A system and method for real-time evaluation of safety state of earth-rock dam based on multi-source data

Abstract

The present application relates to the technical field of hydraulic engineering, and more particularly to a soil and rock dam safety state real-time evaluation system and method based on multi-source data. The method comprises the following steps: performing seepage simulation calculation on the target soil and rock dam based on a preset multi-source working condition parameter combination, constructing a soil and rock dam seepage foundation dataset, and the soil and rock dam seepage foundation dataset comprising a seepage saturation line dataset and a single-width seepage amount dataset; determining the seepage monitoring points of the target soil and rock dam based on the seepage saturation line dataset; and establishing a parameterized relational expression set based on the seepage saturation line dataset and the single-width seepage amount dataset. Through the deviation comparison between real-time monitoring data and the multi-state parameterized model, the present application can accurately distinguish normal water level fluctuations from precursors of anti-seepage body damage, and realize the real-time identification and early warning of early signs of progressive damage.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering technology, and in particular to a real-time safety status assessment system and method for earth-rock dams based on multi-source data. Background Technology

[0002] In the management of medium-sized mountain reservoirs with frequent seasonal water level fluctuations, on-duty personnel collect data from seepage gauges and measuring weirs at regular intervals every day, and judge the safety status by comparing historical curves with empirical thresholds. This traditional method can cope with the normal working conditions of the seepage prevention and drainage system working together, but it is difficult to identify the early signs of gradual failure.

[0003] The main problem with existing technologies lies in the safety monitoring blind spots caused by the structural redundancy of earth-rock dams. When local leakage occurs in the seepage barrier, the drainage system can often temporarily maintain normal seepage discharge, resulting in slight changes in water level and seepage flow at downstream monitoring points, which remain within the traditional early warning threshold range. Under these conditions, the monitoring system has difficulty distinguishing between normal water level fluctuations and precursors to seepage barrier failure, and cannot detect potential safety hazards in a timely manner.

[0004] Another problem lies in the unreasonable layout of monitoring points. Existing monitoring systems rely on discretely deployed piezometers to form monitoring sections, but actual seepage anomalies may not be effectively reflected at the monitoring points, resulting in delayed or inaccurate responses captured by the monitoring system. Summary of the Invention

[0005] Therefore, it is necessary for the present invention to provide a real-time safety status assessment system and method for earth-rock dams based on multi-source data, in order to solve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, a real-time safety assessment method for earth-rock dams based on multi-source data is proposed, comprising the following steps: Step S1: Perform seepage simulation calculations on the target earth-rock dam based on the preset multi-source working condition parameter combination, and construct the earth-rock dam seepage basic dataset. The earth-rock dam seepage basic dataset includes seepage infiltration line dataset and unit width leakage dataset. Step S2: Determine the seepage monitoring points of the target earth-rock dam based on the seepage infiltration line dataset; Step S3: Establish a set of parameterized relational expressions based on the seepage line dataset and the unit width leakage dataset; Step S4: Based on the parameterized relation set, calculate the fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operating conditions, and compare them with the normal operating data in the seepage infiltration line dataset and unit width leakage dataset to obtain the fitting deviation rate set of each seepage monitoring point. Step S5: Collect multi-source monitoring data of the target earth-rock dam in real time; compare the multi-source monitoring data with the calculation results of the parameterized relation set to determine the instantaneous safety state of the target earth-rock dam.

[0007] Preferably, the present invention also provides a real-time safety assessment system for earth-rock dams based on multi-source data, used to execute the real-time safety assessment method for earth-rock dams based on multi-source data as described above. The real-time safety assessment system for earth-rock dams based on multi-source data includes: The simulation calculation module is used to perform seepage simulation calculations on the target earth-rock dam based on a preset multi-source working condition parameter combination, and to construct the earth-rock dam seepage basic dataset, which includes seepage infiltration line dataset and unit width leakage dataset. The point planning module is used to determine the seepage monitoring points of the target earth-rock dam based on the seepage infiltration line dataset; The parametric modeling module is used to establish a set of parametric relationships based on the seepage line dataset and the unit width leakage dataset; The deviation calculation module is used to calculate the fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operation based on the set of parameterized relational formulas, and compare them with the normal operation data in the seepage infiltration line dataset and the unit width leakage dataset to obtain the fitting deviation rate set of each seepage monitoring point. The safety assessment module is used to collect multi-source monitoring data of the target earth-rock dam in real time; compare the multi-source monitoring data with the calculation results of the parameterized relation set to determine the instantaneous safety state of the target earth-rock dam.

[0008] The beneficial effects of this invention are as follows: On the one hand, by constructing a set of parameterized relational formulas through multi-condition seepage simulation, a benchmark data system for the seepage field under three states—normal operation, seepage barrier failure, and drainage failure—is established, breaking the structural redundancy monitoring blind spot formed by the synergistic effect of the seepage barrier and drainage system in traditional monitoring. This system can capture the weak seepage anomalies caused by the drainage system temporarily maintaining discharge when there is local leakage in the seepage barrier. By comparing the deviation between real-time monitoring data and the multi-state parameterized model, it can accurately distinguish between normal water level fluctuations and precursors of seepage barrier failure, achieving immediate identification and early warning of early signs of gradual damage.

[0009] On the other hand, by using spatial variation coefficient analysis and fitting error optimization algorithms based on seepage line datasets, the layout of highly sensitive monitoring points inside the dam body was systematically determined. This overcomes the spatiotemporal mismatch problem caused by the existing discrete deployment of piezometers, ensuring that monitoring points can directly respond to changes in the seepage field in key structural areas, enabling abnormal signals to be effectively captured at the source stage, and significantly improving the spatial coverage rationality and response timeliness of the monitoring system.

[0010] On the other hand, by using a set of parameterized relational formulas to dynamically calculate the deviation of real-time water level and seepage flow data at monitoring points, and by matching the deviation direction characteristics with preset fault modes, the system can immediately determine whether the dam's safety status is normal or not. In abnormal conditions, it can automatically identify the specific type of seepage prevention damage or drainage failure, thus realizing the transformation from single threshold judgment to intelligent assessment based on multi-source data fusion and multi-state pattern recognition, improving the accuracy of safety diagnosis and the appropriateness of emergency response. Attached Figure Description

[0011] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings: Figure 1 A flowchart illustrating the steps of an embodiment of a real-time safety status assessment method for earth-rock dams based on multi-source data is shown.

[0012] Figure 2 A schematic diagram illustrating the statistical analysis of the fitting deviation rate at each monitoring point in an embodiment is shown.

[0013] Figure 3 A schematic diagram of a module of an embodiment of a real-time safety status assessment system for earth-rock dams based on multi-source data is shown. Detailed Implementation

[0014] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0015] 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.

[0016] 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.

[0017] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a method for real-time safety assessment of earth-rock dams based on multi-source data, comprising the following steps: Step S1: Perform seepage simulation calculations on the target earth-rock dam based on the preset multi-source working condition parameter combination, and construct the earth-rock dam seepage basic dataset. The earth-rock dam seepage basic dataset includes seepage infiltration line dataset and unit width leakage dataset. Step S2: Determine the seepage monitoring points of the target earth-rock dam based on the seepage infiltration line dataset; Step S3: Establish a set of parameterized relational expressions based on the seepage line dataset and the unit width leakage dataset; Step S4: Based on the parameterized relation set, calculate the fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operating conditions, and compare them with the normal operating data in the seepage infiltration line dataset and unit width leakage dataset to obtain the fitting deviation rate set of each seepage monitoring point. Step S5: Collect multi-source monitoring data of the target earth-rock dam in real time; compare the multi-source monitoring data with the calculation results of the parameterized relation set to determine the instantaneous safety state of the target earth-rock dam.

[0018] Most importantly, the preset multi-source working condition parameter combinations in step S1 include different dam height parameters, different dam slope parameters, different seepage prevention body layout parameters, different drainage body layout parameters, and different upstream and downstream water level parameters; Among them, different anti-seepage body layout parameters include parameters for local leakage due to anti-seepage body failure, different drainage body layout parameters include parameters for drainage body failure, and different upstream and downstream water level parameters include stable water level conditions and unstable water level conditions.

[0019] Preferably, step S2 includes: Based on the seepage infiltration line dataset, statistical analysis was conducted to determine the locations of seepage monitoring points for the target earth-rock dam. Among them, the seepage monitoring points include at least a first monitoring point, a second monitoring point, and a third monitoring point; The first monitoring point is located on the upstream side of the dam axis and in the upstream area of ​​the core wall seepage prevention body; the second monitoring point is located in the upstream area of ​​the drainage prism or the slope drainage; and the third monitoring point is located at the downstream slope 1 / 2 of the dam height. In this embodiment, the spatial positioning of the first monitoring point needs to follow two related principles: first, it should be located close to the dam axis to ensure that the monitoring data can represent the seepage state of the main structural part of the dam; second, it should be located on the upstream side of the core wall seepage barrier so that it is in the initial influence area of ​​the seepage field and can detect upstream water level fluctuations or possible leakage channels on the upstream side of the seepage barrier as early as possible.

[0020] In addition, when designing the first monitoring point, it is necessary to calculate the spatial coordinates of the dam axis based on the geometric parameters of the dam cross-section, and determine the interface on the upstream side of the core wall by combining the design outline of the core wall seepage prevention body. The monitoring point can be arranged within a range of 0.5 to 2.5 meters horizontally from the dam axis and 0.3 to 1.0 meters from the upstream outline of the core wall.

[0021] When the target earth-rock dam is equipped with a core wall seepage barrier, a fourth monitoring point is added, which is located on the downstream side of the core wall seepage barrier.

[0022] In this embodiment, the setting conditions for the fourth monitoring point depend on whether the dam body is equipped with a core wall seepage prevention structure. When the target earth-rock dam is detected to be a core wall dam or contains a core wall seepage prevention structure, the procedure for adding the fourth monitoring point is automatically triggered. The addition of the fourth monitoring point follows the principle of key defense and redundant monitoring. It does not change the layout of the original three basic monitoring points, but adds a monitoring point for the core wall seepage prevention structure, forming a four-point monitoring network.

[0023] Preferably, based on the seepage infiltration line dataset, statistical analysis reveals that the seepage monitoring points for the target earth-rock dam include: Step S211: Extract water level data at potential locations within the dam body under all working conditions from the seepage line dataset to form a spatial distribution set of water level data; In this embodiment, data extraction and reconstruction are required from the seepage saturation line dataset. The seepage saturation line dataset contains simulation results of various spatial nodes inside the dam body under hundreds of different working condition combinations. Each result records the seepage field distribution under a specific upstream water level, a specific seepage barrier state, and a specific drainage condition. These discrete simulation results are integrated into a unified spatial distribution set of water level data.

[0024] In practice, all working condition entries in the seepage line dataset are traversed. For the pre-divided three-dimensional grid nodes inside the dam body, the water level value of each node under different working conditions is extracted and indexed by both spatial coordinates and working condition number. The three-dimensional coordinate position of each node, the attributes of the dam body structural region to which it belongs, and the combination of working condition parameters are recorded.

[0025] Step S212: Calculate the coefficient of variation of water level at each spatial location point in the spatial distribution set of water level data, and generate a spatial coefficient of variation distribution map; In this embodiment, it is necessary to calculate the ratio of the standard deviation to the mean of the water level data for each spatial node under all operating conditions, and use the calculated ratio as the coefficient of variation. Each node in the spatial distribution set of water level data is traversed sequentially, and all operating condition water level data associated with that node are read. First, the arithmetic mean of these water level data is calculated, then the sum of squares of the deviations of each data point from the mean is calculated, and the standard deviation is obtained. Further, the standard deviation is divided by the mean to obtain the coefficient of variation. The coefficient of variation value of each node is remapped back to its spatial coordinate position, generating a spatial coefficient of variation distribution map consistent with the geometry of the dam body. Different shades of color can be used in the map to represent the magnitude of the coefficient of variation.

[0026] Step S213: Based on the spatial coefficient of variation distribution map, select a set of highly sensitive points with a coefficient of variation higher than the first preset threshold; In this embodiment, a first threshold needs to be preset. This threshold is set between 0.15 and 0.20 based on engineering experience to distinguish between highly sensitive areas and generally sensitive areas. All nodes in the spatial coefficient of variation distribution map are traversed, and the coefficient of variation value of each node is compared with the first threshold. If the coefficient of variation of a node is higher than the threshold, the node is marked as a highly sensitive point and added to the set of highly sensitive points. During the selection process, the spatial coordinates, specific value of the coefficient of variation, and the dam structure area to which the point belongs are recorded for each selected point.

[0027] Step S214: For the set of highly sensitive points, spatial clustering is performed in combination with the structural characteristics of the target earth-rock dam to identify representative points located near the key structural area and form an initial set of monitoring points; wherein, the key structural area includes at least the upstream side of the seepage barrier, the upstream side of the drainage body, and the middle of the downstream slope. In this embodiment, it is necessary to call the structural design parameters of the target earth-rock dam, including key information such as the dam axis position, the geometric contour of the core wall or inclined wall, the size and position of the drainage prism, and the slope ratio of the upstream and downstream dam slopes, to reconstruct the main structural zones of the dam in three-dimensional space. Using the key structural region as a spatial constraint, clustering operations are performed on the set of highly sensitive points, grouping multiple highly sensitive nodes that are physically close and located within the same key structural region into one cluster. Each cluster represents the sensitive response center of a key structural region, and the node with the highest coefficient of variation and closest to the core of the structural region is selected from each cluster as the representative point of that region. The key structural region at least covers three core functional areas: the upstream side of the seepage barrier, the upstream side of the drainage body, and the middle of the downstream slope. The resulting initial monitoring point set contains 3 to 5 representative points.

[0028] Step S215: Based on the seepage infiltration line dataset, calculate the average fitting error between the water level line connecting any three points in the initial monitoring point set and the actual complete infiltration line; In this embodiment, a fitting error analysis method can be used. Three points are arbitrarily selected from the initial monitoring point set. Based on the simulation results of each working condition in the seepage line dataset, the water level values ​​of these three points under each working condition are extracted. A fitted seepage line is generated by connecting the water levels of these three points using a linear interpolation method. This fitted seepage line is compared with the actual complete seepage line obtained from the simulation calculation under the same working condition. The horizontal deviation of the two lines at each elevation position within the cross-section of the dam body is calculated, and the average value of all elevation position deviations is obtained, thus yielding the average fitting error of the three-point combination. The above calculation process is repeated for all possible three-point combinations in the initial monitoring point set to obtain the average fitting error value for each three-point combination, forming a fitting error matrix.

[0029] Step S216: Find the combination of points with the smallest average fitting error from the initial set of monitoring points, and determine the combination as the seepage monitoring points.

[0030] In this embodiment, an optimization algorithm is used to find the optimal combination of monitoring points from the initial set of monitoring points. A traversal optimization calculation is performed, comparing the average fitting error of each three-point combination, and selecting the combination with the smallest error value. This combination is considered the optimal layout that achieves the highest accuracy in reconstructing the seepage field with the fewest number of points. Once the combination with the smallest average fitting error and a reasonable spatial distribution is found, the three points in this combination are determined as the final seepage monitoring points, and the precise spatial coordinates, the key structural region to which they belong, and the corresponding coefficient of variation, etc., of each point are output.

[0031] Of particular importance are the specific steps for obtaining the structural characteristics of the target earth-rock dam, including: Obtain the design drawings and as-built data of the target earth-rock dam, and extract the set of geometric parameters of the dam structure from the design drawings and as-built data; The set of geometric parameters for the dam structure includes the dam axis plane coordinate sequence, dam crest elevation, upstream dam slope ratio and slope segmentation information, and downstream dam slope ratio and slope segmentation information.

[0032] Determine the spatial profile of the seepage prevention structure based on the set of geometric parameters of the dam structure; The spatial outline of the seepage prevention structure is obtained by extracting the top elevation, bottom elevation, and upstream and downstream slope ratio of the core wall or inclined wall from the design drawings, and calculating its three-dimensional spatial boundary based on the dam axis coordinates.

[0033] Determine the spatial profile of the drainage structure based on the set of geometric parameters of the dam structure; The spatial outline of the drainage structure is obtained by extracting the geometric dimensions, top elevation, slope ratio, and positional relationship with the downstream slope toe of the drainage prism, slope drainage, or mattress drainage from the design drawings, and calculating its three-dimensional spatial boundary.

[0034] Based on the obtained dam axis, seepage prevention body outline and drainage body outline, the key structural areas are marked on the dam cross-section diagram. The key structural areas include at least the area adjacent to the upstream side of the seepage barrier, the area adjacent to the upstream side of the drainage body, and the area where half the height of the downstream slope is from the toe to the top of the dam.

[0035] Using the critical structural region as a spatial constraint, the set of highly sensitive points is clustered. Multiple highly sensitive points falling within the same critical structural region or its preset buffer zone are grouped into a cluster. From each cluster, the point closest to the core of the critical structural region and with the highest spatial variation coefficient is selected as the representative point of the region, forming the initial monitoring point set.

[0036] Preferably, step S3 includes: Based on the seepage line dataset and the unit width leakage dataset, parameterized relationships under normal operating conditions are established respectively; In this embodiment, a subset of simulated data marked as operating conditions is selected from the seepage infiltration line dataset and the unit width leakage dataset. Operating conditions refer to all simulated cases where the permeability coefficient of the impermeable body remains at the design value, the drainage body is in good working order, and the upstream and downstream water levels are within a stable or normal scheduling range. For each seepage monitoring point determined in step S2, the water level data sequence and the corresponding unit width leakage data sequence under operating conditions are extracted. Furthermore, the correlation between water level changes at each monitoring point and upstream water level, downstream water level, and dam geometric parameters is analyzed, and an empirical relationship is established using a multiple regression method. Similarly, a quantitative relationship is established between the unit width leakage and parameters such as the upstream and downstream water level difference, dam height, and impermeable body thickness. These relationships are expressed algebraically, allowing direct calculation of the theoretical water level and theoretical leakage at each monitoring point under operating conditions based on the input operating parameters.

[0037] Based on the seepage line dataset and the unit width leakage dataset, parameterized relationships are established for the failure of the impermeable body. In this embodiment, when establishing the parameterized relationship under the seepage barrier failure state, various working condition data simulating local or overall failure of the seepage barrier are extracted from the seepage infiltration line dataset and the unit width leakage data dataset. The seepage barrier failure conditions are mainly simulated by increasing the permeability coefficient of the seepage barrier material by an order of magnitude or by setting up a through-seepage channel. The deviation characteristics of the water level data and unit width leakage data at each seepage monitoring point relative to the normal operating state under these extreme conditions are analyzed. This invention determines through analysis that when the seepage barrier fails, the water level at the first monitoring point will be significantly higher than the theoretical value under normal operating conditions, while the water level at the fourth monitoring point located downstream of the seepage barrier will rise the most. For this abnormal response mode, a specific relationship under the seepage barrier failure state is established.

[0038] Based on the seepage line dataset and the unit width leakage dataset, parameterized relationships are established for drainage failure conditions. In this embodiment, when constructing the parameterized relationship for the drainage body failure state, data simulating the functional degradation or complete failure of the drainage body are retrieved from the seepage infiltration line dataset and the unit width leakage dataset. The drainage body failure condition is simulated by reducing the permeability coefficient of the drainage material or decreasing the effective flow area of ​​the drainage body. It is crucial to analyze the water level response characteristics at each monitoring point during drainage body failure, particularly the second and third monitoring points. This invention determines through analysis that when the drainage body fails, the overall water level inside the dam rises, the infiltration line shows an upward trend, the downstream slope outflow point rises, the water level at the second monitoring point rises significantly, and the pore water pressure at the third monitoring point increases significantly.

[0039] Among them, each parameterized relational expression expresses the mathematical relationship between the water level data and seepage flow data of each seepage monitoring point and the upstream and downstream water level data, dam height data, and dam slope data. The upstream and downstream water level data includes water level change rate data.

[0040] Preferably, step S4 includes: The fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operating conditions are calculated using a set of parameterized relational formulas. In this embodiment, a set of parameterized relational expressions needs to be invoked, and a subset of expressions marked as representing normal operating conditions is selected. Typical parameter combinations for normal operating conditions are extracted from the seepage line dataset, including boundary condition data such as upstream water level, downstream water level, dam height, and dam slope. These parameters are then substituted as input variables into the parameterized relational expressions for each monitoring point. For each monitoring point, a forward calculation is performed to obtain the theoretical water level and theoretical unit width seepage rate under normal operating conditions. These calculated values ​​are referred to as the fitted water level data and fitted seepage rate data. Calculations are performed for each monitoring point under various typical normal operating conditions to form a fitted data matrix covering different operating conditions.

[0041] The fitted water level data and fitted seepage flow data are compared with the normal operation data in the seepage infiltration line dataset and the unit width leakage dataset, respectively. The water level fitting deviation rate and seepage flow fitting deviation rate of each seepage monitoring point are calculated to form a fitting deviation rate set. In this embodiment, simulated normal operating conditions that perfectly correspond to the fitted calculations are extracted from the original seepage line dataset and unit width leakage dataset as the true values. For each monitoring point, the water level value obtained from the fitted calculation is compared with the simulated actual water level value under each operating condition. The percentage of the absolute value of the difference between the two to the actual water level value is calculated to obtain the water level fitting deviation rate for that point. Furthermore, the same method is applied to the unit width leakage to calculate the deviation rate between the fitted seepage flow and the actual seepage flow. Statistical analysis is performed on the deviation rate of each monitoring point under all normal operating conditions, and its mean and standard deviation are calculated. The statistical results of the deviation rate of all monitoring points are then compiled to form a fitting deviation rate set.

[0042] In this case, please refer to Figure 2 The chart shows the statistical analysis of the fitting deviation rate for each monitoring point. It is important to note that: Figure 2 The fitting deviation rate set is calculated based on 120 normal operating conditions. When the measured value exceeds the allowable deviation threshold, an anomaly judgment is triggered.

[0043] Figure 2 A comparative chart of the parametric relationships of water levels at monitoring points under different operating conditions is presented. The chart uses upstream water level (horizontal axis, 25-40 meters) as the independent variable and monitoring point water level (vertical axis, 0-40 meters) as the dependent variable, plotting parametric relationship curves for three operating conditions. Solid curves represent normal operation, dashed curves represent impermeable body failure, and dotted lines represent drainage body failure. The curve for the first monitoring point (upstream side of the core wall) shows that under normal operation, its water level increases linearly with the upstream water level, with a slope of approximately 0.85; when the impermeable body fails, the water level at this point rises significantly, and the curve shifts upwards overall. The fourth monitoring point (downstream side of the core wall) shows a more significant abnormal increase in water level under the impermeable body failure condition, with its curve shifting upwards more significantly than that of the first monitoring point. The second monitoring point (upstream side of the drainage body) shows a significant rise in water level under the drainage body failure condition, indicating that the degradation of the drainage body's function leads to an increase in the water level in the local seepage field. The map clearly marks typical response points when the upstream water level is 34 meters and 37 meters, visually demonstrating the water level differences at monitoring points under different operating conditions. The legend in the upper right corner clearly distinguishes the line markings for the three operating conditions.

[0044] The fitting deviation rate set includes the allowable deviation range of each seepage monitoring point under normal operating conditions.

[0045] Preferably, the comparison between the multi-source monitoring data and the calculation results of the parameterized relation set in step S5 includes: Calculate the deviation between the real-time data of each seepage monitoring point in the multi-source monitoring data and the corresponding theoretical value in the calculation results of the parameterized relation set; In this embodiment, during the real-time evaluation phase, real-time data from each seepage monitoring point is collected through an automated monitoring system. The measured water level value at the current moment for each monitoring point is extracted from the multi-source monitoring data stream, and boundary condition parameters such as upstream and downstream water levels are also obtained. These boundary condition parameters are input into a parameterized relational set, and the relational formula under normal operating conditions is called to calculate the theoretical water level value and theoretical unit width seepage value for each monitoring point under the current operating conditions. Further, deviation calculation is performed: the measured water level value is subtracted from the theoretical water level value to obtain the absolute value of the water level deviation, and then this absolute value is divided by the theoretical water level value to convert it into a relative deviation rate. Further, the same method is applied to the seepage data to calculate the relative deviation rate between the measured seepage and the theoretical seepage. The above calculations are performed for each monitoring point to obtain a data set containing the water level deviation rate and the seepage flow deviation rate.

[0046] When the deviation values ​​of all seepage monitoring points are less than the fitting deviation rate corresponding to the set of fitting deviation rates, the safety state of the target earth-rock dam is determined to be normal. In this embodiment, the fitted deviation rate set is read to obtain the upper limit of the preset allowable deviation range for each monitoring point under normal operating conditions. The actual deviation rate of each point is compared with the corresponding upper limit of the allowable deviation. If the water level deviation rate and seepage flow deviation rate of all monitoring points are lower than their respective upper limits, the seepage behavior of the earth-rock dam is determined to meet the normal operating mode, and the safety assessment result is normal. After being determined to be normal, the current monitoring data acquisition frequency is maintained, no early warning or emergency mechanisms are triggered, the current assessment result is only recorded in the normal operating log, and the data acquisition and assessment process for the next monitoring cycle continues.

[0047] When the deviation value of any seepage monitoring point is greater than the fitting deviation rate corresponding to the fitting deviation rate set, the safety state of the target earth-rock dam is determined to be abnormal. In this embodiment, a veto mechanism is employed during the comparison process. If the real-time deviation rate of any seepage monitoring point exceeds its permissible deviation limit for normal operation, the overall safety status of the target earth-rock dam is deemed abnormal, regardless of whether other points are normal. This judgment logic is based on the principle of the integrity of seepage safety, because the dam body is a continuous hydraulic system, and local anomalies can quickly develop into overall failure. Once an anomaly is determined, an anomaly response process is triggered, updating the safety status marker to an abnormal state, sending emergency alarm information to the monitoring center and maintenance personnel, automatically increasing the monitoring data acquisition frequency from the normal mode to the high-frequency monitoring mode, and packaging all current monitoring data, calculated deviation values, and judgment results into an anomaly status record, which is then stored in the anomaly event database.

[0048] When the safety state of the target earth-rock dam is abnormal, the deviation direction characteristics of the real-time data of the seepage monitoring points are determined based on multi-source monitoring data. The type of safety anomaly of the target earth-rock dam is determined based on the characteristics of the deviation direction.

[0049] Preferably, determining the type of safety anomaly of the target earth-rock dam based on the characteristics of the deviation direction includes: When the deviation direction characteristics are consistent with the parameterized relationship characteristics under the preset seepage barrier failure condition, the safety state of the target earth-rock dam is determined to be seepage barrier failure anomaly. In this embodiment, after identifying the deviation direction features, they are compared with a preset template of abnormal seepage barrier damage features. The preset template of abnormal seepage barrier damage features is a typical feature pattern derived from the parameterized relationship of seepage barrier damage. Its main characteristics are: the first monitoring point upstream of the seepage barrier and the fourth monitoring point downstream of the seepage barrier (if present) both show significant positive deviations, meaning the real-time monitored water level is significantly higher than the theoretical value, and the deviation of the fourth monitoring point is usually greater than that of the first monitoring point; while the second monitoring point upstream of the drainage body has a smaller deviation or is close to the normal range, and the third monitoring point in the middle of the downstream slope may show a slight positive deviation. By calculating the similarity between the currently measured deviation direction feature vector and the seepage barrier damage template, when the similarity exceeds a preset threshold, the two are determined to be consistent.

[0050] When the deviation direction characteristics are consistent with the parameterized relationship characteristics under the preset drainage body failure condition, the safety state of the target earth-rock dam is determined to be drainage body failure anomaly.

[0051] In this embodiment, after the deviation direction feature does not match the impermeable body failure template, it automatically switches to compare with the drainage body failure anomaly feature template. The drainage body failure template is extracted based on the parameterized relational feature of drainage body failure. Its typical manifestations are: the second monitoring point located on the upstream side of the drainage body shows a significant positive deviation, and the real-time water level is significantly higher than the theoretical value; the third monitoring point located in the middle of the downstream slope also shows a significant positive deviation, and the deviation amplitude is usually greater than that of the second monitoring point; while the deviations of the first monitoring point located on the upstream side of the impermeable body and the fourth monitoring point located on the downstream side are relatively small or within the normal range. This is because the failure of the drainage body causes the seepage to be unable to be effectively discharged, causing the water level in the downstream area to rise as a whole, but the impact on the upstream impermeable area is relatively delayed. The similarity between the current feature vector and the drainage body failure template is calculated. When the similarity reaches the judgment standard, the feature is confirmed to be consistent and the drainage body failure is diagnosed as abnormal.

[0052] Preferably, when the safety state of the target earth-rock dam is abnormal, the deviation direction characteristics of the real-time data of the seepage monitoring points determined based on multi-source monitoring data include: For each seepage monitoring point, the difference between its real-time water level data and the theoretical water level data under normal operating conditions calculated by the parameterized relation set is calculated to obtain the water level deviation value of that point; the water level deviation values ​​of all monitoring points are combined in a preset order to form a relative deviation vector. In this embodiment, each established seepage monitoring point is traversed, and the real-time water level monitoring value of that point at the current moment is read. Simultaneously, the relationship formula for normal operating conditions is invoked from the parameterized relationship formula set. The current upstream and downstream water levels, dam parameters, and other input conditions are substituted into the relationship formula to calculate the theoretical water level value that the point should have under the current operating conditions. The water level deviation value is obtained by subtracting the theoretical water level value from the real-time water level value. The sign of the deviation value indicates the direction of deviation, and the absolute value indicates the degree of deviation. Following the preset order of the monitoring point numbers, arranged in the order of first monitoring point, second monitoring point, third monitoring point, and fourth monitoring point, the water level deviation values ​​of each point are arranged sequentially to form a relative deviation vector.

[0053] The water level deviation values ​​of each point in the relative deviation vector are compared with the allowable deviation range corresponding to the fitted deviation rate set. Deviation points that exceed the allowable range are selected, and the deviation direction of each deviation point is recorded, including positive and negative deviations. In this embodiment, the fitted deviation rate set is read to obtain the upper limit of the allowable deviation range for each monitoring point under normal operating conditions. This upper limit is usually stored as a relative percentage and converted into an absolute water level difference. Each component in the relative deviation vector is traversed, and the absolute deviation value of that point is compared numerically with the corresponding allowable upper limit. If the deviation value exceeds the allowable upper limit, the point is marked as a deviation point, and its deviation direction is further recorded according to the sign of the deviation value: a positive value indicates a positive deviation, representing an abnormal rise in water level, and a negative value indicates a negative deviation, representing an abnormal drop in water level. If the deviation value is within the allowable range, the point is marked as a normal point and is not included in subsequent analysis. Two lists are established: one to store the deviation point number and deviation direction, and the other to store the deviation magnitude of the deviation point.

[0054] The spatial location information of all deviation points is combined with their corresponding deviation directions to generate deviation direction features.

[0055] In this embodiment, the spatial location information of each deviation point is read, including the point's three-dimensional coordinates in the dam coordinate system, the type of key structural region it belongs to, and its orientation relative to the dam axis. This spatial location information is paired and combined with the deviation direction marker of the point to form a deviation direction feature description. This description includes not only which points deviated and whether the deviation direction is positive or negative, but also the spatial distribution pattern of these points, such as whether they are concentrated in the seepage prevention area, drainage area, or downstream slope area. This feature is output in a standardized data format as a feature vector or feature matrix containing three elements: point number, spatial coordinates, and deviation direction.

[0056] Preferably, the acquisition of multi-source monitoring data in step S5 includes: An automated monitoring system is deployed around the target earth-rock dam. The automated monitoring system includes rain gauges, water level gauges, seepage flow meters, and seepage pressure gauges. In this embodiment, a complete multi-parameter automated monitoring system is deployed at the target earth-rock dam site. This system consists of four core sensor devices, each corresponding to a different monitored physical quantity. Rain gauges are deployed in open areas of the dam site to monitor the rainfall intensity of the watershed; water level gauges are installed in the upstream reservoir and downstream river channel to obtain the water level at the upstream and downstream boundaries; seepage flow meters are installed at the centralized outlet of the downstream drainage system to measure the total seepage; and piezometers are buried at designated locations inside the dam body according to the monitoring point coordinates determined in step S2.

[0057] In this invention, the monitoring system adopts a distributed architecture. Various sensors are connected to the data acquisition terminal via cables or wireless networks. The terminal equipment has multi-channel synchronous acquisition, data buffering, and remote transmission functions. All sensor devices are calibrated, and their range and accuracy meet the requirements of the earth-rock dam safety monitoring specifications. The installation locations avoid areas with strong electromagnetic interference and mechanical vibration.

[0058] The automated monitoring system collects rainfall data, upstream water level data, downstream water level data, seepage data, and water level data at each seepage monitoring point of the target earth-rock dam at a preset sampling frequency. In this embodiment, data is periodically collected from each sensor at preset time intervals. The sampling frequency is determined based on the rate of change of the monitored parameters and the importance of the project; upstream water level and seepage pressure data are sampled at higher frequencies, while rainfall data are sampled at relatively lower frequencies. The data acquisition terminal triggers acquisition commands at regular intervals according to its internal clock, synchronously reads sensor signals from all channels, converts analog signals into digital quantities, and packages them into timestamped monitoring data frames according to a unified data format.

[0059] The collected monitoring data undergoes quality control and outlier removal to form multi-source monitoring data.

[0060] In this embodiment, after receiving the raw monitoring data, the system performs quality control and data cleaning processes. Quality control includes physical range checks, data continuity checks, correlation checks, and rationality checks. The physical range check determines whether the data is within the sensor's measurement range; the continuity check identifies abrupt changes in the data and removes abnormal jumps that clearly do not conform to the physical change process; the correlation check compares the logical consistency between related parameters such as upstream and downstream water levels, rainfall, and seepage; and the rationality check determines whether the data conforms to the basic laws of the seepage field, such as the seepage pressure level not being higher than the upstream water level. Data points that fail the checks are marked as suspicious data, and a decision is made based on preset rules whether to remove the point or replace it with interpolated data. The valid data after quality control screening is integrated into a multi-source monitoring dataset.

[0061] Preferably, the present invention also provides a real-time safety assessment system for earth-rock dams based on multi-source data, used to execute the real-time safety assessment method for earth-rock dams based on multi-source data as described above. The real-time safety assessment system for earth-rock dams based on multi-source data includes: The simulation calculation module is used to perform seepage simulation calculations on the target earth-rock dam based on a preset multi-source working condition parameter combination, and to construct the earth-rock dam seepage basic dataset, which includes seepage infiltration line dataset and unit width leakage dataset. The point planning module is used to determine the seepage monitoring points of the target earth-rock dam based on the seepage infiltration line dataset; The parametric modeling module is used to establish a set of parametric relationships based on the seepage line dataset and the unit width leakage dataset; The deviation calculation module is used to calculate the fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operation based on the set of parameterized relational formulas, and compare them with the normal operation data in the seepage infiltration line dataset and the unit width leakage dataset to obtain the fitting deviation rate set of each seepage monitoring point. The safety assessment module is used to collect multi-source monitoring data of the target earth-rock dam in real time; compare the multi-source monitoring data with the calculation results of the parameterized relation set to determine the instantaneous safety state of the target earth-rock dam.

[0062] 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.

[0063] 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 present 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 real-time safety assessment of earth-rock dams based on multi-source data, characterized in that, Includes the following steps: Step S1: Perform seepage simulation calculations on the target earth-rock dam based on the preset multi-source working condition parameter combination, and construct the earth-rock dam seepage basic dataset. The earth-rock dam seepage basic dataset includes seepage phreatic line dataset and unit width leakage dataset. Among them, the preset multi-source working condition parameter combination in step S1 includes different dam height parameters, different dam slope parameters, different seepage prevention body layout parameters, different drainage body layout parameters, and different upstream and downstream water level parameters. Among them, different anti-seepage body layout parameters include parameters for local leakage due to anti-seepage body failure, different drainage body layout parameters include parameters for drainage body failure, and different upstream and downstream water level parameters include stable water level conditions and unstable water level conditions. Step S2: Determine the seepage monitoring points of the target earth-rock dam based on the seepage infiltration line dataset; wherein, step S2 includes: Based on the seepage infiltration line dataset, statistical analysis was conducted to determine the locations of seepage monitoring points for the target earth-rock dam. Among them, the seepage monitoring points include at least a first monitoring point, a second monitoring point, and a third monitoring point; The first monitoring point is located on the upstream side of the dam axis and in the upstream area of ​​the core wall seepage prevention body; the second monitoring point is located in the upstream area of ​​the drainage prism or the slope drainage; and the third monitoring point is located at the downstream slope 1 / 2 of the dam height. When the target earth-rock dam is equipped with a core wall seepage barrier, a fourth monitoring point is added, which is located on the downstream side of the core wall seepage barrier. Step S3: Establish a set of parameterized relational expressions based on the seepage line dataset and the unit width leakage dataset; wherein, based on the seepage line dataset and the unit width leakage dataset, establish parameterized relational expressions under normal operating conditions respectively; Based on the seepage line dataset and the unit width leakage dataset, parameterized relationships are established for the failure of the impermeable body. Based on the seepage line dataset and the unit width leakage dataset, parameterized relationships are established for drainage failure conditions. Among them, each parameterized relational expression represents the mathematical relationship between the water level data and seepage flow data of each seepage monitoring point and the upstream and downstream water level data, dam height data, and dam slope data. The upstream and downstream water level data includes water level change rate data. Step S4: Based on the set of parameterized relations, calculate the fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operating conditions, and compare them with the normal operating data in the seepage infiltration line dataset and the unit width leakage dataset to obtain the fitting deviation rate set of each seepage monitoring point. Step S4 includes: The fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operating conditions are calculated using a set of parameterized relational expressions. The fitted water level data and fitted seepage flow data are compared with the normal operation data in the seepage infiltration line dataset and the unit width leakage dataset, respectively. The water level fitting deviation rate and seepage flow fitting deviation rate of each seepage monitoring point are calculated to form a fitting deviation rate set. The fitting deviation rate set includes the allowable deviation range of each seepage monitoring point under normal operating conditions; Step S5: Real-time acquisition of multi-source monitoring data of the target earth-rock dam; comparison of the multi-source monitoring data with the calculation results of the parameterized relation set to determine the instantaneous safety state of the target earth-rock dam. The comparison of the multi-source monitoring data with the calculation results of the parameterized relation set in step S5 includes: Calculate the deviation between the real-time data of each seepage monitoring point in the multi-source monitoring data and the corresponding theoretical value in the calculation results of the parameterized relation set; When the deviation values ​​of all seepage monitoring points are less than the fitting deviation rate corresponding to the set of fitting deviation rates, the safety state of the target earth-rock dam is determined to be normal. When the deviation value of any seepage monitoring point is greater than the fitting deviation rate corresponding to the fitting deviation rate set, the safety state of the target earth-rock dam is determined to be abnormal. When the safety state of the target earth-rock dam is abnormal, the deviation direction characteristics of the real-time data of the seepage monitoring points are determined based on multi-source monitoring data. The type of safety anomaly of the target earth-rock dam is determined based on the characteristics of the deviation direction.

2. The method for real-time safety assessment of earth-rock dams based on multi-source data according to claim 1, characterized in that, Based on the seepage infiltration line dataset, statistical analysis revealed that the seepage monitoring points for the target earth-rock dam include: Step S211: Extract water level data at potential locations within the dam body under all working conditions from the seepage line dataset to form a spatial distribution set of water level data; Step S212: Calculate the coefficient of variation of water level at each spatial location point in the spatial distribution set of water level data, and generate a spatial coefficient of variation distribution map; Step S213: Based on the spatial coefficient of variation distribution map, select a set of highly sensitive points with a coefficient of variation higher than the first preset threshold; Step S214: For the set of highly sensitive points, spatial clustering is performed in combination with the structural characteristics of the target earth-rock dam to identify representative points located near the key structural area and form an initial set of monitoring points; wherein, the key structural area includes at least the upstream side of the seepage barrier, the upstream side of the drainage body, and the middle of the downstream slope. Step S215: Based on the seepage infiltration line dataset, calculate the average fitting error between the water level line connecting any three points in the initial monitoring point set and the actual complete infiltration line; Step S216: Find the combination of points with the smallest average fitting error from the initial set of monitoring points, and determine the combination as the seepage monitoring points.

3. The real-time safety assessment method for earth-rock dams based on multi-source data according to claim 1, characterized in that, Based on the characteristics of the deviation direction, the types of safety anomalies of the target earth-rock dam include: When the deviation direction characteristics are consistent with the parameterized relationship characteristics under the preset seepage barrier failure condition, the safety state of the target earth-rock dam is determined to be seepage barrier failure anomaly. When the deviation direction characteristics are consistent with the parameterized relationship characteristics under the preset drainage body failure condition, the safety state of the target earth-rock dam is determined to be drainage body failure anomaly.

4. The real-time safety assessment method for earth-rock dams based on multi-source data according to claim 1, characterized in that, When the safety state of the target earth-rock dam is abnormal, the deviation direction characteristics of the real-time data at seepage monitoring points, based on multi-source monitoring data, include: For each seepage monitoring point, the difference between its real-time water level data and the theoretical water level data under normal operating conditions calculated by the parameterized relation set is calculated to obtain the water level deviation value of that point; the water level deviation values ​​of all monitoring points are combined in a preset order to form a relative deviation vector. The water level deviation values ​​of each point in the relative deviation vector are compared with the allowable deviation range corresponding to the fitted deviation rate set. Deviation points that exceed the allowable range are selected, and the deviation direction of each deviation point is recorded, including positive and negative deviations. The spatial location information of all deviation points is combined with their corresponding deviation directions to generate deviation direction features.

5. The method for real-time safety assessment of earth-rock dams based on multi-source data according to claim 1, characterized in that, The collection of multi-source monitoring data in step S5 includes: An automated monitoring system is deployed around the target earth-rock dam. The automated monitoring system includes rain gauges, water level gauges, seepage flow meters, and seepage pressure gauges. The automated monitoring system collects rainfall data, upstream water level data, downstream water level data, seepage data, and water level data at each seepage monitoring point of the target earth-rock dam at a preset sampling frequency. The collected monitoring data undergoes quality control and outlier removal to form multi-source monitoring data.

6. A real-time safety status assessment system for earth-rock dams based on multi-source data, characterized in that, For executing the real-time safety state assessment method for earth-rock dams based on multi-source data as described in claim 1, the real-time safety state assessment system for earth-rock dams based on multi-source data comprises: The simulation calculation module is used to perform seepage simulation calculations on the target earth-rock dam based on a preset multi-source working condition parameter combination, and to construct the earth-rock dam seepage basic dataset, which includes seepage infiltration line dataset and unit width leakage dataset. The point planning module is used to determine the seepage monitoring points of the target earth-rock dam based on the seepage infiltration line dataset; The parametric modeling module is used to establish a set of parametric relationships based on the seepage line dataset and the unit width leakage dataset; The deviation calculation module is used to calculate the fitted water level data and fitted seepage flow data of each seepage monitoring point under normal operation based on the set of parameterized relational formulas, and compare them with the normal operation data in the seepage infiltration line dataset and the unit width leakage dataset to obtain the fitting deviation rate set of each seepage monitoring point. The safety assessment module is used to collect multi-source monitoring data of the target earth-rock dam in real time; compare the multi-source monitoring data with the calculation results of the parameterized relation set to determine the instantaneous safety state of the target earth-rock dam.

Images