An airborne L-band underground pipe network hidden danger monitoring and evaluation system and method based on water molecule polarization relaxation correction

Abstract

The application discloses an airborne L-band underground pipe network hidden danger monitoring and evaluation system and method based on water molecule polarization relaxation correction, and belongs to the technical field of underground pipeline remote sensing monitoring. The system as a whole comprises an airborne hardware layer, a ground management and control layer, a backend server processing layer and a business output layer. The application realizes image target registration and deviation correction through unmanned aerial vehicle multi-dimensional space correction, and constructs a unified geographic reference. The 1-2GHz L-band radar can penetrate the soil layer, and the maximum detection depth is 3.5m. The backend processing layer corrects the soil body attenuation and phase noise of the radar signal through the core module water molecule polarization relaxation coefficient and the double-layer model of polarization correction interference phase, and inverses the soil water content and the three-dimensional volume parameters of the underground cavity. The system automatically quantifies hidden dangers in a single area and generates a standardized report, collects the area data to complete global risk clustering evaluation, and matches a hierarchical early warning and data archiving closed-loop mechanism. The application solves the defects of poor penetration, fuzzy positioning and inability to quantify leakage hidden dangers of traditional detection methods, realizes millimeter-level accurate identification and risk grading of pipe network leakage and underground cavity, and provides technical support for urban pipe network operation and maintenance and road collapse prevention and control.

Description

Technical Field

[0001] This invention relates to the fields of urban underground pipeline detection and assessment, and underground cavity hazard monitoring, particularly to a UAV remote sensing monitoring and assessment system equipped with L-band InSAR radar. It is suitable for investigating potential leaks in municipal water supply and drainage networks, gas pipelines, and heating pipelines, and can also identify precursors to road subsidence such as loose soil within the cover, underground cavities, and roadbed settlement. It can be widely applied in urban municipal projects, roads, underground spaces, and the renovation of old pipeline networks, as well as for identifying pipeline leaks and potential hazards. Background Technology

[0002] Urban underground water supply, drainage, gas, and heating pipelines are buried deep beneath roads and green belts, and are prone to long-term pipe corrosion and joint leaks, making water leakage a major urban problem. Meanwhile, continuous leakage can gradually form underground cavities and roadbed subsidence, easily leading to road collapses, pipeline bursts, gas leaks, and other significant safety hazards. Traditional underground pipeline and underground cavity detection technologies have the following shortcomings:

[0003] 1. Manual pipeline detectors and pipeline endoscopes are suitable for short-distance inspections, but their inspection efficiency is low. They cannot be used to quickly identify and promptly investigate safety hazards on main urban roads and in large urban areas.

[0004] 2. Ground-based interferometric radar (Ground-based fixed InSAR) has limited coverage and poor mobility;

[0005] 3. Existing airborne radars mostly use the C / X high-frequency band, and the electromagnetic waves have weak ability to penetrate soil, asphalt, and green belt soil. They can only detect the shallow surface layer and lack an integrated processing flow for spatiotemporal synchronization matching, multi-dimensional comparison of historical images, soil moisture content analysis, and quantitative assessment of leakage.

[0006] 4. Traditional inspection data lacks multi-dimensional geographic and spatiotemporal binding and verification mechanisms, resulting in vague location of potential hazards and making it impossible to conduct horizontal comparisons of historical data from multiple periods. It is also difficult to predict minor and slow settlements and gradual leakage in advance. Purpose of the invention

[0007] To address the shortcomings of existing underground pipe network and underground cavity detection technologies, such as weak penetration, limited inspection range, inaccurate location, difficulty in identifying leaks using conventional methods, and inability to quantify leakage damage, this invention provides an airborne L-band underground pipe network hidden danger monitoring and assessment system and method based on water molecule polarization relaxation correction.

[0008] This invention utilizes a self-developed line drawing processing module for multi-dimensional spatial correction during the UAV flight phase. Leveraging the deep penetration advantage of 1-2GHz L-band radar, it constructs a dual-layer radar signal correction model based on the microscopic loss law of water molecule dipole polarization relaxation. This model is then centrally computed by a GPU computing cluster at the back end, enabling high-precision identification of pipeline leakage and underground cavities within a 3.5m burial depth range, classification of leakage degree, automatic quantification of cavity volume, and meter-level accurate location of potential hazards. This provides a digital and quantitative decision-making basis for urban pipeline operation and maintenance, road collapse risk management, and renovation of old pipelines. Summary of the Invention

[0009] This invention aims to address the shortcomings of existing technologies in the industry, such as severe interference in complex urban areas, inability to identify minor hidden dangers, inability to quantify monitoring results, inaccurate location of monitoring points, and lack of a standardized comprehensive assessment system. It provides an airborne L-band underground pipeline hidden danger monitoring and assessment system and method based on water molecule polarization relaxation correction.

[0010] 1. Overall Architecture of the Monitoring and Evaluation System Based on the complete monitoring business logic and operation process, this invention reconstructs six core functional units, which are as follows: UAV multi-dimensional spatial correction unit, L-band InSAR radar acquisition unit, L-band InSAR radar analysis unit, single-piece hidden danger quantification and single-piece reporting unit, regional assessment unit, and data archiving and hierarchical early warning push unit.

[0011] (1) Unmanned Aerial Vehicle Multidimensional Space Correction Unit This unit is the core unit that ensures the accuracy of system data acquisition. It includes a Beidou positioning module, a nine-axis IMU inertial measurement module, an image line drawing processing module, and a comparison and correction module with buildings or targets above the pipeline.

[0012] (2) L-band InSAR radar acquisition unit This unit is the data acquisition unit for system monitoring and evaluation, including an industrial-grade mapping UAV platform, a 1–2 GHz L-band InSAR radar module, an airborne storage module, and a data transmission module. This unit relies on a calibrated spatial reference to perform data acquisition: L-band low-frequency radar emits penetrating interferometric electromagnetic waves to achieve penetrating detection of asphalt pavement, overlying soil layers, and shallow concrete media, and retrieves basic data. Based on the characteristics of L-band InSAR radar, the maximum effective detection depth can reach 3.5m, which can cover the buried depth range of urban municipal pipelines.

[0013] (3) L-band InSAR radar analytical unit This unit is the core intelligent interpretation unit of the system, deployed on the back-end server, and is responsible for the intelligent processing of raw radar fixed area data (single chip), especially using the basic model of water molecule polarization relaxation absorption coefficient and the polarization correction interference phase solution model; The basic formula for the fundamental model of the polarization relaxation absorption coefficient of water molecules is: Intensity factor ( The : determines the baseline level of absorption intensity and is proportional to the frequency. Linear factor ( The term : describes the shape of the absorption peak. When the incident wave frequency is close to the resonant frequency, this term approaches 1, and the absorption reaches its peak value; the further it deviates, the smaller the absorption becomes. Secondly, a polarization-corrected interferometric phase solution model is used. This model is based on noise-reduced echo to solve for high-precision interferometric phase, and finally corrects the deviation in water content measurement. The formula is: , Radar echo after two stages of noise reduction

[0014] The unit's specific functions include: performing noise reduction filtering on raw radar data, removing atmospheric clutter and surface interference; generating standardized image maps through an adaptive edge feature extraction algorithm, and generating water content, horizontal width of underground cavities, vertical depth, total volume of cavities, and data contour boundary maps, etc.

[0015] (4) Quantification of hidden dangers in a single area and reporting unit for a single area This unit is responsible for conducting precise quantitative analysis of individual data segments that show anomalies after comparison with the analyzed individual data segments. The unit runs an L-band soil moisture content analysis model and compares it with historical detection data to output the locations of suspected leakage points or suspected underground cavities, providing an effective anomaly data source for subsequent quantitative assessment. When anomalies are found in a single area, the risk level of that area is determined by combining parameters such as pipeline burial depth, pipe diameter, and road traffic load. Finally, a standardized single-area monitoring report is automatically generated, including precise geographic coordinates (meter-level), an image above the pipeline network, hazard type, quantitative parameters, and risk level, achieving one quantification per area and one report per hazard.

[0016] (5) Regional assessment unit This unit is responsible for completing the process from point to line, that is, from the assessment of potential hazards in a single area to the assessment and comprehensive analysis of potential hazards in a designated area of ​​the pipeline network. This unit completes the regional leakage distribution statistics, cavity hazard clustering analysis, and risk distribution mapping for all individual hazard data within a designated area. It also conducts a comprehensive, correlational, and regional risk assessment by combining regional geological conditions, pipeline laying years, and road grade and working conditions. The output results include the overall leakage assessment of the regional pipeline network, the distribution trend of underground cavity hazards, and the regional safety level, meeting the overall decision-making needs of urban area pipeline network surveys, old pipeline network renovation, and road safety management.

[0017] (6) Data archiving and hierarchical early warning push unit This unit is the system closed-loop management and on-the-ground operation and maintenance terminal unit. This unit is responsible for storing all raw radar data, standardized line drawing images, soil moisture content data, void parameters, single-area reports, and regional assessment reports collected in this study into the cloud-based electronic ledger of the pipeline network for permanent archiving, serving as benchmark data for the next round of inspection timeline comparison. It also incorporates a tiered early warning logic: continuous tracking of minor hazards in the ledger, and automatic triggering of early warning pushes for moderate and severe high-risk hazards, pushing precise location information, hazard parameters, and maintenance suggestions to the municipal operation and maintenance platform and mobile devices, achieving a fully automated monitoring closed loop with traceable data, trackable trends, early warning of risks, and implementable operation and maintenance. In summary, the six functional units of this invention strictly correspond to the entire system operation process, progressing step by step from pre-calibration, data acquisition, intelligent analysis, single-chip quantification, regional analysis to archiving and early warning, realizing a standardized, automated, quantitative, and digital intelligent monitoring system for monitoring leakage and underground cavities in urban underground pipe networks.

[0018] 2. System Hardware Entities The monitoring system hardware used in this invention is an integrated hardware platform with multiple sensors mounted on a UAV. The entire hardware system is a physical hardware system that can be assembled, mass-produced, and engineered. It mainly includes: a UAV flight platform, L-band InSAR radar equipment, a visible light high-definition image camera, an airborne external real-time processor, a high-precision BeiDou positioning module, a nine-axis IMU inertial measurement unit, an image stabilization and gimbal, an airborne storage unit, an image and data transmission module, a ground industrial control terminal, and a back-end server hardware cluster. The entire hardware system works collaboratively to provide hardware support for the entire process of image acquisition, image preprocessing, spatial correction, radar detection, data transmission, back-end analysis, and risk assessment.

[0019] (1) Unmanned aerial vehicle (UAV) flight platform Employing an industrial-grade compound wing UAV, it features fixed-point hovering, automatic flight path navigation, and attitude stabilization compensation. Equipped with multiple device mounting interfaces and a shock-absorbing frame, it can simultaneously carry radar, cameras, onboard processors, and gimbal equipment. Its load capacity, flight stability, and flight path repeatability meet the requirements of large-scale, high-precision remote sensing inspection operations in cities, providing a stable aerial operating platform for the entire monitoring system.

[0020] (2) L-band InSAR radar hardware The core underground detection hardware of this invention employs an airborne, lightweight 1–2 GHz L-band interferometric synthetic aperture radar. The radar possesses deep soil penetration capabilities, capable of penetrating asphalt pavements, topsoil, and vegetation layers, with a maximum effective detection depth of 3.5 m. The equipment supports continuous interferometric phase acquisition, echo intensity recording, and multi-angle scanning imaging, enabling it to capture minute deformations of underground soil, differences in medium moisture content, and characteristics of loose soil and void structures. It is the core dedicated hardware for detecting leaks in underground pipe networks and underground cavities.

[0021] (3) Visible light high-definition imaging camera Equipped with a high-resolution industrial-grade aerial camera, it is used to simultaneously acquire high-definition real-world images of the ground surface. The camera and radar imaging are synchronized in time, and the image capture is used for multi-dimensional comparison of line drawings and data in later stages, providing accurate location for radar detection.

[0022] (4) Airborne external real-time processor The UAV is equipped with an independent embedded high-speed processor, which serves as the core hardware for real-time preprocessing on the airborne end. This external processor operates independently of the flight control system and is specifically responsible for: real-time caching of raw radar data, processing of camera image line drawings, fusion of flight attitude data, and comparison and correction with overhead buildings or targets. By sharing the workload through independent external computing power, real-time data filtering and preprocessing during flight are achieved, significantly reducing the computational burden on the backend and improving the real-time performance and stability of the entire system.

[0023] (5) High-precision Beidou positioning module Equipped with a BeiDou high-precision positioning module, it outputs latitude, longitude, altitude, flight speed, and positioning accuracy factor data in real time. The positioning module is synchronized with the radar and camera hardware, providing a precise geographic coordinate reference for each frame of radar imagery and each frame of visible light imagery, ensuring the hardware accuracy foundation for subsequent image geographic correction, pixel-level positioning, and hazard coordinate tracing.

[0024] (6) Nine-axis IMU inertial measurement unit Integrating a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer, it acquires real-time pitch, roll, and yaw attitude angular velocities and angles of the UAV. The IMU outputs attitude parameters at high frequency to compensate for radar imaging geometric distortions caused by flight jitter, tilt deviation, and minor deviations in the flight path. It is a key inertial hardware for achieving multi-dimensional spatial correction, image correction, and high-precision registration.

[0025] (7) Anti-shake gimbal The radar and camera are mounted on a high-precision three-axis stabilization gimbal, which can cancel out drone flight disturbances and wind vibrations in real time, ensuring vertical stability of the radar detection angle and no camera jitter in imaging. This ensures that the radar imaging parameters are consistent and the imaging attitude is consistent each time, greatly improving the stability and repeatability of interferometric radar phase data.

[0026] (8) Onboard high-speed storage unit Equipped with airborne solid-state high-speed storage devices, it is used to save original radar interferometric data, aerial images, attitude logs, and positioning data in situ, preventing data loss caused by wireless transmission packet loss, realizing local redundant backup of flight data, and ensuring the integrity and traceability of the entire monitoring data.

[0027] (9) Image transmission and wireless data transmission module Equipped with long-distance 5G+microwave dual-link transmission hardware, it enables real-time transmission of airborne pre-processed data, thumbnail images, and status parameters back to the ground terminal, supporting stable long-distance transmission in large urban areas and ensuring real-time controllability and online data for field operations.

[0028] (10) Ground industrial control terminal hardware It is used for pre-operation route planning, equipment parameter configuration, real-time data reception, on-site data preview, field quality verification, and target calibration parameter input, providing ground control and preliminary data processing hardware support for field operations.

[0029] (11) Backend server hardware cluster The backend is equipped with hardware devices such as GPU computing servers and big data storage disk arrays. It undertakes high-intensity computing tasks such as the analysis of the basic model of water molecule polarization relaxation absorption coefficient and the polarization correction interference phase solution model, multi-phase image comparison, soil moisture content model calculation, void quantification calculation, risk assessment, data archiving, and early warning push, providing stable, efficient, and batch hardware computing support for the entire set of intelligent monitoring algorithms. Overall hardware working relationship The entire set of physical hardware forms a complete hardware system of "UAV onboard multi-sensor acquisition + onboard external preprocessing + ground terminal control + backend computing power analysis". Through hard synchronization circuit, radar, camera, positioning and attitude data are synchronized from the same source. Combined with the onboard independent processor for real-time preprocessing, it can ultimately achieve high-precision, fully automatic, quantifiable and traceable intelligent monitoring of urban underground pipe network leakage, soil moisture content changes and underground cavity hazards. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the overall architecture of the present invention. 1. Airborne hardware acquisition layer: industrial composite wing UAV, L-band InSAR radar, visible light camera, Beidou positioning, nine-axis IMU, image stabilization gimbal, airborne external processor, airborne storage, data transmission module; 2. Ground control layer: Ground industrial control terminal (route planning, on-site calibration, data preview); 3. The backend server processing layer contains six functional units. ①UAV Multi-dimensional Space Correction Unit ②L-band InSAR radar acquisition unit ③ L-band InSAR radar analytical unit (two core models are labeled: water molecule polarization relaxation absorption coefficient model and polarization correction interferometric phase solution model) ④ Single-area hazard quantification and single-point reporting unit ⑤ Regional assessment unit ⑥ Data archiving and hierarchical early warning push unit 4. Business output layer: electronic ledger, municipal operation and maintenance platform, mobile early warning terminal, and on-site verification and maintenance port. Figure 2 This is a flowchart illustrating the comparison between the line drawing and the multidimensional space of this invention. After the airborne visible light image is extracted from the edge to generate a single-line drawing, it is combined with Beidou and IMU multi-dimensional spatial data to complete the contour registration and comparison, eliminating the imaging offset caused by the drone's flight disturbance, providing a unified and accurate geospatial benchmark for subsequent L-band radar detection, and ensuring the meter-level positioning accuracy of radar phase and underground hidden danger points. Figure 3 Schematic diagram of water molecule polarization relaxation correction process 1. The backend GPU computing server receives the raw airborne radar echo dataset aggregated by the ground industrial control terminal; 2. Substitute the values ​​into the polarization relaxation absorption coefficient formula to calculate the soil medium loss value, and perform attenuation correction on the original echo amplitude to obtain the noise-reduced echoes S_{q1} and S_{q2}. 3. Calculate the pure interference phase \Delta\varphi_q after polarization correction using a phase solution model; 4. Based on the corrected phase, complete the global signal noise reduction, atmospheric clutter, and removal of irrelevant surface interference; 5. Generate standardized raster images through adaptive edge feature extraction, and output a dataset of soil moisture content distribution, horizontal width of underground cavities, vertical burial depth, total volume of cavities, and complete contour boundaries. Figure 4 Schematic diagram of the closed-loop implementation of the entire process of underground pipeline network hidden danger monitoring and assessment Pre-operation preparation → Multi-dimensional space reference calibration for flight → L-band radar data acquisition → Back-end radar data analysis → Single-chip hazard quantification and reporting → Regional multi-source data fusion assessment → Data archiving + graded early warning push → On-site verification and maintenance, actual data feedback to iterative model, cycle to the next round of inspection. Detailed Implementation

[0031] This invention relies on an integrated airborne hardware platform and six functional units to form a standardized closed-loop implementation process. The complete implementation steps are as follows:

[0032] Step 1: Preparations for the assignment 1. Conduct on-site surveys of the target monitoring area, import the regional underground pipeline network ledger through the ground control terminal, and enter basic parameters such as pipeline burial depth, pipe diameter, pipe material, laying year, soil type, and soil cover thickness to delineate drone inspection boundaries and no-fly zones. 2. Deploy multiple ground calibration targets in the monitoring area and debug the radar phase corrector; complete the power-on self-test of industrial UAVs, L-band InSAR radar, visible light camera, Beidou, IMU, and airborne external processor hardware, and verify the timing synchronization performance of the shutter hard synchronization circuit. 3. Plan the flight path on the ground control terminal, set the flight altitude, longitudinal and transverse overlap of the flight path, and flight speed, configure radar imaging parameters, image storage, and wireless data transmission parameters, and complete the pre-setting of all parameters before the operation.

[0033] Step 2: Perform in-flight UAV multi-dimensional spatial reference correction After the drone takes off and enters the preset flight path, the onboard multi-dimensional spatial correction unit runs the correction process in real time: 1. The airborne external processor synchronously reads Beidou positioning coordinates and three-axis attitude data from the nine-axis IMU to compensate for imaging geometric distortion caused by flight jitter and fuselage tilt in real time. 2. By combining the preset coordinates of the targets above the underground pipeline with the synchronously acquired visible light real-scene images, the features of surface buildings and targets are matched with the image processing line drawing, and the phase residual caused by the flight path offset is dynamically corrected.

[0034] Step 3: L-band InSAR radar data acquisition 1. Based on the calibrated spatial reference, the UAV cruises automatically along the planned route, and the airborne 1-2GHz L-band InSAR radar continuously emits penetrating interferometric electromagnetic waves to collect raw data on the interference phase and echo intensity of underground soil layers and soil around pipelines. 2. One channel of the collected data can be transmitted back to the ground industrial control terminal in real time via 5G+microwave data transmission, while the other channel can be stored in the airborne high-speed storage unit for local backup.

[0035] Step 4: L-band InSAR radar data analysis The ground terminal uploads the collected dataset to the backend GPU server cluster, where the radar analysis unit performs automated processing. 1. Call the analytical model to perform the solution model of the basic model and polarization correction model of water molecule polarization relaxation absorption coefficient on the original radar data, as well as noise reduction filtering and atmospheric clutter and surface interference separation. 2. Standardized image maps are generated using an adaptive edge feature extraction algorithm, which includes data such as water content, horizontal width of underground cavities, vertical depth, total volume of cavities, and related data outline boundary maps.

[0036] Step 5: Quantification and Reporting of Hidden Dangers in L-Band InSAR Radar Data Pieces For each potential hazard identified in the analysis output, the chip is automatically quantitatively analyzed and assessed. 1. Run the L-band soil moisture content analysis model and call historical detection processing and comparison to output the location of suspected leakage points or suspected underground cavities, providing an effective anomaly data source for subsequent quantitative assessment. 2. When an anomaly is found in a single area, the risk level of that area is determined by combining parameters such as pipeline burial depth, pipe diameter, and road traffic load. Finally, a standardized single-area monitoring report is automatically generated, including precise geographic coordinates (meter-level), an image above the pipeline network, the type of hazard, quantitative parameters, and risk level, achieving one quantification per area and one report per hazard.

[0037] Step 6: Regional Multi-Source Data Fusion Assessment 1. For the identified assessment area, retrieve all relevant data within the area. 2. Based on the data acquisition time, sort the geographic coordinates of the relevant regional single-piece data datasets. 3. Combining temporal and spatial hazard data, complete the regional leakage distribution statistics, cavity hazard clustering analysis, and risk distribution mapping. In conjunction with regional geological conditions, pipeline laying years, road grade and working conditions, conduct overall, correlational, and regional risk assessments, and output the overall leakage assessment results of the regional pipeline network, the distribution trend of underground cavity hazards, and the regional safety level.

[0038] Step 7: Data archiving and tiered early warning push 1. All original radar data, line drawings, moisture content parameters, single-piece reports, and regional assessment reports will be encrypted and permanently archived in the pipeline network electronic ledger, serving as the benchmark data for the next round of inspection timeline comparison. 2. System-based hierarchical handling of different risks and hazards: Minor hazards are only kept in the ledger for continuous tracking; moderate and severe high-risk hazards are automatically triggered and pushed out early warnings. 3. Push early warning information to the municipal operation and maintenance platform and the mobile terminal of the front-line inspection team, along with the location coordinates of the hidden danger, quantitative parameters, risk level and fixed-point maintenance plan.

[0039] Step 8: On-site verification and closed-loop handling Based on the early warning information, maintenance personnel went to the potential hazard locations to conduct on-site verification. They excavated the confirmed points and recorded the on-site verification data, including the pipeline leakage, the actual size of underground cavities, and their burial depth. The maintenance and repair results and the status of hazard elimination were simultaneously entered into the ledger and the pipeline safety database was updated. At the same time, the on-site measured data were fed back to the analytical model to iteratively optimize the accuracy of the water content inversion and cavity calculation algorithms, thus completing the closed loop of the entire process monitoring and maintenance.

Claims

1. An airborne L-band underground pipe network hidden danger monitoring and assessment system based on water molecule polarization relaxation correction, characterized in that, include: Multi-dimensional spatial correction unit, radar data analysis unit, single-piece hidden danger quantitative reporting unit, regional comprehensive assessment unit, and archiving early warning unit; The multi-dimensional spatial correction unit is equipped with an image line drawing processing module and a geographic target comparison module. It combines positioning and attitude sensing data to extract outline lines from aerial visible light images and registers them with fixed targets on the ground to correct deviations, eliminate imaging geometric distortions caused by flight attitude shifts, and output a unified geospatial reference. The radar data analysis unit is deployed on the back-end computing cluster and has a built-in dual-layer correction core model: water molecule polarization relaxation absorption coefficient model and polarization correction interferometric phase solution model. Based on L-band radar echo data under a unified spatial reference, the electromagnetic wave soil attenuation loss is first corrected by the water molecule polarization relaxation absorption coefficient model, and then the phase noise is eliminated by the polarization correction interferometric phase solution model. The atmospheric and surface clutter interference is removed, and the soil moisture content, the three-dimensional size of underground cavities and the cavity volume parameters are inverted. The single-piece hidden danger quantitative reporting unit compares the correction and analysis results with historical monitoring data to identify pipeline leakage and underground cavity hidden dangers, and completes the single-point risk classification by combining pipeline basic parameters, and outputs a standardized single-piece hidden danger report with accurate geographic coordinates. The regional comprehensive assessment unit collects all individual hidden danger data within the region, completes leakage distribution statistics, cavity cluster analysis, and risk thermal mapping, and conducts a comprehensive correlation analysis based on regional geology, pipeline service life, and road conditions, outputting the overall leakage assessment results of the pipeline network in the area and the regional safety level; The archived early warning unit stores all monitoring raw data, line drawings, quantitative parameters, and assessment reports to form a time-series ledger; it classifies and handles hazards according to their severity, and automatically pushes medium and severe hazard warnings and corresponding maintenance suggestions to the operation and maintenance terminal.

2. The airborne L-band underground pipeline hidden danger monitoring and assessment system based on water molecule polarization relaxation correction according to claim 1, characterized in that, It also includes airborne acquisition components, which are equipped with an L-band interferometric radar, a visible light camera, and a positioning and inertial sensing module. These components synchronously acquire radar echoes and surface images in time, providing raw data sources for the multi-dimensional spatial correction unit and the radar data analysis unit.

3. The airborne L-band underground pipeline hidden danger monitoring and assessment system based on water molecule polarization relaxation correction according to claim 1, characterized in that: The water molecule polarization relaxation absorption coefficient model is used to solve the water molecule polarization loss coefficient of soil medium and correct the radar echo amplitude attenuation deviation. The polarization-corrected interferometric phase solution model relies on the noise reduction of multiple radar echoes to solve for the pure interferometric phase, thereby correcting the error in the water content inversion calculation.

4. A method for monitoring and assessing potential hazards in underground pipe networks based on water molecule polarization relaxation correction, characterized in that, The system described in any one of claims 1 to 3 is implemented by comprising the following core steps: S1. Spatial reference unified correction: Simultaneously acquire visible light aerial images of the ground surface, extract standardized outline lines, integrate positioning and inertial attitude data to complete the registration of the outline lines with ground targets, compensate for imaging geometric distortion, and construct a unified geographic reference for the entire region. S2, Radar Data Dual-Layer Polarization Relaxation Correction Analysis: Acquire L-band underground detection echo data collected based on a unified geographic benchmark, and successively call the water molecule polarization relaxation absorption coefficient model and the polarization correction interference phase solution model to complete echo attenuation correction and phase noise elimination; remove atmospheric and surface-independent interference, adaptively extract features, and output soil moisture content, underground cavity length and width, burial depth, and volume quantification dataset. S3. Quantitative assessment of single-area hidden dangers: By comparing the corrected analytical data with historical time-series monitoring data, hidden dangers such as pipeline leakage, loose soil and underground cavities are identified. Combined with pipeline parameters, single-point risk classification is completed, and standardized single-area monitoring reports are automatically generated. S4. Overall Risk Assessment of Regional Pipeline Network: Integrate quantitative data of all individual hidden dangers in the area, conduct leakage distribution statistics, cavity hidden danger clustering, and risk distribution mapping, and complete a comprehensive assessment of the entire region in combination with regional geology and pipeline laying conditions, and output regional leakage assessment results and overall safety level; S5. Data archiving and hierarchical early warning closed loop: All monitoring data and assessment reports are archived and retained as a benchmark for subsequent inspections and comparisons; early warning information is automatically pushed for medium and severe high-risk hazards; on-site measured maintenance data is transmitted back to the dual-layer calibration model to iteratively optimize the accuracy of moisture content and void quantification calculations.

5. The airborne L-band underground pipeline hidden danger monitoring and assessment method based on water molecule polarization relaxation correction according to claim 4, characterized in that, In step S1, the radar phase residual is corrected by matching the line drawing with the geographic target, so as to achieve meter-level accurate geographic positioning of underground hidden dangers.

6. The airborne L-band underground pipeline hidden danger monitoring and assessment method based on water molecule polarization relaxation correction according to claim 4, characterized in that, The 1-2 GHz L-band electromagnetic waves are used to penetrate soil and asphalt pavement for underground detection, with a maximum effective detection depth of 3.5 m.

7. The airborne L-band underground pipe network hidden danger monitoring and assessment method based on water molecule polarization relaxation correction according to claim 4, characterized in that, The tiered early warning rules are as follows: minor hazards are only continuously tracked in the ledger, while moderate and severe hazards are simultaneously pushed to the operation and maintenance platform and mobile terminal with hazard coordinates, quantitative parameters, risk level and fixed-point maintenance plan.

Images