Unmanned aerial vehicle infrared and distributed optical fiber fused dam piping intelligent diagnosis method and system
Patent Information
- Application Number
- CN202510951263.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-10
- Publication Date
- 2025-10-17
AI Technical Summary
Existing technologies for dam piping detection have problems such as high manpower consumption, low accuracy, inability to monitor around the clock, high cost, and difficulty in achieving full coverage and rapid response.
Combining drone infrared imaging with distributed fiber optic sensors, multi-dimensional data fusion is used to achieve full coverage monitoring of the dam surface and deep layers. GPS timing modules are used to synchronize data, and BIM models are used for data visualization and anomaly determination.
It achieves efficient and accurate detection of dam piping, reduces costs, realizes all-weather monitoring and rapid response, improves detection accuracy and detection rate, and reduces dam safety hazards.
Smart Images
Figure CN120800677A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of safety monitoring of hydraulic engineering, and particularly relates to a dam piping intelligent diagnosis method and system based on fusion of unmanned aerial vehicle infrared and distributed optical fiber. BACKGROUND
[0002] The existence of piping phenomenon is an unignorable hidden danger for a dam. Piping phenomenon inevitably occurs during the long-term operation of a dam, and needs to be found and repaired in time to prevent the dam from collapsing. The principle of piping phenomenon detection is to detect abnormal temperature changes. When piping phenomenon occurs in a dam body or a river channel, the temperature of the seepage water is lower than that of the falling rain and the ground temperature. As long as the low-temperature area can be captured, the seepage or piping outlet can be determined by investigation. Traditional monitoring needs a large amount of manpower for regular investigation, which consumes a lot of manpower and time. The existing unmanned aerial vehicle equipped with infrared imaging temperature measurement technology is affected by environmental factors and dam surface material reflection, and is limited to surface monitoring, and the precision needs to be improved. Distributed optical fiber temperature sensors cannot cover positions where optical fibers cannot be laid and have high initial laying costs.
[0003] Therefore, a technology is needed to meet the following requirements: 1. maintaining the life cycle cost of the dam; 2. all-weather monitoring; 3. wide surface coverage for internal monitoring. Based on a multi-source data fusion algorithm, the precision and timeliness of piping risk assessment are improved. Wide-area coverage, accurate positioning and rapid response for dam piping detection are achieved. Compared with traditional technologies, this technology can improve the discovery rate of piping danger, truly achieve early discovery, early disposal and prevention of potential problems. SUMMARY
[0004] To overcome the above technical problems, the application provides a dam piping intelligent diagnosis method and system based on fusion of unmanned aerial vehicle infrared and distributed optical fiber, which combines distributed optical fiber sensors and unmanned aerial vehicle infrared imaging temperature measurement, and realizes multi-dimensional piping detection through “air-ground-internal” cooperation. The system is designed for data analysis to realize real-time display of monitoring data and timely discovery of piping phenomenon, realize real-time and accurate dam hidden danger detection and timely management, and is suitable for rapid positioning and risk assessment of piping phenomenon of earth and rockfill dams, dikes and other hydraulic structures.
[0005] The application adopts the following technical solutions: A dam piping intelligent diagnosis method and system based on fusion of unmanned aerial vehicle infrared and distributed optical fiber; The unmanned aerial vehicle infrared imaging temperature measurement subsystem comprises: A dam piping intelligent diagnosis method and system based on fusion of unmanned aerial vehicle infrared and distributed optical fiber; The unmanned aerial vehicle infrared imaging temperature measurement subsystem comprises: Data preprocessing is carried out, non-uniformity correction (NUC) is used to eliminate infrared sensor noise, and temperature-space mapping is used to generate a thermal map (HSV color space).
[0006] The contrast shooting plane temperature difference situation is used to preliminarily infer the pipe gushing occurrence area, and trigger abnormal early warning.
[0007] The distributed optical fiber temperature sensor temperature and distance measuring subsystem includes: According to the required accuracy and resolution, the optical fiber temperature sensor and the optical fiber type are selected, and the optical fiber is laid in a fence type in the dam body, and the encryption laying is laid in the key area (such as the interface of the sand layer). The distributed optical fiber demodulator is applied to convert the optical signal into temperature data, and the temperature value of each distance point is arranged in spatial order to form a continuous temperature-distance curve. Data preprocessing is carried out, and environmental interference is removed by wavelet denoising.
[0008] Abnormal judgment is carried out according to the set condition threshold, such as local temperature difference ΔT≥0.5℃ for 10 minutes. Time alignment is the core step to ensure the accurate association of multi-source data, and the two subsystems need to be time and space aligned to ensure the detection rate and accuracy of pipe gushing detection.
[0009] Eliminate the time difference of data acquisition to ensure the time sequence consistency of the unmanned aerial vehicle infrared image and the optical fiber temperature sensor data. Since the two subsystems work separately, time synchronization within a small error is required in the time dimension. In terms of hardware, time synchronization can be achieved by using a GPS time module to synchronize, integrating a GPS receiver in the unmanned aerial vehicle and the optical fiber demodulator, obtaining UTC time stamp through satellite signal, and then calibrating the local clock using pulse signal. Then, the interpolation algorithm is used to align the data with different sampling rates, and the low-frequency data is linearly interpolated to match the high-frequency time points.
[0010] A dam BIM model is established, and the unmanned aerial vehicle infrared image and the distributed optical fiber temperature sensor data are superimposed on the dam BIM model to provide temperature field distribution data for the BIM platform, generate a thermal map or a 3D temperature model, locate the temperature anomaly area when the surface low-temperature area and the deep temperature anomaly coincide, prompt the seepage risk, and timely generate a pipe gushing warning report. The report is displayed to the terminal through real-time data communication technology, and the relevant staff receives it in time for management.
[0011] The data acquisition module is used to collect dam surface temperature data and deep temperature change data, which also includes preprocessing of the collected data to maintain data accuracy.
[0012] The data processing module is used for time alignment of data collected by two subsystems respectively by using a GPS timing module in combination with an interpolation algorithm, so as to ensure the detection rate and accuracy of dam piping.
[0013] The visualization module is used for visualizing the temperature abnormal area by combining the collected data with a BIM dam model, so that relevant staff can see the specific position of piping hazard in real time and intuitively and timely manage the piping hazard.
[0014] Compared with the prior art, the present application has the following beneficial effects: The unmanned aerial vehicle infrared and the distributed optical fiber are combined to realize multi-aspect complementation, on the spatial dimension, the surface wide-area coverage of the unmanned aerial vehicle and the internal deep-layer detection of the distributed optical fiber realize full-aspect dead-angle-free detection, on the applicable dimension, the distributed optical fiber compensates for the limitation of weather conditions and the monitoring surface material on the unmanned aerial vehicle temperature measurement, and realizes all-weather monitoring, on the cost, the combination of the low single cost of the unmanned aerial vehicle and the low long-term operation and maintenance cost of the optical fiber realizes the cost optimization of the dam whole life cycle, and on the monitoring efficiency, the periodic inspection of the unmanned aerial vehicle and the real-time continuous monitoring of the distributed optical fiber realize the combination of normal monitoring and emergency response.
[0015] By time alignment of data of two subsystems, the error between the data of the two subsystems is reduced, piping phenomenon is detected together, and the detection rate and accuracy of piping are improved. Then, the data is visualized, so that relevant staff can intuitively observe the real-time change and position of temperature, thereby facilitating the management of dam piping, and reducing the risk of dam collapse and collapse. BRIEF DESCRIPTION OF DRAWINGS
[0016] Figure 1 The figure is a schematic diagram of the functional modules of the present application; Figure 2 The figure is a schematic diagram of the method flow of the present application; Figure 3 The figure is a schematic diagram of the wavelet processing flow of the temperature-distance curve obtained by the distributed optical fiber temperature sensor; Figure 4 The figure is a schematic diagram of the time-space alignment of two subsystems by using a GPS timing module. DETAILED DESCRIPTION
[0017] Embodiments of the present application will be described in detail below, and examples of the embodiments are shown in the drawings, wherein the same or similar reference signs represent the same or similar elements or elements having the same or similar functions throughout. If not specifically stated, the raw materials and equipment used can be purchased from the market or are commonly used in the art. The methods in the embodiments are conventional methods in the art, unless otherwise specified. The embodiments described below by reference to the drawings are exemplary and are used to explain the present application, and cannot be understood as limiting the present application.
[0018] The unmanned aerial vehicle infrared and distributed optical fiber fusion dam piping intelligent diagnosis method realizes the multi-dimensional detection of dam piping "space-ground-internal". The specific steps are as follows: S1: Design the unmanned aerial vehicle to carry the infrared imaging subsystem to patrol the dam surface and the distributed optical fiber temperature sensor subsystem to be laid inside the dam, realize the full coverage of the dam surface and deep layer, and improve the detection rate and accuracy of piping.
[0019] S11: The unmanned aerial vehicle subsystem includes the following steps: S111: According to the dam surface conditions (terrain, surface material, surrounding environment), design the unmanned aerial vehicle patrol route, set the unmanned aerial vehicle patrol cycle according to the demand, and create the patrol task for the unmanned aerial vehicle.
[0020] S112: Use the unmanned aerial vehicle to carry the infrared imaging, block the area about 5 meters above the dam according to the patrol route, and take pictures. Infrared imaging temperature measurement, visible light imaging auxiliary identification environment situation distinguish interference factors.
[0021] S113: Data preprocessing is performed on the collected images, non-uniformity correction (NUC) is used to eliminate infrared sensor noise, and temperature-space mapping is used to generate a thermal map (HSV color space). The obtained data is close to the actual situation, and the error is reduced.
[0022] S114: Combine the obtained thermal map and plan view, compare the temperature difference, and preliminarily infer the piping occurrence area.
[0023] The distributed optical fiber subsystem includes the following steps: S12: The distributed optical fiber temperature sensor temperature measurement and distance measurement subsystem includes: S121: According to the required accuracy and resolution, select the optical fiber temperature sensor and the type of optical fiber. Single-mode bending-resistant optical fiber can be used for long-distance monitoring (such as dam length >1km). According to the daily temperature range of the dam, set the temperature measurement range of the optical fiber temperature sensor. In the dam body, 1.5m deep, along the dam axis, the interval is 5m, the transverse around the dam contour is laid, the interval is 2m, and the key area (such as the sand-soil layer interface) is laid with high density. S122: Apply the distributed optical fiber demodulator to receive the Raman scattering signal (Stokes light and anti-Stokes light) to calculate the temperature of each distance point. The specific calculation formula is: h is the Planck constant, kB is the Boltzmann constant, Δν is the frequency offset, IAS is the anti-Stokes light intensity, IS is the Stokes light intensity, and C is the optical fiber material constant.
[0024] The specific formula for calculating the distance is: Where: c is the speed of light; t is the time difference; n is the fiber refractive index.
[0025] The temperature values of each distance point are arranged in spatial order to form a continuous temperature-distance curve. S123: data preprocessing is performed, and environmental interference is removed by using wavelet denoising. Since the light temperature data is usually a low-frequency slowly varying signal, db5 or sym6 can be selected as the wavelet basis. Day and night temperature fluctuations can be predicted based on a deep learning model to remove the influence of day and night temperature difference.
[0026] S124: abnormality judgment is performed according to the set condition threshold, which can be divided into multiple levels of early warning: primary early warning is local ΔT≥0.5℃ for 10 minutes; senior early warning is ΔT≥1.0℃ and temperature gradient ΔT / Δx≥1℃ / m for 10 minutes.
[0027] S2: time alignment is the core step to ensure accurate association of multi-source data. Time alignment is required for the two subsystems to ensure the detection rate and accuracy of pipe flow detection.
[0028] To ensure data time alignment, the time difference of data acquisition needs to be eliminated to ensure the time sequence consistency of the unmanned aerial vehicle infrared image and the fiber temperature sensor data. Since the two subsystems work separately, time synchronization within a small error is required in the time dimension. In terms of hardware, time synchronization can be achieved by using a GPS time module for synchronization. A GPS receiver is integrated into the unmanned aerial vehicle and the fiber demodulator. UTC timestamp is obtained through satellite signal, and local clock is calibrated using pulse signal. In terms of software, linear interpolation algorithm is used to match the sampling frequencies of the two subsystems. Usually, the sampling frequency of the unmanned aerial vehicle is lower than that of the fiber temperature sensor, so the infrared data of the unmanned aerial vehicle is matched with the sampling rate of the fiber data through linear interpolation. The specific interpolation formula is: The time points of the unmanned aerial vehicle infrared data are ti and ti+1, and the corresponding temperature values are T(ti) and T(ti+1). The temperature increment is calculated as ΔT=T(ti+1)-T(ti), which represents the temperature change between adjacent time points. The time point of the fiber data is t (between ti and ti+1), and the interpolated temperature value Tinterp(t)=T(ti)+α*ΔT is obtained.
[0029] S3: According to the actual dam condition, a dam BIM model is established, which can include material properties, historical temperature data, and then the unmanned aerial vehicle infrared image and distributed optical fiber temperature sensor data are superimposed on the dam BIM model to provide temperature field distribution data for the BIM platform, generate a thermal map or a 3D temperature model, render low-temperature zones (<15℃) as blue and high-temperature zones (>25℃) as red, gradually change according to the temperature gradient, observe whether the surface temperature abnormal area and the deep temperature abnormal area coincide according to the early warning information, locate the temperature abnormal area, prompt the seepage risk, and timely generate a piping phenomenon report for piping warning, display the report to the terminal through real-time data communication technology, and the relevant staff receives and timely manages. Early detection, accurate positioning and fast response of piping risk are achieved.
[0030] The unmanned aerial vehicle infrared and distributed optical fiber fusion dam piping intelligent diagnosis system specifically includes: The data acquisition module is used to collect dam surface temperature data and deep layer temperature change data, and also includes preprocessing of the collected data to maintain data accuracy.
[0031] The data processing module is used to time-align the data collected by the two subsystems by using the GPS timing module and interpolation algorithm to align different sampling frequencies, so as to ensure the detection rate and accuracy of dam piping.
[0032] The visualization module is used to combine the collected data with the BIM dam model to achieve temperature abnormal area visualization, so that the relevant staff can see the specific location of piping hidden danger in real time and intuitively manage in time.
[0033] The system realizes efficient and accurate detection of dam piping, covering full-range, full-weather, and full-life cycle multidimensional detection. The data acquisition realizes full-dimensional coverage of surface-deep layer temperature, supports normal inspection and emergency triggering, and provides accurate and interference-free data for subsequent processing. The data processing solves the time synchronization problem of multi-source data by using GPS timing and interpolation algorithm, so that the data of the two subsystems are time-synchronized, ensuring accurate abnormal detection and discrimination. Visualization enables the relevant staff to intuitively and quickly locate the piping location, improving decision-making and management efficiency. The system comprehensively ensures the safety and stability of dam operation.
[0034] Although the embodiments of the present application have been shown and described, those skilled in the art can understand that various changes, modifications, replacements and deformations can be made to the above embodiments without departing from the principles and purposes of the present application, and the scope of the present application is defined by the claims and their equivalents.
Claims
1. The intelligent diagnosis method for dam piping by integrating UAV infrared and distributed optical fiber is characterized by: include: Design the drone inspection route and inspection cycle, lay distributed fiber optic temperature sensors, set up the drone temperature measurement subsystem and the fiber optic temperature and distance measurement subsystem to pre-process the collected data; The GPS timing modules in the two subsystems are used to time-align the data of the two subsystems, and the linear interpolation algorithm is used to match the sampling frequencies of the two subsystems. By combining the aligned data of the two subsystems on the established dam BIM model, the temperature changes can be visualized, abnormal temperature areas can be located, and timely warnings can be issued.
2. The intelligent diagnosis method for dam piping based on the integration of drone infrared and distributed optical fiber according to claim 1 is characterized in that: The drone inspection work includes: drones taking pictures of the area in blocks, infrared thermal imaging to detect temperature, and visible light imaging to assist in identifying environmental interference; Single-mode bend-resistant optical fiber is used, and the temperature measurement range of the optical fiber temperature sensor is set according to the daily temperature range of the dam; the sensor is arranged longitudinally along the dam axis at a depth of 1.5m inside the dam body and laterally around the dam body contour line.
3. The intelligent diagnosis method for dam piping based on the integration of drone infrared and distributed optical fiber according to claim 2 is characterized in that: The Raman scattering signal is received by a distributed optical fiber demodulator to calculate the temperature at each distance point. The calculation formula is: Where h is Planck's constant, kB is Boltzmann's constant, Δν is the frequency offset, IAS is the anti-Stokes light intensity, IS is the Stokes light intensity, and C is the fiber material constant; The formula for calculating distance is: Where c is the speed of light, t is the time difference, and n is the refractive index of the optical fiber; The temperature values at each distance point are arranged in spatial order to form a continuous temperature-distance curve.
4. The intelligent diagnosis method for dam piping based on the integration of drone infrared and distributed optical fiber according to claim 2 is characterized in that: Data preprocessing is performed on the data collected by the drone. Non-uniformity correction (NUC) is used to eliminate infrared sensor noise, and temperature-space mapping is used to generate thermal maps (HSV color space). For the data of the fiber optic temperature sensor, wavelet noise reduction is used to remove environmental interference.
5. The intelligent diagnosis method for dam piping based on the integration of UAV infrared and distributed optical fiber according to claim 1 is characterized in that: A GPS receiver is integrated into the drone and fiber optic demodulator. After obtaining the UTC timestamp through the satellite signal, the local clock is calibrated using the pulse signal. Then, an interpolation algorithm is used to align data with different sampling rates. The low-frequency drone data is linearly interpolated to match the time point of the fiber optic temperature sensor. The interpolation formula is: The time points of the drone infrared data are ti and ti+1, and the corresponding temperature values are T(ti) and T(ti+1). The temperature increment is calculated as: ΔT=T(ti+1)−T(ti), which represents the temperature change between adjacent time points. The time point of the fiber optic data is t (between ti and ti+1), and the interpolated temperature value is Tinterp(t)=T(ti)+α*ΔT.
6. The intelligent diagnosis method for dam piping based on the integration of drone infrared and distributed optical fiber according to claim 1 is characterized in that: Establish a dam BIM model, overlay drone infrared images and distributed fiber optic temperature sensor data onto the dam BIM model, provide temperature field distribution data for the BIM platform, generate a thermal map or 3D temperature model, and locate the abnormal temperature area when the surface low temperature area coincides with the deep temperature anomaly, prompt the seepage risk, and issue a timely piping warning and generate a piping phenomenon report. The report is displayed to the terminal through real-time data communication technology, and relevant staff receive it and take timely measures.
7. The intelligent dam piping diagnosis system that integrates UAV infrared and distributed optical fiber is characterized by: It includes data acquisition module, data processing module and visualization module; the data acquisition module is used to collect the surface temperature data of the dam and the deep temperature change data, which also includes pre-processing of the collected data to maintain the accuracy of the data; The data processing module uses the GPS timing module in combination with the interpolation algorithm to synchronize the data collected by the two subsystems with different sampling frequencies to ensure the detection rate and accuracy of dam piping; the visualization module is used to combine the collected data with the BIM dam model to achieve visualization of temperature anomaly areas, so that relevant staff can see the specific location of the piping hazard in real time and take timely measures to deal with it.