: 1. Remote sensing monitoring, namely acquiring large-scale earth surface deformation data through satellites (such as high-resolution series and Sentinel-1), wherein the method has the advantages of wide coverage, suitability for regional risk investigation, low spatial resolution (up to 5 meters), and incapability of capturing local micro displacement (such as millimeter-level change of a single slope); 2. The ground monitoring comprises GNSS displacement monitoring, inclinometers (for monitoring underground soil displacement), rain gauges and the like, wherein the precision can reach millimeter level, but the single-equipment monitoring range is small (the effective monitoring radius of GNSS is about 1 km), and a large area can be covered only by densely distributing points; 3. Unmanned aerial vehicle aerial photography, namely, a local terrain three-dimensional model is obtained through carrying a laser radar (LiDAR) on an unmanned aerial vehicle, so that the unmanned aerial vehicle is suitable for post-disaster evaluation, but cannot realize real-time monitoring (single flight duration of about 30 minutes), and is greatly influenced by weather (storm and windy weather cannot take off). The technology is widely applied in the industry, but an integrated solution of 'multi-source data collaboration, real-time analysis, hierarchical early warning and available base layer' is not formed, The following 4 core technical problems exist in the existing geological disaster monitoring technology, and the invention aims to solve the following problems: 1. The existing monitoring is dependent on a single data source (such as a rain gauge or GNSS), a multi-factor induction mechanism of geological disasters cannot be comprehensively reflected (such as the rising of the water content of soil caused by rainfall and the initiation of side slope displacement), and false early warning or missing early warning is easy to occur (such as the existing micro displacement of a mountain in the early stage is ignored only according to rainfall early warning); 2. The early warning is delayed and the response is not timely, the traditional monitoring system needs to transmit data to a remote cloud server for processing, and the data transmission delay can reach 1-2 hours under the scene of weak network signals in remote mountain areas, so that the early warning requirement of geological disasters on the minute level cannot be met; 3. The existing monitoring equipment is mostly fixed-mounted, is easy to damage in extreme environments such as storm, debris flow and the like, needs manual regular field maintenance (such as battery replacement and sensor cleaning), has overlarge single maintenance cost, and is difficult to be sustained for a large-scale monitoring network (such as county level); 4. The data interpretation threshold is high, and the practicability is insufficient, namely original data (such as displacement coordinates and rainfall values) output by the monitoring system are interpreted by professional geological engineers, and basic disaster prevention departments (such as village and town emergency handling) lack of professional capability, so that the monitoring data cannot be quickly converted into disaster prevention decision basis, and therefore, a geological disaster real-time monitoring and early warning system based on multi-source data fusion is provided. Disclosure of Invention The invention aims to provide a geological disaster real-time monitoring and early warning system based on multi-source data fusion so as to solve the problems in the background technology. In order to achieve the above purpose, the present invention provides the following technical solutions: The geological disaster real-time monitoring and early warning system based on multi-source data fusion comprises a field monitoring layer, an edge computing layer, a cloud platform layer and an application layer. Preferably, the in-situ monitoring layer comprises: The multi-parameter sensor group comprises a GNSS displacement sensor, a soil moisture content sensor, an underground water level sensor, an inclination angle