Geological disaster early warning method and system based on big data analysis

Abstract

The invention discloses a geological disaster early warning method and system based on big data analysis, and particularly relates to the technical field of geological disaster early warning. Geological deformation time series data of a target area are obtained through satellite remote sensing monitoring, feature importance is evaluated through a random forest algorithm, and regional geological risk grades are divided; meanwhile, wavelet transformation and long-short-term memory network modeling are carried out on rainfall spatio-temporal data of the meteorological monitoring station, and the potential influence of rainfall is evaluated; high-risk areas are screened, an underground water level change rate and rock-soil body strength correlation model is constructed, and a support vector machine algorithm is adopted to evaluate the landslide probability; and finally, analyzing a geological risk grading result, the potential influence degree of rainfall on the target area and the landslide occurrence probability by adopting an analytic hierarchy process, calculating a comprehensive risk index, comparing the comprehensive risk index with a comprehensive risk threshold value, and judging whether to trigger grading early warning, thereby realizing data fusion and intelligent analysis, early warning in advance, and improving the early warning efficiency. And effective help is provided for disaster prevention and reduction.

Description

Technical Field

[0001] The present invention relates to the field of geological disaster early warning technology, and more specifically, to a geological disaster early warning method and system based on big data analysis. Background Art

[0002] With the rapid development of the economy and society and the acceleration of urbanization, the regional geological environment is becoming increasingly complex and geological disasters occur frequently. Traditional methods have shortcomings in multi-source data fusion, dynamic risk assessment and graded warning. They cannot fully reflect the complexity of the geological environment and the linkage effect of external factors such as rainfall. There are problems such as information islands, data lags and low model accuracy, making it difficult to achieve early and accurate prediction of disasters.

[0003] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a geological disaster early warning method and system based on big data analysis to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A geological disaster early warning method based on big data analysis includes the following steps:

[0007] The geological deformation time series data of the target area is obtained through satellite remote sensing monitoring, and the feature importance of the geological deformation data is evaluated using the random forest algorithm to divide the geological risk level of the target area;

[0008] Wavelet transform analysis was performed on the spatial and temporal distribution data of rainfall at the meteorological monitoring station, and a rainfall trend prediction model was established based on the long short-term memory neural network to evaluate the potential impact of rainfall on the target area.

[0009] Screen high-risk trigger areas based on the geological risk level classification results and the potential impact of rainfall on the target area;

[0010] For high-risk triggering areas, the probability of landslide occurrence is evaluated by using the support vector machine algorithm by constructing a correlation model between the groundwater level change rate and the rock and soil strength parameters;

[0011] The analytic hierarchy process is used to analyze the geological risk level classification results, the potential impact of rainfall on the target area and the probability of landslide occurrence, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal.

[0012] In a preferred embodiment, the geological deformation time series data of the target area is obtained through satellite remote sensing monitoring, and the geological deformation data is evaluated for feature importance using a random forest algorithm to divide the geological risk level of the target area, specifically:

[0013] Acquire multi-source satellite images and extract original observation data;

[0014] De-noising and interpolation of original observation data;

[0015] Select geological attributes and historical disasters and uniformly label features;

[0016] Random forests were used to evaluate feature importance and screen key factors;

[0017] Use convolutional neural networks to identify reflection patterns and interference features;

[0018] Based on the screening factors, the geological risk level of the target area is divided.

[0019] In a preferred embodiment, the temporal and spatial distribution data of rainfall at the meteorological monitoring station is subjected to wavelet transform analysis, and a rainfall trend prediction model is established based on a long short-term memory neural network to evaluate the potential impact of rainfall on the target area, specifically:

[0020] Collect and integrate the temporal and spatial distribution data of rainfall from meteorological monitoring stations in the target area;

[0021] Multi-scale time-frequency decomposition of rainfall data was performed using wavelet transform;

[0022] Construct a long short-term memory network model to train and verify rainfall trend prediction samples;

[0023] Calculate regional rainfall impact coefficient based on forecast results;

[0024] The potential impact of rainfall on the target area is assessed based on the rainfall impact coefficient.

[0025] In a preferred embodiment, the potential impact of rainfall on the target area is evaluated based on the rainfall impact coefficient, specifically:

[0026] Preset the rainfall impact coefficient threshold and compare the rainfall impact coefficient with the rainfall impact coefficient threshold:

[0027] When the rainfall impact coefficient is greater than the rainfall impact coefficient threshold, it indicates that the potential impact of the rainfall event on the target area is high;

[0028] When the rainfall impact coefficient is less than or equal to the rainfall impact coefficient threshold, it means that the potential impact of the rainfall event on the target area is low.

[0029] In a preferred embodiment, based on the geological risk level classification results and the potential impact of rainfall on the target area, high-risk trigger areas are screened, specifically:

[0030] When the geological risk level of the target area is level three, the target area is judged to be a high-risk trigger area; when the geological risk level of the target area is level two and the rainfall impact coefficient is greater than the rainfall impact coefficient threshold, the target area is judged to be a high-risk trigger area; otherwise, the target area is judged to be a non-high-risk trigger area.

[0031] In a preferred embodiment, for high-risk triggering areas, a correlation model between the groundwater level change rate and the rock and soil strength parameters is constructed, and a support vector machine algorithm is used to evaluate the probability of landslide occurrence, specifically:

[0032] Collect groundwater level change rate monitoring data;

[0033] Select the strength factors of rock and soil mass and construct the correlation factors;

[0034] Establish a groundwater level and rock-soil strength correlation model;

[0035] The support vector machine is trained by adopting the kernel function optimization method;

[0036] Input the associated model parameters and evaluate the probability of landslide occurrence.

[0037] In a preferred embodiment, the analytic hierarchy process is used to analyze the geological risk level classification results, the potential impact of rainfall on the target area and the probability of landslide occurrence, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal, specifically:

[0038] Preset the comprehensive risk threshold and compare the comprehensive risk index with the comprehensive risk threshold:

[0039] When the comprehensive risk index is greater than or equal to the comprehensive risk threshold, it means that the comprehensive risk level of the target area is high and a graded warning signal needs to be triggered;

[0040] When the comprehensive risk index is less than the comprehensive risk threshold, it means that the comprehensive risk level of the target area is low and there is no need to trigger a graded warning signal.

[0041] On the other hand, the present invention provides a geological disaster early warning system based on big data analysis, including a risk level classification module, an impact degree assessment module, a regional screening module, a landslide probability assessment module and an early warning signal judgment module;

[0042] The risk level classification module obtains the geological deformation time series data of the target area through satellite remote sensing monitoring, uses the random forest algorithm to evaluate the feature importance of the geological deformation data, and classifies the geological risk level of the target area;

[0043] The impact assessment module performs wavelet transform analysis on the spatial and temporal distribution data of rainfall at the meteorological monitoring station, establishes a rainfall trend prediction model based on the long short-term memory neural network, and assesses the potential impact of rainfall on the target area;

[0044] The regional screening module screens high-risk triggering areas based on the geological risk level classification results and the potential impact of rainfall on the target area;

[0045] For high-risk triggering areas, the landslide probability assessment module uses the support vector machine algorithm to assess the probability of landslide occurrence by constructing a correlation model between the groundwater level change rate and the rock and soil strength parameters;

[0046] The warning signal judgment module uses the hierarchical analysis method to analyze the geological risk level classification results, the potential impact of rainfall on the target area and the probability of landslide occurrence, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal.

[0047] The technical effects and advantages of the geological disaster early warning method and system based on big data analysis of the present invention are as follows:

[0048] By adopting big data fusion technology and comprehensively utilizing satellite remote sensing, meteorological monitoring and sensor data, a multi-level and dynamic assessment of geological disaster risks can be achieved. Through models such as random forests, long short-term memory neural networks and support vector machines, geological deformation, rainfall trends and landslide probabilities are accurately quantified, improving the accuracy of early warning and response speed. The use of hierarchical analysis method to comprehensively analyze various risk indicators can determine in real time whether to trigger graded early warning signals, effectively overcoming the problems of early warning lag and data isolation in traditional methods. It shows excellent robustness and adaptability in complex geological environments and extreme climatic conditions, providing a scientific basis and decision-making support for disaster prevention and mitigation, while promoting the intelligentization and informatization of geological disaster early warning technology, with broad application prospects and social and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of a geological disaster early warning method based on big data analysis according to the present invention;

[0050] Figure 2 This is a structural schematic diagram of a geological disaster early warning system based on big data analysis in the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Embodiment 1:

[0053] Figure 1 The present invention provides a geological disaster early warning method based on big data analysis, which includes the following steps:

[0054] The geological deformation time series data of the target area is obtained through satellite remote sensing monitoring, and the feature importance of the geological deformation data is evaluated using the random forest algorithm to divide the geological risk level of the target area;

[0055] Wavelet transform analysis was performed on the spatial and temporal distribution data of rainfall at the meteorological monitoring station, and a rainfall trend prediction model was established based on the long short-term memory neural network to evaluate the potential impact of rainfall on the target area.

[0056] Screen high-risk trigger areas based on the geological risk level classification results and the potential impact of rainfall on the target area;

[0057] For high-risk triggering areas, the probability of landslide occurrence is evaluated by using the support vector machine algorithm by constructing a correlation model between the groundwater level change rate and the rock and soil strength parameters;

[0058] The analytic hierarchy process is used to analyze the geological risk level classification results, the potential impact of rainfall on the target area and the probability of landslide occurrence, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal.

[0059] Specifically, the geological deformation time series data of the target area is obtained through satellite remote sensing monitoring, and the random forest algorithm is used to evaluate the feature importance of the geological deformation data to divide the geological risk level of the target area, including:

[0060] Acquire multi-source satellite images and extract original observation data: Use satellite remote sensing images from multiple sources to monitor the target area to obtain continuous phase image data. After preprocessing (such as geometric correction, atmospheric correction, etc.), the acquired image data will be used to extract surface deformation information. During the extraction process, high-precision deformation measurement values ​​are obtained through differential interferometric synthetic aperture radar technology. Define the following formula to describe the deformation extraction process: ;in, Represents the difference between the deformation data obtained at two different observation times; Indicates the observation time Corresponding deformation field data; Indicates the observation time The corresponding deformation field data.

[0061] The difference obtained Arranged in chronological order, time series observation data at each spatial position are formed, which are recorded as original observation data.

[0062] De-noising and interpolation of original observation data: The extracted original observation data usually contains noise, missing values ​​and outliers. In order to build stable and reliable time series features, the original observation data must be de-noised and interpolated. The original observation data is defined as ;in and Respectively represent the spatial coordinate index, representing the row and column positions in the image. The following formula is used for denoising filtering: ;in, represents the original observation data after denoising and filtering; represents the denoising filter function; Indicates the observation time.

[0063] After denoising, for data points with missing phases, a dual completion method based on spatial and temporal interpolation is used. The interpolation function is defined as ;in, Represents the interpolated continuous time series deformation data; Represents an interpolation function that combines spatial neighborhood and time series information.

[0064] Select geological attributes and historical disasters, and unify the annotation features: When constructing a complete feature set, in addition to the deformation time series data, the geological attributes and historical disaster records in the target area are also introduced. Geological attributes include stratum type, fault distribution, lithology, etc.; historical disaster data involves historical records of events such as landslides and collapses. In order to unify various types of data, a comprehensive feature vector is constructed for each monitoring point. The comprehensive feature vector of a monitoring point is defined as: ;in, Represents the comprehensive feature vector of the monitoring point; represents the deformation rate extracted from the interpolated continuous time series deformation data; Indicates the local slope value of the target area; The lithology code indicating the target area; It represents the fault density index in the target area; Indicates the frequency of historical disasters.

[0065] All features are normalized to form a unified annotation system to ensure the comparability and effectiveness of each feature in the random forest model.

[0066] Random forests are used to evaluate feature importance and screen key factors: Based on the constructed comprehensive feature vector, a preliminary assessment of the geological risk of the target area is conducted. The random forest algorithm is used to calculate the importance of each feature to the target variable (i.e., risk level) based on its performance in the decision tree. The Gini index decrease is used as the feature importance evaluation indicator. The Gini index before and after the decision tree split is defined as the parent node Gini index , Gini index of left child node , Gini index of right child node , the corresponding node samples are and , then the feature importance of a single split can be expressed as:

[0067] ;in, Indicates the importance contribution value of a feature in a single decision tree splitting process; Represents the Gini index of the parent node, which is used to measure the impurity of the data before splitting; Represents the Gini index of the left child node; Represents the Gini index of the right child node; Indicates the ratio of the number of samples in the left child node to the total number of samples in the parent node; Indicates the ratio of the number of samples in the right child node to the total number of samples in the parent node.

[0068] The random forest model will count the features in all decision trees and obtain the overall feature importance score. By setting a predefined threshold, the features below the predefined threshold are eliminated to screen out the key factors that contribute most to risk assessment. The screening process not only considers deformation data, but also integrates geological attributes and historical disaster information, achieving complementarity and optimization between multi-dimensional features.

[0069] Using convolutional neural networks to identify reflection patterns and interference features: Based on the random forest model, a convolutional neural network is introduced to perform deep feature extraction on the deformable image data. The local features of the signal reflection pattern and interference fringes in the target area are captured through the convolution operation. The mathematical expression of the convolution operation is: ;in, Represents the local features output after the convolution operation; represents the activation function; Represents the convolution kernel, which is a set of trainable weight matrices; Represents the input data matrix, i.e., the continuous time series deformation data after interpolation; Represents the bias term.

[0070] In convolutional neural networks, through multi-layer convolution and pooling operations, reflection patterns and interference features can be effectively extracted, and these deep features can be integrated into the comprehensive feature vector to complement the previously selected key factors, thereby improving the accuracy and robustness of risk assessment.

[0071] Based on the screening factors, the geological risk level of the target area is divided: a risk assessment model is constructed, and the nonlinear mapping function is defined as: ;in, Indicates the geological risk level of the monitoring point; Represents the interpolated continuous time series deformation data; Represents the comprehensive feature vector of the monitoring point; Represents the local features output after the convolution operation; Represents a nonlinear mapping function.

[0072] Mapping Function Combining statistical learning and deep feature fusion, the system outputs a quantitative risk index by weighting the importance of each feature, and then assigns each location in the monitoring area (i.e., the target area) to a different risk level based on the preset risk level classification standard. The classification standard can be calibrated through experimental data. For example, setting a risk threshold and , when the output of the mapping function is less than the risk threshold When , it means that the geological risk level of the target area is level 1; when the output of the mapping function is greater than or equal to the risk threshold , less than the risk threshold When , it means that the geological risk level of the target area is level 2; when the output of the mapping function is greater than or equal to the risk threshold , it means the geological risk level of the target area is level three.

[0073] Specifically, wavelet transform analysis is performed on the temporal and spatial distribution data of rainfall at the meteorological monitoring station, and a rainfall trend prediction model is established based on the long short-term memory neural network to evaluate the potential impact of rainfall on the target area, including:

[0074] Collect and integrate the temporal and spatial distribution data of rainfall from the meteorological monitoring stations in the target area: Collect rainfall information from the meteorological monitoring stations and sensors in the target area to obtain the temporal and spatial distribution data of rainfall in the target area. After the data from each monitoring station is aligned in time and space, it is stored in the form of a digital matrix. Define the rainfall data to be expressed in the form of a function ;in, Represents rainfall data; Indicates the spatial coordinates of the monitoring point in the horizontal direction; Indicates the spatial coordinates of the monitoring point in the vertical direction; Indicates rainfall observation time.

[0075] Using wavelet transform to decompose rainfall data in multi-scale time and frequency: Considering the non-stationary and multi-scale spatiotemporal characteristics of rainfall data, wavelet transform is used to decompose rainfall data in time and frequency to extract the time series characteristics at each scale. Introduce the wavelet transform formula:

[0076] ;in, Represented in space coordinates , rainfall observation time and scale parameter The corresponding wavelet coefficients are: The integration variable, representing continuous time; represents the mother wavelet function, which is used to decompose the input data at different scales; Represents the scale parameter, which controls the resolution of the wavelet transform.

[0077] Through the wavelet transform formula, rainfall data are decomposed into components of multiple scales, providing rich time-frequency information for building prediction models.

[0078] Construct a long short-term memory network model, train and verify the rainfall trend prediction sample: Based on the obtained wavelet decomposition results, use it as the input of the long short-term memory neural network to construct a rainfall trend prediction model. Define the prediction function as ;in, Represented in space coordinates , rainfall observation time of rainfall; Represents the mapping function of the long short-term memory network model. The internal structure includes input gate, forget gate and output gate, which are used to capture long-range dependencies in time series; Indicates the predicted rainfall observation time and the rainfall observation time Different, indicating the future rainfall observation period predicted by the long short-term memory network model.

[0079] In order to train the long short-term memory network model, the mean square error is used as the loss function, and its calculation formula is: ;in, Represents the mean square error value, which is used to measure the gap between the predicted results and the actual observed values; Represents the total number of samples, that is, the number of observation points involved in training and validation; Indicates The actual rainfall of each sample; Represents the long short-term memory network model for the The predicted rainfall for each sample.

[0080] The long short-term memory network model uses the back propagation through time algorithm to update parameters, and uses the adaptive optimization algorithm (such as Adam optimizer) to continuously reduce the mean square error value to achieve a higher prediction accuracy. After sufficient training and verification, the model has good rainfall trend prediction capabilities.

[0081] Calculate the regional rainfall impact coefficient based on the forecast results: Use the rainfall data predicted by the trained long short-term memory network model to evaluate the rainfall impact on the target area. When calculating the rainfall impact coefficient of the target area, set the baseline rainfall As a long-term average, it is quantified by the relative rate of change. The calculation formula for defining the rainfall impact coefficient is:

[0082] ;in, represents the rainfall influence coefficient; Indicates Predicted rainfall at each monitoring point; Indicates the baseline value of rainfall in the target area; Represents the total number of valid monitoring points in the target area.

[0083] By counting the average relative change of rainfall deviation from the baseline value at each monitoring point, the abnormal degree of rainfall events in the entire target area can be intuitively reflected.

[0084] Assess the potential impact of rainfall on the target area based on the rainfall impact coefficient: preset the rainfall impact coefficient threshold, and compare the rainfall impact coefficient with the rainfall impact coefficient threshold:

[0085] When the rainfall impact coefficient is greater than the rainfall impact coefficient threshold, it indicates that the potential impact of the rainfall event on the target area is high; the rainfall amount and spatiotemporal distribution in the target area deviate from the historical norm, and the rainfall is strong and lasts for a long time, indicating that the soil water absorption capacity in the target area may be quickly saturated, the surface runoff increases sharply, and the risk of geological disasters increases;

[0086] When the rainfall impact coefficient is less than or equal to the rainfall impact coefficient threshold, it means that the potential impact of the rainfall event on the target area is low; the rainfall conditions in the target area are basically within the historical normal range, and the rainfall intensity and temporal and spatial distribution have not reached the critical state of causing major disasters, indicating that the rainfall in the target area is relatively uniform, the soil moisture matches the drainage capacity, the hydrological environment is stable, and the overall risk is low.

[0087] The rainfall impact coefficient threshold is determined based on long-term meteorological observation data, historical disaster records, and regional geological and hydrological characteristics through statistical analysis, regression models, and probability calculations. The critical value is selected to accurately distinguish between normal and abnormal rainfall conditions.

[0088] Specifically, based on the geological risk level classification results and the potential impact of rainfall on the target area, high-risk trigger areas are screened;

[0089] The geological risk level classification results for the target area include level one, level two and level three.

[0090] When the geological risk level of the target area is level three, the target area is judged to be a high-risk trigger area; when the geological risk level of the target area is level two and the rainfall impact coefficient is greater than the rainfall impact coefficient threshold, the target area is judged to be a high-risk trigger area; otherwise, the target area is judged to be a non-high-risk trigger area.

[0091] Specifically, for high-risk triggering areas, the probability of landslide occurrence is evaluated by building a correlation model between the groundwater level change rate and the rock and soil strength parameters, using the support vector machine algorithm, including:

[0092] Collecting groundwater level change rate monitoring data: In the landslide monitoring scenario, groundwater level changes often affect the pore water pressure of the rock and soil mass, thereby affecting the stability of the geological mass. In order to accurately obtain the water level change rate at different locations and times, it is necessary to deploy multiple monitoring wells or sensors in high-risk trigger areas to record groundwater level data for continuous periods of time. The original observation data function is defined as: ;in, Indicates the monitoring value of groundwater level change rate; Represents the well point number, which is used to distinguish the locations of different observation points; Represents the layer number, for example, multiple layers of observation points in the same observation well; Represents the time index, reflecting the order at different observation times.

[0093] Select rock and soil strength factors and construct correlation factors: In addition to the rate of change of groundwater level, rock and soil strength has an important influence on the occurrence of landslides. Rock and soil strength usually includes multiple mechanical indicators such as internal friction angle, cohesion, density, etc. In order to fully reflect the properties of rock and soil, different representative factors are integrated and defined as ;in, A comprehensive factor that represents the strength of rock and soil.

[0094] In engineering practice, if a monitoring well corresponds to certain lithological parameters, these parameters are normalized and uniformly indexed to construct a correlation factor, which is recorded as .

[0095] Establish a correlation model between groundwater level and rock-soil strength: There is a nonlinear coupling relationship between the rate of change of groundwater level and the strength of rock-soil mass, and a correlation model needs to be established to obtain a characteristic combination that can represent the risk of landslides. Define the correlation function to describe the correlation model between groundwater level and rock-soil strength:

[0096] ;

[0097] in, It represents the output value of the water level rock soil strength correlation model; It represents the correlation function, which is used to nonlinearly map the groundwater level change rate data with the comprehensive factors of rock and soil mass strength.

[0098] The output of the association model will be used as one of the input features of the support vector machine to characterize the comprehensive state of water level intensity at a certain location at a certain moment.

[0099] Training support vector machines using kernel function optimization: In order to evaluate the probability of landslide occurrence, sample support vector machine classifiers. Since whether a landslide occurs is often a "occurrence / non-occurrence" or "high risk / low risk" model, a binary classification model can be trained first and the probability estimate can be derived. The training sample set is defined as: ;in, Indicates The feature vector of training samples; Indicates The class labels of the training samples (e.g., 1 means there is a landslide in history, 0 means there is no landslide); Represents the total number of training samples.

[0100] The output of the association model is combined with other auxiliary features (such as vegetation coverage, terrain curvature, etc.) to form a feature vector , and annotate the distribution of landslide events in history .

[0101] The support vector machine is trained and its decision function is expressed as:

[0102] ;in, Expressing the training samples (the feature vector is )’s support vector machine decision function output value; Indicates The Lagrange multiplier corresponding to the training samples; Represents the kernel function, which is used to calculate the The training samples and The similarity of training samples in feature space; Represents the bias term, which is a constant term in the support vector machine decision function and is used to adjust the position of the classification boundary.

[0103] The decision function is calculated by accumulating the contributions of all training samples and reflects the The position and distance of each training sample on the classification surface provide a basis for landslide risk determination.

[0104] Input the associated model parameters and evaluate the probability of landslide occurrence: Probabilistically process the classification results, use the Logistic function to map the output value of the support vector machine decision function to the (0,1) interval, and define the probability output as: ;in, Expressing the The predicted value of landslide occurrence probability for training samples; Represents an exponential function.

[0105] The larger the predicted value of the landslide probability, the higher the risk of landslide in the high-risk triggering area; at this time, the geological body is prone to instability, the landslide triggering conditions tend to the limit, and serious geological disasters may occur.

[0106] Specifically, the analytic hierarchy process is used to analyze the geological risk classification results, the potential impact of rainfall on the target area and the probability of landslides, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal, including:

[0107] A hierarchical analysis model is constructed, with the top-level goal being the comprehensive risk level assessment of the target area, the middle level being the three risk factors, and the bottom level being the specific measurement values ​​or quantitative results of each risk factor.

[0108] Risk factors include:

[0109] Results of geological risk classification: The target area is divided into primary, secondary and tertiary levels, reflecting the inherent stability and fragility of the geological body;

[0110] Rainfall impact coefficient: quantifies the potential impact of rainfall on the target area;

[0111] Landslide probability: The predicted value of landslide probability calculated using the support vector machine model indicates the probability of landslide risk.

[0112] In the hierarchical analysis method, it is necessary to compare the risk factors in pairs to determine their relative importance to the overall risk. The following weights are obtained through expert scoring or statistical analysis:

[0113] use Indicates the weight of the geological risk level classification result; Represents the weight of rainfall influence coefficient; Represents the weight of the predicted value of landslide occurrence probability. .

[0114] The geological risk level classification results are converted into quantitative indicators, which are defined as: ;in, It represents the geological risk value of the target area, which is discrete data, such as 1 for level 1, 2 for level 2, and 3 for level 3.

[0115] The comprehensive risk index of the target area is defined as:

[0116] ;in, represents the comprehensive risk index; A numerical value indicating the geological risk of the target area; Expressing the The predicted value of landslide occurrence probability for training samples; represents the rainfall influence coefficient; The weight representing the geological risk value; represents the weight of rainfall impact coefficient; Represents the weight of the predicted value of landslide occurrence probability.

[0117] Preset the comprehensive risk threshold, compare the comprehensive risk index with the comprehensive risk threshold, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal:

[0118] When the comprehensive risk index is greater than or equal to the comprehensive risk threshold, it means that the comprehensive risk level of the target area is high. At this time, it means that all risk indicators in the target area have reached or exceeded the set critical level. The target area is in a high-risk state as a whole. The safety hazards in the target area are obvious. It is necessary to trigger a graded early warning signal and take corresponding preventive and emergency measures to effectively reduce disaster risks.

[0119] When the comprehensive risk index is less than the comprehensive risk threshold, it means that the comprehensive risk level of the target area is low, the geological body is stable under the current environment, and there is no need to trigger a graded warning signal.

[0120] The comprehensive risk threshold is a critical value determined based on long-term historical data, statistical analysis and expert evaluation, and is used to define the risk boundary of geological disasters in the target area.

[0121] Embodiment 2:

[0122] The difference between Example 2 of the present invention and Example 1 is that this example introduces a geological disaster early warning system based on big data analysis.

[0123] Figure 2 A structural schematic diagram of a geological disaster early warning system based on big data analysis of the present invention is given, a geological disaster early warning system based on big data analysis, including a risk level classification module, an impact degree assessment module, a regional screening module, a landslide probability assessment module and an early warning signal judgment module;

[0124] The risk level classification module obtains the geological deformation time series data of the target area through satellite remote sensing monitoring, uses the random forest algorithm to evaluate the feature importance of the geological deformation data, and classifies the geological risk level of the target area;

[0125] The impact assessment module performs wavelet transform analysis on the spatial and temporal distribution data of rainfall at the meteorological monitoring station, establishes a rainfall trend prediction model based on the long short-term memory neural network, and assesses the potential impact of rainfall on the target area;

[0126] The regional screening module screens high-risk triggering areas based on the geological risk level classification results and the potential impact of rainfall on the target area;

[0127] For high-risk triggering areas, the landslide probability assessment module uses the support vector machine algorithm to assess the probability of landslide occurrence by constructing a correlation model between the groundwater level change rate and the rock and soil strength parameters;

[0128] The warning signal judgment module uses the hierarchical analysis method to analyze the geological risk level classification results, the potential impact of rainfall on the target area and the probability of landslide occurrence, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal.

[0129] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0130] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0131] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0132] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0133] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0134] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0135] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0136] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0137] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0138] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A geological disaster early warning method based on big data analysis, characterized in that: The steps include: The geological deformation time series data of the target area is obtained through satellite remote sensing monitoring, and the feature importance of the geological deformation data is evaluated using the random forest algorithm to divide the geological risk level of the target area; Wavelet transform analysis was performed on the spatial and temporal distribution data of rainfall at the meteorological monitoring station, and a rainfall trend prediction model was established based on the long short-term memory neural network to evaluate the potential impact of rainfall on the target area. Screen high-risk trigger areas based on the geological risk level classification results and the potential impact of rainfall on the target area; For high-risk triggering areas, the probability of landslide occurrence is evaluated by using the support vector machine algorithm by constructing a correlation model between the groundwater level change rate and the rock and soil strength parameters; The analytic hierarchy process is used to analyze the geological risk level classification results, the potential impact of rainfall on the target area and the probability of landslide occurrence, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal.

2. A geological disaster early warning method based on big data analysis according to claim 1, characterized in that: The geological deformation time series data of the target area are obtained through satellite remote sensing monitoring. The random forest algorithm is used to evaluate the feature importance of the geological deformation data and divide the geological risk level of the target area into the following categories: Acquire multi-source satellite images and extract original observation data; De-noising and interpolation of original observation data; Select geological attributes and historical disasters and uniformly label features; Random forests were used to evaluate feature importance and screen key factors; Use convolutional neural networks to identify reflection patterns and interference features; Based on the screening factors, the geological risk level of the target area is divided.

3. A geological disaster early warning method based on big data analysis according to claim 2, characterized in that: The temporal and spatial distribution data of rainfall at the meteorological monitoring station were analyzed by wavelet transform, and a rainfall trend prediction model was established based on the long short-term memory neural network to evaluate the potential impact of rainfall on the target area. Specifically: Collect and integrate the temporal and spatial distribution data of rainfall from meteorological monitoring stations in the target area; Multi-scale time-frequency decomposition of rainfall data was performed using wavelet transform; Construct a long short-term memory network model to train and verify rainfall trend prediction samples; Calculate regional rainfall impact coefficient based on forecast results; The potential impact of rainfall on the target area is assessed based on the rainfall impact coefficient.

4. A geological disaster early warning method based on big data analysis according to claim 3, characterized in that: The potential impact of rainfall on the target area is assessed based on the rainfall impact coefficient, specifically: Preset the rainfall impact coefficient threshold and compare the rainfall impact coefficient with the rainfall impact coefficient threshold: When the rainfall impact coefficient is greater than the rainfall impact coefficient threshold, it indicates that the potential impact of the rainfall event on the target area is high; When the rainfall impact coefficient is less than or equal to the rainfall impact coefficient threshold, it means that the potential impact of the rainfall event on the target area is low.

5. A geological disaster early warning method based on big data analysis according to claim 4, characterized in that: Based on the geological risk level classification results and the potential impact of rainfall on the target area, high-risk trigger areas are screened, specifically: When the geological risk level of the target area is level three, the target area is determined to be a high-risk trigger area; When the geological risk level of the target area is level 2 and the rainfall impact coefficient is greater than the rainfall impact coefficient threshold, the target area is determined to be a high-risk trigger area; Otherwise, the target area is determined to be a non-high-risk trigger area.

6. A geological disaster early warning method based on big data analysis according to claim 5, characterized in that: For high-risk triggering areas, the probability of landslide occurrence is evaluated by building a correlation model between the groundwater level change rate and the rock and soil strength parameters, using the support vector machine algorithm, as follows: Collect groundwater level change rate monitoring data; Select the strength factors of rock and soil mass and construct the correlation factors; Establish a groundwater level-rock-soil strength correlation model; The kernel function optimization method is used to train the support vector machine; Input the associated model parameters and evaluate the probability of landslide occurrence.

7. A geological disaster early warning method based on big data analysis according to claim 6, characterized in that: The analytic hierarchy process is used to analyze the geological risk classification results, the potential impact of rainfall on the target area and the probability of landslides, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal. Specifically: Preset the comprehensive risk threshold and compare the comprehensive risk index with the comprehensive risk threshold: When the comprehensive risk index is greater than or equal to the comprehensive risk threshold, it means that the comprehensive risk level of the target area is high and a graded warning signal needs to be triggered; When the comprehensive risk index is less than the comprehensive risk threshold, it means that the comprehensive risk level of the target area is low and there is no need to trigger a graded warning signal.

8. A geological disaster early warning system based on big data analysis, used to implement a geological disaster early warning method based on big data analysis as described in any one of claims 1 to 7, characterized in that: It includes risk level classification module, impact assessment module, area screening module, landslide probability assessment module and early warning signal judgment module; The risk level classification module obtains the geological deformation time series data of the target area through satellite remote sensing monitoring, uses the random forest algorithm to evaluate the feature importance of the geological deformation data, and classifies the geological risk level of the target area; The impact assessment module performs wavelet transform analysis on the spatial and temporal distribution data of rainfall at the meteorological monitoring station, establishes a rainfall trend prediction model based on the long short-term memory neural network, and assesses the potential impact of rainfall on the target area; The regional screening module screens high-risk triggering areas based on the geological risk level classification results and the potential impact of rainfall on the target area; For high-risk triggering areas, the landslide probability assessment module uses the support vector machine algorithm to assess the probability of landslide occurrence by constructing a correlation model between the groundwater level change rate and the rock and soil strength parameters; The warning signal judgment module uses the hierarchical analysis method to analyze the geological risk level classification results, the potential impact of rainfall on the target area and the probability of landslide occurrence, evaluate the comprehensive risk level of the target area, and determine whether to trigger a graded warning signal.

Images