A rural geological disaster prediction method and system based on edge computing

Abstract

This invention discloses a rural geological disaster prediction method and system based on edge computing, which relates to the field of geological disaster prediction technology. By comprehensively calculating the disaster prediction probability coefficient and the dynamic confidence coefficient obtained by dynamic confidence assessment based on the disaster prediction probability coefficient, the probability prediction magnitude is determined and the reliability of the prediction is assessed. This achieves second-level accurate early warning of rural geological disasters and improves the reliability of the early warning.

Description

Technical Field

[0001] This invention relates to the field of geological disaster prediction technology, specifically a rural geological disaster prediction method and system based on edge computing. Background Technology

[0002] Existing methods for detecting geological disasters in rural areas collect data by deploying edge computing nodes and then sending the data back to the cloud for processing. However, because these technologies rely on cloud processing, the large network signal delays during data transmission in rural and mountainous areas prevent them from providing second-level early warnings in the event of sudden disasters. This results in delayed disaster warnings and low warning efficiency. Summary of the Invention

[0003] To address the shortcomings mentioned in the background art, the present invention aims to provide a rural geological disaster prediction method and system based on edge computing, which can determine the accuracy of disaster prediction and early warning status through real-time probability prediction and confidence assessment, thereby improving the efficiency of disaster prediction and early warning.

[0004] Firstly, the objective of this invention can be achieved through the following technical solution: a method for predicting rural geological disasters based on edge computing, the method comprising the following steps: Feature extraction is performed on pre-processed rural geological environment data to obtain predictable feature data. The predictable feature data is then subjected to comprehensive geological disaster prediction processing and calculation to obtain disaster prediction probability coefficients. Among them, rural geological environment data includes surface subsidence data, surface vibration data, precipitation data, and surface deformation data. The probability coefficient of disaster prediction is subjected to dynamic confidence assessment to obtain the dynamic confidence coefficient of probability. When the dynamic confidence coefficient of probability is less than or equal to the preset confidence threshold, disaster prediction is suspended and geological disaster prediction is re-integrated; otherwise, when the probability coefficient of disaster prediction is greater than the preset prediction probability threshold, a local early warning is triggered; otherwise, no local early warning is triggered.

[0005] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the probability dynamic confidence coefficient is dynamically assessed based on the disaster prediction probability coefficient, the process of which is as follows: Based on Monte Carlo Dropout technology, the disaster prediction probability coefficients are subjected to T random forward propagation through the Dropout layer to obtain T disaster prediction probability coefficients Pit; where t is the prediction number index of the disaster prediction probability coefficient, t=1, 2, 3, ..., T; and T is the total number of predictions of the disaster prediction probability coefficient. The prediction uncertainty coefficient U is calculated as follows: In the formula, The average probability coefficient of disaster prediction is obtained through T predictions; The probability dynamic confidence coefficient is calculated based on the prediction uncertainty coefficient U, as follows: In the formula, norm() is the normalization function.

[0006] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the probability dynamic confidence coefficient is used to measure the prediction accuracy of the disaster prediction probability coefficient, and the magnitude of the probability dynamic confidence coefficient is proportional to the prediction accuracy of the disaster prediction probability coefficient.

[0007] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the preset confidence threshold is obtained by taking the average value after calculating multiple probability dynamic confidence coefficients; the preset prediction probability threshold is obtained by taking the average value after calculating multiple disaster prediction probability coefficients.

[0008] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the process of performing comprehensive geological disaster prediction processing and calculation on the feature data to be predicted, comprising the following steps: Settlement characteristic data is labeled Ci, seismic characteristic data is labeled Zi, precipitation characteristic data is labeled Ji, and deformation characteristic data is labeled Xi; where i is the number of characteristic data to be predicted, and i = 1, 2, 3, ..., n, where n is the total number of characteristic data to be predicted; The calculation formula for comprehensive geological disaster prediction and treatment is as follows: In the formula, Pi is the disaster prediction probability coefficient; a is the subsidence characteristic influence coefficient, b is the vibration characteristic influence coefficient, c is the precipitation characteristic influence coefficient, d is the deformation characteristic influence coefficient; k1, k2, k3, and k4 are all preset weight coefficients.

[0009] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the disaster prediction probability coefficient is divided into a geological stability probability coefficient, a geological attention probability coefficient, and a geological early warning probability coefficient according to the probability magnitude; The geological stability probability coefficient is less than the geological attention probability coefficient and less than the geological early warning probability coefficient.

[0010] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the process of feature extraction from the pre-processed rural geological environment data, as follows: The surface subsidence data, surface seismic data, precipitation data, and surface deformation data are arranged according to spatial topological relationships; spatial feature maps of each type are generated as feature data to be predicted. The feature data to be predicted includes settlement feature data, vibration feature data, precipitation feature data, and deformation feature data.

[0011] The beneficial effects of this invention are: This invention uses a disaster prediction probability coefficient calculated through comprehensive geological disaster prediction processing, and a dynamic confidence coefficient obtained through dynamic confidence assessment based on the disaster prediction probability coefficient, to determine the magnitude of the probability prediction and assess the reliability of the prediction; thereby achieving second-level accurate early warning of rural geological disasters and improving the reliability of the early warning. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Example 1: like Figure 1 As shown, a method for predicting rural geological disasters based on edge computing is presented, which includes the following steps: S101: Acquire rural geological environment data, and preprocess the rural geological environment data to obtain processed rural geological environment data; wherein, the rural geological environment data includes surface subsidence data, surface vibration data, precipitation data and surface deformation data; Specifically, in this embodiment, the process of collecting rural geological environment data includes: The process for collecting the surface subsidence data is as follows: By vertically deploying fiber optic grating deep inclinometers along the borehole, the deformation of sliding surfaces at multiple layers is monitored to obtain the displacement of deep soil and rock masses, which serves as surface settlement data. The process of collecting ground motion data is as follows: By deploying a vibration array sensor array, data such as the vibration waveform and vibration frequency of ground vibration are calculated and processed to obtain ground vibration data. The process of collecting precipitation data is as follows: We use the cumulative rainfall data released by the meteorological observatory over the past 10 minutes, 1 hour, and 24 hours as precipitation data. The process of collecting surface deformation data is as follows: A pre-set three-dimensional attitude sensor is installed on the top of the detection pile to detect changes in the three-dimensional deformation of the ground surface; this serves as ground surface deformation data. The process of preprocessing rural geological environment data includes: By deploying all types of data within the rural geological environment data within an edge computing gateway, and preprocessing the data through methods such as filtering and denoising, time alignment, and normalization, standard structured data is obtained; this serves as the processed rural geological environment data. S102: Extract features from the processed rural geological environment data to obtain the feature data to be predicted. Perform comprehensive geological disaster prediction processing and calculation on the feature data to be predicted to obtain the disaster prediction probability coefficient. It should be noted that the process of feature extraction from the processed rural geological environment data is as follows: The surface subsidence data, surface seismic data, precipitation data, and surface deformation data are arranged according to spatial topological relationships; In this embodiment, for example: Displacement sensor features at different depths (0m, 2m, 5m, 10m) on the same monitoring pile are used as one row of an image, and different feature types at the same depth are used as one column of an image, forming a 4×8 two-dimensional matrix; spatial feature maps of each type are generated as feature data to be predicted. The feature data to be predicted includes settlement feature data, vibration feature data, precipitation feature data, and deformation feature data; The process of performing comprehensive geological hazard prediction calculations on the feature data to be predicted includes the following steps: The feature data to be predicted are labeled, with settlement feature data labeled Ci, seismic feature data labeled Zi, precipitation feature data labeled Ji, and deformation feature data labeled Xi; where i is the number of feature data to be predicted, and i = 1, 2, 3, ..., n, and n is the total number of feature data to be predicted. When predicting geological hazards based on the identified feature data, the calculation formula is as follows: In the formula, Pi is the disaster prediction probability coefficient; a is the subsidence characteristic influence coefficient, b is the vibration characteristic influence coefficient, c is the precipitation characteristic influence coefficient, and d is the deformation characteristic influence coefficient; k1, k2, k3, and k4 are all preset weight coefficients, which are derived from the historical weight ratios of the influence of subsidence characteristics, vibration characteristics, precipitation characteristics, and deformation characteristics on the prediction probability. In this embodiment, the subsidence characteristic influence coefficient, vibration characteristic influence coefficient, precipitation characteristic influence coefficient, and deformation characteristic influence coefficient are calculated by comprehensively evaluating the influence of external factors when routinely acquiring surface subsidence data, surface vibration data, precipitation data, and surface deformation data, and performing corresponding feature extraction. These factors include human factors, machine detection, and environmental factors, etc.; human factors refer to those caused by improper human operation or scanning. It should be further explained that the disaster prediction probability coefficient is divided into geological stability probability coefficient, geological attention probability coefficient and geological early warning probability coefficient according to the probability magnitude; wherein, the geological stability probability coefficient is less than the geological attention probability coefficient and less than the geological early warning probability coefficient. Specifically, in this embodiment, the probability coefficient of geological stability is between (0, 0.4); the probability coefficient of geological attention is between (0.4, 0.8); and the probability coefficient of geological early warning is between [0.8, 1). Among them, the geological stability probability coefficient is a probability coefficient reflecting the degree of geological stability. The higher the probability, the higher the degree of geological stability; the geological attention probability coefficient is a probability coefficient reflecting whether geological stability requires attention to potential risks. If potential risks exist, it enters the observation state; the geological early warning probability coefficient is a probability coefficient reflecting whether geological stability requires early warning. If it is a high-risk state, the early warning mechanism is triggered. S103: Perform dynamic confidence assessment on the disaster prediction probability coefficient to obtain the dynamic confidence coefficient. When the dynamic confidence coefficient is less than or equal to the preset confidence threshold, suspend disaster prediction and perform re-integration of geological disaster prediction. Otherwise, when the disaster prediction probability coefficient is greater than the preset prediction probability threshold, trigger a local early warning. Otherwise, do not trigger a local early warning.

[0015] The probability dynamic confidence coefficient is used to measure the reliability of the disaster prediction probability coefficient; and the magnitude of the probability dynamic confidence coefficient is directly proportional to the prediction accuracy of the disaster prediction probability coefficient; the larger the probability dynamic confidence coefficient, the higher the reliability of the disaster prediction probability coefficient and the higher the prediction accuracy. Specifically, in this embodiment, the process of dynamically assessing the confidence level of disaster prediction probability coefficients includes: Based on Monte Carlo Dropout technology, the disaster prediction probability coefficients are subjected to T random forward propagation through the Dropout layer to obtain T disaster prediction probability coefficients Pit; where t is the prediction number index of the disaster prediction probability coefficient, t=1, 2, 3, ..., T; and T is the total number of predictions of the disaster prediction probability coefficient. The prediction uncertainty coefficient U is calculated as follows: In the formula, The average probability coefficient of disaster prediction is obtained through T predictions; The probability dynamic confidence coefficient is calculated based on the prediction uncertainty coefficient U, as follows: In the formula, norm() is the normalization function; Specifically, the present invention will be further illustrated below through embodiments: The probability dynamic confidence coefficient is a value between 0 and 1. When the probability dynamic confidence coefficient is less than or equal to the preset confidence threshold, it is judged as low confidence, indicating that the prediction result of the disaster prediction probability coefficient is not reliable enough and cannot be used to determine whether the geological condition requires an early warning. When the probability dynamic confidence coefficient is greater than the preset confidence threshold, it is judged as high confidence, indicating that the prediction result of the disaster prediction probability coefficient is reliable enough and can be used to determine whether the geological condition requires an early warning. In this embodiment, the preset confidence threshold is 0.5, which can reduce false alarms caused by data anomalies or other improper operations. The preset prediction probability threshold is 0.8, which is within the range of geological early warning probability coefficients. Therefore, when the probability exceeds the preset prediction probability threshold, real-time early warning processing is performed. This enables early warning decisions to be made directly at the edge without waiting for data to be transmitted back to the cloud, and early warning processing is determined in real time, thus improving the efficiency of early warning.

[0016] Example 2: To achieve the above objectives, based on Example 1, this invention discloses a rural geological disaster prediction system based on edge computing, comprising: The data acquisition module 11 is used to acquire rural geological environment data and preprocess the rural geological environment data to obtain processed rural geological environment data; wherein, the rural geological environment data includes surface subsidence data, surface vibration data, precipitation data and surface deformation data. The disaster prediction module 12 is used to extract features from the processed rural geological environment data to obtain the feature data to be predicted, and to perform comprehensive geological disaster prediction processing and calculation on the feature data to be predicted to obtain the disaster prediction probability coefficient. The confidence assessment module 13 is used to perform dynamic confidence assessment on the disaster prediction probability coefficient to obtain the probability dynamic confidence coefficient. When the probability dynamic confidence coefficient is less than or equal to the preset confidence threshold, the disaster prediction is paused and the geological disaster prediction is re-integrated; otherwise, when the disaster prediction probability coefficient is greater than the preset prediction probability threshold, a local early warning is triggered; otherwise, no local early warning is triggered.

[0017] Based on the same inventive concept, this invention also provides a computer device, comprising: one or more processors, and a memory for storing one or more computer programs; the programs include program instructions, and the processor executes the program instructions stored in the memory. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, used to implement one or more instructions, specifically for loading and executing one or more instructions stored in a computer storage medium to implement the above-described method.

[0018] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, performs the above-described method. This storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0019] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.

[0020] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0021] The foregoing has shown and described the basic principles, main features, and advantages of this disclosure. Those skilled in the art should understand that this disclosure is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this disclosure. Various changes and modifications can be made to this disclosure without departing from its spirit and scope, and all such changes and modifications fall within the scope of this disclosure as claimed.

Claims

1. A method for predicting rural geological disasters based on edge computing, characterized in that, The method includes the following steps: Feature extraction is performed on pre-processed rural geological environment data to obtain predictable feature data. The predictable feature data is then subjected to comprehensive geological disaster prediction processing and calculation to obtain disaster prediction probability coefficients. Among them, rural geological environment data includes surface subsidence data, surface vibration data, precipitation data, and surface deformation data. The probability coefficient of disaster prediction is subjected to dynamic confidence assessment to obtain the dynamic confidence coefficient of probability. When the dynamic confidence coefficient of probability is less than or equal to the preset confidence threshold, disaster prediction is suspended and geological disaster prediction is re-integrated; otherwise, when the probability coefficient of disaster prediction is greater than the preset prediction probability threshold, a local early warning is triggered; otherwise, no local early warning is triggered.

2. The method for predicting rural geological disasters based on edge computing according to claim 1, characterized in that, The probability dynamic confidence coefficient is based on the disaster prediction probability coefficient and undergoes dynamic confidence assessment processing, as follows: Based on Monte Carlo Dropout technology, the disaster prediction probability coefficients are subjected to T random forward propagation through the Dropout layer to obtain T disaster prediction probability coefficients Pit; where t is the prediction number index of the disaster prediction probability coefficient, t=1, 2, 3, ..., T; and T is the total number of predictions of the disaster prediction probability coefficient. The prediction uncertainty coefficient U is calculated as follows: In the formula, The average probability coefficient of disaster prediction is obtained through T predictions; The probability dynamic confidence coefficient is calculated based on the prediction uncertainty coefficient U, as follows: In the formula, norm() is the normalization function.

3. The rural geological disaster prediction method based on edge computing according to claim 2, characterized in that, The probability dynamic confidence coefficient is used to measure the prediction accuracy of the disaster prediction probability coefficient, and the magnitude of the probability dynamic confidence coefficient is directly proportional to the prediction accuracy of the disaster prediction probability coefficient.

4. The rural geological disaster prediction method based on edge computing according to claim 3, characterized in that, The preset confidence threshold is obtained by averaging the calculated probability dynamic confidence coefficients from multiple calculations; the preset prediction probability threshold is obtained by averaging the calculated disaster prediction probability coefficients from multiple calculations.

5. The rural geological disaster prediction method based on edge computing according to claim 1, characterized in that, The process of performing comprehensive geological disaster prediction calculations on the feature data to be predicted includes the following steps: Settlement characteristic data is labeled Ci, seismic characteristic data is labeled Zi, precipitation characteristic data is labeled Ji, and deformation characteristic data is labeled Xi; where i is the number of characteristic data to be predicted, and i = 1, 2, 3, ..., n, where n is the total number of characteristic data to be predicted; The calculation formula for comprehensive geological disaster prediction and treatment is as follows: In the formula, Pi is the disaster prediction probability coefficient; a is the subsidence characteristic influence coefficient, b is the vibration characteristic influence coefficient, c is the precipitation characteristic influence coefficient, d is the deformation characteristic influence coefficient; k1, k2, k3, and k4 are all preset weight coefficients.

6. The rural geological disaster prediction method based on edge computing according to claim 5, characterized in that, The disaster prediction probability coefficients are classified into geological stability probability coefficients, geological attention probability coefficients, and geological early warning probability coefficients according to their probability magnitude. The geological stability probability coefficient is less than the geological attention probability coefficient and less than the geological early warning probability coefficient.

7. The rural geological disaster prediction method based on edge computing according to claim 1, characterized in that, The process of feature extraction from the pre-processed rural geological environment data is as follows: The surface subsidence data, surface seismic data, precipitation data, and surface deformation data are arranged according to spatial topological relationships; spatial feature maps of each type are generated as feature data to be predicted. The feature data to be predicted includes settlement feature data, vibration feature data, precipitation feature data, and deformation feature data.

8. A rural geological disaster prediction system based on edge computing, used to execute a rural geological disaster prediction method based on edge computing as described in any one of claims 1-7, characterized in that, include: The disaster prediction module is used to extract features from pre-processed rural geological environment data to obtain the feature data to be predicted, and to perform comprehensive geological disaster prediction processing and calculation on the feature data to be predicted to obtain the disaster prediction probability coefficient; the rural geological environment data includes surface subsidence data, surface vibration data, precipitation data and surface deformation data; The confidence assessment module is used to perform dynamic confidence assessment on the disaster prediction probability coefficient to obtain the dynamic confidence coefficient. When the dynamic confidence coefficient is less than or equal to the preset confidence threshold, the disaster prediction is paused and the geological disaster prediction is re-integrated. Otherwise, when the disaster prediction probability coefficient is greater than the preset prediction probability threshold, a local early warning is triggered. Otherwise, no local early warning is triggered.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, The memory stores a computer program that can run on a processor. When the processor loads and executes the computer program, it employs a rural geological disaster prediction method based on edge computing, as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is loaded and executed by the processor, it employs a rural geological disaster prediction method based on edge computing, as described in any one of claims 1 to 7.

Images