Snow layer stability monitoring method and system

Abstract

This invention proposes a snow layer stability monitoring method and system. The method includes the following steps: acquiring monitoring data transmitted by each multidimensional sensor in a multidimensional sensor array; obtaining snow layer state data based on the monitoring data; and determining disaster risk based on the snow layer state data. By setting up a multidimensional sensor array containing multiple multidimensional sensors, high-precision monitoring of snow layer state can be achieved. Based on the obtained snow layer state data, the snow layer state can be determined to assess disaster risk, providing an all-weather, grid-based, precise prevention and control solution for high-risk avalanche areas.

Description

Technical Field

[0001] This invention relates to the field of meteorological monitoring, and in particular to a method and system for monitoring snow layer stability. Background Technology

[0002] Currently, avalanche disaster prevention and control mainly relies on traditional geological surveys and manual patrols, combined with satellite remote sensing for large-scale monitoring. However, due to factors such as cloud cover and poor cold resistance of equipment, these methods have significant shortcomings in data accuracy and timeliness, making it difficult to meet the needs of refined prevention and control in high-risk areas. Especially in high-altitude and frigid environments, existing monitoring equipment often cannot operate stably for extended periods, resulting in discontinuous and inaccurate data, which in turn affects the timeliness and reliability of avalanche warnings. Summary of the Invention

[0003] The main objective of this invention is to propose a snow layer stability monitoring method and system, which aims to solve the problem of poor reliability in avalanche detection in the prior art.

[0004] To achieve the above objectives, the present invention provides a method for monitoring snow layer stability, the method comprising the following steps:

[0005] Acquire monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array;

[0006] Snow layer status data are obtained based on the monitoring data described above;

[0007] Disaster risk is determined based on the snow layer condition data.

[0008] Optionally, obtaining snow layer state data based on the monitoring data includes:

[0009] Time-aligned data from each monitoring data point is used to obtain time-based monitoring data.

[0010] The spatial monitoring data is obtained by mapping each monitoring data into a preset data space.

[0011] The snow layer state data is determined based on the time monitoring data and the spatial monitoring data.

[0012] Optionally, the snow layer state data includes risk probability, and determining the snow layer state data based on the time monitoring data and the spatial monitoring data includes:

[0013] Obtain a hybrid model, wherein the hybrid model includes a convolutional neural network layer, a long short-term memory network layer, and a fully connected layer;

[0014] The spatial monitoring data is input into the convolutional neural network layer to obtain spatial features;

[0015] The time monitoring data is input into the long short-term memory network layer to obtain time features;

[0016] The spatial features and the temporal features are combined to obtain the fused features;

[0017] The fused features are input into the fully connected layer to obtain the risk probability.

[0018] Optionally, determining the disaster risk based on the snow layer state data includes:

[0019] Determine the comprehensive risk index corresponding to the snow layer state data;

[0020] Obtain the risk classification threshold, and determine the disaster risk corresponding to the comprehensive risk index based on the risk classification threshold.

[0021] Optionally, the snow layer state data includes risk probability and environmental parameters; determining the comprehensive risk index corresponding to the snow layer state data includes:

[0022] Obtain the risk probability and the weights corresponding to the environmental parameters;

[0023] The comprehensive risk index is obtained by weighting the risk probability and the environmental parameters based on the weights.

[0024] Optionally, acquiring the monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array includes:

[0025] Obtain real-time disaster risks;

[0026] Determine the target acquisition frequency corresponding to the real-time disaster risk, wherein the target acquisition frequency is positively correlated with the real-time disaster risk;

[0027] The monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array is acquired at the target acquisition frequency.

[0028] Optionally, the step of determining the disaster risk based on the snow layer state data includes:

[0029] Obtain the terrain model corresponding to the multi-dimensional sensor array;

[0030] The snow layer state data and the disaster risk are mapped onto the terrain model for display.

[0031] Optionally, the step of determining the disaster risk based on the snow layer state data includes:

[0032] Determine whether the disaster risk has reached a preset risk threshold;

[0033] If the disaster risk reaches the preset risk threshold, historical snow cover status data is obtained.

[0034] Predict avalanche prediction time based on the historical snow cover state data;

[0035] A warning operation is executed based on the predicted avalanche time.

[0036] To achieve the above objectives, the present invention also provides a snow layer stability monitoring system, the snow layer stability monitoring system comprising:

[0037] The snow layer stability monitoring system includes a multi-dimensional sensor array and a processing module; the multi-dimensional sensor array includes multiple multi-dimensional sensors, each of which is communicatively connected to the processing module; wherein:

[0038] The multidimensional sensor is used to monitor the snow layer parameters at its location to obtain monitoring data, and then sends the monitoring data to the processing module.

[0039] The processing module is used to receive the monitoring data sent by each of the multi-dimensional sensors, obtain snow layer state data based on the monitoring data, and determine the disaster risk based on the snow layer state data.

[0040] Optionally, the multidimensional sensor includes a longitudinal anchor bolt, on which multiple temperature sensors are mounted, and the temperature sensors are set at preset intervals.

[0041] To achieve the above objectives, the present invention also provides an electronic device, the electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the snow layer stability monitoring method as described above.

[0042] To achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the snow layer stability monitoring method as described above.

[0043] This invention proposes a snow layer stability monitoring method and system, which acquires monitoring data transmitted by each multidimensional sensor in a multidimensional sensor array; obtains snow layer state data based on the monitoring data; and determines disaster risk based on the snow layer state data. By setting up a multidimensional sensor array containing multiple multidimensional sensors, high-precision monitoring of snow layer state can be achieved, and the obtained snow layer state data can be used to determine the snow layer state to assess disaster risk, providing an all-weather, grid-based, precise prevention and control solution for high-risk avalanche areas. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart illustrating the first embodiment of the snow layer stability monitoring method of the present invention;

[0047] Figure 2 This is a three-dimensional view of the snow layer stability monitoring method of the present invention;

[0048] Figure 3 This is an overall flowchart of the snow layer stability monitoring method of the present invention;

[0049] Figure 4 This is a schematic diagram of the module structure of the electronic device of the present invention. Detailed Implementation

[0050] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0051] This invention provides a snow cover stability monitoring method, applied to a snow cover stability monitoring system, with reference to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the snow layer stability monitoring method of the present invention. The method includes the following steps:

[0052] Step S10: Obtain the monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array;

[0053] For ease of explanation, the snow layer stability monitoring system will be described first. The snow layer stability monitoring system includes:

[0054] The snow layer stability monitoring system includes a multi-dimensional sensor array and a processing module; the multi-dimensional sensor array includes multiple multi-dimensional sensors, each of which is communicatively connected to the processing module; wherein:

[0055] The multidimensional sensor is used to monitor the snow layer parameters at its location to obtain monitoring data, and then sends the monitoring data to the processing module.

[0056] The processing module is used to receive the monitoring data sent by each of the multi-dimensional sensors, obtain snow layer state data based on the monitoring data, and determine the disaster risk based on the snow layer state data.

[0057] A multidimensional sensor array is an array composed of multiple multidimensional sensors; the specific multidimensional sensors can be set according to actual needs, such as setting horizontal and vertical intervals, with horizontal multidimensional sensors arranged at horizontal intervals and vertical multidimensional sensors arranged at vertical intervals.

[0058] Each multi-dimensional sensor monitors the snow layer condition at its location to obtain monitoring data. The monitoring data obtained by each multi-dimensional sensor can be used to obtain snow layer condition data, which can reflect the snow layer condition in the area where the multi-dimensional sensor array is located. Based on the snow layer condition data, disaster risk can be analyzed and determined. Disaster risk indicates the probability of an avalanche disaster.

[0059] Multidimensional sensors can monitor different types of parameters, and can be set according to actual needs. For example, the multidimensional sensor includes a longitudinal anchor rod, on which multiple temperature sensors are set at preset intervals.

[0060] Multidimensional sensors are inserted into the snow layer via longitudinal anchors; multiple temperature sensors are installed on the longitudinal anchors at preset intervals in the longitudinal direction, enabling the monitoring of temperature at different depths of the snow layer.

[0061] Besides temperature sensors, other types of sensors can be used to monitor other parameters, such as pressure sensors, snow depth sensors, piezoelectric sensors, and weather instruments. Pressure sensors monitor snow pressure, snow depth sensors monitor snow depth, weather instruments monitor meteorological parameters such as wind speed and precipitation, and piezoelectric sensors monitor minute deformations of the snow layer. It should be noted that the above sensor configurations are merely illustrative; in practical applications, corresponding sensor types can be configured according to the parameters to be monitored.

[0062] Step S20: Obtain snow layer state data based on the monitoring data.

[0063] After determining the monitoring data, the real-time state of the snow layer is determined based on the monitoring data to obtain snow layer state data; the snow layer state data can reflect the snow layer state in the area where the multi-dimensional sensor array is located.

[0064] Snow cover status data can include data indicating different snow cover states, such as environmental parameters, risk probability, regional location information, and the state of the snow cover itself.

[0065] Step S30: Determine the disaster risk based on the snow layer state data.

[0066] After determining the snow cover status data, the disaster risk is determined by comprehensively analyzing the snow cover status data, thereby judging the probability of an avalanche occurring under the current snow cover status.

[0067] This embodiment enables high-precision monitoring of snow layer conditions by setting up a multi-dimensional sensor array containing multiple multi-dimensional sensors. Based on the obtained snow layer condition data, the snow layer condition is determined to assess disaster risk, providing an all-weather, grid-based, precise prevention and control solution for high-risk avalanche areas.

[0068] Furthermore, in the second embodiment of the snow layer stability monitoring method of the present invention based on the first embodiment, step S20 includes the following steps:

[0069] Step S21: Time-align the monitoring data to obtain time monitoring data;

[0070] Step S22: Map each monitoring data to a preset data space to obtain spatial monitoring data;

[0071] Step S23: Determine the snow layer state data based on the time monitoring data and the spatial monitoring data.

[0072] Traditional monitoring methods rely on single-point monitoring, where a single sensor is used to monitor specific parameters at a single location. In contrast, the monitoring data in this embodiment originates from a diverse array of sensors within a multi-dimensional sensor array. Different sensors collect data at different times, locations, and frequencies. Consequently, the spatiotemporal references for different types of data are inconsistent, making it difficult to capture the dynamic correlations of snow cover conditions. This embodiment addresses this by performing spatiotemporal alignment on the monitoring data, thus resolving the spatial fragmentation problem. The time-aligned data possesses temporal and spatial uniformity, thereby reflecting the temporal and spatial distribution of snow cover conditions and providing a consistent dataset for disaster risk assessment.

[0073] Specifically, in this embodiment, the monitoring data is aligned in both time and space.

[0074] Time alignment is used to unify the time of various data points within the monitoring data. In this embodiment, time linear interpolation is specifically used to align heterogeneous data from different sensors within the monitoring data; specifically:

[0075]

[0076] Where x(t) target ) represents the target time interpolation; t target The target time is (t1, x1) and (t2, x2) are different known data. (t1, x1) is the x1 data collected at time t1, and (t2, x2) is the x2 data collected at time t2. It can be understood that (t1, x1) and (t2, x2) are the same type of data detected by the same sensor, the difference being that they were collected at different times. The target time difference obtained from this also reflects the same type of data.

[0077] For each type of data collected by each sensor, data corresponding to the target time is generated through time linear interpolation. This ensures that each type of data collected by each sensor can be obtained at each target time, thus achieving time alignment of various data types. For example, the snow temperature at the deepest position on the first sensor is collected at the first and second moments, while the wind speed on the second sensor is collected at the first, third, and second moments. The third moment is located between the first and second moments. At this time, the snow temperature at the deepest position collected by the first sensor is obtained by time linear interpolation based on the first and second moments to obtain the target time difference corresponding to the third moment. This ensures that the first sensor has corresponding collected data for the snow temperature at the deepest position at the first, third, and second moments, achieving time alignment with the wind speed data on the second sensor.

[0078] Time monitoring data can be obtained by aggregating the time-aligned data; the time monitoring data can be used to track rapid changes in various parameters.

[0079] Spatial alignment is used to spatially unify various data points within the monitoring data. In this embodiment, spatial interpolation is specifically used to align heterogeneous data from different locations and sensors within the monitoring data. Specifically:

[0080]

[0081]

[0082] Where Z(s0) is the target position interpolation; Z(s i ) represents the known parameters corresponding to the i-th point; d id represents the distance between the location of the i-th known sensor and the target location; j λ is the distance between the j-th known sensor position and the target position. i Let be the weight of the i-th point.

[0083] Spatial interpolation calculates the target position interpolation of a new location using data from known points, thereby achieving spatial alignment of different data. The goal of spatial alignment is to map sensor data from different spatial locations onto a unified three-dimensional grid or target point to form a spatially continuous data field, thus reflecting the continuous changes of data in space.

[0084] After monitoring data is collected, noise reduction and temperature drift compensation can be performed to improve data quality; specifically:

[0085]

[0086]

[0087] Where x[n] is the sampled value of the original signal acquired by the sensor at the nth discrete time point; x smooth [n] is the smoothed signal of the original signal x[n] after passing through the moving average filter. smooth [n] can more clearly reflect the sustained vibrational energy of snow layer fracturing events; M is the window size; S comp The compensated signal value; S raw The original signal value is represented by k; the temperature drift coefficient is represented by T. ref This is a reference temperature.

[0088] Noise reduction is achieved by removing high-frequency noise through moving average filtering; temperature drift compensation corrects the drift of the temperature sensor under extremely cold conditions of -25℃ to ensure measurement accuracy.

[0089] Spatial monitoring data can be obtained by combining the spatially aligned data.

[0090] Furthermore, the snow layer state data includes risk probability, and step S23 includes the following steps:

[0091] Step S231: Obtain a hybrid model, wherein the hybrid model includes a convolutional neural network layer, a long short-term memory network layer, and a fully connected layer;

[0092] Step S232: Input the spatial monitoring data into the convolutional neural network layer to obtain spatial features;

[0093] Step S233: Input the time monitoring data into the long short-term memory network layer to obtain time features;

[0094] Step S234: The spatial features and the temporal features are spliced ​​together to obtain the fused features;

[0095] Step S235: Input the fused features into the fully connected layer to obtain the risk probability.

[0096] Time-based monitoring data reflects the continuous changes of different types of data over time; spatial monitoring data reflects the continuous changes of different types of data in space; different types of models have different advantages for different types of monitoring data. Therefore, in this embodiment, a hybrid model integrating different types of models is used to determine the risk probability based on time-based and spatial monitoring data.

[0097] Specifically, the hybrid model includes CNN (Convolutional Neural Networks), LSTM (Long Short-Term Memory) network layers, and fully connected layers.

[0098] Convolutional neural network layers are used to extract the spatial distribution patterns of temperature and pressure fields. Therefore, spatial monitoring data is input into convolutional neural network layers for feature extraction to obtain spatial features. Specifically:

[0099]

[0100] Where W is the convolution kernel, b is the bias, σ is the activation function, and * denotes the convolution operation; after convolution through the convolutional neural network layer, the spatial feature is output. spatial .

[0101] Long Short-Term Memory (LSTM) network layers are used to model the temporal correlation between vibration signal energy and meteorological parameters. Therefore, time-based monitoring data is input into the LSM network layer for feature extraction to obtain temporal features. temporal .

[0102] By extracting features from spatial and temporal monitoring data using convolutional neural network layers and long short-term memory network layers respectively, it is possible to learn complex nonlinear patterns of avalanche precursors from historical data. This fusion mechanism enables the early warning model to reflect physical mechanisms and adapt to dynamic environmental changes, thereby reducing the false alarm rate.

[0103] After obtaining the spatial and temporal features, the temporal features are concatenated to obtain the fused features:

[0104]

[0105] Among them, Feature fusedThis is a feature of fusion.

[0106] The fused features are input into a fully connected layer to output the risk probability.

[0107]

[0108] Among them, P avalanche σ represents the risk probability; σ is the Sigmoid activation function with an output range of [0, 1]; b is a constant.

[0109] The risk probability is the output of the hybrid model used to indicate the probability of an avalanche occurring.

[0110] Understandably, the training and specific structure of hybrid models can be configured based on actual needs.

[0111] Furthermore, in the third embodiment of the snow layer stability monitoring method of the present invention based on the first embodiment, step S30 includes the following steps:

[0112] Step S31: Determine the comprehensive risk index corresponding to the snow layer state data;

[0113] Step S32: Obtain the risk classification threshold, and determine the disaster risk corresponding to the comprehensive risk index based on the risk classification threshold.

[0114] The comprehensive risk index R is used to indicate the degree of disaster risk posed by the snow cover condition.

[0115] Snow cover status data indicates various data reflecting the snow cover status. Therefore, the snow cover status can be reflected through snow cover status data, thereby determining the comprehensive risk index.

[0116] The risk classification threshold is the threshold for classifying disaster risk levels. It is understood that the disaster risk situation corresponding to the snow layer state is different in different application environments. Therefore, in this embodiment, a risk classification threshold is set to specifically determine the relative relationship between the comprehensive risk index and disaster risk in a specific environment, thereby determining the disaster risk corresponding to the comprehensive risk index.

[0117] Risk classification thresholds can be set based on the actual environment, such as building a physical model based on historical data. This physical model is based on existing physical laws, meteorological conditions (such as temperature, wind speed, and snowfall), and topography. By combining historical data and avalanche records, the critical values ​​between different disaster risks are obtained, and these critical values ​​are used as risk classification thresholds. Specifically, this involves analyzing the comprehensive risk index R value corresponding to past disasters under the current environment, such as the current location and weather conditions; obtaining the maximum comprehensive risk index R1 under historical disaster-free conditions, and the comprehensive risk index R2 corresponding to small- to medium-sized avalanche events; using R1 × 95% as the first risk classification threshold T1, and R2 × 75% as the second risk classification threshold T2, where 95% and 75% can be set based on actual needs; the risk classification thresholds can be updated based on actual accumulated historical data. For example, at the beginning and end of the snow season, the thresholds need to be dynamically adjusted according to different snow layer characteristics.

[0118] At this point, T1 and T2 constitute three intervals, each corresponding to a different level of disaster risk, as follows:

[0119]

[0120] By comparing the comprehensive risk index R with the risk classification threshold, the level of the current disaster risk can be determined.

[0121] It is understood that the above risk classification threshold settings are only for illustrative purposes, and the specific number and determination method of risk classification thresholds can be set based on actual needs.

[0122] Furthermore, the snow layer state data includes risk probability and environmental parameters; step S31 includes the following steps:

[0123] Step S311: Obtain the risk probability and the weights corresponding to the environmental parameters;

[0124] Step S312: The comprehensive risk index is obtained by weighting the risk probability and the environmental parameters based on the weights.

[0125] The comprehensive risk index is obtained by combining different types of parameters to fully reflect the snow layer condition.

[0126] Environmental parameters are detected by corresponding sensors, and the data types of these parameters can be set according to actual needs. In this embodiment, the environmental parameters include snow stability factor, temperature gradient, snowfall, wind speed, and slope. Different environmental parameters are assigned corresponding weights based on their impact on snow stability, and a comprehensive risk index is calculated based on these weights.

[0127]

[0128] Among them, S f ∇T is the snow layer stability factor; ∇T is the temperature gradient; Sd is the snowfall amount; Sw is the wind speed; θ is the slope; w1~w6 are the weights of the corresponding parameters.

[0129] Wind speed and snowfall can be detected by corresponding sensors.

[0130] The snow stability factor is used to reflect the stability of the snow layer. Whether the snow layer is stable depends on the shear strength τ and shear stress σ of the snow. s The comparison; the shear strength of snow can be expressed by the Mohr-Coulomb criterion:

[0131]

[0132]

[0133] Where c is the cohesive force, which depends on the type and temperature of the snow; ϕ is the internal friction angle, which depends on the particle characteristics of the snow; σ n It is normal stress, usually generated by the weight of the snow layer itself:

[0134]

[0135] Where ρ is the snow density; ℎ is the snow layer thickness; and 𝜃 is the slope. The slope at the location can be measured and stored when setting up the multi-dimensional sensor array in the early stage, and the stored slope can be retrieved when needed.

[0136] P(z) represents the snow pressure distribution, specifically:

[0137]

[0138]

[0139] Where w is the snow moisture content, obtained through corresponding sensor detection; z is the snow depth, obtained through corresponding sensor detection; P0 is the surface pressure of the snow layer, i.e., atmospheric pressure; g is the acceleration due to gravity; and S is the cross-sectional area of ​​the snow layer, which can be equivalently taken as S = 0.01m. 2 hs is the snow layer thickness; ρ Water This is the density of water.

[0140]

[0141] Shear stress is caused by the gravitational component of the snow layer on the slope, when σ s When the slope is greater than τ, the snow layer may slip, triggering an avalanche. The instability of the snow layer can be assessed using critical slope parameters, such as:

[0142]

[0143] Where, θ c This is the critical slope. If the actual slope θ > θ c The risk of snow layer instability increases.

[0144] Snow layer stability factors include:

[0145]

[0146] Temperature gradients are used to indicate the temperature distribution within a snow layer. Temperature gradients include longitudinal and lateral gradients:

[0147]

[0148]

[0149] Among them, ∇T vertical For the vertical gradient; ∇T horizontal For the lateral gradient; ΔT vertical ΔT represents the temperature difference between different snow depths along the longitudinal direction in the multidimensional sensor array. horizontal Δz represents the temperature difference between different snow depths in the lateral direction of the multidimensional sensor array; Δx represents the distance between the multidimensional sensors in the longitudinal direction of the multidimensional sensor array; and Δx represents the distance between the multidimensional sensors in the lateral direction of the multidimensional sensor array.

[0150] Temperature differences are obtained from longitudinal or transverse temperature data; for example, on a high-risk avalanche slope, several m multidimensional sensors are deployed in a rectangular array, with snow temperature sensors installed on each sensor, and the node spacing is set according to the slope characteristics. A spatial reference is established using the UTM projected coordinate system, and the geographic coordinates of the (i, j)th sensor are (x... i，j y i，j The slope temperature field distribution T(x, y) is generated through spatial interpolation, and the temperature difference in the lateral direction is obtained through T(x, y); where:

[0151]

[0152]

[0153]

[0154] Where Δx and Δy are the spacing between the snow temperature sensors in the horizontal and vertical directions, respectively; λ k These are the weighting coefficients, and the sum of all weighting coefficients is 1.

[0155] In the longitudinal direction, multiple platinum resistance temperature sensors are installed at different depths on the longitudinal anchor bolts of the multidimensional sensor, such as one platinum resistance temperature sensor every 10cm; This yields:

[0156]

[0157] Among them, T h The temperature of the snow layer at depth h is given.

[0158] Temperature gradients affect the snow metamorphism process and are a key factor in snow layer stability. When ∇T > 10℃ / m, the snow layer is prone to forming a fragile layer, which reduces the stability of the snow layer.

[0159] Furthermore, in the fourth embodiment of the snow layer stability monitoring method of the present invention based on the first embodiment, step S10 includes the following steps:

[0160] Step S11: Obtain real-time disaster risk;

[0161] Step S12: Determine the target acquisition frequency corresponding to the real-time disaster risk, wherein the target acquisition frequency is positively correlated with the real-time disaster risk;

[0162] Step S13: Acquire the monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array at the target acquisition frequency.

[0163] Real-time disaster risk refers to the disaster risk level determined in real time. It can be understood that the lower the real-time disaster risk level, the more stable the snow layer condition is. At this time, the frequency of data collection can be reduced, thereby reducing power consumption and maintaining low power consumption operation of the system to extend the battery life. On the other hand, the higher the real-time disaster risk level, the more unstable the snow layer condition is. Therefore, a higher frequency of data collection is required to obtain the snow layer condition more quickly and pay attention to changes in the snow layer condition in order to execute emergency operations.

[0164] The specific correspondence between real-time disaster risk, target acquisition frequency, and emergency operations can be set based on actual needs. For example, when the real-time disaster risk level is "low risk," the system samples data at the set first frequency to maintain low power consumption, continuously monitoring and recording. When the real-time disaster risk level is "medium risk," data is sampled at the second frequency, which is higher than the first frequency, and the emergency department is notified to prepare emergency supplies. When the real-time disaster risk level is "high risk," data is sampled at the third frequency, which is higher than the second frequency, and a comprehensive emergency plan is activated, including evacuating personnel and closing roads.

[0165] Specific emergency operations can be carried out by the monitoring center. For example, the snow layer stability detection system will transmit the real-time disaster risk assessment results and early warning information to the monitoring center and relevant personnel in real time, and link with the emergency system to ensure that the early warning lead time is greater than 2 hours to meet the needs of rapid response. The early warning information includes risk level, probability of occurrence, expected events, etc. It can be transmitted through low power wide area network LoRa or satellite communication.

[0166] Furthermore, in the fifth embodiment of the snow layer stability monitoring method of the present invention based on the first embodiment of the present invention, step S30 is followed by the following step:

[0167] Step S40: Obtain the terrain model corresponding to the multi-dimensional sensor array;

[0168] Step S50: Map the snow layer state data and the disaster risk onto the terrain model for display.

[0169] The terrain model is a model of the terrain at the location of the multi-dimensional sensor array; the specific terrain model can be rendered based on the elevation data of the DEM (Digital Elevation Model) monitored at the location.

[0170] In this embodiment, snow layer state data and disaster risk are mapped onto a terrain model to obtain a three-dimensional view. This enables three-dimensional visualization of the snow layer state, providing an intuitive display of the snow layer state and risk distribution, facilitating monitoring personnel to quickly identify high-risk areas. Specifically:

[0171]

[0172] Here, D(x, y, z) is a three-dimensional tensor corresponding to the coordinates (x, y, z). By obtaining a three-dimensional tensor from different types of data at the same location, it is possible to display the snow layer conditions at a specific location.

[0173] The location corresponding to the disaster risk level in the terrain model can be marked. See [link / reference]. Figure 2 For example, color gradients can be used to represent disaster risk levels, with red indicating high risk, yellow indicating medium risk, and green indicating low risk. Color markings can help observers intuitively understand the snow layer condition and quickly locate high-risk areas. If an area shows a high temperature gradient and high vibration energy, combined with an image marked in red, it indicates that the area has an extremely high risk of avalanches and requires close attention, with preventative measures to be taken in advance.

[0174] Furthermore, in the sixth embodiment of the snow layer stability monitoring method of the present invention based on the first embodiment of the present invention, step S30 is followed by the following step:

[0175] Step S60: Determine whether the disaster risk has reached a preset risk threshold;

[0176] Step S70: If the disaster risk reaches the preset risk threshold, then obtain historical snow layer status data;

[0177] Step S80: Predict the avalanche prediction time based on the historical snow layer state data;

[0178] Step S90: Execute a warning operation based on the avalanche prediction time.

[0179] When the risk of disaster reaches a certain level, it indicates that an avalanche may occur. Therefore, in order to provide early warning of avalanches, this embodiment predicts the avalanche time when there is an avalanche risk, thereby enabling early warning based on the avalanche prediction time.

[0180] The preset risk threshold is used to indicate the threshold at which an avalanche may occur. The specific preset risk threshold can be set based on the actual application scenario. For example, if the preset risk threshold is set to medium risk, when the disaster risk level reaches medium risk, it is considered that there is an avalanche risk, and historical snow layer status data is obtained at this time.

[0181] Historical snow cover data refers to snow cover data obtained from historical monitoring. This data reflects changes in snow cover conditions, and therefore, it allows for the prediction of subsequent changes in snow cover conditions, thus determining the avalanche prediction time t. pred Specifically, in this embodiment, an LSTM model is used to predict the avalanche prediction time.

[0182]

[0183] Among them, X temporal Historical snow cover data; includes:

[0184]

[0185] Among them, t N This indicates the snow layer state data obtained from the Nth acquisition; the number of snow layer state data points included in the historical snow layer state data can be set based on actual needs; T surface E represents the surface temperature of the snow layer. total This is a vibration capability signal;

[0186]

[0187] Among them, E iLet E be the energy of the i-th sensor. Vibration signals generated when the snow layer fractures and undergoes micro-deformation are captured by a piezoelectric sensor. To quantify the intensity of the vibration, the energy of the signal can be calculated. When a high energy value is detected, snow layer fracture may have occurred on the surface. Further analysis of the characteristics of the snow layer fracture vibration signal is needed. Fourier transform can be used to convert the time-domain signal into a frequency-domain signal.

[0188]

[0189] Among them, V 2 (t) is the square of the vibration signal voltage output by the piezoelectric sensor at time t; T is the duration of the vibration signal;

[0190]

[0191] x(t) represents the time-domain vibration signal, X(f) represents the frequency-domain signal, and f represents the frequency. Through spectral analysis, specific frequency components of snow layer fracture can be identified; for example, high-frequency vibration corresponds to rapid fracture, while low-frequency vibration corresponds to slow creep.

[0192] w i These are the corresponding weights;

[0193]

[0194]

[0195]

[0196] Among them, w distance，i For distance weights; d i w is the distance from the i-th sensor to the target area, p is the attenuation coefficient (usually p=2); quality，i For quality weights, Q i The quality index is determined by the signal-to-noise ratio; α and β are adjustment coefficients that reflect the priority between distance and quality. α can be 0.7 and β can be 0.3.

[0197]

[0198] The vibration energy is continuously monitored, and its mean μ and standard deviation σ are calculated within a sliding window to obtain the vibration energy threshold E. threshold， k is an adjustable parameter; when the vibration energy E > E threshold If a fracture event occurs, it is determined that a fracture event has occurred, and then a comprehensive analysis is conducted in conjunction with other parameters.

[0199] After predicting avalanche timing, a RAMMS (Rapid Mass Movements Imaging) dynamic model can be triggered simultaneously. The DEM (Digital Elevation Model) and snow layer state data are input into the RAMMS dynamic model to locate the avalanche initiation zone coordinates based on vibration energy peaks. Snow layer state data can include snow density ρ, cohesion c, and friction angle ϕ. The RAMMS dynamic model outputs the avalanche flow trajectory and maximum impact range in the form of a probability density map, dividing the area into a core zone (p≥10 kPa) and a warning zone (p≥5 kPa) according to the destructive intensity. This allows for the determination of the avalanche's location, facilitating early warning.

[0200] The following is based on Figure 3 The overall implementation process of this invention is described below:

[0201] 1. Obtain snow layer status data and inspect DEM elevation data;

[0202] 2. Perform noise reduction and temperature drift compensation on the snow layer status data;

[0203] 3. Spatiotemporal fusion of the processed snow layer status data and elevation data is performed, specifically by time alignment and spatial alignment to obtain time monitoring data and spatial monitoring data;

[0204] 4. Risk probability is obtained by using a hybrid model based on time-based and spatial-based monitoring data;

[0205] 5. Calculate the risk index based on the risk probability and determine the level of disaster risk;

[0206] 6. Generate a 3D view based on elevation data and issue early warnings based on the level of disaster risk.

[0207] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0208] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0209] This application also provides a snow layer stability monitoring system for implementing the above-described snow layer stability monitoring method. The snow layer stability monitoring system includes a multi-dimensional sensor array and a processing module. The multi-dimensional sensor array includes multiple multi-dimensional sensors, and each multi-dimensional sensor is communicatively connected to the processing module.

[0210] The multidimensional sensor is used to monitor the snow layer parameters at its location to obtain monitoring data, and then sends the monitoring data to the processing module.

[0211] The processing module is used to receive the monitoring data sent by each of the multi-dimensional sensors, obtain snow layer state data based on the monitoring data, and determine the disaster risk based on the snow layer state data.

[0212] This snow stability monitoring system enables high-precision monitoring of snow conditions by setting up a multi-dimensional sensor array containing multiple multi-dimensional sensors. Based on the obtained snow condition data, the system determines the snow condition to assess disaster risk and provides an all-weather, grid-based, precise prevention and control solution for high-risk avalanche areas.

[0213] Furthermore, the multidimensional sensor includes a longitudinal anchor bolt, on which multiple temperature sensors are installed, and the temperature sensors are set at preset intervals.

[0214] This application also provides an electronic device, with reference to... Figure 4 In terms of hardware structure, the electronic device may include components such as a communication module 10, a memory 20, and a processor 30. In the electronic device, the processor 30 is connected to both the memory 20 and the communication module 10. The memory 20 stores a computer program, which is executed by the processor 30. When the computer program is executed, it implements the steps of the above-described method embodiments.

[0215] The communication module 10 can connect to external communication devices via a network. The communication module 10 can receive requests from the external communication devices and can also send requests, instructions, and information to the external communication devices. The external communication devices can be other electronic devices, servers, or IoT devices, such as televisions, etc.

[0216] The memory 20 can be used to store software programs and various data. The memory 20 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as acquiring monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array), etc.; the data storage area may include a database, and may store data or information created based on system usage. Furthermore, the memory 20 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0217] The processor 30 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 20, and by calling data stored in the memory 20, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. The processor 30 may include one or more processing units; optionally, the processor 30 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 30.

[0218] although Figure 4 Not shown, but the above-described electronic device may further include a circuit control module for connecting to a power supply to ensure the normal operation of other components. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0219] The present invention also proposes a computer-readable storage medium having a computer program stored thereon. The computer-readable storage medium may be... Figure 4The memory 20 in the electronic device may also be at least one of ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk, etc. The computer-readable storage medium includes a number of instructions to cause a terminal device with a processor (which may be a television, automobile, mobile phone, computer, server, terminal, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0220] In this invention, the terms "first," "second," "third," "fourth," and "fifth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0221] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0222] Although embodiments of the present invention have been shown and described above, the scope of protection of the present invention is not limited thereto. It is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, and substitutions to the above embodiments within the scope of the present invention, and such changes, modifications, and substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for monitoring snow layer stability, characterized in that, The snow layer stability monitoring method includes: The monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array is obtained. Each multi-dimensional sensor monitors the snow layer state at its location to obtain the monitoring data. Snow layer state data is obtained based on the aforementioned monitoring data. This snow layer state data includes a temperature gradient, which comprises a longitudinal gradient and a lateral gradient. The longitudinal gradient is: The lateral gradient: Among them, ▽T vertical For the vertical gradient; ▽T horizontal For the lateral gradient; ΔT vertical ΔT represents the temperature difference between different snow depths along the longitudinal direction in the multidimensional sensor array. horizontal Δz represents the temperature difference between different snow depths in the lateral direction of the multi-dimensional sensor array; Δx represents the distance between the multi-dimensional sensors in the longitudinal direction of the multi-dimensional sensor array; Δx represents the distance between the multi-dimensional sensors in the lateral direction of the multi-dimensional sensor array. Disaster risk is determined based on the snow layer status data; After determining the disaster risk based on the snow layer state data, the following steps are included: Determine whether the disaster risk has reached a preset risk threshold; If the disaster risk reaches the preset risk threshold, historical snow layer status data is obtained. The historical snow layer status data is the snow layer status data obtained from historical monitoring. The historical snow layer status data reflects the changes in the snow layer status. Predict avalanche prediction time based on the historical snow cover state data; Execute a warning operation based on the predicted avalanche time; The determination of disaster risk based on the snow layer state data includes: Determine the comprehensive risk index corresponding to the snow layer state data; Obtain a risk classification threshold, and determine the disaster risk corresponding to the comprehensive risk index based on the risk classification threshold; the risk classification threshold can be updated based on actual accumulated historical data.

2. The snow layer stability monitoring method as described in claim 1, characterized in that, The snow layer state data obtained based on the monitoring data includes: Time-aligned data from each monitoring data point is used to obtain time-based monitoring data. The spatial monitoring data is obtained by mapping each monitoring data into a preset data space. The snow layer state data is determined based on the time monitoring data and the spatial monitoring data.

3. The snow layer stability monitoring method as described in claim 2, characterized in that, The snow layer state data includes risk probability, and determining the snow layer state data based on the time monitoring data and the spatial monitoring data includes: Obtain a hybrid model, wherein the hybrid model includes a convolutional neural network layer, a long short-term memory network layer, and a fully connected layer; The spatial monitoring data is input into the convolutional neural network layer to obtain spatial features; The time monitoring data is input into the long short-term memory network layer to obtain time features; The spatial features and the temporal features are combined to obtain the fused features; The fused features are input into the fully connected layer to obtain the risk probability.

4. The snow layer stability monitoring method as described in claim 1, characterized in that, The snow layer state data includes risk probability and environmental parameters; determining the comprehensive risk index corresponding to the snow layer state data includes: Obtain the risk probability and the weights corresponding to the environmental parameters; The comprehensive risk index is obtained by weighting the risk probability and the environmental parameters based on the weights.

5. The snow layer stability monitoring method as described in claim 1, characterized in that, The acquisition of monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array includes: Obtain real-time disaster risks; Determine the target acquisition frequency corresponding to the real-time disaster risk, wherein the target acquisition frequency is positively correlated with the real-time disaster risk; The monitoring data sent by each multi-dimensional sensor in the multi-dimensional sensor array is acquired at the target acquisition frequency.

6. The snow layer stability monitoring method as described in claim 1, characterized in that, After determining the disaster risk based on the snow layer state data, the following steps are included: Obtain the terrain model corresponding to the multi-dimensional sensor array; The snow layer state data and the disaster risk are mapped onto the terrain model for display.

7. A snow layer stability monitoring system, characterized in that, The snow layer stability monitoring system includes a multi-dimensional sensor array and a processing module; the multi-dimensional sensor array includes multiple multi-dimensional sensors, and each multi-dimensional sensor is communicatively connected to the processing module; wherein: The multidimensional sensor is used to monitor the snow layer parameters at its location to obtain monitoring data, and then sends the monitoring data to the processing module. The processing module is used to receive the monitoring data sent by each of the multi-dimensional sensors, obtain snow layer state data based on the monitoring data, and determine disaster risk based on the snow layer state data. The snow layer state data includes a temperature gradient, which includes a longitudinal gradient and a lateral gradient. The longitudinal gradient is: The lateral gradient: Among them, ▽T vertical For the vertical gradient; ▽T horizontal For the lateral gradient; ΔT vertical ΔT represents the temperature difference between different snow depths along the longitudinal direction in the multidimensional sensor array. horizontal Δz represents the temperature difference between different snow depths in the lateral direction of the multi-dimensional sensor array; Δx represents the distance between the multi-dimensional sensors in the longitudinal direction of the multi-dimensional sensor array; Δx represents the distance between the multi-dimensional sensors in the lateral direction of the multi-dimensional sensor array. After determining the disaster risk based on the snow layer state data, the following steps are included: Determine whether the disaster risk has reached a preset risk threshold; If the disaster risk reaches the preset risk threshold, historical snow layer status data is obtained. The historical snow layer status data is the snow layer status data obtained from historical monitoring. The historical snow layer status data reflects the changes in the snow layer status. Predict avalanche prediction time based on the historical snow cover state data; Execute a warning operation based on the predicted avalanche time; The determination of disaster risk based on the snow layer state data includes: Determine the comprehensive risk index corresponding to the snow layer state data; Obtain a risk classification threshold, and determine the disaster risk corresponding to the comprehensive risk index based on the risk classification threshold; the risk classification threshold can be updated based on actual accumulated historical data; The multidimensional sensor includes a longitudinal anchor bolt, on which multiple temperature sensors are installed at preset intervals.

Images