An intelligent early warning system for geological disasters based on big data
By designing a geological disaster intelligent warning system based on big data and integrating multi-dimensional data analysis and neural network modules, the problem of difficult to predict and early warning of geological disasters in the existing technology is solved, real-time monitoring and early warning of potential geological disasters is achieved, and the improvement of real-time monitoring and early warning of potential geological disasters is achieved. Prediction accuracy and timeliness, reducing losses and harm.
Patent Information
- Application Number
- CN202410542407.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-30
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-04-30
AI Technical Summary
The existing technology is difficult to effectively predict and early warning of geological disasters, resulting in difficulties in reducing losses and hazards.
Design an intelligent early warning system for geological disasters based on big data, integrates meteorological monitoring, geological monitoring, topography and topography monitoring, image processing, analysis, comparison and neural network modules, and build auxiliary prediction models through multi-dimensional data analysis and deep learning, real-time Monitoring and early warning.
Real-time monitoring and early warning of potential geological disasters is achieved, the accuracy and timeliness of prediction are improved, the losses and harms caused by disasters are reduced, and the reliability of early warnings is enhanced.
Smart Images

Figure CN118486134B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of disaster early warning technology, and more specifically, to an intelligent early warning system for geological disasters based on big data. Background Art
[0002] Geological disasters refer to geological actions or phenomena that are caused by natural or human factors and cause losses to human life and property and damage to the environment. The distribution and change patterns of geological disasters in time and space are subject to both the natural environment and human activities, and are often the result of the interaction between humans and nature. Geological disasters mainly refer to collapse, i.e. dangerous rock masses, landslides, mudslides, karst ground collapses and ground fissures, which are generally recognized to be caused by drastic changes in the original crustal surface geological structure, and are usually considered to be sudden and easy to cause large-scale impacts. It is urgent to carry out comprehensive research on geological disaster risk assessment and occurrence patterns to prevent the occurrence of geological disasters and reduce the impact of losses. Therefore, an intelligent early warning system for geological disasters based on big data is proposed.
[0003] Invention Project
[0004] To achieve the above object, the present invention provides the following technical solutions:
[0005] An intelligent early warning system for geological disasters based on big data, including a meteorological monitoring module, a topographic monitoring module, a geological monitoring module, an image processing module, an analysis module, a comparison module, a neural network module, and an early warning module;
[0006] The meteorological monitoring module is used to monitor the meteorological environment information of the target area and summarize it into a meteorological environment monitoring information set, and then transmit the meteorological environment monitoring information set to the analysis module;
[0007] The geological monitoring module is used to monitor the soil layer information at different depths in the target area and summarize it into a geological monitoring information set, and then transmit the geological monitoring information set to the analysis module;
[0008] The analysis module is used to conduct joint analysis based on the meteorological environment monitoring information set and the geological monitoring information set, and decide whether to perform a prediction operation based on the analysis results;
[0009] When a prediction operation is required, the topography monitoring module collects information about the topography of the target area and transmits it to the image processing module;
[0010] The image processing module is used to process images according to the information collected by the terrain monitoring module, and transmit the processing results to the comparison module;
[0011] The comparison module is used to compare the processed results with all historical disaster comparison sets of the target area for similarity, and transmit the maximum similarity as the comparison result to the neural network module;
[0012] The neural network module is used to construct an auxiliary prediction model based on the meteorological monitoring module, the geological monitoring module, and the comparison results, and transmit the output results of the prediction model to the early warning module when a prediction operation is required;
[0013] The early warning module performs early warning operations on the target area according to the preset early warning strategy.
[0014] In a preferred embodiment, the meteorological environment monitoring information set includes various meteorological environment data, and the geological monitoring information set includes various soil layer fluctuation data of each preset monitoring depth.
[0015] In a preferred embodiment, the analysis module is used to perform joint analysis based on the meteorological environment monitoring information set and the geological monitoring information set, which means:
[0016] Obtain various meteorological environment data in the meteorological environment monitoring information set and various soil layer fluctuation data of the soil layer at each preset monitoring depth in the geological monitoring information set, mark the various meteorological environment data in the meteorological environment monitoring information set as QX i, and mark the various soil layer fluctuation data of the soil layer at each preset monitoring depth in the geological monitoring information set as TC i;
[0017] The meteorological observation values are obtained by calculation: QG i is the meteorological observation value, qi is the preset proportional coefficient corresponding to each meteorological environment data QX i;
[0018] The geological observation value of each observation point is obtained by calculation: DG i is the geological observation value, ti is the preset proportional coefficient corresponding to each soil layer fluctuation data TC i, di is the depth of the observation point, and then the largest geological observation value DG i is screened out and marked as DGmax;
[0019] Perform joint analysis: The joint analysis value is calculated using the following formula: FXi = f1·lnDGmax+f2·lnDGi, where FXi is the joint analysis value, and f1 and f2 are both preset joint coefficients.
[0020] In a preferred embodiment, the analysis module determines whether to perform a prediction operation according to the analysis result, which means:
[0021] The joint analysis value FXi is compared with the preset analysis threshold. If the joint analysis value FXi is greater than or equal to the preset analysis threshold, a prediction operation is performed. If the joint analysis value FXi is less than the preset analysis threshold, no prediction operation is performed.
[0022] In a preferred embodiment, the image processing module is used to perform image processing based on the information collected by the terrain monitoring module, which means:
[0023] While obtaining multi-source satellite images of the monitoring area from the topography monitoring module, the vector file of the monitoring area is imported, the image is cropped by ArcGIS software to obtain the image of the monitoring area, and the cropped image is preprocessed;
[0024] The images of the monitoring area are identified by pattern recognition technology, and the total number of pixels of bare land, the total number of pixels of vegetation, and the total number of pixels of lakes are calculated respectively. Then, the change amount under the acquisition time window of two adjacent images is calculated to obtain the change value of the total number of pixels of bare land, the total number of pixels of vegetation, and the total number of pixels of lakes, which are summarized together to obtain the image change set 1;
[0025] Import the high-precision digital elevation model of the monitoring area, calculate the slope of the hillside and the depth of the canyon in the monitoring area by combining the image of the monitoring area and the high-precision digital elevation model, and then calculate the slope change and depth change in the acquisition time window of two adjacent images to obtain the slope change of each hillside and the depth change of each canyon, and then summarize them to obtain the second image change set;
[0026] The image change set 1 and the image change set 2 are summarized to obtain the total image change set as the processing result.
[0027] In a preferred embodiment, the comparison module is used to compare the processing result with all historical disaster comparison sets of the target area for similarity, and the maximum similarity is transmitted to the neural network module as the comparison result, which means:
[0028] Obtain the total pixel change values of bare land, vegetation and lakes in the total image change set; mark the total pixel change value of bare land as LDi, the total pixel change value of vegetation as ZWi, and the total pixel change value of lakes as HPi, obtain the slope change of each hillside and the depth change of each canyon, and then mark the slope change of each hillside as PD i, and the depth change of each canyon as SD i, then perform similarity comparison between the total image change set and all historical disaster comparison sets, sort the data in the total image change set according to the preset standard to obtain the image change vector A, sort the data of each historical disaster comparison set according to the preset standard to obtain the disaster comparison vector B, then calculate the image change vector A, the disaster comparison vector B, perform similarity calculation, and send the maximum similarity as the comparison result to the neural network module.
[0029] In a preferred embodiment, the neural network module is used to construct an auxiliary prediction model based on the meteorological monitoring module, the geological monitoring module, and the comparison results, and when a prediction operation is required, the output result of the prediction model is transmitted to the early warning module, which means:
[0030] When constructing the auxiliary prediction model, the neural network module receives the meteorological environment monitoring information set, the geological monitoring information set, the comparison results, and the image acquisition time of the monitoring target area, and uses the various meteorological environment data in the meteorological environment monitoring information set, the various soil layer fluctuation data of each preset monitoring depth in the geological monitoring information set, the comparison result, i.e., the maximum similarity, and the image acquisition time of the monitoring target area as the input data for model training, and uses the coordinates of the most likely point of occurrence of geological disasters and the geological disaster level value as the output data for model training. When the performance of the model training meets the preset requirements, the construction of the auxiliary prediction model is completed; when a prediction operation is required, the output results of the prediction model, i.e., the coordinates of the most likely point of occurrence of geological disasters and the geological disaster level value, are transmitted to the early warning module.
[0031] Technical effects and advantages of the present invention:
[0032] The present invention realizes real-time monitoring and early warning of potential geological disasters by integrating multi-dimensional data such as meteorology, geology, topography and image processing. By analyzing meteorological and geological data and combining topographic and geomorphological change information, it can more accurately assess the possibility of geological disasters and take countermeasures in advance. The system can collect and process environmental data in real time. When potential risks are monitored, it can quickly start the early warning process to effectively reduce the losses caused by disasters. The system uses data analysis technologies such as neural networks to integrate various monitoring data for deep learning and pattern recognition, thereby improving the intelligent level of prediction and integrating multiple data sources such as meteorology, geology, topography and satellite images, so that the early warning system can comprehensively analyze the geological environment from different angles, thereby increasing the reliability of early warning.
[0033] By comparing the similarity between the current changes in topography and historical disaster data, the system can discover possible abnormal situations and provide a basis for prevention; the prediction model output provided by the system can serve as an important reference for the government and emergency management departments to formulate preventive measures, which is helpful for scientific decision-making. It not only improves the accuracy and timeliness of geological disaster predictions, but also provides strong data support for related decisions, which is of great significance for reducing casualties and property losses caused by natural disasters. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;
[0035] Figure 1This is a schematic diagram of a geological disaster intelligent early warning system based on big data in the present invention. DETAILED DESCRIPTION
[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0037] Example 1
[0038] A big data-based intelligent early warning system for geological disasters includes a meteorological monitoring module, a topographic monitoring module, a geological monitoring module, an image processing module, an analysis module, a comparison module, a neural network module, and an early warning module; the meteorological monitoring module, the topographic monitoring module, the geological monitoring module, the image processing module, the analysis module, the comparison module, the neural network module, and the early warning module are communicated with each other.
[0039] The meteorological monitoring module is used to monitor the meteorological environment information of the target area and summarize it into a meteorological environment monitoring information set, and then transmit the meteorological environment monitoring information set to the analysis module; the meteorological monitoring module is responsible for collecting and monitoring the meteorological environment information of the target area, such as temperature, humidity, wind speed, rainfall, etc. This information has an important impact on the occurrence of geological disasters. For example, an increase in rainfall may lead to landslides, and changes in wind speed may accelerate soil erosion. The monitored meteorological environment information will be integrated into a meteorological environment monitoring information set, which may include various meteorological data in different time periods, such as temperature, humidity, wind speed, etc. at different times. For example, suppose the meteorological monitoring module monitors the following meteorological data in the target area within one day: at 08:00, the temperature is 20℃, the humidity is 70%, and the wind speed is 5m / s; at 12:00, the temperature is 25℃, the humidity is 60%, and the wind speed is 8m / s. These data will be integrated into a meteorological environment monitoring information set.
[0040] The geological monitoring module is used to monitor the soil layer information at different depths in the target area and summarize it into a geological monitoring information set, and then transmit the geological monitoring information set to the analysis module; the geological monitoring module is responsible for monitoring the soil layer information at different depths in the target area to understand the changes in underground soil, including soil pressure, humidity, displacement, etc. The displacement is decomposed into movements in multiple preset directions, such as the movement distance in the vertical direction and the movement distance in the horizontal direction. The monitored soil layer fluctuation data will be integrated into a geological monitoring information set. This information set includes soil layer data at different depths, such as changes in soil humidity, pressure, and displacement at different depths over time.
[0041] The analysis module is used to perform joint analysis based on the meteorological environment monitoring information set and the geological monitoring information set, and decide whether to perform a prediction operation based on the analysis results; obtain data from the meteorological monitoring module and the geological monitoring module, including various meteorological environment data QXi in the meteorological environment monitoring information set and various soil layer fluctuation data TCi of the soil layers at each preset monitoring depth in the geological monitoring information set; perform joint analysis on the meteorological observation value QG i and the largest screened geological observation value DGmax, and weightedly sum the meteorological observation value and the geological observation value through the preset joint coefficients f1 and f2 to obtain the joint analysis value FXi; the joint analysis value comprehensively considers the influence of meteorological environment information and geological monitoring information, and can be used as a comprehensive indicator to evaluate the possibility of geological disasters in the target area. By weighting the meteorological and geological data and comprehensively considering the influencing factors of both, an assessment of the possibility of geological disasters can be obtained. If the joint analysis value exceeds the preset threshold, it means that the geological environment and meteorological conditions in the target area have potential geological disaster risks and further prediction operations are required. Otherwise, no further prediction operations are required.
[0042] When prediction operations are required, the topography and geomorphology monitoring module collects information on the topography and geomorphology of the target area and transmits it to the image processing module; the topography and geomorphology monitoring module collects topography and geomorphology information of the target area, which includes terrain elevation, geomorphological features, mountain morphology, water system distribution, etc. This information is of great significance to the formation and development of geological disasters.
[0043] The image processing module is used to process images according to the information collected by the terrain and geomorphology monitoring module, and transmit the processing results to the comparison module; the information provided by the terrain and geomorphology monitoring module is used to guide the image processing module to process multi-source satellite images, use ArcG IS software to crop the images to obtain images of the monitoring area, and perform preprocessing, such as removing noise, enhancing contrast, etc. The images are classified and identified through pattern recognition technology, and the total number of pixels of features such as bare land, vegetation, and lakes is calculated. The change in the time window of adjacent images is calculated, and the change in the slope of the hillside and the depth of the canyon are calculated in combination with the digital elevation model. The above change is obtained by subtracting the target type data in the previous image time window from the target type data in the next image time window.
[0044] The comparison module is used to compare the processing results with all historical disaster comparison sets of the target area for similarity, and transmit the maximum similarity as the comparison result to the neural network module; the comparison module compares the processing results with the historical disaster comparison set of the target area for similarity to evaluate the degree of change of the current geological environment and the similarity with historical disasters, and marks the total number of pixel changes of bare land, vegetation, and lakes in the processing results, as well as the slope change of the hillside and the depth change of the canyon as image change vector A, compares the data in the historical disaster comparison set, calculates the disaster comparison vector B, and finds the maximum similarity between the image change vector A and the disaster comparison vector B as the comparison result through similarity calculation such as cosine similarity calculation. The image processing module obtains the change information related to geological disasters by processing and analyzing the satellite images of the monitoring area, including changes in surface features, changes in terrain, etc. The comparison module compares the processing results with historical disasters to determine whether the current geological environment is abnormal or similar to historical disasters, and provides more accurate input data for the prediction model. This process can detect signs of geological disasters in advance by comparing the similarity between the current terrain and landform changes and historical disasters, providing an important basis for the prediction and early warning of geological disasters.
[0045] The neural network module is used to assist in the construction of a prediction model based on the meteorological monitoring module, the geological monitoring module, and the comparison results, and when a prediction operation is required, the output results of the prediction model are transmitted to the early warning module; the neural network module uses the multi-source data provided by the meteorological monitoring module, the geological monitoring module, and the comparison module to assist in the construction of a geological disaster prediction model. The input data of the model includes meteorological environment monitoring information, geological monitoring information, comparison results, and the image acquisition time of the monitoring target area. These data provide a variety of information, including meteorology, geological environment, topography, etc., which helps to improve the accuracy and reliability of the prediction model. When the neural network module uses the provided data for model training, it takes the meteorological environment data, geological monitoring data, comparison results, and the time of monitoring images as input to maximize the probability of geological disasters. The coordinates of the occurrence point and the geological disaster level value are used as outputs. By continuously adjusting the model parameters, the model can more accurately predict the location and level of geological disasters. When geological disaster prediction is needed, the neural network module will apply the trained prediction model to new data to obtain the coordinates of the most likely occurrence point of geological disasters and the disaster level. These prediction results will be sent to the early warning module for formulating corresponding early warning measures to reduce the possible impact of disasters. The neural network module constructs a complex prediction model by comprehensively utilizing data from different sources, which can more accurately predict the occurrence of geological disasters. The establishment and application of the prediction model can improve the accuracy and timeliness of geological disaster predictions, help to take effective prevention and response measures in a timely manner, and reduce the losses and harm caused by geological disasters to people.
[0046] The early warning module performs early warning operations on the target area according to the preset early warning strategy. When the prediction operation is not required, the early warning module does not perform early warning operations. Specifically, the early warning module divides the early warning level into different levels according to the geological disaster prediction results, such as low risk, medium risk, and high risk. When the geological disaster prediction results reach a certain threshold, the corresponding early warning level is triggered. For example, when the geological disaster level is predicted to be medium or above risk, a medium risk early warning is triggered. The early warning module issues corresponding early warning content according to different early warning levels. The early warning content may include information such as the type of geological disaster, possible impact range, and preventive measures. The early warning module will select appropriate early warning methods to issue early warning information to relevant departments and the public, such as through SMS, telephone, television, radio, the Internet and other channels. The early warning module may also include instructing relevant departments and the public to take corresponding emergency measures, such as evacuating dangerous areas, strengthening patrol monitoring, and strengthening protective facilities. Example scenario:
[0047] Suppose that after a certain area is analyzed by a geological disaster prediction model, the prediction results show that there is a possibility of medium or above risk of landslide disaster in the area. According to the warning strategy of the warning module: when the landslide hazard level is predicted to be medium or above risk, a landslide disaster warning is issued, including information such as the areas that may be affected and preventive measures. A warning announcement is issued through the local government. At the same time, warning information is released to the public through text messages, television, radio and other channels. Residents are advised to temporarily leave potential danger areas, strengthen patrol monitoring, remain vigilant, and avoid staying in areas where landslides may occur. Such warning strategies and measures can help communities and the public to make timely prevention and response preparations and reduce the losses and hazards that may be caused by geological disasters.
[0048] The meteorological environment monitoring information set includes various meteorological environment data, and the geological monitoring information set includes various soil layer fluctuation data of each preset monitoring depth.
[0049] The analysis module is used to conduct joint analysis based on the meteorological environment monitoring information set and the geological monitoring information set, which refers to:
[0050] Obtain various meteorological environment data in the meteorological environment monitoring information set and various soil layer fluctuation data of the soil layer at each preset monitoring depth in the geological monitoring information set, mark the various meteorological environment data in the meteorological environment monitoring information set as QX i, and mark the various soil layer fluctuation data of the soil layer at each preset monitoring depth in the geological monitoring information set as TC i;
[0051] The meteorological observation values are obtained by calculation: QG i is the meteorological observation value, qi is the preset proportional coefficient corresponding to each meteorological environmental data QX i, this coefficient is used to weight the meteorological environmental data to consider the impact of different meteorological factors on geological disasters. Each meteorological environmental data may have different importance to the occurrence of geological disasters, so the preset proportional coefficient is used to weight these data;
[0052] The geological observation value of each observation point is obtained by calculation: DG i is the geological observation value, ti is the preset proportional coefficient corresponding to each soil layer fluctuation data TC i, which is similar to the preset proportional coefficient of meteorological environment data. This coefficient is used to weight the soil layer fluctuation data, considering the influence of different soil layer fluctuation factors on geological disasters. Different types of soil layer fluctuation data may have different contributions to the occurrence of geological disasters. Therefore, the preset proportional coefficient is used to weight these data, di is the depth of the observation point, and then the largest geological observation value DG i is screened out and marked as DGmax;
[0053] Conduct joint analysis: The joint analysis value is calculated by the following formula: FXi = f1·lnDGmax+f2·lnDGi, FXi is the joint analysis value, which is the result obtained after joint analysis based on the meteorological environment monitoring information set and the geological monitoring information set. It represents the comprehensive influence of meteorological environment data and soil layer fluctuation data, and reflects the possibility of geological disasters. The calculation of FXi involves the weighted combination of meteorological observation values and geological observation values, as well as the preset joint coefficient. It can help the analysis module determine the risk level of geological disasters in the current environment. Both f1 and f2 are preset joint coefficients, which are used to control the influence of the weights of different data sources in the joint analysis. The preset joint coefficient reflects the assumption or priority setting of the relative importance of meteorological environment data and soil layer fluctuation data in the analysis module. By adjusting the values of f1 and f2, the weights of meteorological environment data and soil layer fluctuation data in the joint analysis can be adjusted to meet different analysis needs and scenarios.
[0054] The analysis module decides whether to perform a prediction operation based on the analysis results:
[0055] The joint analysis value FX i is compared with the preset analysis threshold. If the joint analysis value FX i is greater than or equal to the preset analysis threshold, a prediction operation is performed. If the joint analysis value FX i is less than the preset analysis threshold, no prediction operation is performed.
[0056] The image processing module is used to process images based on the information collected by the terrain monitoring module.
[0057] While obtaining multi-source satellite images of the monitoring area from the topography monitoring module, the vector file of the monitoring area is imported, the image is cropped by ArcGIS software to obtain the image of the monitoring area, and the cropped image is preprocessed;
[0058] The images of the monitoring area are identified by pattern recognition technology, and the total number of pixels of bare land, the total number of pixels of vegetation, and the total number of pixels of lakes are calculated respectively. Then, the change amount under the acquisition time window of two adjacent images is calculated to obtain the change value of the total number of pixels of bare land, the total number of pixels of vegetation, and the total number of pixels of lakes, which are summarized together to obtain the image change set 1;
[0059] Import the high-precision digital elevation model of the monitoring area, calculate the slope of the hillside and the depth of the canyon in the monitoring area by combining the image of the monitoring area and the high-precision digital elevation model, and then calculate the slope change and depth change in the acquisition time window of two adjacent images to obtain the slope change of each hillside and the depth change of each canyon, and then summarize them to obtain the second image change set;
[0060] The image change set 1 and the image change set 2 are summarized to obtain the total image change set as the processing result.
[0061] The comparison module is used to compare the processing results with all historical disaster comparison sets of the target area for similarity, and the maximum similarity is sent to the neural network module as the comparison result, which means:
[0062] Obtain the total pixel change values of bare land, vegetation and lakes in the total image change set; mark the total pixel change value of bare land as LDi, the total pixel change value of vegetation as ZWi, and the total pixel change value of lakes as HPi, obtain the slope change of each hillside and the depth change of each canyon, and then mark the slope change of each hillside as PDi, and the depth change of each canyon as SDi, then perform a similarity comparison between the total image change set and all historical disaster comparison sets, sort the data in the total image change set according to the preset standard to obtain the image change vector A, sort the data of each historical disaster comparison set according to the preset standard to obtain the disaster comparison vector B, then calculate the image change vector A, the disaster comparison vector B, perform similarity calculation, and send the maximum similarity as the comparison result to the neural network module.
[0063] The neural network module is used to construct an auxiliary prediction model based on the meteorological monitoring module, geological monitoring module, and comparison results, and when a prediction operation is required, the output results of the prediction model are transmitted to the early warning module.
[0064] When constructing the auxiliary prediction model, the neural network module receives the meteorological environment monitoring information set, the geological monitoring information set, the comparison results, and the image acquisition time of the monitoring target area, and uses the various meteorological environment data in the meteorological environment monitoring information set, the various soil layer fluctuation data of each preset monitoring depth in the geological monitoring information set, the comparison result, i.e., the maximum similarity, and the image acquisition time of the monitoring target area as the input data for model training, and uses the coordinates of the most likely point of occurrence of geological disasters and the geological disaster level value as the output data for model training. When the performance of the model training meets the preset requirements, the construction of the auxiliary prediction model is completed; when a prediction operation is required, the output results of the prediction model, i.e., the coordinates of the most likely point of occurrence of geological disasters and the geological disaster level value, are transmitted to the early warning module.
[0065] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0066] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0067] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0068] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0069] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0070] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A geological disaster intelligent early warning system based on big data, characterized in that: It includes meteorological monitoring module, topographic monitoring module, geological monitoring module, image processing module, analysis module, comparison module, neural network module and early warning module; The meteorological monitoring module is used to monitor the meteorological environment information of the target area and summarize it into a meteorological environment monitoring information set, and then transmit the meteorological environment monitoring information set to the analysis module; The geological monitoring module is used to monitor the soil layer information at different depths in the target area and summarize it into a geological monitoring information set, and then transmit the geological monitoring information set to the analysis module; The analysis module is used to conduct joint analysis based on the meteorological environment monitoring information set and the geological monitoring information set, and decide whether to perform a prediction operation based on the analysis results; When a prediction operation is required, the topography monitoring module collects information about the topography of the target area and transmits it to the image processing module; The image processing module is used to process images according to the information collected by the terrain monitoring module, and transmit the processing results to the comparison module; The comparison module is used to compare the processed results with all historical disaster comparison sets of the target area for similarity, and transmit the maximum similarity as the comparison result to the neural network module; The neural network module is used to construct an auxiliary prediction model based on the meteorological monitoring module, the geological monitoring module, and the comparison results, and transmit the output results of the prediction model to the early warning module when a prediction operation is required; The early warning module performs early warning operations on the target area according to the preset early warning strategy; The meteorological environment monitoring information set includes various meteorological environment data, and the geological monitoring information set includes various soil layer fluctuation data of each preset monitoring depth; The analysis module is used to conduct joint analysis based on the meteorological environment monitoring information set and the geological monitoring information set, which refers to: Obtain various meteorological environment data in the meteorological environment monitoring information set and various soil layer fluctuation data of the soil layer at each preset monitoring depth in the geological monitoring information set, mark the various meteorological environment data in the meteorological environment monitoring information set as QXi, and mark the various soil layer fluctuation data of the soil layer at each preset monitoring depth in the geological monitoring information set as TCi; The meteorological observation values are obtained by calculation: , QGi is the meteorological observation value, qi is the preset proportional coefficient corresponding to each meteorological environmental data QXi; The geological observation value of each observation point is obtained by calculation: , DGi is the geological observation value, ti is the preset proportional coefficient corresponding to each soil layer fluctuation data TCi, di is the depth of the observation point, and then the largest geological observation value DGi is screened out and marked as DGmax; Perform conjoint analysis: The conjoint analysis value is calculated using the following formula: , FXi is the joint analysis value, f1 and f2 are the preset joint coefficients; The analysis module decides whether to perform a prediction operation based on the analysis results: The joint analysis value FXi is compared with the preset analysis threshold. If the joint analysis value FXi is greater than or equal to the preset analysis threshold, a prediction operation is performed. If the joint analysis value FXi is less than the preset analysis threshold, no prediction operation is performed. The image processing module is used to process images based on the information collected by the terrain monitoring module. While obtaining multi-source satellite images of the monitoring area from the topography and geomorphology monitoring module, the vector file of the monitoring area is imported, the image is cropped using ArcGIS software to obtain the image of the monitoring area, and the cropped image is preprocessed; The images of the monitoring area are identified by pattern recognition technology, and the total number of pixels of bare land, the total number of pixels of vegetation, and the total number of pixels of lakes are calculated respectively. Then, the change amount under the acquisition time window of two adjacent images is calculated to obtain the change value of the total number of pixels of bare land, the total number of pixels of vegetation, and the total number of pixels of lakes, which are summarized together to obtain the image change set 1; Import the high-precision digital elevation model of the monitoring area, calculate the slope of the hillside and the depth of the canyon in the monitoring area by combining the image of the monitoring area and the high-precision digital elevation model, and then calculate the slope change and depth change in the acquisition time window of two adjacent images to obtain the slope change of each hillside and the depth change of each canyon, and then summarize them to obtain the second image change set; The image change set 1 and the image change set 2 are summarized to obtain the total image change set as the processing result; The comparison module is used to compare the processing results with all historical disaster comparison sets of the target area for similarity, and the maximum similarity is sent to the neural network module as the comparison result, which means: Obtain the total number of pixel point changes of bare land, vegetation and lakes in the total image change set; mark the total number of pixel point changes of bare land as LDi, the total number of pixel point changes of vegetation as ZWi, and the total number of pixel point changes of lakes as HPi, obtain the slope change of each hillside and the depth change of each canyon, and then mark the slope change of each hillside as PDi, and the depth change of each canyon as SDi, and then compare the image change set with all historical disaster comparison sets for similarity, sort the data in the image change set according to the preset standard to obtain the image change vector A, sort the data of each historical disaster comparison set according to the preset standard to obtain the disaster comparison vector B, and then calculate the image change vector A and the disaster comparison vector B for similarity calculation, and send the maximum similarity as the comparison result to the neural network module; The neural network module is used to construct an auxiliary prediction model based on the meteorological monitoring module, geological monitoring module, and comparison results, and when a prediction operation is required, the output results of the prediction model are transmitted to the early warning module. When constructing the auxiliary prediction model, the neural network module receives the meteorological environment monitoring information set, the geological monitoring information set, the comparison results, and the image acquisition time of the monitoring target area, and uses the various meteorological environment data in the meteorological environment monitoring information set, the various soil layer fluctuation data of each preset monitoring depth in the geological monitoring information set, the comparison result, i.e., the maximum similarity, and the image acquisition time of the monitoring target area as the input data for model training, and uses the coordinates of the most likely point of occurrence of geological disasters and the geological disaster level value as the output data for model training. When the performance of the model training meets the preset requirements, the construction of the auxiliary prediction model is completed; when a prediction operation is required, the output results of the prediction model, i.e., the coordinates of the most likely point of occurrence of geological disasters and the geological disaster level value, are transmitted to the early warning module.
