A method for early warning and location of dam piping based on satellite and seepage pressure monitoring

Abstract

This invention specifically relates to a method for early warning and location of piping in dams based on satellite and piezometer monitoring. This method acquires satellite imagery data and seepage pressure data of the dam area; after preprocessing and analyzing the data, a correlation model between settlement and seepage intensity is established to understand the distribution and variation of seepage pressure within the dam; satellite data and piezometer data are fused, and a dam seepage monitoring model is constructed using an improved D-S evidence theory to calculate a comprehensive anomaly index; when monitoring data is abnormal, a corresponding level of early warning signal is issued. This method for early warning and location of piping in dams based on satellite and piezometer monitoring has a wide coverage, high monitoring accuracy, and strong timeliness, enabling timely identification of dam seepage anomalies and achieving efficient monitoring and accurate early warning of dam seepage. It is applicable to seepage monitoring of various types of dams.

Description

Technical Field

[0001] This invention relates to the field of satellite remote sensing technology, and in particular to a method for early warning and location of piping in dams based on satellite and seepage pressure monitoring. Background Technology

[0002] Dams, as important water conservancy projects, play a vital role in flood control, power generation, and irrigation. However, during dam operation, seepage is one of the key factors affecting its safety and stability. If seepage is not effectively monitored and controlled, it may lead to seepage damage to the dam body, or even cause serious accidents such as dam failure, resulting in huge casualties and property losses.

[0003] Traditional methods for monitoring dam seepage mainly rely on manual inspections and data collection at a limited number of monitoring points. These methods have significant limitations:

[0004] On the one hand, manual inspections are inefficient and greatly affected by the experience and sense of responsibility of the personnel, making it difficult to achieve comprehensive and continuous monitoring of the dam;

[0005] On the other hand, the number of monitoring points is limited and their distribution is uneven, which can only reflect the seepage situation in local areas and cannot fully grasp the seepage status of the entire dam, making it easy to miss potential seepage hazards.

[0006] In addition, traditional monitoring methods suffer from data processing delays and untimely early warnings, often only being detected when seepage problems become more severe, thus missing the best opportunity for intervention.

[0007] With the continuous expansion of dam construction scale and the increase in operating years, the requirements for seepage monitoring are becoming increasingly stringent. Traditional monitoring methods are no longer sufficient to meet the needs of modern dam safety management. Developing a dam seepage monitoring and early warning technology capable of large-scale, high-precision, and real-time monitoring is of significant practical importance.

[0008] Based on the above, this invention proposes a method for early warning and location of dam piping based on satellite and seepage pressure monitoring. Summary of the Invention

[0009] To overcome the shortcomings of existing technologies, this invention provides a simple and efficient method for early warning and location of dam piping based on satellite and seepage pressure monitoring.

[0010] This invention is achieved through the following technical solution:

[0011] A dam piping early warning and location system based on satellite and seepage pressure monitoring, characterized by the following steps:

[0012] Step S1: Use InSAR and multispectral satellites to acquire satellite image data of the dam area. The resolution of InSAR satellite images shall be no less than 5 meters × 20 meters, and the resolution of multispectral satellite images shall be no less than 10 meters.

[0013] In step S1, the data acquisition cycle of the InSAR satellite is 12 days, the data acquisition cycle of the multispectral satellite is 16 days, and the data acquisition cycle is increased to 5-7 days during the flood season;

[0014] Satellite imagery data of the dam area includes data on surface deformation, water distribution, and vegetation cover.

[0015] Step S2: At key parts of the dam, including the dam body, dam foundation and seepage prevention body, vibrating wire piezometers are installed to collect seepage pressure data in real time through 4G / 5G wireless transmission modules, with a sampling frequency of 1 time / minute.

[0016] Step S3: Preprocess the satellite image data, including:

[0017] Interferogram generation, phase unwrapping, and terrain correction of InSAR satellite data, as well as atmospheric correction, geometric fine correction, and denoising of multispectral satellite data, to improve data quality;

[0018] In step S3, for InSAR satellite data, GAMMA software is used to generate interferograms, Goldstein filtering is used to suppress noise, the filtering window is 32×32, the minimum cost flow method is used for phase unwrapping, and terrain correction is performed in combination with DEM data, with a resolution of 12.5 meters.

[0019] For multispectral satellite data, the FLAASH module of ENVI software was used with a 6S model for atmospheric correction; geometric fine correction was performed using no less than 10 control points with an error not exceeding 5 meters; and wavelet threshold denoising algorithm (db4 wavelet, 3 decomposition layers) was used to remove noise from the image.

[0020] Step S4: Filter the data collected by the vibrating wire piezometer, remove data that exceeds ±5% of the range, and remove outliers after verification.

[0021] In step S4, the data collected by the vibrating wire piezometer is verified by Kalman filtering smoothing, and the process noise variance is 0.01.

[0022] Step S5: Analyze the preprocessed satellite data. Use the threshold method to extract humidity anomaly areas on the dam surface where the Normalized Difference Water Index (NDWI) exceeds 0.3. Use time-series InSAR technology to calculate the cumulative settlement with an accuracy of ±2 mm / year, and obtain the dam settlement time-series curve. Combine the dam survey report and hydrological yearbook data to establish a correlation model (R0) between settlement and seepage intensity.2 ≥0.85);

[0023] The data from the vibrating wire piezometer were analyzed to calculate the daily rate of change and spatial gradient difference of seepage pressure, so as to understand the distribution and variation law of seepage pressure inside the dam.

[0024] If the daily rate of change of seepage pressure exceeds 5%, or the difference between adjacent measuring points exceeds 0.2 MPa, it is considered abnormal.

[0025] In step S5, MATLAB software is used to perform statistical analysis on the data from the vibrating wire piezometer, plot the piezometer-time process line and the piezometer-water level relationship curve, and calculate the characteristic parameters of the seepage pressure, including the average, maximum and minimum values. The t-test method is used with a significance level of 0.05 to analyze whether the trend of the characteristic parameters of the seepage pressure over time is significant.

[0026] Step S6: Import satellite data and seepage pressure data into the data fusion platform, and fuse satellite data and piezometer data based on spatiotemporal alignment. Control the spatial distance between the center point of the satellite pixel and the piezometer deployment point to within 10 meters, and control the time synchronization error to within 1 hour.

[0027] An improved DS evidence theory was used to construct a dam seepage monitoring model, and a comprehensive anomaly index was calculated using the model.

[0028] In step S6, the raster-format satellite data is first converted into vector point data, the spatial coordinates of the seepage pressure data are projected and transformed using UTM, and then imported into a data fusion platform developed based on the GeoPandas library of Python; during the fusion process, the basic probability assignment function is determined by the Gaussian membership function.

[0029] Step S7: Based on the dam seepage monitoring model and the preset three-level early warning threshold, when the monitoring data is abnormal, an early warning signal of the corresponding level is issued to remind the operation and maintenance personnel to conduct on-site verification and handling according to the level of the early warning signal.

[0030] Meanwhile, the processing results are fed back to the dam seepage monitoring model, and the model parameters and early warning thresholds are optimized using a BP neural network; during model optimization, the learning rate is 0.01 and the number of iterations is 1000.

[0031] In step S7, the alarm mechanism of the dam seepage monitoring model is divided into three levels:

[0032] When the abnormal index exceeds 0.6 but is less than 0.8, a Level I warning will be activated.

[0033] When the anomaly index of a Level II warning exceeds 0.8 but is less than 1.0, a Level II warning is activated.

[0034] When the abnormal index of a Level 1 warning exceeds 1.0, a Level 3 warning is activated.

[0035] A dam piping early warning and location system based on satellite and seepage pressure monitoring, used to implement the above method, includes:

[0036] The data acquisition module is responsible for acquiring satellite imagery data of the dam area using InSAR and multispectral satellites. By deploying vibrating wire piezometers at key parts of the dam, including the dam body, dam foundation, and seepage prevention structure, it collects seepage pressure data in real time through a 4G / 5G wireless transmission module at a sampling frequency of 1 time / minute.

[0037] The data preprocessing module is responsible for generating interferograms of InSAR satellite data, performing phase unwrapping and terrain correction; performing atmospheric correction, geometric fine correction and denoising on multispectral satellite data to improve data quality; and screening data collected by vibrating wire piezometers, removing data exceeding ±5% of the range, and removing outliers after verification.

[0038] The data analysis module is responsible for analyzing the preprocessed satellite data. It uses a threshold method to extract areas of abnormal humidity on the dam surface where the Normalized Difference Water Index (NDWI) exceeds 0.3. It then uses time-series InSAR technology to calculate cumulative settlement with an accuracy of ±2 mm / year, obtaining a dam settlement time-series curve. Combining the dam survey report and hydrological yearbook data, it establishes a correlation model between settlement and seepage intensity (R0). 2 ≥0.85);

[0039] Meanwhile, the data from the vibrating wire piezometer were analyzed to calculate the daily rate of change and spatial gradient difference of seepage pressure, so as to understand the distribution and variation law of seepage pressure inside the dam.

[0040] If the daily rate of change of seepage pressure exceeds 5%, or the difference between adjacent measuring points exceeds 0.2 MPa, it is considered abnormal.

[0041] The data fusion and modeling module is responsible for fusing satellite data and piezometer data based on spatiotemporal alignment, controlling the spatial distance between the satellite pixel center point and the piezometer deployment point to within 10 meters, and controlling the time synchronization error to within 1 hour; and constructing a dam seepage monitoring model (model accuracy ≥ 85%) using an improved DS evidence theory, and calculating the comprehensive anomaly index through the model;

[0042] The feedback and optimization module is responsible for issuing warning signals of the corresponding level when monitoring data is abnormal, based on the dam seepage monitoring model and the preset three-level warning thresholds, to remind operation and maintenance personnel to conduct on-site verification and handling according to the warning signal level.

[0043] Meanwhile, the processing results are fed back to the dam seepage monitoring model, and the model parameters and early warning thresholds are optimized using a BP neural network; during model optimization, the learning rate is 0.01 and the number of iterations is 1000.

[0044] A device for early warning and location of dam piping based on satellite and seepage pressure monitoring is characterized by comprising a memory and a processor; the memory is used to store a computer program, and the processor is used to execute the computer program to implement the above-mentioned method steps.

[0045] A readable storage medium, characterized in that: a computer program is stored on the readable storage medium, and the computer program, when executed by a processor, implements the above-described method steps.

[0046] The beneficial effects of this invention are: the dam piping early warning and location method based on satellite and seepage pressure monitoring has a wide coverage, high monitoring accuracy, and strong timeliness. By performing spatiotemporal registration and feature fusion of macroscopic data acquired by satellite and microscopic data collected by piezometer, it can identify dam seepage anomalies in a timely manner, realize efficient monitoring and accurate early warning of dam seepage, provide strong protection for the safe operation of dams, and is not significantly affected by complex terrain and harsh environment, making it suitable for seepage monitoring of various types of dams. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Appendix Figure 1 This is a schematic diagram of the dam piping early warning and location method based on satellite and seepage pressure monitoring according to the present invention. Detailed Implementation

[0049] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions in the embodiments of this invention will be clearly and completely described below in conjunction with the embodiments of this invention. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0050] This method for early warning and location of dam piping based on satellite and seepage pressure monitoring includes the following steps:

[0051] Step S1: Use InSAR and multispectral satellites to acquire satellite image data of the dam area. The resolution of InSAR satellite images shall be no less than 5 meters × 20 meters, and the resolution of multispectral satellite images shall be no less than 10 meters.

[0052] The resolution and shooting cycle of satellite imagery are determined based on the size of the dam and monitoring requirements.

[0053] In step S1, the data acquisition cycle of the InSAR satellite is 12 days, the data acquisition cycle of the multispectral satellite is 16 days, and the data acquisition cycle is increased to 5-7 days during the flood season;

[0054] Satellite imagery data of the dam area includes data on surface deformation, water distribution, and vegetation cover.

[0055] During the data collection process, the cloud cover rate was monitored in real time by meteorological satellites (photographs were taken when the cloud cover was ≤20%) to ensure the acquisition of clear and complete satellite image data.

[0056] Step S2: At key parts of the dam, including the dam body, dam foundation and seepage prevention body, vibrating wire piezometers are installed to collect seepage pressure data in real time through 4G / 5G wireless transmission modules, with a sampling frequency of 1 time / minute.

[0057] Piezometers are deployed in key areas such as the dam body (one layer every 10 meters along the dam height, with 5-8 measuring points per layer), the dam foundation (within 5-10 meters of the dam axis, spaced 10-20 meters apart), and the anti-seepage structure (with denser deployment within 3 meters of the water-facing side, spaced 5 meters apart). Piezometers are arranged according to the "grid + key point" deployment principle to reflect the seepage conditions in different areas (such as the contact surface between the dam body and bedrock, and the joints of the anti-seepage wall).

[0058] The type of piezometer selected is a vibrating wire piezometer (measuring range 0-1.6MPa, resolution 0.001MPa), which is installed by drilling (hole diameter 110mm, burial depth 1 meter below the measuring point). During installation, ensure that the sensor axis is consistent with the water flow direction, and backfill with graded sand and gravel (particle size 2-5mm) for sealing.

[0059] Step S3: Preprocess the satellite image data, including:

[0060] Interferogram generation, phase unwrapping, and terrain correction of InSAR satellite data, as well as atmospheric correction, geometric fine correction, and denoising of multispectral satellite data, to improve data quality;

[0061] In step S3, for InSAR satellite data, GAMMA software is used to generate interferograms, Goldstein filtering is used to suppress noise, the filtering window is 32×32, the minimum cost flow method is used for phase unwrapping, and terrain correction is performed in combination with DEM data, with a resolution of 12.5 meters.

[0062] For multispectral satellite data, the FLAASH module of ENVI software was used with a 6S model for atmospheric correction; geometric fine correction was performed using no less than 10 control points with an error not exceeding 5 meters; and wavelet threshold denoising algorithm (db4 wavelet, 3 decomposition layers) was used to remove noise from the image.

[0063] Step S4: Filter the data collected by the vibrating wire piezometer, remove data that exceeds ±5% of the range, and remove outliers after verification.

[0064] In step S4, the data collected by the vibrating wire piezometer is transmitted to the cloud database (using MySQL cluster storage) via a 4G / 5G module. The data is stored in real time (with a delay of less than 10 seconds), and Kalman filtering is used for smoothing and verification. The process noise variance is 0.01. At the same time, data filtering is performed (data exceeding the 0-1.6MPa range ±5% is removed) to eliminate abnormal data caused by instrument failure, electromagnetic interference, and other factors.

[0065] Step S5: Analyze the preprocessed satellite data. Use the threshold method to extract humidity anomaly areas on the dam surface where the Normalized Difference Water Index (NDWI) exceeds 0.3. Use time-series InSAR technology to calculate the cumulative settlement with an accuracy of ±2 mm / year, and obtain the dam settlement time-series curve. Combine the dam survey report and hydrological yearbook data to establish a correlation model (R0) between settlement and seepage intensity. 2 ≥0.85);

[0066] The data from the vibrating wire piezometer were analyzed to calculate the daily rate of change and spatial gradient difference of seepage pressure, so as to understand the distribution and variation law of seepage pressure inside the dam.

[0067] If the daily rate of change of seepage pressure exceeds 5%, or the difference between adjacent measuring points exceeds 0.2 MPa, it is considered abnormal.

[0068] In step S5, MATLAB software is used to perform statistical analysis on the data from the vibrating wire piezometer, plot the piezometer-time process line and the piezometer-water level relationship curve, and calculate the characteristic parameters of the seepage pressure, including the average, maximum and minimum values. The t-test method is used with a significance level of 0.05 to analyze whether the trend of the characteristic parameters of the seepage pressure over time is significant.

[0069] Step S6: Import satellite data and seepage pressure data into the data fusion platform, and fuse satellite data and piezometer data based on spatiotemporal alignment. Control the spatial distance between the center point of the satellite pixel and the piezometer deployment point to within 10 meters, and control the time synchronization error to within 1 hour.

[0070] An improved DS evidence theory was used to construct a dam seepage monitoring model (model accuracy ≥ 85%), and a comprehensive anomaly index was calculated using the model.

[0071] In step S6, the raster-format satellite data is first converted into vector point data, the spatial coordinates of the seepage pressure data are projected and transformed using UTM, and then imported into a data fusion platform developed based on the GeoPandas library of Python; during the fusion process, the basic probability assignment function is determined by the Gaussian membership function.

[0072] Step S7: Based on the dam seepage monitoring model and the preset three-level early warning threshold, when the monitoring data is abnormal, an early warning signal of the corresponding level is issued to remind the operation and maintenance personnel to conduct on-site verification and handling according to the level of the early warning signal.

[0073] Meanwhile, the processing results are fed back to the dam seepage monitoring model, and the model parameters and early warning thresholds are optimized using a BP neural network; during model optimization, the learning rate is 0.01 and the number of iterations is 1000.

[0074] In step S7, the alarm mechanism of the dam seepage monitoring model is divided into three levels:

[0075] When the abnormal index exceeds 0.6 but is less than 0.8, a Level I warning will be activated.

[0076] When the anomaly index of a Level II warning exceeds 0.8 but is less than 1.0, a Level II warning is activated.

[0077] When the abnormal index of a Level 1 warning exceeds 1.0, a Level 3 warning is activated.

[0078] The warning threshold setting and early warning are based on the dam's design parameters (such as design seepage flow and allowable seepage gradient), historical operating data (maximum seepage pressure value in the past 5 years), and relevant standards and specifications such as the "Design Code for Concrete Gravity Dams" (SL319-2018). Combined with the ROC curve analysis of the dam seepage monitoring model (taking the point with the maximum Youden index as the optimal threshold), the seepage warning thresholds for different regions (dam body, dam foundation, seepage prevention body) and different time periods (flood season and non-flood season) are customized (e.g., the dam foundation seepage pressure threshold during the flood season is increased by 20% compared to the non-flood season).

[0079] When the monitored data exceeds the warning threshold, the monitoring system automatically issues a warning signal and notifies relevant personnel (including dam management personnel, operation and maintenance team, and emergency command center) through SMS (using SMS gateway API interface), audible and visual alarms (loudness of on-site audible and visual alarms ≥85dB), platform pop-ups, etc.

[0080] Based on the warning signal level, maintenance personnel will arrive at the site within 1-6 hours (6 hours for Level I, 3 hours for Level II, and 1 hour for Level III) with drones (equipped with high-definition and thermal infrared cameras) for verification. The drones will first conduct a large-scale rapid survey (flying at an altitude of 100 meters, with a resolution of 0.1 meters) to identify surface seepage traces (such as wet zones and piping outlets). Simultaneously, a portable piezometer (such as the TS-800 model, with a measurement accuracy of ±0.01 MPa) will be used to verify the seepage pressure data. Then, ground-penetrating radar will be used to perform a profile scan of the abnormal area (scanning speed 5 km / h, sampling rate 100 MHz) to determine the location and cause of the seepage anomaly (such as damage to the anti-seepage structure or cracks in the dam body). Combined with ground-penetrating radar (such as the SIR-4000 model, with a detection depth ≥5 meters), the internal structure of the dam body will be probed.

[0081] Appropriate treatment measures were taken. Small leaks were sealed with quick-setting concrete (setting time < 30 minutes), while larger leaks were treated with curtain grouting (cement grout water-cement ratio 1:1-0.5:1) and increased density of drainage holes. After treatment, continuous monitoring was conducted for 30 days to confirm that the seepage index had returned to normal (seepage pressure value stabilized below 80% of the threshold). The treatment results were recorded and fed back to the monitoring system. A BP neural network (5 input layer nodes, 10 hidden layer nodes, and 1 output layer node) was used to optimize the monitoring model and early warning threshold.

[0082] This dam piping early warning and location system, based on satellite and seepage pressure monitoring, is used to implement the above methods, including:

[0083] The data acquisition module is responsible for acquiring satellite imagery data of the dam area using InSAR and multispectral satellites. By deploying vibrating wire piezometers at key parts of the dam, including the dam body, dam foundation, and seepage prevention structure, it collects seepage pressure data in real time through a 4G / 5G wireless transmission module at a sampling frequency of 1 time / minute.

[0084] The data preprocessing module is responsible for generating interferograms of InSAR satellite data, performing phase unwrapping and terrain correction; performing atmospheric correction, geometric fine correction and denoising on multispectral satellite data to improve data quality; and screening data collected by vibrating wire piezometers, removing data exceeding ±5% of the range, and removing outliers after verification.

[0085] The data analysis module is responsible for analyzing the preprocessed satellite data. It uses a threshold method to extract areas of abnormal humidity on the dam surface where the Normalized Difference Water Index (NDWI) exceeds 0.3. It then uses time-series InSAR technology to calculate cumulative settlement with an accuracy of ±2 mm / year, obtaining a dam settlement time-series curve. Combining the dam survey report and hydrological yearbook data, it establishes a correlation model between settlement and seepage intensity (R0). 2 ≥0.85);

[0086] Meanwhile, the data from the vibrating wire piezometer were analyzed to calculate the daily rate of change and spatial gradient difference of seepage pressure, so as to understand the distribution and variation law of seepage pressure inside the dam.

[0087] If the daily rate of change of seepage pressure exceeds 5%, or the difference between adjacent measuring points exceeds 0.2 MPa, it is considered abnormal.

[0088] The data fusion and modeling module is responsible for fusing satellite data and piezometer data based on spatiotemporal alignment, controlling the spatial distance between the satellite pixel center point and the piezometer deployment point to within 10 meters, and controlling the time synchronization error to within 1 hour; and constructing a dam seepage monitoring model (model accuracy ≥ 85%) using an improved DS evidence theory, and calculating the comprehensive anomaly index through the model;

[0089] The feedback and optimization module is responsible for issuing warning signals of the corresponding level when monitoring data is abnormal, based on the dam seepage monitoring model and the preset three-level warning thresholds, to remind operation and maintenance personnel to conduct on-site verification and handling according to the warning signal level.

[0090] Meanwhile, the processing results are fed back to the dam seepage monitoring model, and the model parameters and early warning thresholds are optimized using a BP neural network; during model optimization, the learning rate is 0.01 and the number of iterations is 1000.

[0091] The dam piping early warning and location construction device based on satellite and seepage pressure monitoring includes a memory and a processor; the memory is used to store a computer program, and the processor is used to execute the computer program to implement the above-described method steps.

[0092] The readable storage medium stores a computer program that, when executed by a processor, implements the above-described method steps.

[0093] Compared with existing technologies, this method for early warning and location of dam piping based on satellite and seepage pressure monitoring has the following characteristics:

[0094] 1) Wide-area coverage monitoring of the dam area was achieved through satellite remote sensing technology (InSAR + multispectral) (single scene image coverage ≥100km). 2 This method overcomes the limitations of traditional monitoring methods in terms of their limited monitoring range. It can comprehensively grasp the overall seepage status of the dam (including remote areas such as dam shoulders and bank slopes) and promptly detect potential seepage hazards (such as hidden piping).

[0095] 2) By accurately measuring the seepage pressure inside the dam using a piezometer (vibrating wire type) (accuracy ±0.1% FS), microscopic data support can be provided for seepage analysis. Combined with macroscopic monitoring data (deformation, humidity) from satellites, the accuracy (overall error ≤5%) and precision of seepage monitoring are improved through a multi-source data fusion algorithm (DS evidence theory), making the monitoring results more reliable (accuracy ≥85%).

[0096] 3) Real-time data acquisition (piezometer scans once per minute, satellite updates periodically), transmission (4G / 5G latency <10 seconds), and analysis (cloud server processing speed ≥100MB / minute) are achieved, enabling timely detection of seepage anomalies and issuing early warnings (response time ≤30 minutes). Compared to the lag of traditional monitoring methods (manual inspection cycle ≥7 days), this significantly improves the timeliness of early warnings, providing valuable time for taking corrective measures (anomalies are detected an average of 3-5 days in advance). 4) By combining satellite and piezometer methods, the workload of manual inspections is reduced (from 3 times per week to once per month), lowering monitoring costs (annual maintenance costs reduced by 30-50%). Simultaneously, the automated monitoring and early warning system improves work efficiency (data processing time reduced from 24 hours to 1 hour) and reduces human interference (human error reduced to <3%).

[0097] 5) It can perform long-term and continuous monitoring of dam seepage (system mean time between failures ≥ 10,000 hours), providing comprehensive and systematic monitoring data for dam safety assessment (such as dam stability calculation) and maintenance (such as grouting plan development) (cumulative storage ≥ 10TB / year), which helps to extend the service life of the dam (expected to extend by 5-10 years) and ensure the long-term safe and stable operation of the dam.

[0098] The embodiments described above are merely one specific implementation of the present invention. Ordinary changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for early warning and location of piping in dams based on satellite and seepage pressure monitoring, characterized in that: Includes the following steps: Step S1: Use InSAR and multispectral satellites to acquire satellite image data of the dam area. The resolution of InSAR satellite images shall be no less than 5 meters × 20 meters, and the resolution of multispectral satellite images shall be no less than 10 meters. Step S2: At key parts of the dam, including the dam body, dam foundation and seepage prevention body, vibrating wire piezometers are installed to collect seepage pressure data in real time through 4G / 5G wireless transmission modules, with a sampling frequency of 1 time / minute. Step S3: Preprocess the satellite image data, including: Interferogram generation, phase unwrapping, and terrain correction of InSAR satellite data, as well as atmospheric correction, geometric fine correction, and denoising of multispectral satellite data; Step S4: Filter the seepage pressure data collected by the vibrating wire piezometer, remove data that exceeds ±5% of the range, and remove outliers after verification. Step S5: Analyze the preprocessed satellite data, use the threshold method to extract the humidity anomaly areas on the dam surface where the Normalized Difference Water Index (NDWI) exceeds 0.3, use time-series InSAR technology to calculate the cumulative settlement with an accuracy of ±2 mm / year, and obtain the dam settlement time-series curve; combine the dam survey report and hydrological yearbook data to establish a correlation model between settlement and seepage intensity. The data from the vibrating wire piezometer were analyzed to calculate the daily rate of change and spatial gradient difference of the seepage pressure. If the daily rate of change of seepage pressure exceeds 5%, or the difference between adjacent measuring points exceeds 0.2 MPa, it is considered abnormal. Step S6: Import satellite data and seepage pressure data into the data fusion platform, fuse the satellite data and seepage pressure data based on spatiotemporal alignment, control the spatial distance between the center point of the satellite pixel and the piezometer deployment point to within 10 meters, and control the time synchronization error to within 1 hour. An improved DS evidence theory was used to construct a dam seepage monitoring model, and a comprehensive anomaly index was calculated using the model. Step S7: Based on the dam seepage monitoring model and the preset three-level early warning threshold, when the monitoring data is abnormal, an early warning signal of the corresponding level is issued to remind the operation and maintenance personnel to conduct on-site verification and handling according to the level of the early warning signal. Meanwhile, the processing results are fed back to the dam seepage monitoring model, and the model parameters and early warning thresholds are optimized using a BP neural network; during model optimization, the learning rate is 0.01 and the number of iterations is 1000.

2. The method for early warning and location of dam piping based on satellite and seepage pressure monitoring according to claim 1, characterized in that: In step S1, the data acquisition cycle of the InSAR satellite is 12 days, the data acquisition cycle of the multispectral satellite is 16 days, and the data acquisition cycle is increased to 5-7 days during the flood season; Satellite imagery data of the dam area includes data on surface deformation, water distribution, and vegetation cover.

3. The method for early warning and location of dam piping based on satellite and seepage pressure monitoring according to claim 1, characterized in that: In step S3, for InSAR satellite data, GAMMA software is used to generate interferograms, Goldstein filtering is used to suppress noise, the filtering window is 32×32, the minimum cost flow method is used for phase unwrapping, and terrain correction is performed in combination with DEM data, with a resolution of 12.5 meters. For multispectral satellite data, the FLAASH module of ENVI software was used with a 6S model for atmospheric correction; geometric fine correction was performed using no less than 10 control points with an error of no more than 5 meters; and wavelet threshold denoising algorithm was used to remove noise from the images.

4. The method for early warning and location of dam piping based on satellite and seepage pressure monitoring according to claim 1, characterized in that: In step S4, the seepage pressure data collected by the vibrating wire piezometer is verified by Kalman filtering smoothing, and the process noise variance is 0.

01.

5. The method for early warning and location of dam piping based on satellite and seepage pressure monitoring according to claim 1, characterized in that: In step S5, MATLAB software is used to perform statistical analysis on the seepage pressure data, plot the seepage pressure-time process line and the seepage pressure-water level relationship curve, and calculate the characteristic parameters of the seepage pressure, including the average value, maximum value and minimum value. The t-test was used with a significance level of 0.05 to analyze whether the characteristic parameters of seepage pressure changed significantly over time.

6. The method for early warning and location of dam piping based on satellite and seepage pressure monitoring according to claim 1, characterized in that: In step S6, the raster-format satellite data is first converted into vector point data, the spatial coordinates of the seepage pressure data are projected and transformed using UTM, and then imported into a data fusion platform developed based on the GeoPandas library of Python; during the fusion process, the basic probability assignment function is determined by the Gaussian membership function.

7. The method for early warning and location of dam piping based on satellite and seepage pressure monitoring according to claim 1, characterized in that: In step S7, the alarm mechanism of the dam seepage monitoring model is divided into three levels: When the comprehensive anomaly index exceeds 0.6 but is less than 0.8, a Level I warning will be activated. When the comprehensive anomaly index exceeds 0.8 but is less than 1.0, a Level II warning will be activated. When the comprehensive anomaly index exceeds 1.0, a Level III warning will be activated.

8. A dam piping early warning and location system based on satellite and seepage pressure monitoring, characterized in that: To implement the method according to any one of claims 1 to 7, comprising: The data acquisition module is responsible for acquiring satellite imagery data of the dam area using InSAR and multispectral satellites. By deploying vibrating wire piezometers at key parts of the dam, including the dam body, dam foundation, and seepage prevention structure, it collects seepage pressure data in real time through a 4G / 5G wireless transmission module at a sampling frequency of 1 time / minute. The data preprocessing module is responsible for generating interferograms of InSAR satellite data, performing phase unwrapping and terrain correction; performing atmospheric correction, geometric fine correction and noise reduction on multispectral satellite data; and screening seepage pressure data collected by vibrating wire piezometers, removing data exceeding ±5% of the range, and removing outliers after verification. The data analysis module is responsible for analyzing the preprocessed satellite data, using the threshold method to extract the humidity anomaly areas on the dam surface with a normalized water index (NDWI) exceeding 0.3, using time-series InSAR technology to calculate the cumulative settlement with an accuracy of ±2 mm / year, and obtaining the dam settlement time-series curve; combined with the dam survey report and hydrological yearbook data, a correlation model between settlement and seepage intensity is established. Simultaneously, the seepage pressure data were analyzed to calculate the daily rate of change and spatial gradient difference of the seepage pressure. If the daily rate of change of seepage pressure exceeds 5%, or the difference between adjacent measuring points exceeds 0.2 MPa, it is considered abnormal. The data fusion and modeling module is responsible for fusing satellite data and seepage pressure data based on spatiotemporal alignment, controlling the spatial distance between the satellite pixel center point and the piezometer deployment point to within 10 meters, and controlling the time synchronization error to within 1 hour; and using the improved DS evidence theory to construct a dam seepage monitoring model, and calculating the comprehensive anomaly index through the model. The feedback and optimization module is responsible for issuing warning signals of the corresponding level when monitoring data is abnormal, based on the dam seepage monitoring model and the preset three-level warning thresholds, to remind operation and maintenance personnel to conduct on-site verification and handling according to the warning signal level. Meanwhile, the processing results are fed back to the dam seepage monitoring model, and the model parameters and early warning thresholds are optimized using a BP neural network; during model optimization, the learning rate is 0.01 and the number of iterations is 1000.

9. A dam piping early warning and location construction device based on satellite and seepage pressure monitoring, characterized in that: It includes a memory and a processor; the memory is used to store a computer program, and the processor is used to execute the computer program to implement the method according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1 to 7.

Images