A method and system for analyzing hidden dangers of embankments based on physical detection
By constructing a segmented spatial mapping system in the levee safety monitoring system, collecting and integrating multi-source data, identifying seepage status and generating graded early warnings, the problem of the lack of segmented data chains in the existing system is solved, realizing refined and dynamic management of levee hazards, and improving the accuracy of risk prediction and the efficiency of emergency response.
Patent Information
- Application Number
- CN202611140061.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-30
- Publication Date
- 2026-11-10
AI Technical Summary
The existing levee safety monitoring system lacks a segmented data chain based on the levee's spatial axis, making it difficult to continuously identify and determine abnormal sections, and unable to form a closed-loop analysis from spatial positioning to risk classification. This fails to meet the needs of refined, structured, and dynamic hazard investigation of levees.
By constructing a segmented spatial mapping system with the dike axis coordinate system as the core, installing physical detectors to collect multi-source data, performing data fusion, identifying seepage status, assessing seepage stability, dividing potential seepage channel segments, recording particle migration intensity, and generating graded early warning and control instructions, a closed-loop control from risk identification to management reporting is achieved.
It improves spatial consistency and anomaly location accuracy under multi-point and multi-depth conditions, enhances early warning capability and risk prediction accuracy of dike hazards, forms an intelligent monitoring system, reduces the risk of seepage instability and piping, and meets the needs of refined, digital and intelligent management of water conservancy projects.
Smart Images

Figure CN123070686A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of safety monitoring technology, and in particular to a method and system for investigating and analyzing potential hazards in dikes based on physical detection. Background Technology
[0002] In existing levee safety monitoring and patrol management, the operational status of levees is typically assessed through manual inspections, fixed-point instrument monitoring, or single physical parameter acquisition. Existing monitoring systems often operate modules such as topographic mapping, seepage monitoring, and deformation monitoring in a fragmented manner, lacking an organizational method that constructs a segmented data chain based on the levee's spatial axis. This results in a lack of unified mapping between monitoring data and specific levee sections, making it difficult to continuously identify and determine abnormal sections under multi-point and multi-depth conditions, and failing to form a closed-loop analysis mechanism from spatial location to risk classification. Current technologies mostly remain at the stage of single hydraulic gradient threshold judgment or surface anomaly identification, failing to further link and execute water content fusion calculations, pore structure change analysis, and particle migration rate quantitative modeling in abnormal seepage sections. They lack the ability to predict seepage evolution trends and piping risk levels in a graded manner, making it difficult to trigger timely graded early warnings and precise control measures, and failing to meet the needs of refined, structured, and dynamic hazard investigation of levees. Summary of the Invention
[0003] Therefore, it is necessary for the present invention to provide a method and system for investigating and analyzing potential hazards in dikes based on physical detection, so as to solve at least one of the above-mentioned technical problems.
[0004] To achieve the above objectives, a method for investigating and analyzing potential hazards in dikes based on physical detection includes the following steps: Step S1: Obtain the outline of the dike and deploy physical detectors according to the outline; collect the dike detection data of the physical detectors and perform data fusion based on the dike detection data to form a dike physical dataset; Step S2: Identify the seepage state of the dike based on the dam physical dataset; assess the seepage stability based on the seepage state; and delineate potential seepage channel sections based on the seepage stability. Step S3: Perform particle loss trend analysis in the leakage channel section and record the particle migration intensity; classify the piping risk level according to the particle migration intensity; Step S4: Generate a graded early warning and control instruction based on the piping risk level; trigger an early warning response using the graded early warning and control instruction, and simultaneously generate a levee hazard investigation and management report.
[0005] Preferably, this specification also provides a physical detection-based levee hazard investigation and analysis system for performing the above-mentioned physical detection-based levee hazard investigation and analysis method. This physical detection-based levee hazard investigation and analysis system includes: The physical detection module is used to acquire the outline of the dike and deploy physical detectors according to the outline; it collects the dike detection data from the physical detectors and performs data fusion based on the dike detection data to form a dike physical dataset; The seepage stability assessment module is used to identify the seepage state of the embankment based on the embankment physical dataset; assess the seepage stability based on the seepage state of the embankment; and delineate potential seepage channel sections based on the seepage stability. The piping identification module is used to perform particle loss trend analysis in the seepage channel section, record the particle migration intensity, and classify the piping risk level according to the particle migration intensity. The early warning and control module is used to generate graded early warning and control instructions based on the risk level of piping; to trigger early warning responses using the graded early warning and control instructions, and to simultaneously generate a report on the investigation and management of hidden dangers in the dike.
[0006] The beneficial effects of this invention are as follows: (1) By constructing a segmented spatial mapping system with the levee axis coordinate system as the core, and uniformly projecting the physical detector installation coordinates onto the levee segment number, the structured binding between pore water pressure data, soil moisture content data and resistivity distribution data is realized. This makes the multi-source physical monitoring data no longer exist in the form of isolated measuring points, but form a segmented data chain that corresponds one-to-one with the levee spatial structure, thereby improving the spatial consistency and anomaly positioning accuracy under multiple measuring points and multiple depths, and enhancing the system's collaborative analysis capability under long-distance levees and complex geological conditions.
[0007] (2) At the level of seepage state identification and evolution trend analysis, the continuous identification and connection judgment of abnormal seepage sections are realized, so that the seepage instability judgment is upgraded from a single threshold judgment to a multi-parameter coupled structural evolution analysis process. It can identify the continuous seepage enhancement trend before piping forms, and improve the early warning capability and risk prediction accuracy of dike hazards.
[0008] (3) By establishing a quantitative model of soil particle migration rate and mapping the particle migration rate to the preset migration intensity grading threshold, the risk of piping is transformed from qualitative judgment to quantitative grading. At the same time, the piping risk level is bound to the graded early warning and control instructions, forming a closed-loop control mechanism from risk identification, grade classification to early warning response and management report generation. This upgrades the dike hidden danger investigation system from the traditional passive inspection mode to an intelligent monitoring system with dynamic evolution analysis and proactive risk control capabilities, reducing engineering losses caused by seepage instability and sudden piping risks, improving the level of dike safety operation and emergency response efficiency, and meeting the development needs of refined, digital and intelligent management of water conservancy projects. Attached Figure Description
[0009] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the steps in the method for investigating and analyzing potential hazards in dikes based on physical detection according to the present invention. Figure 2 This is a schematic diagram illustrating the implementation of the physical detection-based embankment hidden danger investigation and analysis system of the present invention; Figure 3 This is a schematic diagram of the layout and section numbering of the embankment profile detector according to the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0010] The technical method of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.
[0011] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.
[0012] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.
[0013] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a method for investigating and analyzing potential hazards in dikes based on physical detection. The method includes the following steps: Step S1: Obtain the outline of the dike and deploy physical detectors according to the outline; collect the dike detection data of the physical detectors and perform data fusion based on the dike detection data to form a dike physical dataset; In one embodiment, a drone lidar (point density) is employed. The entire dike was scanned, and a 3D contour model with centimeter-level accuracy was obtained using RTK measurements. The dike crest line, upstream slope line, and downstream slope line were extracted as reference axes. A segment unit was divided every 20m along the dike axis. Within each segment, physical detectors were deployed in a three-layer structure: slope surface – slope toe – dike interior. These included resistivity electrode arrays (1m spacing), soil moisture sensors (buried at depths of 0.5m and 1.5m), and pore water pressure gauges (buried at a depth of 2m). Resistivity, moisture content, and pore pressure data were collected synchronously at a 10-minute sampling period. Multi-source data fusion was performed through timestamp alignment and spatial coordinate mapping to form a dike physical dataset containing spatial coordinates, time series, and physical parameters.
[0014] In another embodiment, assuming the dike is 5km long, scanning generates approximately 12 million three-dimensional point cloud data points. The dike is divided into 200 segments, each 25m long. Each segment is equipped with 16 electrodes, 4 moisture content sensors, and 2 pore pressure gauges, resulting in a total of 4400 detection nodes. The sampling period is set to 5 minutes, with continuous data collection for 48 hours, yielding approximately 2.5 million sets of raw data. After outlier removal (approximately 3%) and weighted fusion processing, a standardized dike physical dataset is generated, including resistivity compensation values, volumetric moisture content, and pore pressure gradient indices.
[0015] Step S2: Identify the seepage state of the dike based on the dam physical dataset; assess the seepage stability based on the seepage state; and delineate potential seepage channel sections based on the seepage stability. In one embodiment, a seepage identification index is constructed based on the fused resistivity compensation value and the rate of change in water content. When the resistivity continues to decrease and the rate of increase in water content exceeds a preset threshold, it is determined to be an abnormal seepage state. The seepage stability index is calculated in conjunction with the change in pore water pressure gradient and is distinguished into three levels: stable, critical, and unstable. Spatial connectivity analysis is performed on continuous sections with stability indices at the critical and unstable levels to classify them as potential seepage channel sections, and their starting and ending station numbers and length range are recorded.
[0016] In another embodiment, it is assumed that among 200 sections, 32 sections exhibit abnormal changes with a resistivity decrease exceeding 15% and a moisture content increase exceeding 8%; of these, 18 sections have a stability index below 0.6 and are classified as critical, while 6 sections have a stability index below 0.4 and are classified as unstable. After spatial cluster analysis, three continuous potential leakage channel sections are formed, with lengths of 75m, 110m, and 45m, respectively.
[0017] Step S3: Perform particle loss trend analysis in the leakage channel section and record the particle migration intensity; classify the piping risk level according to the particle migration intensity; In one embodiment, continuous temporal resistivity gradient change data and pore pressure fluctuation data are retrieved within the potential leakage channel section to analyze the particle migration trend inside the soil. By statistically analyzing the resistivity fluctuation frequency and pore pressure pulsation amplitude per unit time, a particle migration intensity index is constructed and divided into three levels: low intensity, medium intensity, and high intensity. When the particle migration intensity continues to increase and shows an accelerating trend, it is determined to be a high-risk piping state.
[0018] In another embodiment, assuming that the number of abnormal local resistivity fluctuations in a certain 110m leakage channel section increases from 5 to 18 times within 24 hours, the pore pressure pulsation amplitude increases from 2kPa to 7kPa, and the corresponding particle migration intensity index increases from 0.35 to 0.82; according to the preset classification standard (0–0.4 is low risk, 0.4–0.7 is medium risk, and above 0.7 is high risk), this section is identified as a high-level piping risk section.
[0019] Step S4: Generate a graded early warning and control instruction based on the piping risk level; trigger an early warning response using the graded early warning and control instruction, and simultaneously generate a levee hazard investigation and management report.
[0020] In one embodiment, the piping risk level is matched with a preset early warning strategy library. When the risk level is medium, an inspection enhancement instruction and a data encryption collection instruction are generated. When the risk level is high, on-site verification, load reduction control and emergency response preparation instructions are generated, and early warning information is pushed to the management terminal through the monitoring platform. At the same time, the section number, risk level, trend curve and suggested measures are automatically summarized to generate a dike hidden danger investigation and management report.
[0021] In another embodiment, assuming that a total of 2 high-risk areas and 5 medium-risk areas are identified, the system triggers 7 warning commands respectively. Among them, the high-risk areas send red warning information to 3 management terminals within 10 minutes and generate 2 PDF reports each containing risk evolution curves, spatial distribution maps and disposal suggestions; the medium-risk areas generate yellow warning notices and 5 inspection task sheets, realizing closed-loop control of risk identification, graded response and management archiving.
[0022] It should be noted that you should refer to [link / reference]. Figure 2The diagram illustrates the data flow of the four functional modules: sensors input data into the physical detection module, which constructs a dam physical dataset before transmitting it to the seepage stability assessment module for seepage identification and channel division. The data then flows into the piping identification module to complete particle loss analysis and risk level classification. Finally, the early warning and control module generates tiered early warning instructions and outputs two results: an early warning response and a hidden danger investigation report. This achieves fully automated processing from data acquisition and seepage identification to piping early warning. Each module has a clear responsibility and a sequential connection, effectively avoiding the subjectivity and lag of manual judgment. It can trigger tiered responses at the early stage of hidden dangers, significantly improving the timeliness and accuracy of dam safety investigation.
[0023] It should be noted that you should refer to [link / reference]. Figure 3 The diagram shows that the embankment is divided into five equal-length sections along its axis. Within each section, three types of sensors are deployed from top to bottom: a pore water pressure sensor (P) for calculating the hydraulic gradient and identifying abnormal seepage zones; a moisture content sensor (W) for assessing the loosening trend of the soil's pore structure; and a resistivity detection electrode (R) for detecting changes in porosity caused by particle migration. These three types of sensors work together to collect data and fuse information from multiple dimensions, enabling a comprehensive understanding of the embankment's seepage status.
[0024] Preferably, in step S1, the dam detection data collected by the physical detectors and the data fusion performed based on the dam detection data to form a dam physical dataset include: Collect dam detection data from physical detectors, including pore water pressure data, soil moisture content data, and resistivity distribution data. In one embodiment, pore water pressure gauges, soil moisture sensors, and multi-channel resistivity electrode arrays are deployed at different burial depths on the upstream slope, downstream slope, and inside the dike. The pore water pressure gauges are buried at depths of 1.5m to 3.0m, with a sampling frequency of once every 5 minutes; the soil moisture sensors are buried at depths of 0.5m and 1.5m, with a sampling frequency of once every 10 minutes; the resistivity array uses an automatic scanning method, completing a two-dimensional profile acquisition every 15 minutes. All detectors are connected to a unified data acquisition module via a field data acquisition terminal. After analog-to-digital conversion, raw dike detection data with device numbers and timestamps is generated and uploaded to the monitoring server via a wireless transmission module.
[0025] In another embodiment, assuming a section of the embankment is 2 km long, 60 pore water pressure gauges, 80 soil moisture sensors, and 240 resistivity measuring points are deployed. After 72 hours of continuous operation, approximately 86,400 sets of pore pressure data, approximately 34,560 sets of moisture content data, and approximately 5,760 sets of resistivity profile data are collected. The pore water pressure data fluctuates between 5 kPa and 28 kPa, the soil volumetric moisture content ranges from 18% to 36%, and the resistivity distribution ranges from... All data includes a millisecond-level timestamp and a corresponding device number for subsequent unified processing.
[0026] The dam detection data is used to perform timestamp synchronization processing to form a unified time series; the installation coordinates of the corresponding physical detectors are read from the unified time series; the installation coordinates of the physical detectors are mapped to the dam outline to generate section numbers; and the dam physical dataset is constructed based on the section numbers.
[0027] In one embodiment, data uploaded by different types of sensors are timestamped and calibrated. Data records with an alignment error within ±1 second using a unified standard time source are established, and a unified time series is created with a time step of 5 minutes. Subsequently, the pre-stored three-dimensional installation coordinate information (including station position, lateral distance, and burial depth parameters) of each physical detector is read, and this coordinate is projected onto the established three-dimensional contour model of the dike. According to the dike axis mileage division rules (e.g., every 20m is a segment), corresponding segment numbers are generated, and the unified time series data is bound to the segment numbers to form a dike physical dataset containing a three-dimensional structure of "time-segment-physical parameters".
[0028] In another embodiment, it is assumed that there is a maximum time deviation of 3 seconds for various types of data before time synchronization, and the error is controlled within 0.5 seconds after correction by the synchronization algorithm; the unified time step is set to 10 minutes, and a total of 432 time nodes are generated within 72 hours. The dike is divided into sections of 25 meters each, with a total of 80 section numbers; through coordinate mapping, 380 effective detection nodes are assigned to the corresponding sections, and one section contains 2 pore pressure gauges, 3 moisture content sensors, and 8 resistivity measuring points. In the final constructed dataset, each time node corresponds to 80 section parameter records, with a total of approximately 34,560 data entries, forming a structured and traceable dike physical dataset.
[0029] Preferably, the installation coordinates of the physical detectors are mapped to the dike outline to generate section numbers, including: Obtain the axis coordinate data of the embankment outline and establish the embankment axis coordinate system based on the axis coordinate data; transform the installation coordinates of the physical detector to the embankment axis coordinate system and calculate the projection distance of the installation coordinates of the physical detector in the embankment axis direction in the embankment axis coordinate system. In one embodiment, three-dimensional point cloud data of the levee surface is acquired through UAV lidar scanning and RTK measurement. A curve fitting algorithm (such as cubic spline fitting) is used to extract the centerline of the levee crest as the levee axis, forming an axis coordinate sequence sorted in ascending order of mileage. Taking the origin O as the starting point of the axis, the direction along the extension of the dike is defined as... Establish a coordinate system for the dike axis along the axial direction. ,in The coordinates are along the axis. This is the lateral offset distance. This refers to elevation coordinates. The three-dimensional coordinates of the installation of each physical detector are also considered. By mapping the coordinates to the coordinate system of the dike axis through coordinate translation and rotation, the corresponding coordinates are obtained. Calculate the projected distance of each detector along the axial direction. That is, in The coordinate values on the axis are used to characterize the actual mileage position of the detector in the direction of the dike axis.
[0030] In another embodiment, assuming a section of the dike is 3.6 km long, 720 discrete points along the axis are obtained through point cloud processing, with an average spacing of 5 m. An axis coordinate system is established with the starting point (0,0,0) as the origin. The original coordinates of a certain physical detector are (356.2 m, 128.5 m, 12.4 m), which are obtained after coordinate transformation. This indicates that the detector is 1250.6m from the starting point of the dike, with a lateral offset of 3.2m and an elevation of 11.8m. After converting all 180 detectors, a projected distance sequence with a range of 0 to 3580m is obtained.
[0031] The projection distance is divided according to the preset section length to determine the range of the dike section; the dike section is marked with numbers to generate section numbers.
[0032] In one embodiment, based on the required length of the dike and the monitoring accuracy, the preset section length is 20m. The dike is divided into equidistant segments according to the S-axis projection distance, dividing the interval [0,L] into several continuous segments. Each segment is 20m long. Each segment is then assigned a unique number (e.g., D001, D002, etc.), and all physical detectors within that segment are... The value is associated with the corresponding segment number to achieve a one-to-one mapping between the detector and the segment.
[0033] In another embodiment, assuming the total length of the dike is 3600m and the preset section length is 25m, it is divided into 144 sections; the sections are numbered sequentially from Q001 to Q144. A certain detector's... If the value is 1250.6m, it falls within the range of [1225m, 1250m) to [1250m, 1275m), and is assigned to segment Q051 after boundary determination. Statistical results show that among the 180 detectors, each segment has an average of 1-2 detector nodes, with some key segments having more than 3 nodes. Through the above numbering and marking, a structured correspondence is formed between "axis projection distance - segment number - detector number".
[0034] Preferably, step S2, which involves identifying the seepage state of the dike based on the dike physical dataset, includes: Pore water pressure parameters for each section are extracted based on the dam physical dataset; hydraulic gradient values between adjacent depths are calculated based on the pore water pressure parameters; if the hydraulic gradient value is greater than a preset hydraulic stability threshold, it is determined to be an abnormal seepage section. In one embodiment, pore water pressure time series data for each section are read from a pre-constructed dam physical dataset and sorted by burial depth (e.g., 0.5m, 1.5m, 2.5m). The pore water pressure difference between adjacent depths within the same section is calculated, and the hydraulic gradient value is obtained by combining the corresponding depth intervals. A preset hydraulic stability threshold is used as the criterion (e.g., a safe gradient range determined based on historical operational data). When the hydraulic gradient value of a section exceeds this threshold over multiple consecutive time steps, the section is marked as an abnormal seepage section, and the duration of the abnormality and the peak gradient are recorded.
[0035] In another embodiment, suppose a dike is divided into 60 sections, with three layers of pore pressure gauges installed in each section at 1.0m depth intervals. In a certain section, the pressure difference between the upper and lower pore pressure layers increases from 3kPa to 9kPa over six consecutive time points, corresponding to a hydraulic gradient value increasing from 0.3 to 0.9. If the preset hydraulic stability threshold is 0.7, then this section exceeds the threshold in the latter three time points and is identified as an abnormal seepage section. Statistical results show that the maximum hydraulic gradient values of eight sections are between 0.75 and 1.05, and these sections are included in the abnormal monitoring range.
[0036] Calculate the soil moisture content in the abnormal seepage zone; assess the trend of continuous seepage enhancement based on the soil moisture content; determine the seepage state of the dike based on the trend of continuous seepage enhancement.
[0037] In one embodiment, data from the corresponding moisture content sensor is retrieved within the identified abnormal seepage section. The average volumetric moisture content and rate of change at different depths are calculated, and their time-series trends are analyzed. When the moisture content shows a continuous upward trend within a certain time window, and the rate of increase exceeds a preset growth threshold, it is determined that the section has a continuous trend of enhanced seepage. Combining the hydraulic gradient changes, the seepage state of the dike is divided into stable seepage state, enhanced seepage state, and unstable seepage state, and corresponding state identifiers are output.
[0038] In another embodiment, it is assumed that in 5 out of 8 abnormal seepage zones, the moisture content of the zones increases from 22% to 31%, with an average growth rate of [missing information]. The duration exceeds 10 hours; if the preset growth threshold is... The five sections were determined to have a continuous increasing seepage trend. Among them, the peak water content of three sections exceeded 30% and was accompanied by a continuous increase in the hydraulic gradient, and was identified as an unstable seepage state; the remaining two sections, although the water content increased, the hydraulic gradient tended to be stable, and were identified as an enhanced seepage state.
[0039] Of particular importance is that the calculation of soil moisture content in abnormal seepage zones includes: Resistivity data and moisture sensor sampling data were collected in the abnormal seepage zone; temperature compensation was performed based on the resistivity data to generate compensated resistivity data. In one embodiment, resistivity electrode arrays and moisture sensors are deployed in layers along the depth direction within the identified abnormal seepage zone, simultaneously collecting resistivity data and moisture sensing sampling data at corresponding time points. The resistivity sampling cycle is set to once every 10 minutes, and the moisture sensing sampling cycle is set to once every 5 minutes, with data alignment achieved through timestamps. Since soil resistivity is significantly affected by temperature, soil temperature data is acquired simultaneously with resistivity data collection. The original resistivity is then compensated based on a preset temperature correction model to eliminate the influence of temperature fluctuations on resistivity, resulting in compensated resistivity data. After compensation, outliers are filtered and smoothed to obtain a stable compensated resistivity time series.
[0040] In another embodiment, suppose sensors are deployed at three depths (0.5m, 1.0m, 1.5m) in an abnormal section, and the original resistivity data are as follows: , , The corresponding temperatures are 25℃, 27℃, and 30℃. After temperature compensation, the compensated resistivity data are corrected to... , , Meanwhile, the original sampled values of volumetric water content collected by the moisture sensor were 22%, 28%, and 33%, respectively. After time alignment of the above data, three sets of spatially synchronized data were formed.
[0041] The first moisture content value is calculated based on the compensated resistivity data; the second moisture content value is calculated based on the moisture sensor sampling data; the first moisture content value and the second moisture content value are weighted and merged, and recorded as the soil moisture content.
[0042] In one embodiment, based on compensated resistivity data The mapping relationship model between the soil electrical parameters and the soil moisture content was used to calculate the first moisture content value. Simultaneously, the second moisture content value is obtained by directly converting or calibrating the moisture sensor sampling data. To improve the stability of the results, a weighted fusion model was constructed. and A weighted average is performed, where the weighting coefficients are determined based on the sensor's accuracy level or historical stability, for example, by setting... For resistivity weighting, For moisture sensing weights, and After fusion, the overall soil moisture content W is obtained, and longitudinal analysis is performed on data at different depths to identify abnormal gradient changes in moisture content.
[0043] In another embodiment, assuming a depth of 0.5m, the first water content value is calculated using the compensated resistivity. The second moisture content value was calculated from the moisture sensor sampling, which yielded a value of 24%. It is 22%; at a depth of 1.0m. 30%, It is 28%; at a depth of 1.5m. 35%, It is 33%. Set the weight. , The water content of the soil after fusion is as follows: at 0.5m ; 1.0m 1.5m If, after 6 hours of continuous monitoring, the moisture content at 1.5m increases from 34.2% to 38.5%, while at 0.5m it only increases from 23.2% to 24.0%, then it can be determined that the increase in moisture content in deeper soil layers is more significant, and there may be a trend of concentrated and enhanced seepage.
[0044] Preferably, the trend of continuous increase in permeability assessed based on soil moisture content includes: Extract the rate of increase of soil moisture content based on soil moisture content; determine the degree of change of soil porosity based on the rate of increase of soil moisture content; calculate the looseness coefficient of soil pore structure based on the degree of change of soil porosity. In one embodiment, continuous time-series soil moisture content data is retrieved within the identified abnormal seepage zone. The moisture content difference between adjacent time nodes is calculated, and a moisture content growth rate curve is obtained by combining the sampling time interval. The growth rate is smoothed using a sliding window, and the peak growth rate and average growth rate are extracted as trend features. Subsequently, based on the baseline porosity parameters of the soil type, the moisture content growth rate is correlated with the compaction coefficient of the corresponding depth layer to determine the degree of porosity change. When the growth rate is continuously increasing, it is determined that the porosity has an expansion trend. The porosity change amplitude is further normalized with the original porosity to construct a pore structure looseness coefficient, which is used to characterize the degree of soil skeleton transformation from a dense to a loose state.
[0045] In another embodiment, assuming that the moisture content of an abnormal section increases from 24% to 32% within 8 hours, with an average growth rate of 1.0% / h and a peak growth rate of 1.5% / h; the baseline porosity is 0.38, corresponding to a calculated porosity change of 0.05 (i.e., from 0.38 to 0.43); after normalization, the pore structure looseness coefficient is obtained as 0.13. If the looseness coefficients at multiple monitoring depths are 0.10, 0.13, and 0.16 respectively, then the weighted average of 0.13 can be taken as the comprehensive pore structure looseness coefficient for this section.
[0046] The permeability enhancement index is evaluated based on the soil pore structure looseness coefficient; the soil stability decay index is calculated based on the permeability enhancement index and preset pore water pressure parameters; and the continuous permeability enhancement trend is evaluated based on the soil stability decay index and the permeability enhancement index.
[0047] In one embodiment, the obtained pore structure looseness coefficient is input into the permeability mapping model and coupled with the original soil permeability coefficient to calculate the permeability enhancement index; the larger the looseness coefficient, the higher the enhancement index. Then, combined with the pore water pressure parameters of the monitoring section, including the current pore pressure value and its rate of change, a stability decay assessment model is constructed, outputting a soil stability decay index. A two-dimensional trend analysis is performed on the permeability enhancement index and the stability decay index. When both rise simultaneously within a continuous time window and exceed a preset synergistic threshold, it is determined that the section has a continuous permeability enhancement trend, and a risk level is output.
[0048] In another embodiment, assuming the pore structure loosening coefficient of a certain section is 0.13, the corresponding permeability enhancement index is calculated to be 0.21; the preset pore water pressure parameter is 85 kPa, which is 12 kPa higher than the benchmark pressure, and the pressure increase rate is 2 kPa / h. The soil stability decay index is calculated to be 0.18. Within six consecutive time nodes, if the permeability enhancement index increases from 0.15 to 0.23 and the stability decay index increases from 0.10 to 0.20, both exceeding the preset synergistic threshold of 0.17, then the section is determined to be in a state of continuous permeability enhancement. If either index falls below the threshold in subsequent time nodes, the trend level is adjusted from "high risk" to "medium risk" or "controllable risk". Through the above linkage analysis, the dynamic trend judgment of the seepage enhancement process of the embankment soil is realized.
[0049] Preferably, step S2, which involves dividing potential leakage channel sections based on seepage stability, includes: Extract the unstable segment numbers based on seepage stability; sort the unstable segment numbers in ascending order and record the unstable segment sequence numbers; mark consecutive unstable segments based on the unstable segment sequence numbers; In one embodiment, after completing the seepage stability analysis, the stability evaluation results of each monitoring section are classified. When the stability level of a section reaches a preset "instability level" threshold, the corresponding section number is extracted as a set of instability level section numbers. Then, the instability level section numbers are sorted in ascending order according to their spatial location to generate a list of instability level section sequence numbers. The list of sequence numbers is further traversed; when the difference between two adjacent section numbers is 1, they are determined to be consecutive instability sections, and multiple consecutive numbers are combined and marked as the same consecutive instability section. When the number interval is greater than 1, it is considered a boundary point between different instability sections, thereby completing the automatic identification and segmentation marking of consecutive instability sections.
[0050] In another embodiment, assuming a dike is divided into 80 monitoring sections, after seepage stability analysis, sections numbered 3, 4, 5, 9, 10, 18, 19, 20, and 21 are determined to be unstable. These sections are then sorted in ascending order to obtain the sequence numbers [3,4,5,9,10,18,19,20,21]. Based on the difference rule between adjacent numbers, these sections can be divided into three consecutive unstable segments: the first consecutive unstable segment is 3–5 (3 segments in total), the second consecutive unstable segment is 9–10 (2 segments in total), and the third consecutive unstable segment is 18–21 (4 segments in total).
[0051] Calculate the length and width of the continuous unstable section; calculate the area of the unstable section based on the length and width; mark potential leakage channels based on the area of the unstable section.
[0052] In one embodiment, the total length of the continuous unstable segment is calculated based on the number of segments contained in each continuous unstable segment and the standard length parameter of a single segment; simultaneously, the average width of the continuous unstable segment is determined by combining the lateral influence range or seepage diffusion width parameter of each segment. Subsequently, the area of the unstable segment is calculated based on the length and width, and compared with a preset area threshold. When the area of the unstable segment exceeds the threshold, the continuous unstable segment is marked as a potential leakage channel segment, and corresponding risk identification information is generated.
[0053] In another embodiment, it is assumed that the standard length of a single segment is 10m and the lateral influence width is 4m. For the first continuous instability segment (3–5), the length is 3×10m=30m, and the average width is 4m, so the area of the instability segment is 120m²; the second continuous instability segment (9–10) has a length of 20m and an area of 80m²; the third continuous instability segment (18–21) has a length of 40m and an area of 160m². If the preset potential leakage channel area threshold is 100m², then the first segment (120m²) and the third segment (160m²) are marked as potential leakage channel segments, while the second segment (80m²) is determined to be a local instability segment but does not form a channel-level risk.
[0054] Preferably, the potential leakage channel sections are marked according to the area of the unstable section, including: Determine the spatial coordinates of the unstable segment based on its area; perform path sorting based on the spatial coordinates of the unstable segment to form a connection diagram of the unstable elements; calculate the length of the continuous path based on the connection diagram of the unstable elements. In one embodiment, after calculating the area of the unstable segment, the spatial positioning information corresponding to each unstable segment is retrieved, including station coordinates, lateral offset coordinates, and elevation coordinates, to construct a set of spatial coordinates for the unstable segments. For each unstable segment, its geometric center point is used as a representative spatial coordinate point, and multiple center points are sorted according to the direction along the embankment axis. Subsequently, a connection diagram of unstable units is constructed based on the spatial distance and relative positional relationship between adjacent unstable segments. When the distance between two adjacent unstable segments is less than a preset connection threshold, a connection edge is established in the diagram, forming several continuous paths. The length of each continuous path is calculated by summing the results.
[0055] In another embodiment, assume four unstable segments are detected, with their center spatial coordinates as follows: S1 (chainage K10+020, lateral 2m, elevation 15m), S2 (chainage K10+045, lateral 3m, elevation 15.2m), S3 (chainage K10+080, lateral 2.5m, elevation 15.1m), and S4 (chainage K10+200, lateral 4m, elevation 14.9m). Setting the spatial connection threshold to 40m, a connection is established between S1 and S2 (25m distance), S2 and S3 (35m distance), and S3 and S4 (120m distance). This forms a continuous path S1–S2–S3 with a total path length of 25m + 35m = 60m; S4 forms a separate independent node with a path length of 0m. This 60m continuous path can be the focus of analysis.
[0056] When the length of a continuous path exceeds a preset breakthrough threshold, extract the spatial boundary information of the section; mark potential leakage channel sections based on the spatial boundary information of the section.
[0057] In one embodiment, the length of a continuous path is compared with a preset breakthrough threshold. When the path length exceeds the threshold, the path is determined to have a breakthrough risk. Then, the spatial boundary information of the section covered by the path is extracted, including the start and end station range, the lateral influence width, and the depth influence range, forming a three-dimensional spatial boundary box. Based on this spatial boundary box, the dike sections are marked, and sections within the boundary range are defined as potential seepage channel sections, generating risk level identifiers.
[0058] In another embodiment, assuming a preset breakthrough threshold of 50m, and the continuous path length formed in step S1 is 60m, it is determined that it exceeds the breakthrough threshold. The corresponding spatial boundary information is: starting station K10+020, ending station K10+080, with a total axial range of 60m; a maximum lateral expansion width of 4m; and a depth influence range of 3m below the surface of the embankment. Based on this, the system generates a spatial boundary box (60m×4m×3m) and uniformly marks the sections within this range as potential seepage channel sections. If the path length of another independent node S4 is 0m, which does not exceed the 50m threshold, it is only marked as a local instability point and does not constitute a breakthrough channel risk.
[0059] Preferably, in step S3, particle loss trend analysis is performed in the leakage channel section, and the particle migration intensity is recorded, including: The resistance change rate is detected in the seepage channel section; the change in soil porosity is assessed based on the resistance change rate; and the soil particle migration rate is calculated based on the change in soil porosity. In one embodiment, a resistivity monitoring electrode array is deployed within a marked seepage channel section. Soil resistivity data is collected at set time intervals, and the resistance values at adjacent time points are differentially processed to extract the resistance change rate curve. After denoising and smoothing the resistance change rate, the magnitude and trend of resistance change are evaluated in conjunction with the initial soil resistivity benchmark value. Based on the correspondence model between resistivity and porosity, the resistance change rate is mapped to the change in soil porosity; when resistivity continuously decreases, it is determined that the porosity is increasing. Combining soil particle size distribution parameters and seepage direction information, the soil particle migration rate is calculated based on the change in porosity.
[0060] In another embodiment, it is assumed that the resistivity of a certain leakage channel section changes from [value missing] to [value missing] within 6 hours. Descending to The rate of change of average resistance is The reference resistivity is The initial porosity is 0.36. Based on the mapping relationship, the porosity change is calculated to be +0.04 (i.e., increasing from 0.36 to 0.40). Under this porosity change condition, combined with particle size distribution parameters, the soil particle migration rate is calculated to be 0.8 mm / h. If the migration rates measured at different depths are 0.6 mm / h, 0.8 mm / h, and 1.1 mm / h, then the maximum value of 1.1 mm / h can be taken as the particle migration characteristic value for this section.
[0061] The migration rate of soil particles is compared with a preset migration intensity grading threshold, and the particle migration intensity is recorded simultaneously.
[0062] In one embodiment, a particle migration intensity grading standard is pre-constructed, dividing the migration rate into low, medium, and high migration level ranges. The soil particle migration rate is compared with the corresponding grading threshold. When the migration rate falls within different ranges, a corresponding migration intensity level identifier is generated, and the time, spatial location, and duration of the migration are recorded, enabling dynamic tracking of the degree of fine particle loss within the seepage channel.
[0063] In another embodiment, it is assumed that the preset migration intensity classification thresholds are: low migration (0–0.5 mm / h), medium migration (0.5–1.0 mm / h), and high migration (>1.0 mm / h). If the calculated particle migration rate of a certain segment is 1.1 mm / h, it is determined to be high migration intensity; if the migration rate of another adjacent segment is 0.7 mm / h, it is determined to be medium migration intensity. The system records the results as follows: segment A (1.1 mm / h, high migration, lasting 4 hours), segment B (0.7 mm / h, medium migration, lasting 6 hours). When the duration of high migration intensity exceeds the preset duration threshold (e.g., 3 hours), a further structural safety assessment or reinforcement warning instruction can be triggered, thereby realizing the graded control of the risk of particle loss inside the leakage channel.
[0064] Of particular importance is the comparison of soil particle migration rates with preset migration intensity grading thresholds, simultaneously recording particle migration intensity, including: When the migration rate of soil particles is less than 0.0005 m³ / h, it is judged as low-intensity migration; When the soil particle migration rate is between 0.0005 m³ / h and 0.005 m³ / h, it is considered to be of medium intensity. When the migration rate of soil particles is greater than 0.005 m³ / h, it is judged as high-intensity migration; In one embodiment, after obtaining soil particle migration rate data for each monitoring section, the migration rate is compared with a preset strength grading threshold. The system pre-sets three threshold intervals in the database: the lower strength threshold has an upper limit of [missing information]. The medium intensity range is The lower limit of the high-intensity threshold is Migration rate for each monitoring segment. Perform interval determination and generate corresponding intensity labels. .when When the intensity is less than the low-intensity threshold, output a "low-intensity migration" flag; when... When Vi is in the medium range, output a "medium intensity migration" flag; when Vi exceeds the high intensity threshold, output a "high intensity migration" flag. Simultaneously record the determination time, segment number, and duration.
[0065] In another embodiment, assume that the particle migration rates measured for five abnormal segments are 0.0003 m³ / h, 0.0008 m³ / h, 0.0025 m³ / h, 0.006 m³ / h, and 0.012 m³ / h, respectively. Based on a preset grading standard, segment A (0.0003 m³ / h) is classified as low-intensity migration; segment B (0.0008 m³ / h) as medium-intensity migration; segment C (0.0025 m³ / h) as medium-intensity migration; segment D (0.006 m³ / h) as high-intensity migration; and segment E (0.012 m³ / h) as high-intensity migration. Assuming that segment D maintains a rate above 0.006 m³ / h for 4 hours, and segment E increases from 0.004 m³ / h to 0.012 m³ / h within 2 hours, then segment E can be labeled as a "rapidly enhanced high-intensity migration segment."
[0066] Low-intensity migration, medium-intensity migration, and high-intensity migration are used as particle migration intensities.
[0067] In one embodiment, after completing the migration rate classification, the corresponding classification results are uniformly mapped to a particle migration intensity index S, and expressed in numerical grade form, for example, low intensity is set as S1, medium intensity as S2, and high intensity as S3. The system binds the migration intensity index with the spatial location and time axis of the section to construct a particle migration intensity distribution map, realizing intensity comparison and trend analysis between different sections. This particle migration intensity can be used as an input parameter for subsequent seepage risk level assessment, structural stability determination, and early warning triggering conditions.
[0068] In another embodiment, assuming a certain embankment section has 10 monitoring zones, 3 zones are identified as low-intensity migration (S1), 4 zones as medium-intensity migration (S2), and 3 zones as high-intensity migration (S3). Further assuming the high-intensity zones are concentrated in the downstream area of the embankment toe, and after 6 hours of continuous monitoring, 2 of these zones still maintain an S3 level, then the overall particle migration intensity of the embankment section can be evaluated as "locally concentrated high-intensity." If, during subsequent monitoring, a medium-intensity zone increases from 0.004 m³ / h to 0.007 m³ / h, its intensity upgrades from S2 to S3, achieving dynamic updating of the migration intensity level.
[0069] Preferably, calculating the soil particle migration rate based on the change in soil porosity includes: The porosity increment is calculated based on the change in soil porosity and the preset initial porosity; the particle volume reduction in the corresponding section is determined based on the porosity increment. In one embodiment, within the identified abnormal seepage zone, the initial porosity of that zone is retrieved. (Determined from embankment construction data or preliminary testing), combined with porosity changes obtained from real-time monitoring. Calculate the current porosity The porosity increment is obtained by comparing it with the initial porosity. Subsequently, based on the relationship between the section volume V and the initial solid particle volume fraction of the soil, the porosity increment was mapped to the particle volume reduction. Specifically, when the porosity increases, the volume ratio of solid particles decreases, thereby calculating the reduction in particle volume per unit segment and recording the corresponding spatial location and time point.
[0070] In another embodiment, assuming the volume V of an abnormal section is 10 m³, and the initial porosity is... The value is 0.35, and the current porosity is obtained through monitoring. If the value is 0.41, then the porosity increment is... Based on the volume distribution relationship, the volume of solid particles in this section decreased from 6.5 m³ to 5.9 m³. Therefore, the reduction in particle volume is... If another adjacent section has a volume of 8 m³, and its porosity increases from 0.33 to 0.37, then the particle volume reduction is 0.32 m³. By calculating the particle volume reduction for multiple sections separately, a spatial distribution map can be generated.
[0071] The particle reduction time is calculated based on the particle volume reduction; the soil particle migration rate is calculated based on the particle volume reduction and particle reduction time.
[0072] In one embodiment, the change curve of particle volume reduction in the same segment is recorded within a continuous monitoring period to identify the time interval in which the particle volume continues to decrease, and this time interval is defined as the particle reduction time. If the reduction in particle volume occurs in stages, then the duration of each stage should be recorded separately. The amount of particle volume reduction should be... The ratio is calculated with the corresponding particle reduction time T to obtain the particle volume loss per unit time. This value is then combined with the cross-sectional area of the section or the migration path length to convert it into a soil particle migration rate index.
[0073] In another embodiment, assuming the above-mentioned segment The volume reduction occurred within a continuous 12-hour monitoring period, therefore the particle reduction time T = 12 hours, and the particle volume loss per unit time is 0.05 m³ / h. If the effective seepage cross-sectional area of this section is 2 m², then the particle migration rate can be calculated to be 0.025 m / h. Another adjacent section... If the process is completed within 8 hours, the volume loss per unit time is 0.04 m³ / h, corresponding to a migration rate of 0.02 m / h. By comparing the migration rates of multiple sections, the development speed of particle loss and the potential piping risk level can be determined.
[0074] Preferably, this specification also provides a physical detection-based levee hazard investigation and analysis system for performing the above-mentioned physical detection-based levee hazard investigation and analysis method. This physical detection-based levee hazard investigation and analysis system includes: The physical detection module is used to acquire the outline of the dike and deploy physical detectors according to the outline; it collects the dike detection data from the physical detectors and performs data fusion based on the dike detection data to form a dike physical dataset; The seepage stability assessment module is used to identify the seepage state of the embankment based on the embankment physical dataset; assess the seepage stability based on the seepage state of the embankment; and delineate potential seepage channel sections based on the seepage stability. The piping identification module is used to perform particle loss trend analysis in the seepage channel section, record the particle migration intensity, and classify the piping risk level according to the particle migration intensity. The early warning and control module is used to generate graded early warning and control instructions based on the risk level of piping; to trigger early warning responses using the graded early warning and control instructions, and to simultaneously generate a report on the investigation and management of hidden dangers in the dike.
[0075] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.
[0076] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.
Claims
1. A method for investigating and analyzing potential hazards in dikes based on physical detection, characterized in that, Includes the following steps: Step S1: Obtain the outline of the dike and deploy physical detectors according to the outline of the dike; Collect dam detection data from physical detectors, and perform data fusion based on the dam detection data to form a dam physical dataset; Step S2: Identify the seepage state of the dike based on the dam physical dataset; assess the seepage stability based on the seepage state; and delineate potential seepage channel sections based on the seepage stability. Step S3: Perform particle loss trend analysis in the leakage channel section and record the particle migration intensity; Piping risk levels are classified based on particle migration intensity; Step S4: Generate a graded early warning and control instruction based on the piping risk level; trigger an early warning response using the graded early warning and control instruction, and simultaneously generate a levee hazard investigation and management report.
2. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 1, characterized in that, Step S1 involves collecting dam detection data from physical detectors and performing data fusion based on this data to form a dam physical dataset, which includes: Collect dam detection data from physical detectors, including pore water pressure data, soil moisture content data, and resistivity distribution data. The dam detection data is used to perform timestamp synchronization processing to form a unified time series; the installation coordinates of the corresponding physical detectors are read from the unified time series; the installation coordinates of the physical detectors are mapped to the dam outline to generate section numbers; and the dam physical dataset is constructed based on the section numbers.
3. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 2, characterized in that, Mapping the installation coordinates of the physical detectors to the dike outline generates section numbers including: Obtain the axis coordinate data of the embankment outline and establish the embankment axis coordinate system based on the axis coordinate data; transform the installation coordinates of the physical detector to the embankment axis coordinate system and calculate the projection distance of the installation coordinates of the physical detector in the embankment axis direction in the embankment axis coordinate system. The projection distance is divided according to the preset section length to determine the range of the dike section; the dike section is marked with numbers to generate section numbers.
4. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 1, characterized in that, Step S2, which involves identifying the seepage state of the dam based on the dam physical dataset, includes: Pore water pressure parameters for each section are extracted based on the dam physical dataset; hydraulic gradient values between adjacent depths are calculated based on the pore water pressure parameters; if the hydraulic gradient value is greater than a preset hydraulic stability threshold, it is determined to be an abnormal seepage section. Calculate the soil moisture content in the abnormal seepage zone; assess the trend of continuous seepage enhancement based on the soil moisture content; determine the seepage state of the dike based on the trend of continuous seepage enhancement.
5. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 4, characterized in that, Based on soil moisture content assessment, the trend of continued increased permeability includes: Extract the rate of increase of soil moisture content based on soil moisture content; determine the degree of change of soil porosity based on the rate of increase of soil moisture content; calculate the looseness coefficient of soil pore structure based on the degree of change of soil porosity. The permeability enhancement index is evaluated based on the soil pore structure looseness coefficient; the soil stability decay index is calculated based on the permeability enhancement index and preset pore water pressure parameters; and the continuous permeability enhancement trend is evaluated based on the soil stability decay index and the permeability enhancement index.
6. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 1, characterized in that, Step S2, which involves delineating potential leakage pathways based on seepage stability, includes: Extract the unstable segment numbers based on seepage stability; sort the unstable segment numbers in ascending order and record the unstable segment sequence numbers; mark consecutive unstable segments based on the unstable segment sequence numbers; Calculate the length and width of the continuous unstable section; calculate the area of the unstable section based on the length and width; mark potential leakage channels based on the area of the unstable section.
7. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 6, characterized in that, Potential leakage pathways are identified based on the area of the unstable segment, including: Determine the spatial coordinates of the unstable segment based on its area; perform path sorting based on the spatial coordinates of the unstable segment to form a connection diagram of the unstable elements; calculate the length of the continuous path based on the connection diagram of the unstable elements. When the length of a continuous path exceeds a preset breakthrough threshold, extract the spatial boundary information of the section; mark potential leakage channel sections based on the spatial boundary information of the section.
8. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 1, characterized in that, In step S3, particle loss trend analysis is performed in the leakage channel section, and the particle migration intensity is recorded, including: The resistance change rate is detected in the seepage channel section; the change in soil porosity is assessed based on the resistance change rate; and the soil particle migration rate is calculated based on the change in soil porosity. The migration rate of soil particles is compared with a preset migration intensity grading threshold, and the particle migration intensity is recorded simultaneously.
9. The method for investigating and analyzing potential hazards in dikes based on physical detection according to claim 8, characterized in that, Calculating soil particle migration rate based on changes in soil porosity includes: The porosity increment is calculated based on the change in soil porosity and the preset initial porosity; the particle volume reduction in the corresponding section is determined based on the porosity increment. The particle reduction time is calculated based on the particle volume reduction; the soil particle migration rate is calculated based on the particle volume reduction and particle reduction time.
10. A system for investigating and analyzing potential hazards in dikes based on physical detection, characterized in that, For executing the physical detection-based levee hazard investigation and analysis method as described in claim 1, the physical detection-based levee hazard investigation and analysis system includes: The physical detection module is used to acquire the outline of the dike and deploy physical detectors according to the outline; it collects the dike detection data from the physical detectors and performs data fusion based on the dike detection data to form a dike physical dataset; The seepage stability assessment module is used to identify the seepage state of the embankment based on the embankment physical dataset; assess the seepage stability based on the seepage state of the embankment; and delineate potential seepage channel sections based on the seepage stability. The piping identification module is used to perform particle loss trend analysis in the seepage channel section, record the particle migration intensity, and classify the piping risk level according to the particle migration intensity. The early warning and control module is used to generate graded early warning and control instructions based on the risk level of piping; to trigger early warning responses using the graded early warning and control instructions, and to simultaneously generate a report on the investigation and management of hidden dangers in the dike.
