Mountain crack geological disaster information monitoring and early warning method

Abstract

The invention provides a mountain fissure geological disaster information monitoring and early warning method, which comprises the steps that fissure form data of a mountain fissure area is acquired, the fissure form data comprises the length, width, depth and trend of a fissure, and real-time acquisition is performed through a preset sensor network; geologic characteristic parameters of the crack area are obtained, the geologic characteristic parameters comprise lithology, structure and stress states, and data acquisition is carried out through a pre-arranged geologic sensor network; transmitting fracture form data and geological characteristic parameters acquired in the field to a monitoring platform in real time; according to the dynamically updated crack division result, designing a visual display module, and generating a crack spatial distribution map and an attribute feature map; and carrying out correlation analysis on the visually displayed crack division result and the rainfall, the surface displacement and other monitoring information in the monitoring platform, and generating a disaster early warning report.

Description

Technical Field

[0001] The present invention relates to the field of geological disasters, and in particular to a method for monitoring and early warning of geological disaster information of mountain cracks. Background Art

[0002] In the monitoring of mountain crack geological disasters, crack division is a key technology. Due to the complex and changeable morphology of mountain cracks and the significant differences in geological characteristics in different regions, traditional crack division methods are difficult to meet the needs of refined analysis. In addition, the dynamic and uncertain nature of crack expansion makes it difficult to update the division results in a timely manner, affecting the timeliness of disaster warnings. In order to achieve accurate division and real-time updating of cracks, it is necessary to break through the following technical bottlenecks: First, how to design a scientific and reasonable division index system based on the morphological characteristics of cracks, such as length, width, depth, direction, etc., and establish corresponding division models and algorithms to achieve automatic classification and regional division of cracks; second, how to obtain the geological characteristic parameters of the fracture area, such as lithology, structure, stress state, etc., and combine them with the morphological characteristics of the fractures for comprehensive analysis to improve the accuracy and reliability of the division; third, how to develop a set of efficient data transmission and processing technologies to transmit the massive monitoring data collected in the field to the monitoring platform in real time, and dynamically update the division results to ensure the synchronization and consistency of the monitoring data and the division results; finally, how to design a set of visual division result display methods to intuitively present the spatial distribution and attribute characteristics of the fractures, and conduct correlation analysis with other monitoring information to provide an intuitive and reliable basis for disaster warning and decision-making. To solve the above technical problems, it is necessary to comprehensively apply multidisciplinary knowledge, such as geology, rock and soil mechanics, computer science, communication engineering, etc., and carry out interdisciplinary collaborative research to break through the technical bottleneck of accurate crack division and real-time transmission, and provide more intelligent and efficient technical support for mountain crack geological disaster monitoring. Summary of the invention

[0003] The present invention provides a method for monitoring and early warning of geological disaster information of mountain cracks, which mainly includes:

[0004] Acquire crack morphology data of the mountain crack area, wherein the crack morphology data includes crack length, width, depth and direction, and is collected in real time through a preset sensor network;

[0005] According to the fracture morphology data, a classification index system is established, wherein the classification index system includes thresholds of fracture length, width, depth and direction, and a clustering algorithm is used to classify the fractures to obtain a preliminary fracture classification result;

[0006] Acquiring geological characteristic parameters of the fracture area, the geological characteristic parameters including lithology, structure and stress state, and collecting data through a pre-arranged geological sensor network;

[0007] Combining the geological characteristic parameters with the fracture morphology data, using a support vector machine algorithm to construct a classification model, comprehensively analyzing the fractures, and optimizing the fracture division results;

[0008] Transmit the fracture morphology data and geological characteristic parameters collected in the field to the monitoring platform in real time;

[0009] On the monitoring platform, the received fracture morphology data is processed in real time using stream data processing technology, and the fracture division result is dynamically updated in combination with the geological characteristic parameters;

[0010] According to the dynamically updated fracture division results, a visualization display module is designed to generate fracture spatial distribution maps and attribute feature maps;

[0011] The visually displayed crack division results are correlated and analyzed with monitoring information such as rainfall and surface displacement in the monitoring platform to generate a disaster warning report.

[0012] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0013] The present invention discloses a method for monitoring and early warning of geological disaster information of mountain cracks. The method collects crack morphology data and geological characteristic parameters in real time through a sensor network, and classifies and comprehensively analyzes the cracks using a clustering algorithm and a support vector machine algorithm. On a monitoring platform, a stream data processing technology is used to process the received data in real time, and the crack division results are dynamically updated. The present invention also designs a visualization display module to generate a crack spatial distribution map and an attribute characteristic map, and associates and analyzes the crack division results with monitoring information such as rainfall and surface displacement to generate a disaster warning report. This method can realize real-time monitoring, dynamic analysis and visualization of mountain cracks, provides effective technical support for early warning of geological disasters such as landslides, improves the accuracy and timeliness of geological disaster early warnings, and is of great significance to protecting the lives and property of residents in mountainous areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flow chart of a method for monitoring and early warning of geological disaster information of mountain cracks. DETAILED DESCRIPTION

[0015] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0016] like Figure 1 As shown, a method for monitoring and early warning of geological disaster information of mountain cracks in this embodiment may specifically include:

[0017] S101, obtaining crack morphology data of a crack area in a mountain, wherein the crack morphology data includes crack length, width, depth and direction, and is collected in real time through a preset sensor network.

[0018] A preset sensor network layout scheme is obtained, wherein the sensor network layout scheme includes the layout positions of multiple sensor nodes; according to the sensor network layout scheme, multiple sensor nodes are deployed in the crack area of ​​the target mountain to form a sensor network covering the crack area; image data and depth data of the crack area are collected by the sensor nodes; the image data and depth data collected by the sensor nodes are transmitted to a data processing center; at the data processing center, image segmentation processing is performed on the image data to obtain a crack area image; according to the crack area image, the length and width of the crack are calculated to obtain the size data of the crack; according to the depth data, the depth value of each position of the crack area is determined to obtain the depth data of the crack; the crack area image is processed by using a Hough transform algorithm to calculate the strike angle of the crack to obtain the strike data of the crack; the size data, depth data and strike data of the crack are integrated to obtain crack morphology data characterizing the crack morphology; according to the crack morphology data, the crack morphology changes in the mountain crack area are monitored in real time.

[0019] Specifically, obtaining a preset sensor network layout plan is the first step in the mountain crack monitoring system. This plan is usually jointly formulated by geological experts and engineers, taking into account factors such as mountain topography and crack distribution. For example, in a mountain crack area of ​​about 5,000 square meters, 50 sensor nodes may be deployed to form a grid distribution. This layout can ensure comprehensive coverage of the entire area.

[0020] When placing sensor nodes in the target area according to the layout plan, the communication distance and power supply between nodes need to be considered. Solar power supply and wireless communication technology can be used to improve the reliability and durability of the system. Each node may be equipped with a high-resolution camera and a laser rangefinder to collect image and depth data.

[0021] Data transmission to the processing center is completed through a wireless network. To ensure the security and real-time nature of data transmission, encrypted transmission and edge computing technologies can be used. In the data processing center, the image is first preprocessed, such as denoising and enhancement, and then a deep learning algorithm is used to perform image segmentation and accurately extract the crack area.

[0022] The calculation of crack size involves the conversion of pixels to actual distance. For example, if the image resolution is 0.1mm / pixel, then a 100-pixel wide crack is actually 10mm wide. The processing of depth data needs to take into account the accuracy of the laser rangefinder and the influence of environmental factors, such as temperature and humidity. The Hough transform algorithm is used to calculate the crack direction and can effectively identify straight line features. In this way, the angle of the crack relative to the north direction, such as 30°, can be obtained. This data is crucial for analyzing the development trend of the crack.

[0023] By integrating size, depth and strike data, a 3D model of the crack can be constructed. This comprehensive data not only provides an intuitive display of the current state of the crack, but also predicts its future development trend through time series analysis. For example, if it is found that the crack width has increased by 5mm, the depth has increased by 10mm, and the strike angle has changed by 2° within a week, this may indicate that the risk of landslide is increasing.

[0024] Real-time monitoring of crack morphology changes is the core goal of the entire system. By setting warning thresholds, such as when the crack width increases by more than 1mm / day, the system can automatically issue an alarm. This continuous monitoring and timely warning mechanism can provide valuable decision-making support for geological disaster prevention and control, and effectively reduce the risk of disasters such as landslides.

[0025] S102: establishing a classification index system based on the fracture morphology data, wherein the classification index system includes thresholds of fracture length, width, depth and direction, and classifying the fractures using a clustering algorithm to obtain a preliminary fracture classification result.

[0026] Acquire crack morphology data, the crack morphology data including crack length, width, depth and direction parameters; set a corresponding threshold range for the crack length, width, depth and direction division indicators, and the threshold range is used as a judgment basis for the clustering algorithm; adopt a K-means clustering algorithm, take the crack morphology data as input, perform clustering analysis according to the threshold setting of the division indicator system, and obtain a preliminary crack classification result, in which cracks of different categories have similar morphological characteristics; evaluate the preliminary division results to determine the accuracy and rationality of the preliminary division results; if the preliminary division results meet the requirements, determine the final crack division scheme; if the preliminary division results do not meet the requirements, adjust the threshold setting of the division indicator system, return to execute the step of using the K-means clustering algorithm, and re-perform clustering analysis until a crack classification result that meets the requirements is obtained; formulate a targeted crack treatment and maintenance plan according to the final crack division scheme.

[0027] Specifically, the acquisition of crack morphology data is the key to mountain stability assessment. The crack length, width, depth and strike parameters collected by the sensor network provide the basis for subsequent analysis. Setting threshold ranges for these parameters can better distinguish different types of cracks. For example, length can be divided into short (<1m), medium (1-5m) and long (>5m); width can be divided into narrow (<1cm), medium (1-5cm) and wide (>5cm); depth can be divided into shallow (<10cm), medium (10-50cm) and deep (>50cm); strike can be divided according to the angle with the principal stress direction.

[0028] The application of K-means clustering algorithm in crack classification reflects the data-driven analysis method. The algorithm classifies cracks with similar characteristics into one category through iterative calculation. In specific implementation, the crack morphology data can be standardized as the input vector, and the appropriate initial cluster center can be selected, and classification can be performed by calculating the Euclidean distance. For example, assuming there are 100 crack data, 3-5 main categories may be obtained, and each category represents a group of cracks with similar morphological characteristics. Evaluation of the preliminary division results is an important step to ensure the accuracy of classification. The silhouette coefficient can be used to measure the clustering effect. The closer the coefficient value is to 1, the better the clustering effect.

[0029] In addition, the clustering results can be intuitively displayed through visualization methods such as scatter plots or heat maps. If it is found that some cracks are misclassified or the boundaries between categories are not clear, it is necessary to adjust the threshold setting and re-cluster. Threshold adjustment is an iterative optimization process. For example, if it is found that cracks with a length of 4.5-5.5m are frequently divided back and forth between two categories, you can consider adjusting the medium range of length to 1-6m. Similarly, the thresholds of other parameters can also be fine-tuned according to actual conditions. This process may need to be repeated many times until a stable and reasonable classification result is obtained.

[0030] The final crack classification scheme provides a basis for formulating targeted treatment measures. For example, for long and deep cracks, grouting reinforcement technology may be required; for wide and shallow cracks, surface plugging can be considered; and for cracks that are consistent with the direction of the principal stress, comprehensive reinforcement treatment may be required. Through this classification method, resources can be allocated more effectively and the efficiency and effectiveness of mountain reinforcement can be improved.

[0031] This data-driven crack classification method can not only objectively reflect the actual situation of mountain cracks, but also continuously optimize the classification standards as data accumulates. It provides a scientific basis for mountain stability assessment and early warning, and helps to improve the accuracy and effectiveness of geological disaster prevention and control. At the same time, this method also has good scalability and can be appropriately adjusted according to the geological characteristics of different regions, so that it can be applied in a wider range.

[0032] S103, obtaining geological characteristic parameters of the fracture area, wherein the geological characteristic parameters include lithology, structure and stress state, and collecting data through a pre-arranged geological sensor network.

[0033] The original geological data of the fracture area is obtained, and the original geological data comes from a pre-arranged geological sensor network; the original geological data is preprocessed and feature extracted to obtain characteristic parameters characterizing the geological characteristics of the fracture area, and the characteristic parameters include lithology, structure and stress state; the characteristic parameters are analyzed by a clustering algorithm, and the fracture area is divided into at least two geological units according to the similarity of the characteristic parameters; for each of the at least two geological units, a mapping relationship between the characteristic parameters and the degree of fracture development is established by a support vector machine algorithm to obtain the fracture development probability of each geological unit; if the fracture development probability of the geological unit exceeds a preset threshold, the geological unit is determined as a high-risk fracture area; otherwise, the geological unit is determined as a low-risk fracture area; according to the characteristic parameters and the fracture development probability, a decision tree algorithm is used to generate decision rules for fracture prevention and control measures; a prediction model between the characteristic parameters and the fracture extension trend is established by a neural network algorithm to predict the development of fractures in a future period of time; the division result of the fracture area, the fracture development probability, the decision rule and the prediction result are visualized.

[0034] Specifically, obtaining the original geological data of the crack area is the basis for crack analysis. For example, in a mountain tunnel project, a geological sensor network including strain gauges, inclinometers and crack meters was deployed. These sensors monitor rock deformation, ground stress changes and crack opening and closing in real time, providing a rich data source for subsequent analysis.

[0035] Preprocessing and feature extraction of raw data are key steps to improve analysis efficiency. The raw data can be cleaned by filtering, denoising, outlier processing and other methods, and then characteristic parameters such as lithology (such as rock type, weathering degree), structure (such as fault, fold distribution) and stress state (such as principal stress direction, stress concentration area) can be extracted. The characteristic parameters are analyzed by clustering algorithm, which can effectively identify areas with similar geological characteristics. The characteristic parameters are analyzed using K-means clustering algorithm to divide three geological units: hard rock area, broken zone area and weak interlayer area. This division lays the foundation for subsequent fracture risk assessment.

[0036] The support vector machine algorithm can establish a mapping relationship between characteristic parameters and the degree of fracture development. For each geological unit, characteristic parameters such as lithology, structure and stress state are input, and the support vector machine model obtained through training can output the probability of fracture development of the unit. For example, due to its lower rock strength and higher stress concentration, the fracture development probability of the broken zone may reach 0.8, exceeding the preset threshold of 0.7, and is therefore judged as a high-risk fracture area.

[0037] The decision tree algorithm can generate decision rules for crack prevention measures based on characteristic parameters and crack development probability. In this example, the decision tree may generate the following rules: if the rock mass strength is less than 30MPa and the stress concentration factor is greater than 2, then adopt system anchor support; if the crack development probability is greater than 0.7 and the groundwater is abundant, then implement curtain grouting, etc. These rules provide engineers with a clear basis for decision-making. The neural network algorithm performs well in predicting the trend of crack expansion.

[0038] By establishing the relationship between characteristic parameters and crack expansion rate, the development of cracks in the future can be predicted. For example, in the weak interlayer area, the neural network model predicts that the length of the main cracks may increase by 20% and the width may increase by 15% in the next three months. This prediction result provides an important reference for timely prevention and control measures. Finally, the visualization of the analysis results can intuitively present the risk distribution and development trend of the crack area. For example, GIS technology can be used to superimpose information such as the division of geological units along the tunnel, the probability distribution of crack development, and recommended prevention and control measures on the three-dimensional geological model, and the prediction results of crack expansion can be displayed through dynamic charts. This visualization method helps engineers quickly understand complex geological conditions and potential risks, so as to formulate more scientific and reasonable construction plans and monitoring plans.

[0039] S104, combining the geological characteristic parameters with the fracture morphology data, using a support vector machine algorithm to construct a classification model, performing a comprehensive analysis on the fractures, and optimizing the fracture division results.

[0040] The method comprises the following steps: obtaining geological characteristic parameters and fracture morphological data, fusing the geological characteristic parameters and the fracture morphological data to obtain a comprehensive characteristic data set; using data preprocessing technology to clean and normalize the comprehensive characteristic data set to obtain a standardized characteristic vector; using a feature selection algorithm to extract a feature subset with a correlation with fracture classification higher than a preset threshold for the standardized characteristic vector; inputting the feature subset into a support vector machine algorithm, obtaining a fracture classification model through training, and optimizing the fracture classification model by a cross-validation method; obtaining fracture data to be analyzed, extracting geological characteristic parameters and morphological data of the fracture data to be analyzed, and constructing a characteristic vector to be analyzed; inputting the characteristic vector to be analyzed into the fracture classification model to obtain a classification result of the fracture data to be analyzed; optimizing and adjusting the preliminary fracture division result according to the classification result and in combination with geological knowledge and experience to obtain a final comprehensive fracture analysis result.

[0041] Specifically, obtaining geological characteristic parameters and fracture morphology data is the basis for analyzing fractures. Geological characteristic parameters include lithology, structure, and stress state, which can be collected through a geological sensor network. Fracture morphology data includes fracture length, width, and orientation, which can be obtained through image recognition technology. Fusion processing of these two types of data can obtain a more comprehensive and integrated feature data set.

[0042] Data preprocessing is a key step to improve the quality of analysis. Cleaning the comprehensive feature data set can remove outliers, such as treating data with a crack width of more than 10 cm as abnormal and removing them. Normalization processing unifies data of different dimensions to the same scale, such as scaling the rock hardness index and crack length to the range of 0-1, which is convenient for subsequent analysis.

[0043] Feature selection algorithms help extract the most relevant subset of features. For example, by calculating the correlation coefficient between each feature and fracture classification, select features with a correlation coefficient greater than 0.7. This may reveal that lithology and stress state have a greater impact on fracture classification, while some minor factors can be ignored.

[0044] Support vector machine is a powerful classification algorithm. The optimal hyperplane is constructed by training data to separate different types of cracks. Cross-validation methods such as K-fold cross-validation can be used to optimize model parameters and improve classification accuracy.

[0045] Feature extraction and classification of the fracture data to be analyzed are the key to model application. For example, for a newly discovered fracture, the lithology (such as sandstone) and stress state (such as compressive stress of 20 MPa) of the area where it is located are extracted to construct a feature vector. The vector is input into the trained model to obtain the classification result of the fracture, such as judging it as a high-risk fracture.

[0046] Finally, it is very important to optimize and adjust the classification results based on geological knowledge and experience. For example, if the model determines that a certain fracture is of low risk, but geological experts believe that its actual risk is higher based on the historical data and structural characteristics of the area, the results should be adjusted accordingly.

[0047] This human-machine combined method can make full use of the efficiency of the algorithm and the experience of experts to obtain more reliable comprehensive crack analysis results. This series of steps forms a complete crack analysis process, from data acquisition, preprocessing, feature extraction to model training and application, and finally optimization combined with expert knowledge. This method not only improves the accuracy and efficiency of the analysis, but also provides reliable decision support for crack prevention and control.

[0048] S105, transmitting the fracture morphology data and geological characteristic parameters collected in the field to the monitoring platform in real time.

[0049] Acquire crack morphology data and geological characteristic parameters to obtain original monitoring data; use data preprocessing technology to remove noise interference and outliers from the original monitoring data to obtain cleaned valid data; package and encapsulate the cleaned valid data according to the preset data transmission protocol and format specification to form a standardized transmission data packet; use an encrypted transmission method through the constructed wireless transmission channel to send the standardized transmission data packet to the remote monitoring platform in real time; after receiving the transmission data packet, the monitoring platform uses data parsing technology to extract the crack morphology characteristics and geological parameters therein to obtain structured monitoring data; input the structured monitoring data into a pre-built machine learning model, use a support vector machine algorithm to perform feature analysis, and judge the degree of danger of the crack morphology; according to the analysis results of the degree of danger of the crack morphology, combined with the preset warning threshold, realize automatic warning of crack disasters through a decision tree algorithm, and push the warning information to relevant departments.

[0050] Specifically, obtaining crack morphology data and geological characteristic parameters is a key step in monitoring rock cracks. These data may include morphological characteristics such as crack length, width, and depth, as well as geological parameters such as rock type, geological structure, and stress state. For example, in a mountain quarry, a laser scanner can be used to obtain the three-dimensional morphology of the crack, and a strain gauge can be used to measure the stress state of the rock mass.

[0051] The original monitoring data often contains noise and outliers, which require preprocessing. The median filter can be used to remove impulse noise, and the moving average method can be used to smooth the data curve. For outliers, a threshold can be set to eliminate them. For example, in a certain monitoring, the crack width data shows a significant deviation, which may be caused by equipment failure and should be eliminated. After the data is cleaned, it needs to be packaged in a standard format. The JSON format can be used to organize the crack morphology and geological parameters into structured data. For example,

[0052] {"crack_length":2.5,"crack_width":0.03,"rock_type":"granite","stress_state":"compressive"}. This format facilitates data transmission and subsequent processing.

[0053] Data transmission is encrypted to ensure security. The AES encryption algorithm can be used to encrypt data packets using a pre-shared key. The transmission channel can choose 4G / 5G network or satellite communication to achieve remote real-time monitoring. After the monitoring platform receives the data, it needs to be parsed and structured. The JSON parser can be used to extract various parameters and store them in a relational database for subsequent analysis and query. The machine learning model is used to analyze the degree of crack hazard. The support vector machine algorithm can effectively process high-dimensional features and establish a mapping relationship between crack features and hazard levels by training historical data. For example, the length, width, depth, etc. of the crack can be used as feature input to output a hazard level score.

[0054] Finally, disaster warning is carried out based on the analysis results. Multi-level warning thresholds can be set, such as a yellow warning triggered by a hazard score exceeding 0.7 and a red warning triggered by a hazard score exceeding 0.9. The decision tree algorithm can make a comprehensive judgment based on multiple factors, such as considering the hazard level of cracks, rainfall, seismic activity, etc., to generate more accurate warning information. Warning information can be pushed to relevant departments via SMS, email, etc. to ensure that protective measures are taken in time. This system realizes intelligent monitoring and early warning of rock cracks through the entire process of data collection, processing, transmission, and analysis, effectively improving the ability to prevent and control geological disasters.

[0055] S106. On the monitoring platform, the received fracture morphology data is processed in real time using a stream data processing technology, and the fracture division result is dynamically updated in combination with the geological characteristic parameters.

[0056] Acquire the fracture morphology data received on the monitoring platform, extract the morphological characteristic parameters for each fracture morphology data, the morphological characteristic parameters include the length, width, depth and direction of the fracture, and obtain the fracture morphology characteristic vector; acquire the geological characteristic parameters corresponding to the fracture morphology data, the geological characteristic parameters include lithology, stratigraphic age and fault distribution, fuse the geological characteristic parameters with the fracture morphology characteristic vector, and construct a fracture morphology-geological characteristic fusion vector; classify the fracture morphology-geological characteristic fusion vector according to the pre-established fracture division model using the support vector machine algorithm, divide the fracture into tension fractures and shear fractures, and obtain the initial fracture division result; in the process of stream data processing, continuously receive the latest fracture morphology transmitted by the monitoring platform The data is extracted, and its morphological characteristic parameters are merged with the corresponding geological characteristic parameters to obtain a new fracture morphology-geological characteristic fusion vector; the new fracture morphology-geological characteristic fusion vector is input into the fracture division model, and the fracture division model is updated in real time by using an incremental learning algorithm to obtain a dynamically adjusted fracture division result; a statistical analysis is performed on the dynamically adjusted fracture division result, the number and length of each type of fracture are calculated, a fracture distribution statistical report is generated, and the statistical report is sent to the monitoring platform for display; if a sudden increase in the number or length of fractures in a certain area is monitored, an early warning mechanism is triggered, and a time series analysis algorithm is used to predict the fracture development trend in the area, to evaluate the potential geological disaster risk, and to push the early warning information to relevant departments.

[0057] Specifically, the acquisition of crack morphology data is a key link in geological disaster monitoring. By extracting the morphological characteristic parameters of the crack, such as length, width, depth and direction, a crack morphology characteristic vector can be constructed. For example, a crack with a length of 10 meters, a width of 0.5 meters, a depth of 2 meters and a direction of 45 degrees northeast was monitored in a mountainous area. These data constitute the morphological characteristic vector of the crack.

[0058] The fusion of geological characteristic parameters can provide more comprehensive information for fracture analysis. Factors such as lithology, stratigraphic age and fault distribution directly affect the formation and development of fractures. For example, in a Mesozoic stratum composed of sandstone and shale, if there is a north-south fault, this information is combined with the fracture morphology characteristic vector to form a richer fracture morphology-geological characteristic fusion vector.

[0059] The application of support vector machine algorithm in crack classification reflects the importance of machine learning in geological disaster early warning. By training the model to identify the characteristics of tensile cracks and shear cracks, newly observed cracks can be quickly classified. For example, if a crack has a wide opening and is perpendicular to the principal stress direction, it is likely to be classified as a tensile crack; if the crack shows shear displacement, it may be judged as a shear crack.

[0060] The introduction of the incremental learning algorithm enables the fracture classification model to be continuously optimized. With the continuous input of new data, the model can adjust its classification boundaries and improve accuracy. For example, if the initial model incorrectly classifies certain cracks as tensile, through incremental learning, the model can gradually correct this deviation and improve the recognition rate of shear cracks.

[0061] The generation of fracture distribution statistics provides decision makers with an intuitive risk assessment tool. By calculating the number and total length of each type of fracture, geological stability can be quickly determined. For example, if the total length of tensile fractures suddenly increases from 100 meters to 500 meters in a 1 square kilometer area, this may indicate that the geological structure of the area is undergoing significant changes.

[0062] Time series analysis algorithms play an important role in predicting the development trend of cracks. By analyzing the change patterns of crack parameters over time, the possible development of cracks in the future can be estimated. For example, if the width of a crack is observed to expand at a rate of 0.5 cm per day in the past week, through time series analysis, the width that the crack may reach in the next week can be predicted, thereby assessing the potential risk of landslides or collapses. This comprehensive crack monitoring and analysis system can greatly improve the accuracy and timeliness of geological disaster warnings. Through real-time data collection, intelligent algorithm analysis and dynamic model updates, the system can identify danger signals before a disaster occurs, providing a valuable time window for relevant departments to take preventive measures, thereby effectively reducing casualties and property losses caused by geological disasters.

[0063] S107. Based on the dynamically updated crack division results, a visualization display module is designed to generate a crack spatial distribution map and an attribute feature map.

[0064] The method comprises the steps of: obtaining a crack division result, wherein the crack division result includes spatial coordinate information and attribute information of the crack; storing the crack division result in a database; obtaining the spatial coordinate information and attribute information of the crack from the database; generating a three-dimensional spatial distribution model of the crack using a three-dimensional modeling technology according to the spatial coordinate information of the crack; generating an attribute feature map of the crack using a chart visualization technology according to the attribute information of the crack, wherein the attribute feature map includes a statistical distribution map of the length, width and depth of the crack; superimposing and displaying the three-dimensional spatial distribution model of the crack and the attribute feature map of the crack to generate a comprehensive visualization display interface of the crack; grouping the cracks using a clustering algorithm to obtain a classification result of the cracks; associating the classification result of the cracks with the comprehensive visualization display interface of the cracks, identifying different categories of cracks by different colors or patterns, and generating a classification visualization display interface of the cracks.

[0065] Specifically, the acquisition of fracture delineation results is a key step in geological analysis, which includes spatial coordinates and attribute information. Spatial coordinates may include longitude, latitude and depth, while attribute information may include fracture type, length and direction. These data are stored in a database, such as using

[0066] PostgreSQLwithPostGIS extension can efficiently manage geospatial data.

[0067] 3D modeling technology plays an important role in visualizing the distribution of cracks. For example, using graphics libraries such as OpenGL or WebGL, a 3D model can be constructed based on the spatial coordinate information of the cracks. This model can intuitively show the distribution of cracks underground, which helps geological engineers assess potential risk areas. The generation of attribute feature maps relies on chart visualization technology. For example, using the D3.js library, you can create interactive statistical distribution maps. A specific example is to draw a histogram of crack length, where the horizontal axis represents the length range (such as 0-5m, 5-10m, etc.) and the vertical axis represents the frequency, which can clearly show the distribution characteristics of crack length.

[0068] The generation of a comprehensive visualization display interface is a process of superimposing the three-dimensional spatial distribution model and the attribute feature map. This can be achieved through the combination of WebGL and SVG technology. For example, clickable hotspots are added to the three-dimensional model, and the attribute feature map of the corresponding crack pops up after clicking. This interactive display method can help analysts understand the crack characteristics more comprehensively.

[0069] Clustering algorithms play an important role in fracture classification. Algorithms such as K-means or DBSCAN can group fractures based on their spatial distribution and attribute characteristics. For example, fractures may be classified into categories such as "shallow short fractures" and "deep long fractures". This classification helps identify potential geological structural units or risk areas.

[0070] The generation of the classification visualization display interface is a process of combining clustering results with comprehensive visualization. Different colors can be used to identify different categories of cracks, such as red for high-risk cracks and green for low-risk cracks. This visualization method can intuitively display the spatial distribution and risk level of cracks, which helps to quickly identify areas that need to be focused on. Through this series of steps, complex crack data can be transformed into intuitive and easy-to-understand visualizations. This not only helps geological engineers better understand underground structures, but also provides decision makers with an intuitive basis for risk assessment, so as to formulate more targeted disaster prevention and mitigation measures. For example, if a large number of deep and long cracks are found in a certain area, it may be necessary to strengthen monitoring efforts or take reinforcement measures in that area.

[0071] S108, performing correlation analysis on the visually displayed crack division results and monitoring information such as rainfall and surface displacement in the monitoring platform to generate a disaster warning report.

[0072] Obtain crack division results and monitoring information of rainfall and surface displacement in the monitoring platform; perform data preprocessing on the crack division results and the monitoring information to obtain preprocessed crack division results and monitoring information, wherein the data preprocessing includes data cleaning and data normalization; construct an association analysis model based on the preprocessed crack division results and monitoring information, wherein the association analysis model uses a time series analysis algorithm to analyze the correlation between the crack division results and the rainfall and surface displacement; obtain association rules and association strengths between the crack division results and the rainfall and surface displacement through the association analysis model; determine whether the current crack division results and monitoring information meet preset disaster warning conditions based on the association rules and association strengths; if the disaster warning conditions are met, determine the disaster warning level based on the preset warning level standard and generate a disaster warning report; display the disaster warning report in a visual manner, and push the warning information to relevant departments and personnel.

[0073] Specifically, obtaining crack demarcation results and monitoring information is the key starting point for analyzing geological disaster risks. For example, a monitoring station in a mountainous area recorded that the crack width increased from 2mm to 5mm recently, while the rainfall reached 150mm / day and the surface displacement reached 20cm. These raw data need to be preprocessed before they can be used for subsequent analysis.

[0074] Data preprocessing includes cleaning outliers and normalization. During the cleaning process, you may find that the rainfall on a certain day is abnormally high at 1000mm. This is obviously erroneous data caused by equipment failure and needs to be eliminated. Normalization unifies data of different dimensions into the range of 0-1 for easy comparison and analysis. For example, a crack width of 5mm is converted to 0.5, a rainfall of 150mm is converted to 0.6, and a surface displacement of 20cm is converted to 0.4.

[0075] Constructing an association analysis model is the key to revealing the interaction between various factors. The time series analysis algorithm can capture the law of data changes over time. For example, through analysis, it was found that the crack width was positively correlated with the cumulative rainfall in the previous three days, with a correlation coefficient of 0.8; it was also positively correlated with the surface displacement, with a correlation coefficient of 0.7. This means that rainfall and surface displacement may be important factors leading to crack expansion. The acquisition of association rules and association strength helps to quantify the impact between various factors. For example, the following rule may be derived: when the cumulative rainfall in three days exceeds 200mm and the surface displacement exceeds 15cm, the probability of crack width increase reaches 90%. This quantitative relationship provides a scientific basis for early warning.

[0076] The judgment of early warning conditions requires comprehensive consideration of multiple factors. Assume that the preset disaster warning conditions are: the cumulative rainfall in 3 days exceeds 250mm, and the surface displacement exceeds 25cm, and the crack width increases by more than 50%. When the monitoring data meets these conditions, the system will trigger an early warning. The determination of the early warning level is usually divided into multiple levels. For example, four levels of early warning can be set: blue (mild), yellow (moderate), orange (severe) and red (particularly severe). When the crack width increases by more than 100%, a red warning may be triggered, indicating that the disaster risk is extremely high and emergency measures need to be taken immediately.

[0077] The generation and visualization of early warning reports are important links in conveying information to decision makers. The report can include current monitoring data, warning levels, risk assessments, and recommended measures. The visualization can be in the form of a dashboard to intuitively display the real-time status and warning levels of various indicators. At the same time, the early warning information is pushed to relevant departments and personnel through text messages, emails, etc. to ensure that the information is conveyed in a timely manner. The establishment of this system helps to improve the accuracy and timeliness of geological disaster early warnings. Through data-driven methods, potential risks can be discovered earlier, providing scientific basis and decision-making support for disaster prevention and mitigation. However, the effectiveness of the system also requires continuous data accumulation and model optimization to adapt to the geological conditions and climate characteristics of different regions.

[0078] Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if these modifications and variations of the embodiments of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for monitoring and early warning of geological disaster information of mountain cracks, characterized in that: The method comprises: Acquire crack morphology data of the mountain crack area, wherein the crack morphology data includes crack length, width, depth and direction, and is collected in real time through a preset sensor network; According to the fracture morphology data, a classification index system is established, wherein the classification index system includes thresholds of fracture length, width, depth and direction, and a clustering algorithm is used to classify the fractures to obtain a preliminary fracture classification result; Acquiring geological characteristic parameters of the fracture area, the geological characteristic parameters including lithology, structure and stress state, and collecting data through a pre-arranged geological sensor network; Combining the geological characteristic parameters with the fracture morphology data, using a support vector machine algorithm to construct a classification model, comprehensively analyzing the fractures, and optimizing the fracture division results; Transmit the fracture morphology data and geological characteristic parameters collected in the field to the monitoring platform in real time; On the monitoring platform, the received fracture morphology data is processed in real time using stream data processing technology, and the fracture division result is dynamically updated in combination with the geological characteristic parameters; According to the dynamically updated fracture division results, a visualization display module is designed to generate fracture spatial distribution maps and attribute feature maps; The visually displayed crack division results are correlated and analyzed with monitoring information such as rainfall and surface displacement in the monitoring platform to generate a disaster warning report.

2. The method according to claim 1, characterized in that The method of obtaining crack morphology data of the mountain crack area, wherein the crack morphology data includes crack length, width, depth and direction, is collected in real time through a preset sensor network, including: Acquire a preset sensor network layout scheme, wherein the sensor network layout scheme includes layout positions of multiple sensor nodes; According to the sensor network layout plan, multiple sensor nodes are deployed in the crack area of ​​the target mountain to form a sensor network covering the crack area; Collecting image data and depth data of the crack area through the sensor node; Transmitting the image data and depth data collected by the sensor node to a data processing center; In the data processing center, the image data is subjected to image segmentation processing to obtain a crack region image; Calculating the length and width of the crack according to the crack region image to obtain the size data of the crack; According to the depth data, determining the depth value of each position of the crack area to obtain the depth data of the crack; The image of the crack region is processed by using a Hough transform algorithm to calculate the strike angle of the crack and obtain the strike data of the crack; The size data, depth data and direction data of the fracture are integrated to obtain fracture morphology data representing the fracture morphology; According to the crack morphology data, the changes in the crack morphology of the mountain crack area are monitored in real time.

3. The method according to claim 1, characterized in that According to the fracture morphology data, a division index system is established, the division index system includes thresholds of fracture length, width, depth and direction, and a clustering algorithm is used to classify the fractures to obtain preliminary fracture division results, including: Acquiring fracture morphology data, wherein the fracture morphology data includes fracture length, width, depth and strike parameters; According to the fracture length, width, depth and direction classification index, a corresponding threshold range is set, and the threshold range is used as a judgment basis for the clustering algorithm; Using a K-means clustering algorithm, taking the crack morphology data as input, performing cluster analysis according to the threshold setting of the classification index system, and obtaining a preliminary crack classification result, in which cracks of different categories have similar morphological characteristics; Evaluate the preliminary division results to determine their accuracy and rationality; If the preliminary division results meet the requirements, the final crack division scheme is determined; If the preliminary classification result does not meet the requirements, the threshold setting of the classification index system is adjusted, and the step of using the K-means clustering algorithm is returned to execute, and cluster analysis is performed again until a crack classification result that meets the requirements is obtained; According to the final crack division plan, a targeted crack treatment and repair plan is formulated.

4. The method according to claim 1, characterized in that: The method of obtaining geological characteristic parameters of the fracture area, wherein the geological characteristic parameters include lithology, structure and stress state, and collecting data through a pre-arranged geological sensor network, includes: Acquiring original geological data of the fracture area, wherein the original geological data comes from a pre-arranged geological sensor network; Preprocessing and feature extraction are performed on the original geological data to obtain characteristic parameters characterizing geological characteristics of the fracture area, wherein the characteristic parameters include lithology, structure and stress state; The characteristic parameters are analyzed by using a clustering algorithm, and the fracture area is divided into at least two geological units according to the similarity of the characteristic parameters; For each of the at least two geological units, a mapping relationship between the characteristic parameter and the degree of fracture development is established by a support vector machine algorithm to obtain the fracture development probability of each geological unit; If the probability of fracture development in the geological unit exceeds a preset threshold, the geological unit is determined to be a high-risk fracture area; Otherwise, the geological unit is determined to be a low-risk fracture zone; According to the characteristic parameters and the crack development probability, a decision tree algorithm is used to generate decision rules for crack prevention and control measures; Establishing a prediction model between the characteristic parameters and the crack expansion trend through a neural network algorithm to predict the development of the cracks in the future; The division result of the fracture area, the fracture development probability, the decision rule and the prediction result are visualized.

5. The method according to claim 1, characterized in that The method combines the geological characteristic parameters with the fracture morphology data, constructs a classification model using a support vector machine algorithm, performs a comprehensive analysis on the fractures, and optimizes the fracture division results, including: Acquiring geological characteristic parameters and fracture morphology data, and fusing the geological characteristic parameters and the fracture morphology data to obtain a comprehensive characteristic data set; According to the comprehensive feature data set, a data preprocessing technology is used to clean and normalize the comprehensive feature data set to obtain a standardized feature vector; For the standardized feature vector, a feature selection algorithm is used to extract a feature subset whose correlation with crack classification is higher than a preset threshold; Inputting the feature subset into a support vector machine algorithm, obtaining a fracture classification model through training, and optimizing the fracture classification model by using a cross-validation method; Acquire fracture data to be analyzed, extract geological characteristic parameters and morphological data of the fracture data to be analyzed, and construct a characteristic vector to be analyzed; Inputting the feature vector to be analyzed into the crack classification model to obtain the classification result of the crack data to be analyzed; According to the classification results, combined with geological knowledge and experience, the preliminary fracture division results are optimized and adjusted to obtain the final fracture comprehensive analysis results.

6. The method according to claim 1, characterized in that The real-time transmission of the fracture morphology data and geological characteristic parameters collected in the field to the monitoring platform includes: Obtain fracture morphology data and geological characteristic parameters to obtain original monitoring data; For the original monitoring data, data preprocessing technology is used to remove noise interference and abnormal values ​​therein to obtain effective data after cleaning; According to the preset data transmission protocol and format specification, the cleaned valid data is packaged and encapsulated to form a standardized transmission data packet; The standardized transmission data packets are sent to the remote monitoring platform in real time through the constructed wireless transmission channel and in an encrypted transmission manner; After receiving the transmission data packet, the monitoring platform uses data analysis technology to extract the fracture morphological characteristics and geological parameters therein to obtain structured monitoring data; Inputting the structured monitoring data into a pre-built machine learning model, using a support vector machine algorithm to perform feature analysis to determine the degree of danger of the crack morphology; According to the analysis results of the danger level of the crack morphology, combined with the preset warning threshold, the automatic warning of crack disasters is realized through the decision tree algorithm, and the warning information is pushed to the relevant departments.

7. The method according to claim 1, characterized in that The monitoring platform uses a stream data processing technology to process the received fracture morphology data in real time, combines the geological characteristic parameters, and dynamically updates the fracture division results, including: Acquire the crack morphology data received on the monitoring platform, extract the morphology characteristic parameters therein for each piece of the crack morphology data, wherein the morphology characteristic parameters include the length, width, depth and direction of the crack, and obtain the crack morphology characteristic vector; Acquire geological characteristic parameters corresponding to the fracture morphology data, the geological characteristic parameters including lithology, formation age and fault distribution, fuse the geological characteristic parameters with the fracture morphology characteristic vector, and construct a fracture morphology-geological characteristic fusion vector; According to the pre-established fracture division model, the fracture morphology-geological characteristic fusion vector is classified by using a support vector machine algorithm, and the fractures are divided into tension fractures and shear fractures, so as to obtain an initial fracture division result; During the stream data processing, the latest crack morphology data transmitted by the monitoring platform is continuously received, and its morphological characteristic parameters are extracted. And it is fused with the corresponding geological characteristic parameters to obtain a new fracture morphology-geological characteristic fusion vector; The new fracture morphology-geological feature fusion vector is input into the fracture division model, and the fracture division model is updated in real time by using an incremental learning algorithm to obtain a dynamically adjusted fracture division result; Performing statistical analysis on the dynamically adjusted crack division results, calculating the number and length of each type of cracks, generating a crack distribution statistical report, and sending the statistical report to the monitoring platform for display; If a sudden increase in the number or length of cracks in a certain area is detected, the early warning mechanism will be triggered, and a time series analysis algorithm will be used to predict the development trend of cracks in the area, assess the potential geological disaster risks, and push the early warning information to relevant departments.

8. The method according to claim 1, characterized in that According to the dynamically updated crack division results, a visualization display module is designed to generate a crack space distribution map and an attribute feature map, including: Acquire a crack division result, wherein the crack division result includes spatial coordinate information and attribute information of the crack; Storing the crack division results in a database; Acquire spatial coordinate information and attribute information of the crack from the database; According to the spatial coordinate information of the crack, a three-dimensional spatial distribution model of the crack is generated by using a three-dimensional modeling technology; According to the attribute information of the crack, a graph visualization technology is used to generate an attribute characteristic diagram of the crack, wherein the attribute characteristic diagram includes a statistical distribution diagram of the length, width and depth of the crack; The three-dimensional spatial distribution model of the cracks is superimposed and displayed with the attribute characteristic map of the cracks to generate a comprehensive visualization display interface of the cracks; The cracks are grouped using a clustering algorithm to obtain a classification result of the cracks; The classification result of the cracks is associated with the comprehensive visualization display interface of the cracks, and different types of cracks are marked by different colors or patterns to generate a classification visualization display interface of the cracks.

9. The method according to claim 1, characterized in that: The visually displayed crack division results are correlated and analyzed with monitoring information such as rainfall and surface displacement in the monitoring platform to generate a disaster warning report, including: Obtain crack delineation results and rainfall and surface displacement monitoring information in the monitoring platform; Performing data preprocessing on the fracture division result and the monitoring information to obtain preprocessed fracture division result and monitoring information, wherein the data preprocessing includes data cleaning and data normalization; According to the pre-processed crack division results and monitoring information, a correlation analysis model is constructed, wherein the correlation analysis model adopts a time series analysis algorithm to analyze the correlation between the crack division results and the rainfall and surface displacement; By using the correlation analysis model, the correlation rules and correlation strength between the crack division results and the rainfall and surface displacement are obtained; According to the association rules and the association strength, judging whether the current crack division results and monitoring information meet the preset disaster warning conditions; If the disaster warning conditions are met, the disaster warning level is determined according to the preset warning level standard, and a disaster warning report is generated; The disaster warning report is displayed in a visual manner, and the warning information is pushed to relevant departments and personnel.

Images