Dam Monitoring Abnormal Early Warning Method and System Based on Optical Fiber Sensing

Abstract

The present invention relates to the technical field of water conservancy projects and structural health monitoring, and particularly to a method and system for abnormal early warning of dam monitoring based on optical fiber sensing. The method includes: collecting stress and strain data of different position areas of the dam to be monitored to determine the stress distribution inside the dam body; collecting the temperature of the water surface on the inner side of the dam to be monitored to determine the influence of water on the dam stress; collecting the actual stress generated by vehicles on the dam when vehicles pass above the dam to be monitored; applying simulated seismic vibrations to the test model to obtain simulated seismic vibration stress; applying vehicle loads to the test model to obtain vehicle load simulated stress; constructing a comprehensive feature vector of the dam; inputting the comprehensive feature vector of the dam into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the health degree of the dam. The present invention accurately evaluates the health state of the dam and timely discovers potential structural hidden dangers.

Description

Technical Field

[0001] The present invention relates to the technical field of hydraulic engineering and structural health monitoring, and particularly to a method and system for abnormal early warning of dam monitoring based on optical fiber sensing. Background Art

[0002] As an important hydraulic engineering facility, a reservoir dam undertakes multiple key tasks such as regulating water sources, flood control and drought relief, power generation and water supply. However, the long-term safe operation of the dam faces challenges from various natural and human factors, such as water pressure, seepage pressure, earthquakes, settlement, vehicle loads, etc. These external factors will affect the structure of the dam. In particular, the internal stress and strain changes of the dam will gradually accumulate, and may ultimately lead to structural damage or collapse. Once a dam instability or dam break accident occurs, the consequences will be unimaginable, possibly causing huge property losses and casualties. Therefore, the safety monitoring and early warning of the dam are of great significance.

[0003] Traditional dam monitoring technologies mainly rely on the monitoring of water level and flow rate. By setting a water level warning line, flood discharge operations are carried out when the warning water level is exceeded, and water storage is carried out when the water level is below the warning water level. However, relying solely on water level monitoring cannot comprehensively reflect the stress state inside the dam. Over time, the internal stress and strain of the dam body will gradually accumulate under the long-term impact of water flow, and at the same time, under the combined action of various factors such as vehicle loads, earthquakes and leakage, minor structural changes may occur inside the dam. These changes are usually difficult to detect by traditional monitoring means, but they may bring potential damage risks to the dam, such as piping and crack propagation. Some sudden accidents are often due to the stress or strain accumulated over a long time not being detected and processed in time, resulting in sudden instability.

[0004] In order to make up for the deficiencies of traditional monitoring means, in recent years, dam health monitoring systems based on optical fiber sensing technology have gradually become a research hotspot. Optical fiber sensors can monitor multiple physical quantities such as stress, strain, temperature, and vibration with high precision, and have the advantages of corrosion resistance, high temperature resistance, and electromagnetic interference resistance, making them suitable for long-term operation in harsh environments. By embedding optical fiber sensors in the internal structure of the dam, the stress and strain changes inside the dam can be monitored in real time, and structural abnormalities can be detected in a timely manner. Optical fiber sensors can also monitor the water level in combination with the change of water temperature, deduce the change of water pressure near the dam body through the distribution of the temperature field, and then analyze the safety state of the dam body through a multi-dimensional stress-strain model. In addition, by arranging optical fiber sensors on the surface of the dam, the immediate impact of dynamic external forces such as vehicle passage and earthquakes on the dam structure can also be monitored.

[0005] The current optical fiber monitoring technology is still less applied to the multi-dimensional monitoring of dams, and there is a lack of system design for collaborative monitoring with multiple physical quantities such as temperature, water level, vibration, and stress. Therefore, there is an urgent need for a multi-dimensional monitoring system for dams that combines optical fiber sensors, which can cover the needs of internal stress and strain monitoring of dams, temperature and water level back-calculation, vehicle load and vibration monitoring, etc., and realize the real-time evaluation and safety warning of dam structures. Summary of the Invention

[0006] In order to solve the deficiencies of the prior art, the present invention provides a method and system for abnormal warning of dam monitoring based on optical fiber sensing;

[0007] On the one hand, a method for abnormal warning of dam monitoring based on optical fiber sensing is provided, including:

[0008] Collect stress and strain data of different position areas of the dam to be monitored, and based on the stress and strain data, determine the stress distribution inside the dam body; collect the temperature of the water surface on the inner side of the dam to be monitored, and based on the water surface temperature, determine the influence of water on the dam body stress; collect the actual stress generated by the vehicle on the dam body when the vehicle passes above the dam to be monitored.

[0009] Construct a test model of the dam to be monitored, apply simulated seismic vibrations to the test model to obtain simulated seismic vibration stress; apply vehicle loads to the test model to obtain vehicle load simulated stress;

[0010] Based on the stress distribution inside the dam body, the influence of water on the dam body stress, the actual stress generated by the vehicle on the dam body, the simulated seismic vibration stress, and the vehicle load simulated stress, construct a comprehensive feature vector of the dam;

[0011] Input the comprehensive feature vector of the dam into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the health degree of the dam.

[0012] On the other hand, a system for abnormal warning of dam monitoring based on optical fiber sensing is provided, including:

[0013] A collection module, which is configured to: collect stress and strain data of different position areas of the dam to be monitored, and based on the stress and strain data, determine the stress distribution inside the dam body; collect the temperature of the water surface on the inner side of the dam to be monitored, and based on the water surface temperature, determine the influence of water on the dam body stress; collect the actual stress generated by the vehicle on the dam body when the vehicle passes above the dam to be monitored.

[0014] A simulation module, which is configured to: construct a test model of the dam to be monitored, apply simulated seismic vibrations to the test model to obtain simulated seismic vibration stress; apply vehicle loads to the test model to obtain vehicle load simulated stress;

[0015] A building module, configured to: construct a comprehensive feature vector of the dam based on the stress distribution inside the dam body, the influence of water on the dam stress, the actual stress generated by vehicles on the dam body, the simulated seismic vibration stress, and the vehicle load simulation stress;

[0016] An output module, configured to: input the comprehensive feature vector of the dam into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the health degree of the dam.

[0017] The above technical solution has the following advantages or beneficial effects:

[0018] The advantages of the present invention are that by arranging fiber optic sensors inside and outside the dam to monitor physical parameters such as stress, strain, temperature, and vibration in real time, the health status of the dam can be accurately evaluated, and potential structural hidden dangers can be discovered in time. The fiber optic sensors have the characteristics of high sensitivity, anti-interference, and high temperature resistance, and are suitable for working in complex environments for a long time. Combining with the multi-dimensional data analysis model, the system realizes the comprehensive monitoring and early warning of the dam structure, greatly improving the safety and operation efficiency of the dam. Description of the Drawings

[0019] The accompanying drawings forming a part of this invention are used to provide a further understanding of the invention. The schematic embodiments and descriptions thereof of the invention are used to explain the invention and do not constitute an improper limitation of the invention.

[0020] Figure 1 It is the method flow chart of Embodiment 1.

[0021] Figure 2 It is the overall system structure diagram of Embodiment 1.

[0022] Figure 3 It is the schematic diagram of the monitoring module for the internal stress and strain of the dam in Embodiment 1.

[0023] Figure 4 It is the schematic diagram of the sensing device for the surface stress of the dam in Embodiment 1. Detailed Description of the Invention

[0024] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0025] Term Explanation:

[0026] 1. Fiber Bragg Grating (FBG) sensor: A fiber optic sensor used to detect changes in physical quantities such as internal stress, strain, and temperature of a structure.

[0027] 2. Stress-strain model: By analyzing the fiber optic strain data, calculate the distribution of internal forces within the structure.

[0028] 3. Temperature-water level inversion model: Based on the temperature data measured by the fiber optic, combined with the heat conduction equation, calculate the real-time water level of the reservoir and inversely deduce the stress on the dam.

[0029] 4. Vibration monitoring: Used to monitor the vibration response of the dam surface and interior under seismic or vehicle loads, and evaluate the dynamic safety of the dam structure.

[0030] Example 1

[0031] This example provides a method for abnormal early warning of dam monitoring based on fiber optic sensing;

[0032] As Figure 1 shown, the method for abnormal early warning of dam monitoring based on fiber optic sensing includes:

[0033] S101: Collect stress-strain data of different position areas of the dam to be monitored. Based on the stress-strain data, determine the stress distribution inside the dam body; collect the temperature of the water surface on the inner side of the dam to be monitored. Based on the water surface temperature, determine the influence of water on the dam body stress; collect the actual stress generated by the vehicle on the dam when the vehicle passes above the dam to be monitored.

[0034] S102: Construct a test model of the dam to be monitored, apply simulated seismic vibrations to the test model to obtain simulated seismic vibration stress; apply vehicle loads to the test model to obtain vehicle load simulated stress.

[0035] S103: Based on the stress distribution inside the dam body, the influence of water on the dam body stress, the actual stress generated by the vehicle on the dam, the simulated seismic vibration stress, and the vehicle load simulated stress, construct a comprehensive feature vector of the dam.

[0036] S104: Input the comprehensive feature vector of the dam into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the dam health degree.

[0037] Furthermore, the S101: Collect stress-strain data of different position areas of the dam to be monitored. Based on the stress-strain data, determine the stress distribution inside the dam body, includes:

[0038] As Figure 2 、 Figure 3 and Figure 4 shown, during the construction of the dam, fiber optic sensors are arranged on the surface of the dam body close to the water side; fiber optic sensors are buried on the upper surface of the dam body.

[0039] Collect stress and strain data at different positions of the dam through fiber optic sensors;

[0040] ; (1)

[0041] Among them, is the strain at time moment, is the change in the reflected wavelength measured by the fiber optic sensor, that is, the change in the reflected wavelength in the fiber Bragg grating sensor, is the initial wavelength of the fiber optic sensor (the wavelength when there is no stress or strain).

[0042] Calculate the stress distribution inside the dam body. The stress calculation formula is:

[0043] , (2)

[0044] Among them, is the stress at time moment, is the elastic modulus, is the strain value.

[0045] Furthermore, the S101: Collect the temperature of the water surface on the inner side of the dam to be monitored, and based on the water surface temperature, determine the influence of water on the stress of the dam body, including:

[0046] Use the fiber optic temperature sensor on the surface of the dam close to the water to obtain the water temperature data at different depths and times in real time. The water temperature change formula is:

[0047] , (3)

[0048] Among them, is the water temperature at depth and time moment, is the initial water temperature, is the heat source intensity, is the thermal conductivity, is the thermal diffusion length;

[0049] Combine the water temperature and the strain data of the dam body, and calculate the influence of the water level on the stress of the dam body through the formula.

[0050] The influence of the water level on the stress of the dam body The calculation formula is as follows:

[0051] , (4)

[0052] Among them, represents the water level height calculated through the water temperature change and the temperature coefficient ; It represents calculating water pressure using the water level height, which is to convert water pressure into dam body stress, is the vertical height of the dam body, is the transverse thickness of the dam body; is the elastic modulus of the dam body material; is the strain of the dam body, reflecting the deformation of the dam body material under the action of external forces, represents the density of water, usually taking the value of , represents the acceleration due to gravity, usually taking the value of .

[0053] Furthermore, for the step S101: When collecting the actual stress generated by a vehicle on the dam body when the vehicle passes above the dam to be monitored, it includes:

[0054] Install fiber optic sensors above the dam (on the road surface). Through the fiber optic sensors on the road surface, record the strain and stress generated by the passing vehicle on the dam body. The stress calculation formula is:

[0055] , (5)

[0056] where, represents the stress on the dam body surface, is the vehicle load, is the contact area between the vehicle and the dam body, is the load direction angle. The load direction angle is the angle between the action direction of the vehicle load and the normal line of the dam body surface. The load direction angle determines how the load is distributed on the dam body surface and affects the stress calculation.

[0057] Furthermore, for the step S102: Constructing a test model of the dam to be monitored, it includes:

[0058] During the production of the test model, construct a test model with the same structure as the actual dam according to a ratio of 1:10;

[0059] The material of the test model is selected as concrete, with the model number C60, used to simulate the bearing capacity and stress distribution of the dam body. The layout position of the fiber optic sensors is the same as that of the actual dam;

[0060] During the test process, simulate different water level conditions. By gradually adjusting the water level to the critical height of the model, and simultaneously applying preset stress conditions and piping scenarios to observe the structural response and stress changes.

[0061] The test model uses permeable materials to simulate the dam body leakage and piping conditions. The permeable material is selected as geotextile (model: TX160) to cover the bottom of the dam body to ensure that the influence of leakage and stress on the dam body is truly reflected in the test.

[0062] It should be understood that by arranging sensors on the test model for monitoring, data on the stress, strain, and water pressure changes of the dam under different working conditions can be obtained, providing a basis for verifying the effectiveness of the sensor arrangement and the monitoring model.

[0063] Further, the step S102: applying simulated seismic vibrations to the test model to obtain simulated seismic vibration stress includes:

[0064] In the test model, simulated seismic vibrations are applied to simulate the influence of an earthquake on the dam body through periodic vibrations. The formula for applying the vibration stress is:

[0065] , (6)

[0066] where is the stress borne by the dam body at time , is the vibration coefficient, is the vibration amplitude, is the vibration frequency, is the length of the dam body.

[0067] By adjusting the amplitude and frequency of the vibration, the influence of earthquakes of different intensities on the stress distribution and strain changes of the dam body structure is simulated, and then the seismic performance and stability of the dam body are evaluated.

[0068] Further, the step S102: applying vehicle loads to the test model to obtain vehicle load simulated stress includes:

[0069] Periodic pressure is applied to the surface of the dam. The stress caused by the vehicle load is calculated by the following formula:

[0070] , (7)

[0071] where is the stress caused by the vehicle load at time , is the vehicle weight, is the acceleration due to gravity, is the contact area between the vehicle tire and the dam body surface, is the pressure transfer angle when the vehicle is moving; the pressure transfer angle when the vehicle is moving is the angle between the contact pressure between the vehicle tire and the dam body surface along the contact surface and the normal line of the dam body surface during vehicle movement.

[0072] This formula simulates the dynamic stress generated on the surface and internal structure of the dam body when a heavy vehicle is moving.

[0073] Further, in step S103, based on the stress distribution inside the dam body, the influence of water on the dam stress, the actual stress generated by vehicles on the dam body, the simulated seismic vibration stress, and the vehicle load simulation stress, a comprehensive characteristic vector of the dam is constructed, where the comprehensive characteristic vector of the dam has the following expression:

[0074] .

[0075] Further, in step S104, the comprehensive characteristic vector of the dam is input into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the health degree of the dam. Among them, for the trained dam risk prediction model, the training process includes:

[0076] Construct a dam risk prediction model; the dam risk prediction model is implemented by an LSTM model;

[0077] Construct a training set, where the training set is the comprehensive vector of the dam with known dam risk scores;

[0078] Input the training set into the dam risk prediction model to train the model. When the total loss function value of the model no longer decreases, stop the training to obtain the trained dam risk prediction model.

[0079] Further, for the total loss function of the model, its specific expression is:

[0080] ;

[0081] where is the error term (mean square error), which is used to measure the difference between the model prediction value and the true value , is the smoothing regularization term, which is a time series gradient constraint term used to limit the fluctuation amplitude of the model prediction value in the time dimension to ensure that the prediction results change smoothly between adjacent time points.

[0082] Real-time monitoring of physical parameters such as stress, strain, temperature, and vibration inside and outside the dam. This type of monitoring is mainly applied to large-scale water conservancy facilities such as reservoir dams and levees to ensure their long-term safe operation. With the increasing dependence of modern society on water conservancy facilities, dam safety monitoring has become a key technical link in fields such as flood control, power generation, and agricultural irrigation. Dams are under long-term pressure from multiple aspects such as water flow, water pressure, settlement, seepage, and earthquakes. Any minor structural change may lead to serious consequences. Therefore, how to conduct real-time monitoring of the internal stress distribution, surface deformation, and dynamic changes of the surrounding water body of the dam has become an important technical requirement for ensuring dam safety.

[0083] Traditional monitoring methods usually focus on the monitoring of water level and flow rate, and cannot comprehensively cover the detailed changes in the mechanical state, material fatigue, settlement and displacement inside the dam. Existing reservoir management systems pay more attention to the warning water level, but pay insufficient attention to the long-term accumulated stress, strain, vehicle traffic load, temperature effect, etc. inside the dam. Therefore, the present invention proposes an integrated monitoring system based on fiber optic sensing technology. By arranging fiber optic sensor arrays inside and outside the dam, stress, strain, temperature, vibration and other data inside and outside the dam are collected in real time, and calculations are carried out through physical models to timely discover possible structural hidden dangers and provide scientific warning information.

[0084] Fiber optic sensors have the advantages of high sensitivity, high temperature resistance, electromagnetic interference resistance, and long-term usability, and are suitable for use in complex environments such as dams. The present invention uses fiber Bragg grating (FBG) sensors to monitor the stress and strain inside the dam with high precision. At the same time, the change of water level is deduced by measuring the temperature change, and the health state of the dam body is evaluated in combination with the stress-strain model and vibration monitoring model. This technology provides a new means of guarantee for the long-term safe operation of the dam and has important technical value and social significance.

[0085] The purpose of the present invention is to provide a multi-dimensional monitoring system and method for dams based on fiber optic sensing technology. The system buries fiber optic sensors inside and on the surface of the dam to monitor key data such as stress, strain, water temperature, and vibration of the dam structure in real time, so as to effectively detect the health state of the dam and give early warnings of possible hidden dangers. The system is verified by combining model experiments, and can simulate the dynamic effects of earthquakes and vehicles on the dam, and evaluate the safety of the dam under different conditions.

[0086] Strain, stress and temperature sensors are arranged inside the dam to monitor the changes of strain, stress and water temperature inside the dam body in real time. At the same time, fiber optic sensors are arranged on the surface of the dam body and on the road surface above to monitor the surface strain, temperature change and the influence of vehicle passage on the dam body.

[0087] For the surface area, the optical fibers are evenly arranged along the surface of the dam body. Especially in areas that may be greatly affected by external factors (such as the dam top and the water-facing surface), the arrangement density is appropriately increased (one optical fiber is arranged every 20 cm) to accurately capture the surface strain and temperature changes.

[0088] On the road surface above, the optical fibers are installed by embedding (one optical fiber is buried every 30 cm) to monitor the dynamic load effect caused by vehicle passage in real time. All fiber optic sensors are connected to the central data processing system through connecting optical cables to ensure the efficient collection and transmission of data. Through this precise and comprehensive layout scheme, a stable and reliable data basis is provided for subsequent monitoring.

[0089] The fiber optic sensor system collects data inside, on the surface of the dam, and on the road surface in real time. Internal sensors monitor strain and stress, and analyze the structural state of the dam through strain sensors and stress sensors. Surface sensors record the strain and temperature changes on the dam surface, while road surface sensors capture the dynamic impact on the dam caused by vehicle passage. The data from all sensors is aggregated into a central data processing system for comprehensive analysis to evaluate the health status and safety of the dam under different working conditions.

[0090] By establishing a scaled-down model similar to the actual dam structure, tests are conducted under various working conditions, including different water levels and earthquake simulations. Fiber optic sensors are installed in the scaled-down model to record the strain, stress, and vibration responses of the model under different conditions. The test data is compared with the theoretical prediction results to verify the effectiveness of the sensor layout and data analysis model, and the system is optimized based on the test results to ensure accurate assessment of the dam's stability and safety in practical applications.

[0091] Fiber optic cables are installed on the water-facing side surface of the dam. The fiber optic cables are installed in hoses with apertures smaller than a set size, which can achieve reservoir water flow cleaning while preventing foreign objects from entering. This further ensures the detection and sensing accuracy and service life of the fiber optic cables.

[0092] During the construction of the dam, fiber optic sensors are embedded in the core area of the dam. The fiber optic sensors include strain sensors, stress sensors, and temperature sensors. The strain sensors are arranged in the main stress-bearing areas of the dam to monitor the internal strain changes of the dam. The stress sensors are arranged at key positions on the surface layer of the dam to capture the stress distribution of the dam structure. The temperature sensors are installed near the water surface to measure the water temperature changes in real time.

[0093] Fiber optic sensors are installed on the surface of the dam to monitor the strain and temperature changes on the surface. The fiber optic sensors should be evenly arranged to ensure coverage of all key areas on the dam surface.

[0094] Fiber optic sensors are buried in the road surface area above the dam to monitor the dynamic impact on the dam caused by vehicle passage. The sensors should be arranged at the center of the lane and under the wheel paths to capture the stress and strain caused by vehicle loads.

[0095] Data recording: Record the fiber optic sensor data of the model under different working conditions in real time, including internal stress, strain, water level, temperature, and vibration response. Analyze the data to evaluate the stability and stress distribution of the model.

[0096] Verification and optimization: Compare the test data with the theoretical model prediction results to verify the effectiveness of the fiber optic sensor layout scheme. Optimize the sensor layout strategy and data analysis model based on the test results to improve the accuracy and reliability in practical applications.

[0097] To meet the special requirements of dam monitoring, the following improvements can be made to the PSO algorithm:

[0098] The inertia weight in the original PSO algorithm is usually fixed, which easily leads to slow convergence speed of the algorithm or getting trapped in local optima. Improvement method: Use an adaptive inertia weight, and the weight is dynamically adjusted with the number of iterations:

[0099]

[0100] Among them, and are the maximum and minimum values of the inertia weight respectively, is the current number of iterations, is the maximum number of iterations.

[0101] After laying optical fibers on the surface of the dam on the water side, install a hose with a small aperture to prevent foreign objects from entering the optical fiber area and ensure the normal operation of the optical fibers.

[0102] Design the hose to use the reservoir water flow to clean the optical fibers and keep the optical fibers in a clean state. The hose should ensure that the water flow evenly covers the surface of the optical fibers and effectively removes dirt and sediments.

[0103] Through the implementation of this system, the detection accuracy of the optical fiber sensors can be effectively guaranteed, their service life can be extended, and stable support can be provided for the structural health monitoring of the dam.

[0104] Build a scaled-down model similar to the actual dam to simulate the geometric structure and material properties of the actual dam. Optical fiber sensors are arranged in the model to monitor changes in stress, strain, water temperature, etc. Apply vibrations through a seismic simulation device to simulate the effects of earthquakes of different intensities on the dam and record the changes in stress and strain. Simulate the dynamic effects of vehicle loads in the model, and the sensors record the dynamic stress changes caused by vehicle loads. Compare the test data with the predicted results of the theoretical model to verify the effectiveness of the sensor layout and data analysis model, optimize the design of the actual dam monitoring system, and ensure its reliability and accuracy.

[0105] Arrange optical fiber sensors inside and outside the dam to monitor physical quantities such as stress, strain, temperature, and vibration in real time.

[0106] Analyze the health status of the dam structure through a stress-strain model and combine it with a vibration monitoring model to evaluate the effects of earthquakes or vehicle loads. The stress-strain model is a mechanical model that describes the deformation behavior of materials under external forces. By analyzing the relationship between stress (the internal force per unit area caused by external forces) and strain (the degree of deformation of materials), the health status of materials or structures is evaluated.

[0107] Collect and analyze multi-dimensional data of the dam in real time, detect structural anomalies in a timely manner, issue early warning information, and ensure the long-term safe operation of the dam.

[0108] Embodiment 2

[0109] This embodiment provides a dam monitoring anomaly early warning system based on fiber optic sensing, including:

[0110] An acquisition module, which is configured to: acquire stress and strain data of different position areas of the dam to be monitored, and determine the stress distribution inside the dam body based on the stress and strain data; acquire the temperature of the water surface on the inner side of the dam to be monitored, and determine the influence of water on the stress of the dam body based on the water surface temperature; acquire the actual stress generated by vehicles on the dam body when vehicles pass above the dam to be monitored.

[0111] A simulation module, which is configured to: construct a test model of the dam to be monitored, apply simulated seismic vibrations to the test model to obtain simulated seismic vibration stress; apply vehicle loads to the test model to obtain vehicle load simulation stress.

[0112] A construction module, which is configured to: construct a comprehensive feature vector of the dam based on the stress distribution inside the dam body, the influence of water on the stress of the dam body, the actual stress generated by vehicles on the dam body, the simulated seismic vibration stress, and the vehicle load simulation stress.

[0113] An output module, which is configured to: input the comprehensive feature vector of the dam into the trained dam risk prediction model to obtain the anomaly risk score of the dam to be monitored. The higher the anomaly risk score, the lower the health degree of the dam.

[0114] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. The abnormal early warning method for dam monitoring based on optical fiber sensing is characterized by: include: Collect stress and strain data at different locations of the dam to be monitored, and determine the stress distribution inside the dam body based on the stress and strain data; The temperature of the water surface inside the dam to be monitored is collected, and based on the water surface temperature, the influence of water on the stress of the dam body is determined, including: using the optical fiber temperature sensor on the surface of the dam close to the water side to obtain real-time water temperature data at different depths and times. The water temperature change formula is: ； in, For Depth and time The water temperature at is the initial water temperature, is the intensity of the heat source, is the thermal conductivity, is the heat diffusion length; Effect of water level on dam stress The calculation formula is as follows: ； in, Indicates the change in water temperature and temperature coefficient Calculate the water level, Indicates that water pressure is calculated using water level height. It converts water pressure into dam body stress. is the vertical height of the dam, is the transverse thickness of the dam body; is the elastic modulus of the dam material; It is the strain of the dam body, reflecting the deformation of the dam material under the action of external force. Represents the density of water, usually taken as , Represents the acceleration due to gravity, usually taken as ; Collect the actual stress generated by vehicles on the dam body when vehicles pass over the dam to be monitored, including: Fiber optic sensors are buried above the dam. The fiber optic sensors on the road surface record the strain and stress caused by the passage of vehicles on the dam body. The stress calculation formula is: ； in, represents the surface stress of the dam body, is the vehicle load, is the contact area between the vehicle and the dam, is the load direction angle, which is the angle between the direction of the vehicle load and the normal line of the dam surface; Constructing a test model of the dam to be monitored, applying simulated earthquake vibration to the test model, and obtaining simulated earthquake vibration stress, including: In the test model, simulated earthquake vibration is applied to simulate the impact of earthquake on the dam body through periodic vibration. The formula for applying vibration stress is: ； in, For time The stress on the dam body is is the vibration coefficient, is the vibration amplitude, is the vibration frequency, is the length of the dam body; a vehicle load is applied to the test model to obtain a vehicle load simulation stress; Based on the stress distribution inside the dam body, the influence of water on the stress of the dam body, the actual stress caused by vehicles on the dam body, simulated earthquake vibration stress and vehicle load simulation stress, a comprehensive characteristic vector of the dam is constructed; The comprehensive feature vector of the dam is input into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the health of the dam.

2. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 1 is characterized in that: Collect stress and strain data at different locations of the dam to be monitored, and determine the stress distribution inside the dam body based on the stress and strain data, including: During the dam construction process, optical fiber sensors are laid on the surface of the dam body close to the water; optical fiber sensors are buried on the upper surface of the dam body; and stress and strain data at different positions of the dam are collected through optical fiber sensors. ； in, For time The adaptation of the moment, is the change in reflected wavelength measured by the optical fiber sensor, that is, the change in reflected wavelength in the optical fiber Bragg grating sensor. is the initial wavelength of the optical fiber sensor; Calculate the stress distribution inside the dam body. The stress calculation formula is: ； in, For time The stress of the moment, is the elastic modulus, is the strain value.

3. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 1 is characterized in that: Construct a test model of the dam to be monitored, including: During the test model making process, a test model with the same structure as the actual dam was constructed at a scale of 1:

10. The material of the test model was concrete to simulate the bearing capacity and stress distribution of the dam body. The fiber optic sensor was arranged in the same position as the actual dam. During the test, different water levels were simulated by gradually adjusting the water level to the critical height of the model while applying preset stress conditions and piping scenarios to observe structural responses and stress changes.

4. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 1 is characterized in that: Applying a vehicle load to the test model to obtain a vehicle load simulation stress includes: A periodic pressure is applied to the dam surface and the stress caused by vehicle loads is calculated using the following formula: ； in, For time The stress caused by vehicle load is is the vehicle weight, is the acceleration due to gravity, is the contact area between the vehicle tire and the dam surface, It is the pressure transfer angle when the vehicle is driving; the pressure transfer angle when the vehicle is driving is the angle between the angle at which the contact pressure between the vehicle tire and the dam surface is transferred along the contact surface when the vehicle is driving and the normal line of the dam surface.

5. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 4 is characterized in that: Based on the stress distribution inside the dam body, the influence of water on the dam body stress, the actual stress generated by vehicles on the dam body, simulated earthquake vibration stress and vehicle load simulated stress, a comprehensive characteristic vector of the dam is constructed. The expression is: 。 6. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 5 is characterized in that: The comprehensive feature vector of the dam is input into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the health of the dam. The training process of the trained dam risk prediction model includes: Constructing a dam risk prediction model; the dam risk prediction model is implemented by an LSTM model; Constructing a training set, wherein the training set is a comprehensive dam vector of known dam risk scores; Input the training set into the dam risk prediction model, train the model, and stop training when the total loss function value of the model no longer decreases, to obtain the trained dam risk prediction model; The total loss function of the model is specifically expressed as: ； in, is the error term, which is used to measure the model prediction value With the true value The difference between is a smoothness regularization term, which is a time series gradient constraint term used to limit the predicted value of the model The fluctuation range in the time dimension ensures that the forecast results change smoothly between adjacent time points.

7. The dam monitoring abnormality early warning system based on optical fiber sensing is characterized by: include: A collection module is configured to: collect stress and strain data of different locations of the dam to be monitored, and determine the stress distribution inside the dam body based on the stress and strain data; The temperature of the water surface inside the dam to be monitored is collected, and based on the water surface temperature, the influence of water on the stress of the dam body is determined, including: using the optical fiber temperature sensor on the surface of the dam close to the water side to obtain real-time water temperature data at different depths and times. The water temperature change formula is: ； in, For Depth and time The water temperature at is the initial water temperature, is the intensity of the heat source, is the thermal conductivity, is the heat diffusion length; Effect of water level on dam stress The calculation formula is as follows: ； in, Indicates the change in water temperature and temperature coefficient Calculate the water level, Indicates that water pressure is calculated using water level height. It converts water pressure into dam body stress. is the vertical height of the dam, is the transverse thickness of the dam body; is the elastic modulus of the dam material; It is the strain of the dam body, reflecting the deformation of the dam material under the action of external force. Represents the density of water, usually taken as , Represents the acceleration due to gravity, usually taken as ; Collect the actual stress generated by vehicles on the dam body when vehicles pass over the dam to be monitored, including: Fiber optic sensors are buried above the dam. The fiber optic sensors on the road surface record the strain and stress caused by the passage of vehicles on the dam body. The stress calculation formula is: ； in, represents the surface stress of the dam body, is the vehicle load, is the contact area between the vehicle and the dam, is the load direction angle, which is the angle between the direction of the vehicle load and the normal line of the dam surface; The simulation module is configured to: construct a test model of the dam to be monitored, apply simulated earthquake vibration to the test model, and obtain simulated earthquake vibration stress, including: In the test model, simulated earthquake vibration is applied to simulate the impact of earthquake on the dam body through periodic vibration. The formula for applying vibration stress is: ； in, For time The stress on the dam body is is the vibration coefficient, is the vibration amplitude, is the vibration frequency, is the length of the dam body; a vehicle load is applied to the test model to obtain a vehicle load simulation stress; A construction module is configured to: construct a comprehensive characteristic vector of the dam based on stress distribution inside the dam body, influence of water on the stress of the dam body, actual stress generated by vehicles on the dam body, simulated earthquake vibration stress and simulated stress of vehicle load; The output module is configured to: input the comprehensive feature vector of the dam into the trained dam risk prediction model to obtain the abnormal risk score of the dam to be monitored. The higher the abnormal risk score, the lower the health of the dam.

Images