Full-scale earthen site top edge sliding accumulated snow sudden melting monitoring device and method
By building an experimental platform in an open area with climatic conditions consistent with the site, and using ancient ramming technology and a variety of sensor monitoring equipment, the problem of real simulation and intelligent response of the sudden melting of snow at the earthen site was solved, and high-precision identification of slip precursors and active intervention were achieved.
Patent Information
- Application Number
- CN202510784922.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-12
- Publication Date
- 2025-09-23
AI Technical Summary
Existing technologies make it difficult to accurately simulate the sudden melting of snow at earthen sites in real environments, and lack high-precision dynamic monitoring and intelligent response capabilities of multiple physical fields, making it impossible to effectively identify slip precursors and conduct active intervention.
An experimental platform was built in an open area that matches the climatic conditions of the site. The experimental walls were constructed using ancient rammed earth techniques, and a variety of sensors and monitoring equipment were deployed. Data integration and analysis were carried out through a cloud platform to achieve synchronous monitoring and intelligent response of multiple physical fields.
It has achieved realistic simulation and high-precision monitoring of the sudden melting process of snow at earthen ruins, can identify slip precursors and take active intervention, and improve the accuracy of disease identification and response efficiency.
Smart Images
Figure CN120685889A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of cultural heritage protection, and in particular to a device and method for monitoring the sudden melting of snow at the top edge of a Zuchitu site. Background Art
[0002] Cultural relics sites primarily constructed from rammed earth are widely distributed in Northwest my country. Many of these sites are located in arid, cold climates and exhibit significant environmental sensitivity. Statistics show that over 70% of these sites are located in permafrost zones. Seasonal snow cover, diurnal temperature swings, and frequent freeze-thaw cycles are the primary environmental factors contributing to wall structural degradation. In particular, with heavy winter snow cover and rapid spring warming, walls experiencing prolonged freezing are susceptible to sudden snow melt, triggering a series of structural failures. Research has shown that snow cover significantly alters the thermal and moisture boundary conditions of the soil. Its sublimation, compaction, and sudden thawing processes influence the temperature gradient, moisture migration, and strength evolution of the surface layer of the earthen sites, often leading to destructive behaviors such as frost heave, loosening, water saturation, crack expansion, and strength degradation in rammed earth structures. The top edge of the wall, due to its fragile structure and drastic temperature and humidity fluctuations, is particularly susceptible to catastrophic failures characterized by "slip and collapse," known as top edge slip. This disease is mostly caused by the concentrated melting of snow on the top. The meltwater penetrates into the weak surface of the wall along the freeze-thaw interface or the rammed layer interface, causing local softening. Once the stress exceeds the structural strength limit, it triggers the overall sliding of the top rammed layer along the potential slip surface. In severe cases, it can lead to wall loss, bedding separation, and even chain collapse, posing a direct threat to the safety of the site.
[0003] Current research on freeze-thaw damage at earthen sites focuses on two main categories. The first involves laboratory-based freeze-thaw simulations, which employ dry-wet freeze-thaw cycles and coupled thermal-hydraulic-mechanical experiments under controlled temperature and humidity conditions, supplemented by microstructural analysis and mechanical testing, to reveal the frost heave deformation, crack evolution, and structural degradation of rammed earth. These methods offer advantages in quantitative analysis and variable control, making them suitable for extracting single-factor patterns and evaluating material properties. However, these methods are limited by their small experimental scale, simplified environmental boundaries, and idealized external force loading, making them incapable of accurately reflecting the multi-field interactions and dynamic responses of earthen sites under complex climatic conditions. The second category involves in-situ monitoring techniques based on field conditions. These utilize non-contact methods such as laser scanning, 3D point cloud modeling, thermal infrared imaging, and drone photography to characterize wall surface changes. Furthermore, these techniques utilize embedded sensors for temperature, humidity, conductivity, stress, and strain to construct a multi-layered monitoring network focused on the site itself. These methods offer the advantages of a long monitoring period, diverse data sources, and a lack of structural disturbance, effectively capturing the responses of site walls under natural conditions. In recent years, researchers have introduced algorithms such as image recognition, time series comparison, and deep learning to automatically identify the evolutionary characteristics of wall defects. However, limited by data integration capabilities, response speed, and the adaptability of monitoring platforms, these methods still face bottlenecks in multi-source information linkage, emergency event capture, and intelligent response control. In particular, they lack sufficient recognition accuracy and response efficiency for rapid loading events such as sudden snowmelt.
[0004] Against this backdrop, researchers have attempted to construct full-scale experimental walls around the site, recreating a true-to-life rammed earth structure using a replica of ancient rammed earth construction techniques. These walls, along with sensor arrays and remote sensing imaging equipment, are designed to capture freeze-thaw response processes in a multi-physics environment. Preliminary results indicate that these experimental walls can effectively replicate the temperature gradients, moisture migration pathways, and structural stress evolution of rammed earth walls under realistic climate conditions, providing crucial data for analyzing the mechanisms of freeze-thaw damage. However, these studies still have several limitations: First, they lack the ability to simulate realistic snow conditions, particularly extreme boundary conditions such as sudden snow melt and rapid hydraulic infiltration. Second, monitoring systems primarily focus on thermal and hygrometric parameters, lacking comprehensive integration and spatiotemporal coupling of physical quantities such as temperature, moisture content, salinity, conductivity, and stress and strain. Third, the response mechanism remains passive, lacking active identification and feedback mechanisms, hindering automated identification and intervention of slip precursors.
[0005] Therefore, there is an urgent need to develop an experimental technology platform capable of reproducing real-world environments, simultaneously monitoring multiple physical fields, and intelligent response control. This platform should be able to comprehensively capture the coupled water-salt-heat-mechanical response of earthen site walls throughout the entire process of snow accumulation, frost heave, thaw, and slip under realistic or quasi-realistic climatic conditions, thereby revealing the inducing mechanisms, triggering pathways, and critical conditions of slip. This technology has important theoretical and practical significance for establishing slip response models, supporting disaster prediction and early warning, and improving site protection intervention strategies. Summary of the Invention
[0006] The present invention aims to provide a device and method for monitoring the sudden melting of snow and sliding along the top edge of a full-scale earthen ruin to solve the problems raised in the above background technology. Specific technical problems include:
[0007] How to realistically simulate the environmental and structural conditions of freeze-thaw slippage at the site to address the issues of laboratory environmental distortion and the limited working conditions of full-scale experimental walls;
[0008] How to achieve high-precision dynamic monitoring and coupled analysis of multi-physics fields to solve the problems of data isolation and weak spatiotemporal correlation;
[0009] How to intelligently identify slip precursors and proactively intervene to address the problems of delayed warnings and insufficient system reliability.
[0010] To achieve the above-mentioned object, the present invention provides a method for monitoring the sudden melting of snow at the top edge of a full-scale earthen site, comprising the following steps:
[0011] S1. Construct the experimental platform in an open area with the same climatic characteristics as the site. The experimental platform should be located in a representative, open area immediately adjacent to the site, ensuring that the terrain is flat and free of tall buildings or vegetation. This open environment can avoid interference with natural factors such as wind speed and direction, ensuring that the observed meteorological conditions are authentic and representative. Furthermore, the orientation of the experimental wall should be consistent with the original site wall (e.g., maintaining the same south or north orientation) to ensure that influencing factors such as sunlight and prevailing wind direction match, thereby ensuring that the environmental stresses exposed during the experiment are as similar as possible to the actual site conditions.
[0012] The experimental platform has reliable power supply and data transmission conditions to support the continuous operation of the monitoring system and real-time data upload. If there is mains power access on site, voltage stabilization and backup power supplies should be configured to prevent power outages. If mains power is unavailable, off-grid power supply solutions such as solar panels and battery packs or small wind power generation should be used, and backup power supplies should be set up to ensure normal operation of the equipment in extreme weather such as severe cold and heavy snow. For data transmission, wired broadband or fiber optic networks should be used first. 4G / 5G cellular networks or satellite communication devices should be deployed under field conditions to achieve high-speed connectivity between monitoring data and the cloud platform.
[0013] The climatic conditions of the selected experimental location should be highly consistent with those of the site, including the average annual temperature, diurnal temperature difference, sunshine intensity, precipitation / snowfall, and wind speed and direction, so that the environmental effects experienced by the experimental wall (such as dry-wet cycles, freeze-thaw cycles) are the same as those of the site, ensuring that the experimental results have reference value for the site. Before selecting the site, the historical meteorological data of the area should be investigated to confirm that the lowest winter temperature is equivalent to that of the site, so as to simulate the frost heave stress and snowmelt scouring that the site walls may suffer during the freezing period. In addition, the altitude and terrain of the experimental site should be as close as possible to the site to avoid microclimate deviations due to differences in air pressure or terrain.
[0014] S2. Reconstruct the slope morphology before the slide based on the original three-dimensional model of the site. Select soil with physical properties similar to the rammed earth at the site. Determine the optimal moisture content and dry density through compaction experiments. Use traditional ramming techniques and a formwork system to construct the experimental wall in layers. The thickness of the rammed layers is determined based on archaeological data.
[0015] The experimental wall's shape was determined based on an on-site survey of the damaged wall at the site, clarifying the characteristics and extent of surface or internal slippage damage. A detailed 3D model of the damaged wall segment was obtained using 3D laser scanning or multi-view image reconstruction technology. The original slope morphology before the wall's failure was then inferred through geological analysis. By comparing geometric deformation before and after the slip, parameters such as the inclination and curvature of the wall's front slope were determined, and a cross-sectional model of the wall close to its pre-slip state was reconstructed as the target shape for the experimental reconstruction. This 3D modeling-based morphological restoration method ensures that the experimental wall's shape is consistent with its true pre-damage morphology, facilitating accurate simulation of stress distribution and failure mechanisms.
[0016] By conducting geotechnical tests on site soil samples, such as particle grading, clay content, and liquid and plastic limits, the soil type (e.g., silty clay, sandy clay, etc.) is determined. Based on this, soil with similar composition is selected or mixed nearby as experimental soil. The sampled soil undergoes a standard compaction test in the laboratory to determine its optimal moisture content and maximum dry density. During the compaction test, by controlling the compaction work at different moisture contents, a compaction curve is drawn to find the moisture content at which the dry density reaches its peak. For example, if the experimental soil reaches its maximum dry density at a moisture content of approximately 16% (hypothetical value), the soil moisture content should be adjusted to this optimal value during on-site construction and compaction should be performed. At the same time, the compaction curve should be recorded for on-site verification of the compaction effect to ensure that the density and mechanical strength of the experimental wall are consistent with the design requirements.
[0017] The experimental wall was rammed layer by layer using the ancient traditional ramming technology to reproduce the historical construction method. During construction, wooden formwork (hoops) were used to limit the cross-sectional dimensions of the wall, and the rammer was manually controlled to tamp the wall layer by layer. Key process parameters such as soil thickness, number of tamping times, and hammer lifting height were optimized through preliminary small-scale trial pit tamping experiments. For example, loose soil layers of different thicknesses of 5 cm and 10 cm can be tried first, and each layer is tamped 20 times and the density is measured to determine the thickness of a single layer of soil. At the same time, the weight and drop height of the rammer are adjusted so that the impact energy of each hit is close to the strength of the historical artificial rammed earth. Through multiple rounds of field experiments, the optimal combination of parameters such as the thickness of each layer of soil (such as 8 to 10 cm), the number of tamping times (such as 4 times per layer, each time containing several tamping times), and the drop height of the rammer (such as 50 cm) are finally determined to take into account both construction efficiency and tamping quality. The determined parameters must be strictly followed during the tamping process, and each layer must be tamped until it is dense and not loose, so as to obtain a uniform density distribution and mechanical properties.
[0018] The thickness of each layer of rammed earth in the experimental wall and the tamping process should refer to the original wall structure information of the site obtained through archaeological excavations. For example, if archaeological records show that the ancient wall was composed of several rammed earth layers approximately 10 cm thick, with a small amount of grass-mixed mud interposed between each layer for reinforcement, then similar layer thickness and interlayer materials should be followed during the experiment; in terms of the tamping sequence, the historical construction process should be restored: including the cyclical process of taking soil, spreading, wetting, tamping until dense, and then laying the next layer; the shape and weight of the rammer used in construction should be imitated based on unearthed cultural relics or historical records to ensure that the tamping method and the energy transferred to the soil are consistent with ancient construction methods; through the above measures, the constructed experimental wall is close to the original site wall in terms of layer structure, material density and overall appearance, making the data obtained from subsequent monitoring more authentic and comparable.
[0019] S3. Integrated multifunctional sensors and stress and strain gauge sensors are deployed in the rammed layer inside the experimental wall, distributed in different sections to avoid interference and form complementarity; external monitoring includes a weather station, time-lapse photography system, drone airfield system and multispectral drone;
[0020] Various types of sensing and monitoring equipment were deployed inside and outside the experimental wall to form a multi-field coupled monitoring system to simultaneously acquire environmental parameters and wall response data. This integrated monitoring revealed the interactions between multiple physical fields, such as temperature, humidity, water and salt migration, and stress and strain, enabling in-depth analysis of the occurrence and development mechanisms of slip failure.
[0021] Temperature sensors, moisture content sensors, conductivity sensors, and strain gauge sensors are buried inside the experimental wall to obtain real-time data on changes in the wall's internal environment and mechanical state. The sensor locations must cover key areas where damage occurs (such as near potential slip surfaces) and cross-sectional locations at different heights and depths. For example, a group of temperature, humidity, and strain sensors can be buried every 10 cm along the wall's height and every 5 cm in its depth. These sensors can be placed at the wall's foot, middle, and top to monitor temperature gradients, moisture content distribution, and differences in stress states at various locations. The installation of each sensor should avoid mutual interference: for example, the strain gauge should be kept at an appropriate distance from the humidity probe to avoid interference during installation. Hardware affects the local strength of the soil structure. Conductivity probes are placed away from metal strain gauges to reduce electromagnetic noise coupling. Furthermore, diverse sensor types are used to achieve data complementarity. Temperature and humidity sensors reveal environmental factors, while stress and strain gauges reflect structural responses. The combination of these two fully describes the changing state of the wall's interior. All sensors must be calibrated before installation to ensure measurement accuracy meets requirements (e.g., temperature accuracy of ±0.1°C and strain accuracy of 10⁻⁶). The sampling frequency for routine monitoring is set to meet the needs of slowly changing processes (e.g., every 5 minutes) to obtain sufficient data resolution. Furthermore, room is reserved for increasing the sampling frequency to enable high-frequency monitoring when anomalies occur.
[0022] The weather station is used to collect meteorological elements such as wind speed, wind direction, temperature, humidity, and radiation in the experimental area: micro-weather stations and image recording systems are built around the experimental site to monitor the impact of the external environment on the wall; the weather station should be placed in an open area several times the height of the experimental wall, away from buildings and the experimental wall itself, and equipped with a cup anemometer, wind vane, temperature and humidity probe, rain and snow gauge, and solar radiation sensor to record meteorological data of the microenvironment of the experimental wall around the clock; to avoid interference, meteorological sensors should meet the standard installation height (such as anemometer 10 meters from the ground) and be calibrated regularly.
[0023] The time-lapse photography system is used to periodically record the evolution of the wall surface state. In terms of image monitoring, a time-lapse camera is set up at an appropriate distance and fixedly aimed at the key parts of the experimental wall (such as slip-sensitive areas), and one frame of image is captured every 5 minutes at 4K resolution. The time-lapse image can be used to observe the surface changes of the wall (such as cracks, bulges or signs of collapse) with the naked eye, and the images of adjacent time series are compared through computer vision algorithms to automatically identify anomalies. The camera needs to be equipped with an all-weather cover and cleaning device (dust-proof and snow-proof), and be connected to the network to transmit images to cloud storage in real time. In addition, a safety inspection route is planned around the experimental wall to facilitate operators to regularly use handheld devices to conduct on-site inspections and compare the accuracy of sensor readings.
[0024] The drone field system is used to deploy automated drones; and multispectral drones for high-frequency infrared and multispectral image acquisition; the drone field system regularly conducts aerial monitoring of the experimental wall; a drone field can be set up in the experimental site, with a built-in quadcopter drone equipped with visible light, infrared and multispectral cameras; according to a preset plan (such as once a day at noon and once in the evening) or manual remote command, the drone automatically takes off and flies around the wall to obtain high-definition images and infrared thermal imaging data from all angles of the wall facade; infrared imaging can reveal the temperature field distribution on the wall surface, which is used to determine which areas may have water anomalies (wetness) The drone can also be used to monitor the surface temperature of certain areas (such as areas with low surface temperatures due to evaporative cooling) or to monitor the heat dissipation uniformity of the wall at night; multispectral imaging is used to identify material differences, weathered areas, and early biological attachment on the wall surface; the data from drone inspections is transmitted back in real time via wireless networks and spliced and analyzed by software on ground stations or cloud platforms to generate a three-dimensional point cloud model of the wall and a temperature distribution map to intuitively present the spatial changes in the health status of the wall; considering the winter climate and battery performance, drones should be low-temperature resistant models, automatically check the power and wind conditions before flight, and suspend flights for safety reasons when wind speed exceeds the limit or there is heavy snowfall.
[0025] S4. Remotely collect multi-source monitoring data through the cloud data collection and processing platform and upload it to the cloud;
[0026] S5. The cloud-based data acquisition and processing platform couples and analyzes multi-source monitoring data, combining image recognition, stress response, and temperature and humidity change characteristics to identify slip precursors and response thresholds, and construct a slip triggering model and evolution path identification mechanism.
[0027] All sensor and image data are aggregated through the edge computing gateway and uploaded to the cloud data collection and processing platform in real time. The cloud data collection and processing platform has a built-in automatic difference comparison algorithm, which determines whether there is an abnormal state according to the manually preset threshold. The image processing module determines whether there is an image anomaly based on indicators such as pixel change rate and boundary deformation.
[0028] When data or image anomalies are identified, a red path response operation will be executed. The cloud-based data acquisition and processing platform will feed back the identified abnormal event code (including sensor ID, anomaly type, location, and value) to the remote control terminal through the data scheduling module. The device control module will then issue commands, including automatically increasing the sensor sampling frequency in the abnormal area (for example, to 1 minute / time), shortening the time interval for time-lapse photography, and pushing warning information to researchers. If the image anomaly is manually confirmed, the operator can remotely issue a drone takeoff command to perform a multispectral cruise mission in the target area and obtain high-resolution images for further identification.
[0029] The cloud-based data acquisition and processing platform integrates multivariate fusion and dynamic response analysis of various physical field data (temperature, humidity, conductivity, stress and strain) to identify coupling patterns and response paths before slip occurs. It supports the use of deep learning models (such as convolutional neural network time series analysis) to identify image anomalies and jointly model with sensor data to extract key feature point change curves and establish slip cause-response relationships.
[0030] During the three critical periods of early freezing, extreme freezing and early thawing, the sampling frequency, image processing frequency and analysis priority are automatically increased to the highest level. The meteorological station data and wall response are analyzed in time series pairing, and the freeze-thaw disaster mechanism evolution curve is drawn to provide direct judgment criteria for wall strength decline and slip triggering.
[0031] The second object of the present invention is to provide a device for monitoring the sudden melting of snow and sliding on the top edge of a full-scale earthen site, which includes an environmental simulation platform construction module, a slope morphology reconstruction and tamping experiment module, a multi-dimensional monitoring network deployment module, a remote monitoring module and a model construction module.
[0032] The environmental simulation platform construction module constructs an experimental platform in an open area with the same climatic characteristics as the site;
[0033] The slope morphology reconstruction and ramming experiment module reconstructs the pre-slip slope morphology based on a 3D model of the original site. Soil with physical properties similar to the rammed earth at the site is selected, and compaction experiments are conducted to determine the optimal moisture content and dry density. The experimental walls are then rammed in layers using traditional ancient ramming techniques and a formwork system. The thickness of the rammed layers is determined based on archaeological data.
[0034] The multi-dimensional monitoring network deployment module deploys integrated multi-function sensors and stress and strain gauge sensors in the rammed layer inside the experimental wall, distributed in different cross-sections to avoid interference and form complementarity; external monitoring includes a weather station, time-lapse photography system, drone airfield system and multispectral drones;
[0035] The remote monitoring module remotely collects multi-source monitoring data through the cloud data collection and processing platform and uploads it to the cloud;
[0036] The cloud-based data acquisition and processing platform of the model building module couples and analyzes multi-source monitoring data, combines image recognition, stress response, and temperature and humidity change characteristics, identifies slip precursors and response thresholds, and constructs a slip triggering model and evolution path identification mechanism.
[0037] Compared with the prior art, the present invention has the following beneficial effects:
[0038] The simulation environment of this invention is realistic and reliable. By constructing a full-scale experimental wall in a field area that matches the climatic conditions of the original site, strictly controlling the site selection conditions, and ensuring that the environmental factors such as snowfall, sunshine, wind speed and direction to which the wall is subjected are consistent with those of the actual site, the multi-field coupling process under the sudden snowmelt environment is realistically reproduced, providing a highly reliable simulation environment and experimental data support for the study of the occurrence mechanism of slip diseases.
[0039] The multi-physics field monitoring system of the present invention has a high degree of integration and establishes a multi-field coupling monitoring system that coordinates the interior and exterior of the wall. The internal system includes multiple sensors such as temperature, humidity, conductivity, stress and strain, and the external system is equipped with equipment such as a weather station, time-lapse photography, and multispectral drones. The scientific and reasonable layout ensures the integrity, continuity, and spatial coverage accuracy of data collection, and can fully reflect the development process of freeze-thaw damage and slip diseases.
[0040] The present invention has a strong ability to capture and analyze the entire process of slippage failure. For the special type of disease, slippage at the top edge of earthen ruins, the system can continuously record the entire process from potential, development to failure by accurately monitoring key parameters such as local stress concentration, moisture migration and surface deformation of the wall. It can also combine image recognition and stress response linkage analysis to identify slip precursors, failure locations and response thresholds, thereby improving the accuracy and foresight of disease identification.
[0041] The system of the present invention is highly intelligent and has significant engineering application value. The data platform of the present invention has functions such as remote call, abnormal warning, and multi-source linkage. It supports dynamic monitoring throughout the year, especially strengthens high-frequency sampling and in-depth analysis of freezing cycles, and has dynamic adaptation and intelligent response capabilities.
[0042] The method of the present invention is applicable to different types of earthen sites and freeze-thaw environmental conditions, has good replicability and regional adaptability, can be widely used in cold-region cultural heritage protection, environmental disaster assessment and structural intervention optimization, and has good promotion value and engineering application prospects. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 It is an experimental flow chart of the device of the present invention;
[0044] Figure 2 This is a layout diagram of the experimental wall monitoring system of the device of the present invention;
[0045] Figure 3 This is a flow chart of the slip disease response linkage of the device of the present invention;
[0046] Figure 4 Schematic diagram of the module of the device of the present invention.
[0047] In the figure: 1. Experimental wall; 2. Rammed layer; 3. Integrated multifunctional sensor; 4. Stress and strain gauge sensor; 5. Weather station; 6. Time-lapse photography system; 7. UAV airfield system; 8. Multispectral UAV. DETAILED DESCRIPTION
[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.
[0049] Next, see Figure 1 One of the purposes of this embodiment is to provide a method for monitoring the sudden melting of snow at the top edge of a full-scale earthen site, including the following method steps:
[0050] S1. First, during the experimental preparation phase, an open area with a high degree of similarity to the original site environment (ensuring no interference with wind direction, snow accumulation, temperature differences, and terrain) should be selected as the test site based on the climate and disease distribution characteristics of the site area. This area must have stable power supply and network conditions to provide a foundation for subsequent remote monitoring and response control.
[0051] The design of S2, Experimental Wall 1, was based on a 3D laser scanning model of the original site and the results of a slip hazard investigation. The slope morphology and inclination before the slip occurred were reconstructed to ensure that the simulated wall had realistic physical geometric boundaries. The experimental soil was optimized for moisture content and dry density through particle grading analysis and compaction testing. During construction, an antique rammer and formwork system was used, and the soil was compacted layer by layer according to the historical rammed earth thickness (e.g., 10 cm), forming a layered rammed earth structure consistent with the site.
[0052] S3. During the tamping process of the experimental wall 1, integrated multifunctional sensors 3, including temperature, humidity (water content), conductivity and strain sensors, should be embedded simultaneously. The layout of the integrated multifunctional sensors 3 is based on the axis of "potential slip zone - wall center - top edge", forming a densely distributed observation profile in the vertical and horizontal directions. The integrated multifunctional sensors 3 must be calibrated for accuracy before, during and after tamping to ensure that the temperature accuracy is ±0.1℃ and the humidity is 10℃. -6 Strain resolution and supports a minimum sampling frequency of 5 minutes / time;
[0053] See also Figure 2 Among them: integrated multifunctional sensors 3 and stress strain gauge sensors 4 are arranged in the rammed layer 2 inside the experimental wall 1, distributed in different sections to avoid interference and form complementarity; external monitoring includes a weather station 5, a time-lapse photography system 6, a drone airport system 7 and a multispectral drone 8.
[0054] S4. Remotely collect multi-source monitoring data through the cloud data collection and processing platform and upload it to the cloud;
[0055] S5. A high-precision monitoring system is deployed outside the experimental wall 1, including a weather station 5 (collecting wind speed, wind direction, radiation, temperature, etc.) and a time-lapse photography system 6. The time-lapse photography system 6 uses a 4K image sensor, which is fixed on the main slope of the experimental wall 1 and has a set shooting cycle (e.g., every 5 minutes). The images are uploaded to the cloud for time series comparison. The multispectral drone 8 is dispatched and managed by the drone airport system 7 and performs wall patrol missions at regular intervals to capture thermal field anomalies and material changes of the wall.
[0056] Steps S1 to S5 belong to the deployment of the monitoring system, and the work tasks mainly include the entire process of experimental platform deployment, experimental wall 1 construction, and multi-source monitoring system deployment.
[0057] See also Figure 3 ,During the S5 process, the cloud data acquisition and ,processing platform couples and analyzes multi-source monitoring data, ,combining image recognition, stress response, and temperature and humidity change characteristics, ,identifies slip precursors and response thresholds, and constructs a slip triggering model and ,evolution path identification mechanism;
[0058] All sensor and image data are aggregated through the edge computing gateway and uploaded to the cloud data collection and processing platform in real time. The platform has a built-in automatic difference comparison algorithm, which determines whether there is an abnormal state based on manually preset thresholds (such as temperature change > 0.5°C / 5 minutes, moisture content increase > 2%). It also uses indicators such as pixel change rate and boundary deformation to determine whether there are image anomalies (such as snow cover area change exceeding 0.01mm2, boundary change exceeding 15 pixels). Among them:
[0059] Abnormal identification is based on two mechanisms: sensor data abnormality and image timing abnormality. Among them, sensor data abnormality refers to the difference thresholds of monitoring quantities such as temperature, strain, and moisture content preset by staff in the system (such as temperature difference ±0.5℃ / 5 minutes, strain difference 100με / 1 minute, etc.), and the data difference between two adjacent sampling periods is calculated. If it exceeds the threshold, it is judged as abnormal; image timing abnormality refers to the pixel difference analysis and boundary contour comparison after the time-lapse photography image is uploaded. If the image change index exceeds the preset range (such as the snow cover area changes by more than 0.01mm 2 , the boundary changes by more than 15 pixels), an image anomaly is triggered.
[0060] When data or image anomalies are identified, a "red path" response operation will be executed. The cloud platform will feedback the identified abnormal event code (including sensor ID, anomaly type, location and value) to the remote control terminal through the data scheduling module. The device control module will complete the command issuance, including automatically increasing the sensor sampling frequency in the abnormal area (for example, to 1 minute / time), shortening the time interval for time-lapse photography, and sending warning information to researchers. If the image anomaly is manually confirmed, the operator can remotely issue a drone takeoff command to perform a multispectral patrol mission in the target area to obtain high-resolution imagery for further identification. Among them:
[0061] When any anomaly is confirmed, the "red path" linkage response is automatically executed. The red path linkage response means that when the above data anomaly or image anomaly occurs, the system automatically executes the multi-source response strategy according to the personnel preset program, including:
[0062] Start the sensor frequency increase mechanism (such as increasing from 5 minutes / time to 1 minute / time); start the image enhancement monitoring mode (such as increasing from 5 minutes / time to 1 minute / time); push early warning prompts to scientific research management personnel through email, text messages, platform APP, etc.; if the image abnormality is manually confirmed, the staff can remotely start the drone patrol mission in the field to achieve high-frequency, multi-angle infrared and multi-spectral image acquisition in key areas.
[0063] The cloud-based data collection and processing platform module is responsible for the unified scheduling of multi-source sensor data and image data, unified modeling based on "timestamp + spatial coordinate + type", performing operations such as sliding average, trend analysis, and mutation identification, and comparing with the platform's built-in rule base to determine whether there are trend anomalies;
[0064] When the cloud-based data collection and processing platform determines that the system has entered a "potential slippage risk" or "entered a high-sensitivity phase," it will send instructions to the data scheduling and analysis module via a standard API interface. This module includes a device control module (which performs actions such as sensor frequency upscaling, drone wake-up, and image enhancement), a user interaction module (which pushes events to the management terminal), and a policy library management module (which supports manual intervention, rule modification, and policy coverage updates).
[0065] Based on the multi-source data graphic recognition, anomaly comparison and level judgment output by the data scheduling and analysis module, when the monitoring indicators continue to deviate from the safe range and exceed the preset warning threshold (such as temperature change exceeding 0.5℃ / 5 minutes), the system will issue an intelligent warning to the remote control terminal and feedback the "sensor ID, anomaly type, data difference, time, spatial location, recommended response strategy level" and other content to the cloud data collection and processing platform.
[0066] The cloud-based data acquisition and processing platform integrates multivariate fusion and dynamic response analysis of various physical field data (temperature, humidity, conductivity, and stress and strain) to identify the coupling mode and response path before slip occurs. The system supports the use of deep learning models (such as convolutional neural network time series analysis) to identify image anomalies and jointly model with sensor data to extract the change curve of key feature points and establish the slip cause-response relationship.
[0067] During the three critical periods of early freezing, extreme freezing and early thawing, the sampling frequency, image processing frequency and analysis priority are automatically increased to the highest level. The meteorological station data and wall response are analyzed in time series pairing, and the freeze-thaw disaster mechanism evolution curve is drawn to provide direct judgment criteria for wall strength decline and slip triggering.
[0068] From the above description, it can be seen that the method for monitoring the sudden melting of snow at the top edge of a 100-foot-high earthen site provided in this embodiment has the following technical effects:
[0069] By constructing a full-scale experimental wall that is both authentic, stable and controllable, the entire process of the top edge sliding of the earthen site under the sudden melting of snow is reproduced. In terms of technical path, it breaks through the limitations of traditional experimental methods in scale similarity, loading field coupling and data acquisition integrity. It can be widely used in cold region site protection research, disaster warning model construction and key environmental factor identification, and has significant theoretical value and engineering application prospects.
[0070] See also Figure 4 The second purpose of this embodiment is to provide a device for monitoring the sudden melting of snow and sliding on the top edge of a full-scale earthen site, which includes an environmental simulation platform construction module, a slope morphology reconstruction and tamping experiment module, a multi-dimensional monitoring network deployment module, a remote monitoring module and a model construction module.
[0071] The environmental simulation platform construction module constructs an experimental platform in an open area with the same climatic characteristics as the site;
[0072] The slope morphology reconstruction and ramming experiment module reconstructs the pre-slip slope morphology based on the original 3D model of the site. Soil with physical properties similar to the rammed earth at the site is selected, and compaction experiments are conducted to determine the optimal moisture content and dry density. The experimental wall 1 is rammed layer by layer using traditional ancient ramming techniques and a formwork system. The thickness of the rammed layers is determined based on archaeological data.
[0073] The multi-dimensional monitoring network deployment module deploys integrated multi-function sensors 3 and stress and strain gauge sensors 4 in the rammed layer 2 inside the experimental wall 1, distributed in different cross-sections to avoid interference and form complementarity. External monitoring includes a weather station 5, a time-lapse photography system 6, a drone airfield system 7, and a multispectral drone 8.
[0074] The remote monitoring module remotely collects multi-source monitoring data through the cloud data collection and processing platform and uploads it to the cloud;
[0075] The cloud-based data acquisition and processing platform of the model building module couples and analyzes multi-source monitoring data, combines image recognition, stress response, and temperature and humidity change characteristics, identifies slip precursors and response thresholds, and constructs a slip triggering model and evolution path identification mechanism.
[0076] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for monitoring the sudden melting of snow and sliding at the top edge of a full-scale earthen site, characterized in that: The method comprises the following steps: S1. Build an experimental platform in an open area with the same climatic characteristics as the site; S2. Reconstruct the slope morphology before the slide based on the original three-dimensional model of the site, select soil with physical properties similar to the rammed earth of the site, determine the optimal moisture content and dry density through compaction experiments, and use the ancient traditional ramming technology and formwork system to ram the experimental wall in layers (1); S3. An integrated multifunctional sensor (3) and a stress and strain gauge sensor (4) are arranged in the rammed layer (2) inside the experimental wall (1), distributed in different sections to avoid interference and form complementarity; external monitoring includes a weather station (5), a time-lapse photography system (6), a drone airport system (7) and a multispectral drone (8); S4. Remotely collect multi-source monitoring data through the cloud data collection and processing platform and upload it to the cloud; S5. The cloud-based data collection and processing platform couples and analyzes multi-source monitoring data, combines image recognition, stress response, and temperature and humidity change characteristics, identifies slip precursors and response thresholds, and constructs a slip triggering model and evolution path identification mechanism.
2. The method for monitoring the sudden melting of snow and sliding at the top edge of a 90-foot-high earthen site according to claim 1 is characterized in that: The tamping of the experimental wall (1) adopts the ancient traditional tamping technology, and the construction parameters include the thickness of the soil, the number of tamping times and the height of the hammer, wherein the height of the hammer is determined after the on-site preliminary test, and the construction sequence is restored according to the historical tamping process.
3. The method for monitoring sudden snowmelt caused by top edge sliding of a 90-foot-high earthen site according to claim 1 is characterized in that: The data acquisition frequency of the integrated multifunctional sensor (3) is at least 5 minutes per time, wherein the temperature measurement accuracy of the integrated multifunctional sensor (3) meets ±0.1°C; and the stress and strain measurement accuracy of the stress and strain gauge sensor (4) meets the requirements for freeze-thaw disease interpretation.
4. The method for monitoring sudden snowmelt caused by top edge sliding of a 90-foot earthen site according to claim 1 is characterized in that: The weather station (5) includes functions for measuring wind speed, wind direction, temperature and humidity, snowfall, and solar radiation; the time-lapse photography system (6) includes time-lapse photography equipment with a resolution of not less than 4K; and the drone airport system (7) includes drones capable of collecting infrared and multispectral images.
5. The method for monitoring sudden snowmelt caused by top edge sliding of a 90-foot earthen site according to claim 1 is characterized in that: The cloud-based data acquisition and processing platform has cloud-based data access, adjustable acquisition frequency, image recognition linkage control, and abnormality warning functions. It supports the fusion analysis of multi-source sensor data and image data, and automatically increases the sampling frequency and triggers drone aerial photography tasks when slip precursor signals are detected.
6. The method for monitoring sudden snowmelt caused by top edge sliding of a Zuchitu site according to claim 1 is characterized in that: The identification of slip precursors adopts a combination of image time series comparison, crack recognition algorithm and sensor data threshold judgment to extract slip causes, development stages and trigger response characteristics, and build a response model that can be used for early warning and intervention.
7. The method for monitoring sudden snowmelt caused by top edge sliding of a 90-foot-high earthen site according to claim 5, characterized in that: The cloud-based data acquisition and processing platform supports high-frequency dynamic monitoring of the entire freeze-thaw cycle, with a high-priority sampling mechanism set for the three key stages: the early freezing stage, the freezing extreme stage, and the early stage of sudden thaw, to capture the dynamic changes in wall stress and strain and slip triggering characteristics.
8. The method for monitoring sudden snowmelt caused by top edge sliding of a 90-foot-high earthen site according to claim 1 is characterized in that: The infrared temperature images and time-lapse image data of the top and edge areas of the experimental wall (1) can be used to train a disease recognition model based on deep learning and to construct an intelligent recognition system for slip risk.
9. The method for monitoring sudden snowmelt caused by top edge sliding of a 9-foot-high earthen site according to claim 1, characterized in that: The multi-source monitoring data is used to construct a multi-field coupling numerical simulation boundary conditions and constitutive parameter library for wall slip disasters caused by sudden melting of snow.
10. A device using the method for monitoring the sudden melting of snow at the top edge of a full-scale earthen site according to any one of claims 1 to 9, characterized in that: It includes an environmental simulation platform construction module, a slope morphology reconstruction and tamping experiment module, a multi-dimensional monitoring network deployment module, a remote monitoring module and a model construction module. The environmental simulation platform construction module constructs an experimental platform in an open area with the same climatic characteristics as the site; The slope morphology reconstruction and ramming experiment module reconstructs the slope morphology before the sliding based on the original three-dimensional model of the site. Soil with physical properties similar to the rammed earth of the site is selected, and the optimal moisture content and dry density are determined through compaction experiments. The ancient traditional ramming technology and template system are used to ram the experimental wall in layers (1). The thickness of the rammed layer is set according to archaeological data. The multi-dimensional monitoring network deployment module deploys integrated multifunctional sensors (3) and stress and strain gauge sensors (4) in the internal rammed layer (2) of the experimental wall (1), distributed in different sections to avoid interference and form complementarity; external monitoring includes a weather station (5), a time-lapse photography system (6), a drone airport system (7) and a multispectral drone (8); The remote monitoring module remotely collects multi-source monitoring data through the cloud data collection and processing platform and uploads it to the cloud; The cloud-based data acquisition and processing platform of the model building module couples and analyzes multi-source monitoring data, combines image recognition, stress response, and temperature and humidity change characteristics, identifies slip precursors and response thresholds, and constructs a slip triggering model and evolution path identification mechanism.