Dam monitoring abnormity early warning method and system based on optical fiber sensing

Abstract

The invention relates to the technical field of hydraulic engineering and structure health monitoring, in particular to a dam monitoring abnormity early warning method and system based on optical fiber sensing, and the method comprises the steps: collecting stress-strain data of different position areas of a to-be-monitored dam, and determining the stress distribution in a dam body; collecting the temperature of the water surface on the inner side of the dam to be monitored, and determining the influence of water on dam body stress; collecting actual stress generated by a vehicle on the dam body when the vehicle passes above the dam to be monitored; applying simulated seismic vibration to the test model to obtain simulated seismic vibration stress; applying a vehicle load to the test model to obtain a vehicle load simulation stress; constructing a comprehensive feature vector of the dam; and inputting the comprehensive feature vector of the dam into the trained dam risk prediction model to obtain an abnormal risk score of the to-be-monitored dam, and the higher the abnormal risk score is, the lower the health degree of the dam is. The health state of the dam is accurately evaluated, and potential structural hidden dangers are found in time.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy engineering and structural health monitoring, and in particular to a dam monitoring abnormality early warning method and system based on optical fiber sensing. Background Art

[0002] As an important water conservancy project facility, reservoir dams carry many key tasks such as regulating water sources, flood control and drought relief, power generation and water supply. However, the long-term safe operation of dams faces challenges from various natural and human factors, such as water pressure, seepage pressure, earthquakes, subsidence, vehicle loads, etc. These external factors will have an impact on the structure of the dam, especially the internal stress and strain changes of the dam will gradually accumulate, which may eventually lead to structural damage or collapse. Once the dam is unstable or breaks, the consequences will be disastrous, which may cause huge property losses and casualties. Therefore, safety monitoring and early warning of dams are of great significance.

[0003] Traditional dam monitoring technology mainly relies on monitoring water levels and flows. By setting water level warning lines, flood discharge is carried out when the water level exceeds the warning level, and water is stored when the water level is below the warning level. However, relying solely on water level monitoring cannot fully reflect the stress conditions inside the dam. Over time, the stress and strain inside the dam body will gradually accumulate under the impact of long-term water flow. At the same time, under the combined effects of various factors such as vehicle loads, earthquakes and leakage, small structural changes may occur inside the dam. These changes are usually difficult to detect by traditional monitoring methods, but they may bring potential damage risks to the dam, such as pipe bursts and crack expansion. Some sudden accidents are often caused by the failure to monitor and deal with the long-term accumulated stress or strain in time, which eventually leads to sudden instability.

[0004] In order to make up for the shortcomings of traditional monitoring methods, in recent years, dam health monitoring systems based on fiber optic sensing technology have gradually become a research hotspot. Fiber optic sensors can monitor a variety of physical quantities such as stress, strain, temperature, vibration, etc. with high precision, and have the advantages of corrosion resistance, high temperature resistance, and anti-electromagnetic interference, and are suitable for long-term work in harsh environments. By burying fiber optic 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 discovered in time. Fiber optic sensors can also monitor water levels in combination with changes in water temperature, deduce changes in water pressure near the dam body through the distribution of the temperature field, and then analyze the safety status of the dam body through a multi-dimensional stress-strain model. In addition, by deploying fiber optic sensors on the surface of the dam, it is also possible to monitor the immediate impact of dynamic external forces such as vehicle traffic and earthquakes on the dam structure.

[0005] At present, fiber optic monitoring technology is still rarely used in multi-dimensional monitoring of dams, and there is a lack of system design for coordinated monitoring of multiple physical quantities such as temperature, water level, vibration, and stress. Therefore, there is an urgent need for a multi-dimensional dam monitoring system combined with fiber optic sensors that can cover the needs of internal stress and strain monitoring of dams, temperature and water level inversion, vehicle load and vibration monitoring, and realize 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 dam monitoring abnormality early warning method and system based on optical fiber sensing; On the one hand, a dam monitoring abnormality early warning method based on optical fiber sensing is provided, including: 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; collect the temperature of the water surface inside the dam to be monitored, and determine the impact of water on the stress of the dam body based on the water surface temperature; collect the actual stress generated by vehicles on the dam body when vehicles pass over the dam to be monitored; Constructing a test model of the dam to be monitored, applying simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress; applying vehicle load to the test model to obtain vehicle load simulated 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.

[0007] On the other hand, a dam monitoring abnormality early warning system based on optical fiber sensing is provided, including: The acquisition 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; collect the temperature of the water surface inside the dam to be monitored, and determine the influence of water on the stress of the dam body based on the water surface temperature; collect the actual stress generated by vehicles on the dam body when vehicles pass above the dam to be monitored; The simulation module is configured to: construct a test model of the dam to be monitored, apply simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress; apply vehicle load to the test model to obtain vehicle load simulated 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.

[0008] The above technical solution has the following advantages or beneficial effects: The advantage of the present invention is that by deploying optical fiber sensors inside and outside the dam, real-time monitoring of physical parameters such as stress, strain, temperature, and vibration can be performed, and the health status of the dam can be accurately assessed, and potential structural hazards can be discovered in a timely manner. Optical fiber sensors have the characteristics of high sensitivity, anti-interference, and high temperature resistance, and are suitable for long-term operation in complex environments. Combined with a multi-dimensional data analysis model, the system realizes comprehensive monitoring and early warning of the dam structure, greatly improving the safety and operation efficiency of the dam. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0010] Figure 1 This is a flow chart of the method of embodiment 1.

[0011] Figure 2 This is a diagram of the overall structure of the system of Embodiment 1.

[0012] Figure 3 This is a schematic diagram of a monitoring module for internal stress and strain of a dam according to Embodiment 1.

[0013] Figure 4 Schematic diagram of the sensing device for dam surface stress in Example 1. DETAILED DESCRIPTION

[0014] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0015] Terminology explanation: 1. Fiber Bragg Grating (FBG) sensor: a fiber optic sensor used to detect changes in physical quantities such as stress, strain, and temperature inside a structure.

[0016] 2. Stress-strain model: By analyzing the optical fiber strain data, the distribution of internal forces in the structure is calculated.

[0017] 3. Temperature-water level inverse model: The temperature data measured by optical fiber is combined with the heat conduction equation to calculate the real-time water level of the reservoir and inversely deduce the stress on the dam.

[0018] 4. Vibration monitoring: used to monitor the vibration response of the dam surface and interior under earthquake or vehicle loads, and evaluate the dynamic safety of the dam structure.

[0019] Embodiment 1 This embodiment provides a dam monitoring abnormality early warning method based on optical fiber sensing; like Figure 1 As shown, the dam monitoring abnormality early warning method based on optical fiber sensing includes: S101: 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; collect the temperature of the water surface inside the dam to be monitored, and determine the influence of water on the stress of the dam body based on the water surface temperature; collect the actual stress generated by vehicles on the dam body when vehicles pass above the dam to be monitored; S102: constructing a test model of the dam to be monitored, applying simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress; applying vehicle load to the test model to obtain vehicle load simulated stress; S103: Construct a comprehensive characteristic 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, simulated earthquake vibration stress and vehicle load simulated stress; S104: 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.

[0020] Furthermore, the step S101: collecting stress and strain data of different locations of the dam to be monitored, and determining the stress distribution inside the dam body based on the stress and strain data, includes: like Figure 2 , Figure 3 and Figure 4 As shown, during the construction of the dam, optical fiber sensors are arranged on the surface of the dam body close to the water side; optical fiber sensors are buried on the upper surface of the dam body; The stress and strain data at different locations of the dam are collected through optical fiber sensors; ; (1) 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 fiber Bragg grating sensor. is the initial wavelength of the optical fiber sensor (the wavelength when there is no stress or strain).

[0021] Calculate the stress distribution inside the dam body. The stress calculation formula is: , (2) in, For time The stress of the moment, is the elastic modulus, is the strain value.

[0022] Furthermore, the step S101: collecting the temperature of the water surface inside the dam to be monitored, and determining the influence of water on the stress of the dam body based on the water surface temperature, includes: Using the optical fiber temperature sensor on the surface of the dam close to the water, the water temperature data at different depths and times are obtained in real time. The water temperature change formula is: , (3) 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; Combining the water temperature and the strain data of the dam body, the influence of the water level on the stress of the dam body is calculated through a formula.

[0023] Effect of water level on dam stress The calculation formula is as follows: , (4) 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 .

[0024] Furthermore, the S101: collecting the actual stress generated by vehicles on the dam body when vehicles pass over the dam to be monitored, includes: Fiber optic sensors are buried above the dam (on the road). 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: , (5) 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. It determines how the load is distributed on the dam surface and affects the calculation of stress.

[0025] Furthermore, the S102: constructing a test model of the dam to be monitored includes: During the test model making process, a test model identical to the actual dam structure was constructed at a scale of 1:10; The material of the test model is concrete, model C60, which is used to simulate the bearing capacity and stress distribution of the dam body. The fiber optic sensor is 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.

[0026] The test model uses permeable materials to simulate the leakage and piping of the dam body. The permeable material uses geotextile (model: TX160) to cover the bottom of the dam body to ensure that the impact of leakage and stress on the dam body is truly reflected in the test.

[0027] It should be understood that by deploying sensors on the test model for monitoring, it is possible to obtain data on the stress, strain and water pressure changes of the dam under different working conditions, and provide a basis for verifying the effectiveness of sensor deployment and monitoring models.

[0028] Further, the S102: applying simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress includes: 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: , (6) 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.

[0029] By adjusting the amplitude and frequency of vibration, the effects of earthquakes of different intensities on the stress distribution and strain changes of the dam structure are simulated, thereby evaluating the seismic performance and stability of the dam.

[0030] Furthermore, the S102: 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: , (7) 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.

[0031] This formula simulates the dynamic stress on the dam surface and internal structure caused by heavy-loaded vehicles.

[0032] Further, the S103: construct a comprehensive characteristic vector of the dam 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, the simulated earthquake vibration stress and the vehicle load simulated stress, wherein the comprehensive characteristic vector of the dam The expression is: .

[0033] Further, the S104: 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, wherein the higher the abnormal risk score, the lower the health of the dam; wherein 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; The training set is input into the dam risk prediction model to train the model. When the total loss function value of the model no longer decreases, the training is stopped to obtain the trained dam risk prediction model.

[0034] Furthermore, the total loss function of the model is specifically expressed as: ; in, is the error term (mean square error), 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.

[0035] Real-time monitoring of physical parameters such as stress, strain, temperature, vibration, etc. inside and outside the dam. This type of monitoring is mainly used in 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 the fields of flood control, power generation, agricultural irrigation, etc. Dams have long been under pressure from water flow, water pressure, sedimentation, infiltration, earthquakes and other aspects. Any minor structural changes may lead to serious consequences. Therefore, how to monitor the internal stress distribution, surface deformation, and dynamic changes of the surrounding water bodies of the dam in real time has become an important technical requirement for ensuring the safety of the dam.

[0036] Traditional monitoring methods usually focus on monitoring water level and flow, and cannot fully cover the detailed changes in the mechanical state, material fatigue, settlement and displacement inside the dam. The existing reservoir management system pays more attention to the warning water level, but not enough attention to the long-term accumulated stress, strain, vehicle traffic load, temperature effect, etc. inside the dam. Therefore, the present invention proposes a comprehensive monitoring system based on optical fiber sensing technology, which collects stress, strain, temperature, vibration and other data inside and outside the dam in real time by arranging optical fiber sensor arrays inside and outside the dam, and calculates through physical models to timely discover possible structural hazards and provide scientific early warning information.

[0037] Optical fiber sensors have the advantages of high sensitivity, high temperature resistance, anti-electromagnetic interference, and long-term use, 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, and at the same time derives the change in water level by measuring temperature changes, and combines stress-strain models and vibration monitoring models to evaluate the health status of the dam. This technology provides a new means of protection for the long-term safe operation of the dam, and has important technical value and social significance.

[0038] The purpose of the present invention is to provide a multi-dimensional monitoring system and method for dams based on optical fiber sensing technology. The system uses optical fiber sensors buried 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, thereby effectively detecting the health status of the dam and providing early warnings for potential hidden dangers. The system is verified by combining model experiments, and can simulate the dynamic impact of earthquakes and vehicles on the dam, and evaluate the safety of the dam under different conditions.

[0039] Strain, stress and temperature sensors are installed inside the dam to monitor the strain, stress and water temperature changes inside the dam in real time. At the same time, fiber optic sensors are installed on the surface of the dam and the road surface above to monitor surface strain, temperature changes and the impact of vehicle traffic on the dam.

[0040] For the surface area, the optical fiber is evenly laid along the surface of the dam body. In particular, in areas that may be greatly affected by external factors (such as the dam top and the water-facing surface), the laying density is appropriately increased (one optical fiber is laid at an interval of 20 cm) to accurately capture surface strain and temperature changes.

[0041] On the road surface above, optical fibers are pre-buried (one optical fiber is buried every 30 cm) to monitor the dynamic load effects caused by vehicle traffic in real time. All optical fiber sensors are connected to the central data processing system through optical cables to ensure efficient data collection and transmission. This precise and comprehensive layout provides a stable and reliable data foundation for subsequent monitoring.

[0042] The fiber optic sensor system collects data from the inside, surface and road surface of the dam in real time. Internal sensors monitor strain and stress, and the structural state of the dam is analyzed 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 of vehicle traffic on the dam. Data from all sensors are aggregated to the central data processing system for comprehensive analysis to evaluate the health and safety of the dam under different working conditions.

[0043] By building a scaled model similar to the actual dam structure, tests were conducted under various working conditions, including different water levels and earthquake simulations. Fiber optic sensors were deployed in the scaled model to record the strain, stress and vibration response of the model under different conditions. The test data were compared with the theoretical prediction results to verify the effectiveness of the sensor deployment and data analysis model, and the system was optimized based on the test results to ensure that the stability and safety of the dam can be accurately evaluated in practical applications.

[0044] Optical fibers are laid on the surface of the water side of the dam and installed in a hose with a hole diameter smaller than the set size, which can prevent foreign matter from entering while achieving reservoir water cleaning, thereby ensuring the detection sensing accuracy and service life of the optical fiber.

[0045] During the construction of the dam, fiber optic sensors were embedded in the core area of ​​the dam body. Fiber optic sensors include strain sensors, stress sensors and temperature sensors. Strain sensors are arranged in the main stress-bearing areas of the dam body to monitor the strain changes inside the dam body. Stress sensors are arranged in key parts of the surface of the dam body to capture the stress distribution of the dam structure. Temperature sensors are arranged near the water surface to measure water temperature changes in real time.

[0046] Fiber optic sensors are deployed on the surface of the dam to monitor surface strain and temperature changes. The fiber optic sensors should be evenly arranged to ensure that all key areas on the dam surface are covered.

[0047] Fiber optic sensors are buried in the road surface area above the dam to monitor the dynamic impact of vehicle traffic on the dam. The sensors should be placed in the center of the lane and under the wheel path to capture the stress and strain caused by vehicle loads.

[0048] Data logging: Real-time recording of fiber optic sensor data of the model under different operating conditions, including internal stress, strain, water level, temperature, and vibration response. Analyze the data to evaluate the stability and stress distribution of the model.

[0049] Verification and optimization: Compare the test data with the predicted results of the theoretical model to verify the effectiveness of the fiber optic sensor deployment scheme. Optimize the sensor deployment strategy and data analysis model based on the test results to improve the accuracy and reliability in practical applications.

[0050] To meet the special needs of dam monitoring, the PSO algorithm can be improved as follows: The inertia weight in the original PSO algorithm is usually fixed, which can easily lead to slow convergence of the algorithm or fall into local optimum. Improvement method: Use adaptive inertia weight, and the weight is dynamically adjusted with the number of iterations:

[0051] in, and are the maximum and minimum values ​​of the inertia weight, is the current iteration number, is the maximum number of iterations.

[0052] After laying the optical fiber on the surface of the water side of the dam, install a hose with a small aperture to prevent foreign matter from entering the optical fiber area and ensure the normal operation of the optical fiber.

[0053] The hose is designed to use the water flow from the reservoir to clean the optical fiber and keep it clean. The hose should ensure that the water flow evenly covers the surface of the optical fiber and effectively removes dirt and sediment.

[0054] Through the implementation of this system, the detection accuracy of the optical fiber sensor can be effectively guaranteed, its service life can be extended, and stable support can be provided for the structural health monitoring of the dam.

[0055] A scaled model similar to the actual dam was built to simulate the geometry and material properties of the actual dam. Fiber optic sensors were placed in the model to monitor changes in stress, strain, and water temperature. Vibrations were applied through an earthquake simulator to simulate the effects of earthquakes of different intensities on the dam, and changes in stress and strain were recorded. The dynamic effects of vehicle loads were simulated in the model, and sensors recorded dynamic stress changes caused by vehicle loads. The test data was compared 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.

[0056] Fiber optic sensors are deployed inside and outside the dam to monitor physical quantities such as stress, strain, temperature and vibration in real time.

[0057] The health status of the dam structure is analyzed through the stress-strain model, and the impact of earthquakes or vehicle loads is evaluated in combination with the vibration monitoring model. The stress-strain model is a mechanical model that describes the deformation behavior of materials under external forces. It evaluates the health status of materials or structures by analyzing the relationship between stress (internal force per unit area caused by external forces) and strain (the degree of deformation of the material).

[0058] 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.

[0059] Embodiment 2 This embodiment provides a dam monitoring abnormality early warning system based on optical fiber sensing, including: The acquisition 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; collect the temperature of the water surface inside the dam to be monitored, and determine the influence of water on the stress of the dam body based on the water surface temperature; collect the actual stress generated by vehicles on the dam body when vehicles pass above the dam to be monitored; The simulation module is configured to: construct a test model of the dam to be monitored, apply simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress; apply vehicle load to the test model to obtain vehicle load simulated 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.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in 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; Collect the temperature of the water surface inside the dam to be monitored, and determine the effect of water on the stress of the dam body based on the water surface temperature; Collect the actual stress generated by vehicles on the dam body when vehicles pass over the dam to be monitored; Constructing a test model of the dam to be monitored, applying simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress; applying vehicle load to the test model to obtain vehicle load simulated 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 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 2 is characterized in that: 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 body material under the action of external force. Represents the density of water, usually taken as , Represents the acceleration due to gravity, usually taken as .

4. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 3 is characterized in that: 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 vehicle load 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: 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.

6. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 5 is characterized in that: Applying simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress includes: 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.

7. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 6 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.

8. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 7 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: 。 9. The optical fiber sensing-based dam monitoring abnormality early warning method according to claim 8 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.

10. 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; Collect the temperature of the water surface inside the dam to be monitored, and determine the effect of water on the stress of the dam body based on the water surface temperature; Collect the actual stress generated by vehicles on the dam body when vehicles pass over the dam to be monitored; The simulation module is configured to: construct a test model of the dam to be monitored, apply simulated earthquake vibration to the test model to obtain simulated earthquake vibration stress; apply vehicle load to the test model to obtain vehicle load simulated 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