Deep and far sea small island health monitoring system and method

Abstract

The invention relates to the technical field of deep and far sea small island health monitoring, and discloses a deep and far sea small island health monitoring system and method. A non-intrusive data acquisition system composed of a sensor network, a shore-based camera device and a vibration sensor is adopted, and total-factor real-time monitoring of open sea environment parameters, reef images and the vibration state of a protective building is achieved. Through joint operation of a wave field distribution model and a structural damage prediction model, actually measured data and numerical simulation are combined, and the temporal-spatial resolution of wave parameters of the reef and the evaluation precision of the health state of a protective building are improved; the early warning threshold value is dynamically adjusted through deviation analysis of real-time association of historical disaster data and the current monitoring value, and the timeliness and accuracy of early warning of extreme events are improved; finally, through an automatic triggering mechanism of a multi-stage early warning signal, all-weather and high-reliability safety monitoring guarantee is provided for deep and far sea islands, meanwhile, the manual maintenance cost is reduced, and a remarkable technical synergistic effect is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of health monitoring for small islands in the deep and far - reaching sea, and particularly to a health monitoring system and method for small islands in the deep and far - reaching sea. Background Art

[0002] Small islands in the deep and far - reaching sea usually refer to islands or reefs that are more than 200 nautical miles away from the continental coastline and are isolated in the open sea area. Such islands are mostly composed of coral reefs, volcanic rocks or sedimentary rocks and have a small area.

[0003] Due to the special geographical location of small islands in the deep and far - reaching sea, these islands are long - term exposed to high - intensity wind and waves, extreme meteorological events (such as super typhoons, tsunamis) and complex ocean current environments, resulting in significant dynamic changes in topography and shoreline erosion. At the same time, the surrounding waters of small islands in the deep and far - reaching sea are often hotspots for marine biodiversity, and their health status directly affects the stability of coral reef ecosystems, fishery resources and habitats of endangered species. The water depth and waves in the deep and far - reaching sea are large, and the marine environment is very complex. The risk of damage to small islands and their buildings under sudden disasters such as earthquakes, typhoon waves and storm surges is high.

[0004] However, the water depth and waves in the deep and far - reaching sea are large, and the marine environment is very complex. At the same time, there is a lack of long - term marine observation data, the coupling mechanism between island water - related buildings and strong hydrodynamic loads is unclear, and the risk of damage to small islands and their buildings under sudden disasters such as earthquakes, typhoon waves and storm surges is high. Once the protection project is unstable, it will cause irreparable strategic losses.

[0005] Therefore, there is an urgent need for a health monitoring system and method for small islands in the deep and far - reaching sea to achieve rapid and accurate monitoring of the safety of small islands in the deep and far - reaching sea. Summary of the Invention

[0006] To solve the above - mentioned technical problems, on the one hand, the present invention provides a health monitoring system for small islands in the deep and far - reaching sea, including a data collection module, a data simulation module, a machine learning module and an early - warning execution module; The data collection module is used to collect real - time offshore marine environment data of the island through a sensor network, obtain video image data of the reef flat area through a shore - based camera device, and obtain vibration data of the protective building through a vibration sensor; The data simulation module is connected to the data collection module and includes a wave field distribution model and a structural damage prediction model. The wave field distribution model is used to generate reef flat wave parameters according to the offshore marine environment data, and the structural damage prediction model is used to output the health information of the protective building by combining the reef flat wave parameters and the vibration data of the protective building; The machine learning module, connected to the data simulation module, is used to dynamically generate an early warning threshold range based on historical monitoring data and real-time monitoring data within a preset time window. The early warning threshold range includes an effective wave height threshold, a main vibration frequency offset threshold, and a foundation settlement rate threshold; An early warning execution module, connected to the machine learning module, is used to trigger multi-level early warning signals when the reef flat wave parameters or the health information of the protective building exceed the early warning threshold range.

[0007] Further, the data collection module includes: A vibration sensor array buried inside the protective building to form a self-organizing network; A shore-based panoramic camera device deployed at the highest point of the island and equipped with a salt fog resistant protective cover; A buoy array deployed in the offshore area of the island and integrated with a Doppler current meter and a temperature-salinity-depth sensor.

[0008] Further, the operation of the wave field distribution model includes: Obtain wind field data, input the wind field data into a wave numerical simulation model to obtain an offshore wave spectrum; Take the offshore wave spectrum as a boundary condition and input it into a nearshore wave-current coupling model to calculate the reef flat wave parameters. The reef flat wave parameters include the wave height, period, and velocity distribution in the reef flat area.

[0009] Further, a wave-foundation coupled dynamic response model is integrated between the wave field distribution model and the structural damage prediction model. The wave-foundation coupled dynamic response model is used for: Obtain the foundation stress distribution according to the reef flat wave parameters and the foundation geological parameters; And input the foundation stress distribution into the structural damage prediction model, and combine the reef flat wave parameters and the vibration data of the protective building to obtain the health information of the protective building.

[0010] Further, the data simulation module also includes a GPU parallel computing cluster for performing the following operations: Jointly compile the wave numerical model, the nearshore wave-current coupling model, and the wave-foundation coupled dynamic response model; When the typhoon path prediction data is updated, preferentially allocate the grid computing tasks within the preset radius area of the typhoon center to the GPU parallel computing cluster.

[0011] Further, the specific steps for the machine learning module to dynamically generate the early warning threshold range include: Obtain a historical monitoring data set within a preset time window. The data set includes an effective wave height sequence, a main vibration frequency offset sequence, and a foundation settlement rate sequence; Calculate the historical mean and standard deviation of each parameter, and determine the basic threshold based on the mean plus N times the standard deviation, where N is a preset safety factor; Collect the current effective wave height, vibration main frequency deviation, and foundation settlement rate monitoring values in real time. When any parameter deviates from the historical mean by more than a preset percentage, increase or decrease the corresponding threshold in proportion to the deviation degree; Compare the dynamically adjusted threshold with the design specification threshold, and take the smaller value of the two as the final warning threshold; When the real-time monitoring data exceeds the final warning threshold, activate the warning signal of the corresponding level.

[0012] Furthermore, the data collection module further includes: A bionic unmanned aerial vehicle unit, which adopts a flapping wing silent structure and starts when the shoreline is abnormal, and its flight trajectory avoids ecological sensitive areas; A bionic coral camera unit, which includes a 3D printed base and an internal waterproof component. The surface texture of the base is consistent with the surrounding coral morphology and integrates a wireless charging module.

[0013] Furthermore, the system further includes a visualization module for performing three-dimensional dynamic display by integrating real-time monitoring data, simulation results, and warning information.

[0014] Furthermore, the visualization module includes: A digital twin engine, which integrates multi-temporal remote sensing data and real-time monitoring results to construct a three-dimensional island model; A disaster deduction interface, which supports interactive modification of the typhoon path and real-time prediction of the shoreline evolution trend; A structural health monitoring dashboard, which displays the vibration spectrum and damage heat map of the protective building.

[0015] On the other hand, the present invention also provides a health monitoring method for small deep-sea islands, which is implemented based on any one of the above-mentioned health monitoring systems for small deep-sea islands. The method specifically includes: Data collection: Real-time collect the offshore marine environment data of the island through a sensor network, obtain the video image data of the reef flat area through a shore-based camera device, and obtain the vibration data of the protective building through a vibration sensor; Data simulation: Generate reef flat wave parameters according to the offshore marine environment data through a wave field distribution model, and output the health information of the protective building by combining the reef flat wave parameters and the vibration data of the protective building through a structural damage prediction model; Dynamically generate a warning threshold range, and dynamically generate a warning threshold range based on the historical monitoring data and real-time monitoring data within a preset time window. The warning threshold range includes an effective wave height threshold, a vibration main frequency deviation threshold, and a foundation settlement rate threshold; Early warning is triggered. When the reef flat wave parameters or the health information of the protective building exceed the early warning threshold range, multi-level early warning signals are triggered.

[0016] The embodiments of the present invention have the following technical effects: A health monitoring system for small islands in the deep sea and far sea provided by the present invention adopts a non-invasive data acquisition system composed of a sensor network, a shore-based camera device and a vibration sensor, realizing the all-element real-time monitoring of offshore environmental parameters, reef flat images and the vibration state of protective buildings, and avoiding the interference of manual monitoring on the fragile island ecosystem; secondly, through the joint operation of the wave field distribution model and the structural damage prediction model, combining the measured data with numerical simulation, significantly improving the spatio-temporal resolution of reef flat wave parameters and the evaluation accuracy of the health state of protective buildings; furthermore, based on the machine learning algorithm, the early warning threshold is dynamically adjusted, breaking through the limitation that the traditional fixed threshold cannot adapt to the changes of complex marine environments. Through the deviation analysis of real-time correlation between historical disaster data and current monitoring values, the intelligent optimization of early warning indicators is realized, greatly improving the timeliness and accuracy of extreme event early warning; finally, through the automatic triggering mechanism of multi-level early warning signals, a closed-loop system from data acquisition, model deduction to risk decision-making is formed, providing all-weather and highly reliable safety monitoring guarantee for deep sea and far sea islands, while reducing the manual maintenance cost, and having significant technical synergy effects. In short, by integrating multi-source data acquisition and model-based analysis and processing, the problems of low accuracy, ecological damage and high cost existing in traditional monitoring technologies are effectively solved. Description of the Drawings

[0017] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0018] Figure 1 It is a module schematic diagram of a health monitoring system for small islands in the deep sea and far sea provided by the embodiments of the present invention; Figure 2 It is a relationship schematic diagram of each simulation model in the data simulation module provided by the embodiments of the present invention; Figure 3 It is a step flow chart of a health monitoring method for small islands in the deep sea and far sea provided by the embodiments of the present invention. Specific Embodiments

[0019] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0020] In order to achieve rapid and accurate monitoring of the security of small islands in the deep and far seas, on the one hand, the present invention provides a health monitoring system for small islands in the deep and far seas, as Figure 1 shown, including a data collection module, a data simulation module, a machine learning module, and an early warning execution module; The data collection module is used to collect the offshore marine environment data of the island in real time through a sensor network, obtain the video image data of the reef flat area through a shore-based camera device, and obtain the vibration data of the protective building through a vibration sensor; The data collection module is responsible for integrating multi-type sensor networks and using vibration sensors, shore-based camera devices, etc. to collect environmental parameters in different dimensions.

[0021] In some embodiments, the data collection module includes: A vibration sensor array buried inside the protective building to form a self-organizing network; A shore-based panoramic camera device deployed at the highest point of the island and equipped with an anti-salt fog protective cover; A buoy array deployed in the offshore area of the island and integrated with a Doppler current meter and a CTD sensor.

[0022] The vibration sensors are embedded inside the protective building, and the vibration state of the building is transmitted in real time through a self-organizing network. Exemplarily, a distributed deployment strategy is adopted, and micro accelerometers are buried at the key load-bearing nodes of the protective building, and multi-node data synchronization collection and redundant transmission are realized through wireless ad hoc network technology. The shore-based camera device is installed at the highest point of the island and equipped with an anti-salt fog protective cover to cope with the harsh marine climate, ensuring long-term stable operation in a strongly corrosive environment and continuously capturing video images of the reef flat area. Exemplarily, a high-resolution optical lens is selected. The buoy array is distributed in the offshore area, integrated with a Doppler current meter and a CTD sensor, and periodically transmits back the marine environment data. Exemplarily, it is arranged radially in the offshore area of the island, and each buoy is integrated with a Doppler current meter, a CTD sensor, and a Beidou positioning module, and transmits back real-time marine parameters through a satellite link. The buoys form a cooperative observation network through underwater acoustic communication. When a certain buoy is displaced or fails, the adjacent buoys automatically adjust the observation frequency to ensure data continuity. This design realizes the all-round three-dimensional monitoring of marine environmental parameters through the spatial optimization configuration of multi-source heterogeneous sensors.

[0023] In some embodiments, the data collection module further includes: A bionic unmanned aerial vehicle unit, which adopts a flapping-wing silent structure and is activated when the shoreline is abnormal. Its flight trajectory avoids ecologically sensitive areas. The body is equipped with a multi-spectral imager and a lidar, and can real-time identify shoreline erosion features through edge computing.

[0024] A bionic coral camera unit, which includes a 3D printed base and internal waterproof components. The surface texture of the base is consistent with the surrounding coral morphology and integrates a wireless charging module. The 3D printed base uses biocompatible materials. When surrounding organisms approach, the camera unit automatically enters the sleep mode to reduce ecological interference. This bionic monitoring device maximally maintains the balance of the original ecosystem while ensuring the quality of data collection.

[0025] A data simulation module, which is connected to the data collection module, includes a wave field distribution model and a structural damage prediction model. The wave field distribution model is used to generate reef flat wave parameters according to the offshore marine environment data, and the structural damage prediction model is used to output the health information of the protective building by combining the reef flat wave parameters and the vibration data of the protective building; In some embodiments, the operation of the wave field distribution model includes: Obtain wind field data, input the wind field data into the wave numerical simulation model, and obtain the offshore wave spectrum; Take the offshore wave spectrum as the boundary condition, input it into the nearshore wave-current coupling model, and calculate the reef flat wave parameters. The reef flat wave parameters include the wave height, period, and flow velocity distribution in the reef flat area.

[0026] The present invention directly adopts the existing wave numerical simulation model without improvement. The input parameter wind field data includes wind speed, wind direction, fetch length, duration, air pressure, etc., and the output is the reef flat wave parameters.

[0027] In some embodiments, a wave-soil coupling dynamic response model is integrated between the wave field distribution model and the structural damage prediction model, as Figure 2 shown. The wave-soil coupling dynamic response model is used for: Obtain the soil stress distribution according to the reef flat wave parameters and the soil geological parameters. Exemplarily, the wave-soil coupling dynamic response model can be implemented by finite element simulation.

[0028] And input the soil stress distribution into the structural damage prediction model, and combine the reef flat wave parameters and the vibration data of the protective building to obtain the health information of the protective building.

[0029] The vibration data of the protective building is a direct measurement of the structural response, which can verify the prediction results of the model and make real-time adjustments. For example, changes in the vibration spectrum may indicate a loss of structural stiffness or loosening of the connection nodes. However, the foundation stress distribution reflects the impact of wave loads on the foundation, which is an important factor in structural damage. Using vibration data alone may only detect the immediate response of the structure, but combining it with foundation stress can predict long-term cumulative damage. For example, stress concentration in the foundation may lead to uneven settlement of the structural foundation, which in turn causes structural damage. Incorporating the foundation stress distribution into the structural damage prediction model can improve the accuracy of damage detection, early warning capabilities, reduce false alarms and missed detections, extend the service life of the structure, and reduce maintenance costs. For example, through the analysis of foundation stress, fatigue damage can be predicted before visible cracks appear in the structure, enabling early maintenance.

[0030] The data simulation module significantly reduces the dependence on traditional intensive on-site monitoring by jointly operating the wave field distribution model and the structural damage prediction model. The wave field model uses limited wind field data collected by offshore buoys to simulate and invert the wave parameter distribution across the entire reef flat, replacing the traditional direct measurement method of deploying wave gauges at key points on the reef flat, reducing the number of underwater sensor deployments and maintenance costs. The structural damage model combines the reef flat wave parameters and the vibration data of the protective building, eliminating the need to densely install accelerometers on the entire surface of the building. While ensuring the evaluation accuracy, it greatly reduces the difficulty of equipment deployment and data collection costs in the harsh deep-sea and far-sea environments, extends the non-intervention operation cycle of the monitoring system, and reduces the environmental damage caused by human activities.

[0031] In some embodiments, the data simulation module further includes a GPU parallel computing cluster for performing the following operations: Jointly compile the wave numerical model, the nearshore wave-current coupling model, and the wave-foundation coupled dynamic response model; When the typhoon path prediction data is updated, preferentially allocate the grid computing tasks within the preset radius area centered on the typhoon center to the GPU parallel computing cluster. When the typhoon path changes, the GPU cluster quickly recalculates the wave parameters to ensure the timeliness of the early warning. In addition, jointly compiling multiple models means that these models can work efficiently and collaboratively on the same cluster, reducing data transmission delays and improving the overall computing efficiency.

[0032] The GPU parallel computing cluster is a hardware architecture that utilizes multiple GPUs for parallel processing to accelerate computing tasks. It jointly compiles and optimizes wave numerical models, nearshore wave-current coupling models, and wave-foundation coupling dynamic response models to achieve dynamic load balancing of computing tasks. When typhoon path prediction data is updated, the system automatically identifies the computing grids within the influence range of the typhoon center and preferentially assigns high-priority tasks to the GPU cluster for accelerated computing. The parallel computing framework adopts an unstructured grid partitioning strategy combined with a dynamic memory allocation algorithm, significantly improving the computing efficiency of large-scale wave field simulations. This design ensures real-time computing requirements under extreme weather conditions such as typhoons through intelligent task scheduling and hardware acceleration.

[0033] The machine learning module is connected to the data simulation module and is used to dynamically generate an early warning threshold range based on historical monitoring data and real-time monitoring data within a preset time window. The early warning threshold range includes the significant wave height threshold, the main vibration frequency offset threshold, and the foundation settlement rate threshold. In some embodiments, the specific steps for the machine learning module to dynamically generate the early warning threshold range include: Obtain the historical monitoring data set within the preset time window. The data set includes the significant wave height sequence, the main vibration frequency offset sequence, and the foundation settlement rate sequence. Calculate the historical mean and standard deviation of each parameter, and determine the basic threshold based on the mean plus N times the standard deviation, where N is a preset safety factor. Real-time collect the current significant wave height, main vibration frequency offset, and foundation settlement rate monitoring values. When any parameter deviates from the historical mean by more than a preset percentage, increase or decrease the corresponding threshold in proportion to the deviation degree. Compare the dynamically adjusted threshold with the design specification threshold, and take the smaller value of the two as the final early warning threshold. When the real-time monitoring data exceeds the final early warning threshold, activate the corresponding level of early warning signal.

[0034] The dynamic early warning threshold generation mechanism adopts a sliding time window technology to continuously update the historical monitoring database. Perform time series analysis on parameters such as significant wave height, main vibration frequency offset, and foundation settlement rate. When the real-time monitoring value shows a trend deviation, the system automatically adjusts the threshold sensitivity, effectively improving the environmental adaptability of the early warning system, enhancing the early warning accuracy, to adapt to the complex and changeable marine environment of small islands in the deep sea and far sea, reducing maintenance costs, and extending the autonomous operation time of the system. For example, during the typhoon season, the significant wave height may be higher than normal, and static thresholds may not be applicable. Therefore, it is necessary to dynamically adjust the early warning threshold and may optimize the early warning sensitivity according to the current situation.

[0035] The early warning execution module is connected to the machine learning module and is used to trigger multi-level early warning signals when the reef flat wave parameters or the health information of the protective building exceed the early warning threshold range.

[0036] In some embodiments, the system further includes a visualization module for performing three-dimensional dynamic display by integrating real-time monitoring data, simulation results, and early warning information.

[0037] In some embodiments, the visualization module includes: A digital twin engine that integrates multi-temporal remote sensing data and real-time monitoring results to construct a three-dimensional island model; A disaster deduction interface that supports interactive modification of typhoon paths and real-time prediction of shoreline evolution trends; A structural health monitoring dashboard that displays the vibration spectrum and damage heat map of protective buildings.

[0038] Exemplarily, the visualization module can also adopt a three-dimensional geographic information system engine to integrate multi-temporal remote sensing images and real-time monitoring data to construct a digital twin. The disaster deduction function supports interactive modification of typhoon paths and real-time prediction of shoreline evolution trends under different scenarios. The data dashboard provides multi-dimensional statistical analysis charts and supports playback of the development process of historical disaster events along the time axis. The virtual reality interface can generate an immersive observation environment to assist management personnel in intuitively evaluating the state of protective engineering structures.

[0039] The deep-sea small island health monitoring system provided by the present invention adopts a non-intrusive data acquisition system composed of a sensor network, a shore-based camera device, and vibration sensors, realizing all-element real-time monitoring of offshore environmental parameters, reef flat images, and the vibration state of protective buildings, and avoiding the interference of manual monitoring on the fragile island ecosystem; secondly, through the joint operation of the wave field distribution model and the structural damage prediction model, combining measured data with numerical simulation, significantly improving the spatio-temporal resolution of reef flat wave parameters and the evaluation accuracy of the health state of protective buildings; furthermore, dynamically adjusting the early warning threshold based on machine learning algorithms, breaking through the limitation that traditional fixed thresholds cannot adapt to complex marine environment changes, and realizing intelligent optimization of early warning indicators through real-time correlation analysis of the deviation between historical disaster data and current monitoring values, greatly improving the timeliness and accuracy of extreme event early warning; finally, through the automatic triggering mechanism of multi-level early warning signals, a closed-loop system from data acquisition, model deduction to risk decision-making is formed, providing all-weather and highly reliable safety monitoring guarantee for deep-sea islands, while reducing the manual maintenance cost, and having significant technical synergistic effects. In short, by integrating multi-source data acquisition and model-based analysis and processing, the problems of low accuracy, ecological damage, and high cost existing in traditional monitoring technologies are effectively solved.

[0040] On the other hand, the present invention also provides a deep-sea small island health monitoring method, which is implemented based on the deep-sea small island health monitoring system according to any one of the above, as Figure 3 shown, and the method specifically includes: S1: Data collection, real-time collection of offshore marine environment data of the island through a sensor network, acquisition of video image data of the reef flat area through a shore-based camera device, and acquisition of vibration data of the protective building through a vibration sensor; S2: Data simulation: Generate reef flat wave parameters according to the offshore marine environment data through a wave field distribution model, and output the health information of the protective building by combining the reef flat wave parameters and the vibration data of the protective building through a structural damage prediction model; S3: Dynamically generate the early warning threshold range, dynamically generate the early warning threshold range based on the historical monitoring data and real-time monitoring data within a preset time window, and the early warning threshold range includes the significant wave height threshold, the dominant frequency deviation threshold of vibration, and the foundation settlement rate threshold; S4: Early warning trigger, when the reef flat wave parameters or the health information of the protective building exceed the early warning threshold range, trigger a multi-level early warning signal.

[0041] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A health monitoring system for small islands in the deep sea, characterized in that: It includes data collection module, data simulation module, machine learning module and early warning execution module; The data collection module is used to collect the offshore marine environment data of the island in real time through the sensor network, obtain the video image data of the reef flat area through the shore-based camera device, and obtain the vibration data of the protective building through the vibration sensor; The data simulation module is connected to the data collection module and includes a wave field distribution model and a structural damage prediction model. The wave field distribution model is used to generate reef flat wave parameters according to offshore marine environment data. The structural damage prediction model is used to combine the reef flat wave parameters and the protective building vibration data to output protective building health information; The machine learning module is connected to the data simulation module and is used to dynamically generate a warning threshold range based on historical monitoring data and real-time monitoring data within a preset time window, wherein the warning threshold range includes a significant wave height threshold, a vibration main frequency deviation threshold, and a foundation settlement rate threshold; The early warning execution module is connected to the machine learning module and is used to trigger a multi-level early warning signal when the reef flat wave parameters or the health information of the protective building exceed the early warning threshold range.

2. The deep sea small island health monitoring system according to claim 1 is characterized in that: The data collection module comprises: An array of vibration sensors is buried inside the protective building to form a self-organizing network; The shore-based panoramic camera device is deployed at the commanding heights of the island and equipped with an anti-salt spray protective cover; The buoy array is deployed in the offshore area of ​​the island and integrates Doppler current meters and temperature, salinity and depth sensors.

3. The deep sea small island health monitoring system according to claim 1 is characterized in that: The operation of the wave field distribution model includes: Acquiring wind field data, inputting the wind field data into a wave numerical simulation model, and obtaining an offshore wave spectrum; The offshore wave spectrum is used as a boundary condition and input into a nearshore wave-current coupling model to calculate reef flat wave parameters, which include wave height, period and flow velocity distribution in the reef flat area.

4. The deep sea small island health monitoring system according to claim 3 is characterized in that: A wave-foundation coupled dynamic response model is integrated between the wave field distribution model and the structural damage prediction model. The wave-foundation coupled dynamic response model is used to: According to the reef flat wave parameters and foundation geological parameters, the foundation stress distribution is obtained; The foundation stress distribution is input into the structural damage prediction model, and the health information of the protective building is obtained by combining the reef flat wave parameters and the vibration data of the protective building.

5. The deep sea small island health monitoring system according to claim 3 is characterized in that: The data simulation module also includes a GPU parallel computing cluster for performing the following operations: Jointly compiling the wave numerical model, the nearshore wave-current coupling model and the wave-foundation coupling dynamic response model; When the typhoon path prediction data is updated, grid computing tasks within a preset radius area of ​​the typhoon center are preferentially allocated to the GPU parallel computing cluster.

6. The deep sea small island health monitoring system according to claim 1 is characterized in that: The specific steps of dynamically generating the warning threshold range by the machine learning module include: Acquire a historical monitoring data set within a preset time window, wherein the data set includes an effective wave height sequence, a vibration main frequency offset sequence, and a foundation settlement rate sequence; Calculate the historical mean and standard deviation of each parameter, and determine the basic threshold based on the mean plus N times the standard deviation, where N is the preset safety factor; Real-time collection of current effective wave height, vibration main frequency deviation and foundation settlement rate monitoring values. When any parameter deviates from the historical mean by more than a preset percentage, the corresponding threshold is increased or decreased in proportion to the deviation. Compare the dynamically adjusted threshold with the design specification threshold, and take the smaller value as the final warning threshold; When the real-time monitoring data exceeds the final warning threshold, the corresponding level of warning signal is activated.

7. The deep sea small island health monitoring system according to claim 1 is characterized in that: The data collection module also includes: The bionic drone unit adopts a flapping-wing silent structure and is activated when the shoreline is abnormal, and its flight trajectory avoids ecologically sensitive areas; The bionic coral camera unit includes a 3D printed base and internal waterproof components. The surface texture of the base is consistent with the shape of the surrounding corals and has an integrated wireless charging module.

8. The deep sea small island health monitoring system according to claim 1 is characterized in that: The system also includes a visualization module for integrating real-time monitoring data, simulation results and early warning information for three-dimensional dynamic display.

9. The deep sea small island health monitoring system according to claim 8, characterized in that: The visualization module comprises: Digital twin engine, integrating multi-temporal remote sensing data with real-time monitoring results to build a three-dimensional island model; Disaster simulation interface, supporting interactive modification of typhoon paths and real-time prediction of coastline evolution trends; The structural health monitoring dashboard displays the vibration spectrum and damage heat map of the protective building.

10. A method for monitoring the health of small islands in the deep sea, implemented based on a system for monitoring the health of small islands in the deep sea as claimed in any one of claims 1 to 9, characterized in that: The method specifically comprises: Data collection: collect offshore marine environment data of the island in real time through sensor networks, obtain video image data of the reef area through shore-based cameras, and obtain vibration data of protective buildings through vibration sensors; Data simulation: Generate reef flat wave parameters based on offshore marine environment data through wave field distribution model, and output protective building health information through structural damage prediction model combined with reef flat wave parameters and protective building vibration data; Dynamically generate a warning threshold range based on historical monitoring data and real-time monitoring data within a preset time window, wherein the warning threshold range includes a significant wave height threshold, a vibration main frequency deviation threshold, and a foundation settlement rate threshold; Early warning triggering: when the reef flat wave parameters or the health information of the protective building exceed the early warning threshold range, a multi-level early warning signal is triggered.

Images