Ground disaster monitoring and early warning system and method based on multi-source fusion

Abstract

The invention relates to the technical field of automatic monitoring and early warning, in particular to a ground disaster monitoring and early warning system and method based on multi-source fusion, and the system comprises a data collection module, a background service module, a ground disaster early warning module, an early warning release module and a data visualization module. The data acquisition module is used for acquiring GNSS displacement data, sensor data and meteorological product data, and integrating various data for ground disaster monitoring and early warning. Secondly, a ground disaster early warning module capable of flexibly replacing an early warning algorithm model is designed, the module carries out model training by utilizing historical monitoring data, two optimal model files and two optimal parameter files generated by training are used for realizing rapid modeling and carrying out displacement prediction, and if the predicted displacement exceeds a threshold value, the early warning algorithm is executed; if yes, the predicted value and the early warning result are processed into a fixed JSON format and sent to the RabbitMQ to wait for consumption data of the early warning issuing module. And the early warning publishing module splices the notification content and sends the notification content to the contact person according to a mode specified by the configuration information.

Description

Technical Field

[0001] The present invention relates to the field of automatic monitoring and early warning technology, and in particular to a geological disaster monitoring and early warning system and method based on multi-source fusion. Background Art

[0002] Landslide is a common and harmful geological disaster, accompanied by heavy rain, floods, and mudslides. The occurrence of landslides not only threatens people's lives and property, but also damages the ecological environment. At present, domestic deformation monitoring technology has reached a new height in the professional field. In particular, relying on the high-precision positioning technology of my country's Beidou navigation system, the deformation monitoring system has achieved a monitoring accuracy of millimeter level.

[0003] However, there is a common challenge in current geohazard monitoring systems, that is, most systems rely on only a single data source, which may limit the comprehensiveness and accuracy of the monitoring system. Summary of the invention

[0004] The purpose of the present invention is to provide a geological disaster monitoring and early warning system and method based on multi-source fusion, aiming to solve the problem of reduced comprehensiveness and accuracy of the monitoring system caused by reliance on a single data source in the existing geological disaster monitoring system.

[0005] To achieve the above-mentioned purpose, the present invention proposes a geological disaster monitoring and early warning system based on multi-source fusion, including a data acquisition module, a background business module, a geological disaster early warning module, an early warning release module, and a data visualization module;

[0006] The data acquisition module is used for receiving, preprocessing and storing data; the background business module is used to provide query, download and business management function support for the front-end program, and provide users with data access and login permission control; the geological disaster warning module is used to operate according to pre-set rules and configuration information in the database, and issue warnings when geological disaster risks are found; the warning release module is used to promptly convey warning information and equipment status information to relevant personnel; the data visualization module displays the information of the selected hidden danger points, including the spatial distribution of monitoring equipment, real-time status, warning logs and data curves.

[0007] The data received by the data acquisition module include GNSS monitoring data, sensor data and meteorological product data, among which the GNSS monitoring data is displacement monitoring data. The sensor data of the inclinometer, rain gauge, crack meter and soil moisture meter are solved by the solution server and then transmitted through the message middleware Rabbitmq. The meteorological product data is provided by the meteorological monitoring department, including the actual and forecast precipitation data of area A and area B.

[0008] Among them, the background business module adopts two methods to meet the needs of data query. The first method is to use the MVC model and MyBatis framework to realize front-end request parsing, data query and data echo; the second method allows users to enter SQL statements on the front end and then perform parsing and query in the background.

[0009] Among them, the geological disaster warning module provides parameter configuration, warning log recording and warning services, among which the warning services include predictive warning and real-time warning. The predictive warning algorithm is developed based on Python's Django framework, and the real-time warning is a multi-parameter real-time warning, including key indicators of displacement, water content, cumulative displacement and displacement change rate.

[0010] The warning release module loads notification object information, monitors the RabbitMQ queue to obtain warning information and device status information, and has two message notification methods: SMS and email.

[0011] Among them, the data visualization module is developed based on the Vue framework, supports Chinese and English versions, and uses ECharts components to realize the intuitive display of monitoring data and statistics; the data visualization module displays the information of hidden danger points on the high-definition map layer, including the spatial distribution of monitoring equipment, real-time status, early warning log and deformation data curve; at the same time, it supports meteorological data visualization, and has the functions of real-time precipitation, accumulated rainfall query and precipitation forecast display.

[0012] The present invention also proposes a geological disaster monitoring and early warning method based on multi-source fusion, which adopts the geological disaster monitoring and early warning system based on multi-source fusion, including the following steps:

[0013] Step 1: The GNSS monitoring station installed on the landslide collects displacement monitoring data and transmits it to the solution server in real time for solution; the multi-source auxiliary sensor collects the tilt angle, rainfall and crack changes and uploads them to the server in real time;

[0014] Step 2: Calculate and obtain the comprehensive deformation information of the site based on the solved displacement data and sensor data; input the comprehensive displacement information into the early warning algorithm to obtain the early warning risk value;

[0015] Step 3: Based on the comprehensive deformation and warning risk value information, conduct real-time monitoring and early warning of the status of the potential danger point where the monitoring equipment is located.

[0016] Optionally, the comprehensive deformation information in step 2 is the displacement value D total , the expression is as follows:

[0017] D total =E n +P s ·S+Pc ·C+P r ·R+P m ·M

[0018] Among them, E n is the displacement monitoring data collected by the monitoring station, P s is the weight of the soil moisture meter, S is the soil moisture content, P c is the weight of the crack meter, C is the width of the soil crack, P r is the weight of the rain gauge, R is the rainfall at the monitoring station, P m is the weight of the mud level meter, and M is the soil silt thickness.

[0019] Optionally, the step of performing early warning in step 3 is different from the step of real-time monitoring, and includes the following steps:

[0020] Step 3.1: The scheduled task module obtains the algorithm configuration information in MySQL;

[0021] Step 3.2: The data acquisition and processing module obtains all required historical monitoring data from the InfluxDB database according to the configuration information, and performs data preprocessing and grouping;

[0022] Step 3.3: Add the training processes of all devices to the model training process queue in sequence; each device process contains asynchronous training processes for horizontal and vertical data, and calls the algorithm model training modules to complete the model training;

[0023] Step 3.4: After the training is completed, the training process of each device will generate two optimal model files and two optimal parameter files, corresponding to the prediction models in the horizontal and vertical directions respectively;

[0024] Step 3.5: Call the optimal model file and optimal parameter file generated by the prediction model training module to achieve rapid modeling, and then calculate the comprehensive prediction displacement through the horizontal and vertical prediction results;

[0025] Step 3.6: Compare the calculated comprehensive predicted displacement with the threshold. If it exceeds the threshold, the predicted value and warning result are processed into a fixed JSON format and sent to RabbitMQ, waiting for the warning publishing module to consume the data;

[0026] Step 3.7: The warning publishing module loads the notification object information, monitors the RabbitMQ queue to obtain the notification message, splices the notification content and sends it to the contact in the manner specified by the configuration information. The warning has two message notification methods: SMS and email.

[0027] The present invention provides a geological disaster monitoring and early warning system and method based on multi-source fusion. The system includes a data acquisition module, a background business module, a geological disaster early warning module, an early warning release module, and a data visualization module. First, the data acquisition module is used to obtain GNSS displacement data, sensor data, and meteorological product data, and integrate various data for geological disaster monitoring and early warning. Secondly, a geological disaster early warning module that can flexibly replace the early warning algorithm model is designed. The module uses historical monitoring data to train the model. The two optimal model files and two optimal parameter files generated by the training are used to achieve rapid modeling and displacement prediction. If the predicted displacement exceeds the threshold, the predicted value and the early warning result are processed into a fixed JSON format and sent to RabbitMQ, waiting for the early warning release module to consume the data. The early warning release module splices the notification content and sends it to the contact person in the manner specified by the configuration information. The present invention more accurately assesses the potential risks of geological disasters through multi-source fusion, timely warns, and takes corresponding preventive measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 It is a schematic diagram of the architecture of a geological disaster monitoring and early warning system based on multi-source fusion of the present invention.

[0030] Figure 2 It is a schematic diagram of the flow of meteorological product parsing and warehousing procedures of a geological disaster monitoring and early warning system based on multi-source fusion according to the present invention.

[0031] Figure 3 It is a schematic diagram of model training and prediction and early warning of a geological disaster monitoring and early warning system based on multi-source fusion of the present invention. DETAILED DESCRIPTION

[0032] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0033] See also Figure 1 ,The present invention provides a geological disaster monitoring and early warning system based on multi-source fusion, including a data acquisition module, a background business module, a geological disaster early warning module, an early warning release module, and a data visualization module;

[0034] The data acquisition module is used for receiving, preprocessing and storing data; the background business module is used to provide query, download and business management function support for the front-end program, and provide users with data access and login permission control; the geological disaster warning module is used to operate according to pre-set rules and configuration information in the database, and issue warnings when geological disaster risks are found; the warning release module is used to promptly convey warning information and equipment status information to relevant personnel; the data visualization module displays the information of the selected hidden danger points, including the spatial distribution of monitoring equipment, real-time status, warning logs and data curves.

[0035] Specifically, the data acquisition module receives GNSS monitoring data, sensor data and meteorological product data, and stores them in the database after analysis and processing.

[0036] The backend business module provides query, download and business management support for the frontend program, and provides data access and login permission control for users. In terms of data query, two methods are used to meet different needs. The first method is to use the MVC mode and MyBatis framework to implement frontend request parsing, data query and data echo. The second method allows users to enter SQL statements on the frontend and then perform parsing queries on the backend. Business information management is mainly about the management of potential danger points and equipment, including the core functions of adding, modifying and deleting potential danger points and equipment. The user login control function is mainly responsible for verifying the identity of system users and completing login jumps.

[0037] The core functions of the geological disaster warning module include parameter configuration, warning log recording and warning services. Parameter configuration is mainly carried out in the system background. Each time the warning algorithm runs, the configuration information in the database will be loaded first. Even if the configuration information changes, these updates will automatically take effect the next time the algorithm is executed. The warning algorithm software includes two types: predictive warning algorithm and real-time warning algorithm. The predictive warning algorithm is developed based on Python's Django framework. The multi-parameter real-time warning model has the ability to monitor and analyze multiple data types in real time, including displacement, water content, cumulative displacement and displacement change rate key indicators.

[0038] The warning publishing module loads notification object information, monitors the RabbitMQ queue to obtain warning information and device status information, and has two message notification methods: SMS and email to meet the needs and scenarios of different users.

[0039] The data visualization module is developed based on the Vue framework, supports Chinese and English versions, and uses ECharts components to achieve intuitive display of monitoring data and statistics. It can display information of hidden danger points on the high-definition map layer, including the spatial distribution of monitoring equipment, real-time status, warning logs, and deformation data curves. The module also supports meteorological data visualization, with real-time precipitation, cumulative rainfall query and precipitation forecast display functions.

[0040] Furthermore, the data acquisition module receives the displacement monitoring data. After the sensor data of the inclinometer, rain gauge, crack meter, and soil moisture meter are solved by the solving server, they are transmitted to the geological disaster monitoring and early warning system through the message middleware Rabbitmq. The meteorological product data is provided by the meteorological monitoring department, including the actual and forecast precipitation data of area A and area B. These data are regularly sent by the meteorological department to the FTP server of the monitoring and early warning system. The preprocessing program is required to parse and store the files and transmit them to the data processing platform.

[0041] The multi-source fusion-based geological disaster monitoring and early warning system uses MongoDB shard clusters for long-term storage of massive monitoring data. The time series characteristics of the InfluxDB database are used to store monitoring data for the past 1-2 years to support visual monitoring. A table partitioning strategy is also adopted to optimize access efficiency. The original product files of meteorological products are saved in the file directory of the FTP server, and the parsed data is stored in the Redis cache database to accelerate the visualization of meteorological cloud maps. Considering the timeliness of meteorological products, only the meteorological data of the last 5 days is retained in the Redis database.

[0042] At the same time, the system visualizes the on-site information of the monitored area. Its presentation form is combined with the map component and is designed to be stored in the form of map tiles. Nginx is used to build a map tile service, data files are saved on the server, and map services are provided to the outside world.

[0043] The present invention also proposes a geological disaster monitoring and early warning method based on multi-source fusion, which adopts the geological disaster monitoring and early warning system based on multi-source fusion, including the following steps:

[0044] Step 1: The GNSS monitoring station installed on the landslide collects displacement monitoring data and transmits it to the solution server in real time for solution; the multi-source auxiliary sensor collects the tilt angle, rainfall and crack changes and uploads them to the server in real time;

[0045] Step 2: Calculate and obtain the comprehensive deformation information of the site based on the solved displacement data and sensor data; input the comprehensive displacement information into the early warning algorithm to obtain the early warning risk value;

[0046] Step 3: Based on the comprehensive deformation and warning risk value information, conduct real-time monitoring and early warning of the status of the potential danger point where the monitoring equipment is located.

[0047] The following is further described in conjunction with specific embodiments and execution steps:

[0048] Specifically, the geological disaster monitoring and early warning method based on multi-source fusion includes two parts: geological disaster monitoring and geological disaster early warning. The steps of the geological disaster monitoring method include:

[0049] Step 1: The displacement monitoring data, inclinometer, rain gauge, crack meter, and soil moisture meter sensor data are transmitted to the solution server through TCP. After the data solution is completed, it is forwarded to the monitoring and early warning module;

[0050] Step 2: The original product files of meteorological products are saved in the file directory of the FTP server. The meteorological product parsing and storage program obtains the meteorological product files released by the meteorological department from the FTP server, and then performs data parsing operations on these files. After the parsing is completed, the program stores the processed data in the Redis database. The operation process of the entire program is as follows Figure 2 As shown, the flowchart shows the data transmission and processing flow in detail.

[0051] Step 3: Take the weighted average of the displacement monitoring data and sensor data, and compare the obtained displacement value with the set threshold to determine whether displacement has occurred and the degree of displacement. Displacement value D total =E n +P s ·S+P c ·C+P r ·R+P m ·M, where E n is the displacement monitoring data collected by the monitoring station, P s is the weight of the soil moisture meter, S is the soil moisture content, P c is the weight of the crack meter, C is the width of the soil crack, P r is the weight of the rain gauge, R is the rainfall at the monitoring station, P m is the weight of the mud level meter, and M is the soil silt thickness.

[0052] Step 4: Transfer the displacement monitoring data obtained in step 1 and the displacement value obtained in step 3 to the data visualization module to display the distribution and real-time status of all potential risk points. On the potential risk point monitoring page, the displacement value is displayed in the form of a data curve, and the spatial distribution, real-time status, and early warning log information of the monitoring equipment are displayed.

[0053] Step 5: The meteorological data obtained in step 2 is transferred to the data visualization module, and the meteorological conditions are displayed on the map layer using intelligent grid forecasting technology. It mainly includes two core parts: actual rainfall visualization and forecast rainfall visualization. According to the specific monitoring range, it is further subdivided into rainfall data in area B and rainfall data in area A. In addition, the color division of rainfall levels strictly follows the professional rainfall classification standards provided by the meteorological department.

[0054] The process of geological disaster early warning method is as follows Figure 3 As shown, the steps are:

[0055] Step 1: The data acquisition and processing process is consistent with steps 1 to 3 of the geological disaster monitoring method.

[0056] Step 2: The scheduled task module first obtains the algorithm configuration information in MySQL.

[0057] Step 3: The data acquisition and processing module obtains all required historical monitoring data from the InfluxDB database according to the configuration information, and performs data preprocessing and grouping.

[0058] Step 4: Add the training processes of all devices to the model training process queue in sequence. Each device process contains asynchronous training processes for horizontal and vertical data, and calls the algorithm model training modules to complete the model training.

[0059] Step 5: After the training is completed, the training process of each device will generate two optimal model files and two optimal parameter files, corresponding to the prediction models in the horizontal and vertical directions respectively.

[0060] Step 6: Call the optimal model file and optimal parameter file generated by the prediction model training module to achieve rapid modeling, and then calculate the comprehensive predicted displacement through the horizontal and vertical prediction results.

[0061] Step 7: Finally, the calculated comprehensive predicted displacement is compared with the threshold. If it exceeds the threshold, the predicted value and warning result are processed into a fixed JSON format and sent to RabbitMQ, waiting for the warning publishing module to consume the data.

[0062] Step 8: The warning release module loads the notification object information and monitors the RabbitMQ queue to obtain the notification message. It splices the notification content and sends it to the contact in the manner specified by the configuration information. The warning has two message notification methods: SMS and email.

[0063] In summary, the present invention has the following advantages:

[0064] 1. It integrates information from different sensors and data sources, including satellite positioning data, sensor data, and meteorological data, to provide more comprehensive monitoring information. Moreover, the integration of multiple data sources improves the accuracy of the monitoring system. By comprehensively utilizing various advantages and making up for their respective limitations, it achieves refined and high-resolution monitoring of geological disasters and improves the reliability of monitoring results. At the same time, the multi-source fusion system can provide more timely and comprehensive monitoring information, help relevant departments respond and make decisions more quickly, improve the effectiveness of geological disaster warnings, and reduce possible losses.

[0065] 2. A flexible and effective data storage strategy is adopted. By combining the MongoDB sharding cluster and the InfluxDB database, the system can store a large amount of high-frequency, multi-device monitoring data for a long time, and ensure the importance of historical data for model prediction and visualization. At the same time, the use of a table partitioning strategy and the method of cutting data tables by monitoring area and month improves access efficiency and data management flexibility.

[0066] 3. A multi-parameter real-time early warning model was established, using a modular design. The model’s early warning effect can be adjusted by configuring parameters, and it has the ability to monitor and analyze multiple data types in real time, including displacement, moisture content, cumulative displacement, and displacement change rate indicators. By analyzing the real-time monitoring data of multiple sensors and combining the corresponding relationship between each parameter’s early warning threshold and early warning level, the model can quickly output multi-parameter early warning results.

[0067] What is disclosed above is only a preferred embodiment of the present invention, and it certainly cannot be used to limit the scope of rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A geological disaster monitoring and early warning system based on multi-source fusion, characterized in that: Including data collection module, background business module, geological disaster warning module, warning release module, data visualization module; The data acquisition module is used for receiving, preprocessing and storing data; the background business module is used to provide query, download and business management function support for the front-end program, and provide data access and login authority control for users; the geological disaster warning module is used to operate according to the pre-set rules and configuration information in the database, and issue a warning when a geological disaster risk is found; The warning release module is used to promptly convey warning information and equipment status information to relevant personnel; The data visualization module displays the information of the selected potential risk points, including the spatial distribution of monitoring equipment, real-time status, early warning logs, and data curves.

2. The geological disaster monitoring and early warning system based on multi-source fusion according to claim 1, characterized in that: The data received by the data acquisition module include GNSS monitoring data, sensor data and meteorological product data, among which GNSS monitoring data is displacement monitoring data. The sensor data of the inclinometer, rain gauge, crack meter and soil moisture meter are solved by the solution server and then transmitted through the message middleware Rabbitmq. The meteorological product data is provided by the meteorological monitoring department, including the actual and forecast precipitation data of area A and area B.

3. The geological disaster monitoring and early warning system based on multi-source fusion as claimed in claim 2, characterized in that: The backend business module adopts two methods to meet the needs of data query. The first method is to realize front-end request parsing, data query and data echo through MVC mode and MyBatis framework; the second method allows users to enter SQL statements on the front end and then perform parsing and query in the background.

4. The geological disaster monitoring and early warning system based on multi-source fusion as claimed in claim 3 is characterized in that: The geological disaster warning module provides parameter configuration, warning log recording and warning services, wherein the warning service includes predictive warning and real-time warning. The predictive warning algorithm is developed based on Python's Django framework, and the real-time warning is a multi-parameter real-time warning, including key indicators of displacement, water content, cumulative displacement and displacement change rate.

5. The geological disaster monitoring and early warning system based on multi-source fusion according to claim 4 is characterized in that: The warning release module loads notification object information, monitors the RabbitMQ queue to obtain warning information and device status information, and has two message notification methods: SMS and email.

6. The geological disaster monitoring and early warning system based on multi-source fusion according to claim 5 is characterized in that: The data visualization module is developed based on the Vue framework, supports Chinese and English versions, and uses ECharts components to achieve intuitive display of monitoring data and statistics; the data visualization module displays information on hidden danger points on a high-definition map layer, including the spatial distribution of monitoring equipment, real-time status, early warning logs, and deformation data curves; it also supports meteorological data visualization, and has real-time precipitation, accumulated rainfall query, and precipitation forecast display functions.

7. A geological disaster monitoring and early warning method based on multi-source fusion, using the geological disaster monitoring and early warning system based on multi-source fusion as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: Step 1: The GNSS monitoring station installed on the landslide collects displacement monitoring data and transmits it to the solution server in real time for solution; the multi-source auxiliary sensor collects the tilt angle, rainfall and crack changes and uploads them to the server in real time; Step 2: Calculate and obtain the comprehensive deformation information of the site based on the solved displacement data and sensor data; input the comprehensive displacement information into the early warning algorithm to obtain the early warning risk value; Step 3: Based on the comprehensive deformation and warning risk value information, conduct real-time monitoring and early warning of the status of the potential danger point where the monitoring equipment is located.

8. The method for monitoring and early warning of geological disasters based on multi-source fusion according to claim 7, characterized in that: The comprehensive deformation information in step 2 is the displacement value D total , the expression is as follows: D total =E n +P s ·S+P c ·C+P r ·R+P m ·M Among them, E n is the displacement monitoring data collected by the monitoring station, P s is the weight of the soil moisture meter, S is the soil moisture content, P c is the weight of the crack meter, C is the width of the soil crack, P r is the weight of the rain gauge, R is the rainfall at the monitoring station, P m is the weight of the mud level meter, and M is the soil silt thickness.

9. The method for monitoring and early warning of geological disasters based on multi-source fusion according to claim 8, characterized in that: The steps for early warning in step 3 are different from those for real-time monitoring and include the following steps: Step 3.1: The scheduled task module obtains the algorithm configuration information in MySQL; Step 3.2: The data acquisition and processing module obtains all required historical monitoring data from the InfluxDB database according to the configuration information, and performs data preprocessing and grouping; Step 3.3: Add the training processes of all devices to the model training process queue in sequence; each device process contains asynchronous training processes for horizontal and vertical data, and calls the algorithm model training modules to complete the model training; Step 3.4: After the training is completed, the training process of each device will generate two optimal model files and two optimal parameter files, corresponding to the prediction models in the horizontal and vertical directions respectively; Step 3.5: Call the optimal model file and optimal parameter file generated by the prediction model training module to achieve rapid modeling, and then calculate the comprehensive prediction displacement through the horizontal and vertical prediction results; Step 3.6: Compare the calculated comprehensive predicted displacement with the threshold. If it exceeds the threshold, the predicted value and warning result are processed into a fixed JSON format and sent to RabbitMQ, waiting for the warning publishing module to consume the data; Step 3.7: The warning publishing module loads the notification object information, monitors the RabbitMQ queue to obtain the notification message, splices the notification content and sends it to the contact in the manner specified by the configuration information. The warning has two message notification methods: SMS and email.

Images