Dike hidden danger identification method based on multi-source data fusion

Abstract

The invention relates to the technical field of embankment management, in particular to an embankment hidden danger recognition method based on multi-source data fusion, which comprises the following steps: dividing an embankment structure into a plurality of monitoring control sections according to hydrogeological units, and arranging a multi-order sensing array in each control section; acquiring time-space sequence data of each sensing array in real time, and constructing a field parameter functional including a displacement field, a strain field, an osmotic pressure field and a temperature field; the strain energy spectrum entropy value of the field parameter functional is extracted through space-time singular value decomposition, and the energy distribution disorder degree of the dike body is reflected; based on a stress seepage coupling constitutive equation, a coupling degree index is obtained through calculation, and the soil-water interaction strength is represented; and inputting the strain energy spectrum entropy value, the coupling degree index and a predetermined hydrodynamic load correction factor into an embankment body structure analysis model to obtain the safety reliability of the embankment body. The comprehensiveness, timeliness and accuracy of dike hidden danger identification can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of embankment management, and particularly relates to a method for identifying embankment hidden dangers based on multi-source data fusion. BACKGROUND

[0002] As an important water conservancy project for resisting floods and protecting the safety of life and property in the coastal area, the structural stability of embankment is directly related to the reliability of the flood control system. However, in the long-term operation process, the embankment is easily affected by multiple factors such as changes in hydrological and meteorological conditions, geological structure movement and human activities, and is prone to cracks, piping, landslides and other hidden dangers. If not identified and disposed in time, it may cause major disasters such as embankment collapse.

[0003] The existing method for identifying embankment hidden dangers mainly relies on a single type of sensor for monitoring, so it can only obtain local data of a certain physical field of the embankment. The monitoring data of a single physical field is difficult to reflect the complex coupling effect of the embankment under multiple physical fields, resulting in one-sided and lagging hidden danger identification results. Especially in extreme working conditions such as sudden water level changes and continuous heavy rainfall, the existing method lacks sensitivity to initial weak hidden dangers, which easily leads to false positives and missed judgments, making it difficult to identify embankment hidden dangers in time and accurately. SUMMARY

[0004] The present application provides a method for identifying embankment hidden dangers based on multi-source data fusion, which can improve the comprehensiveness, timeliness and accuracy of embankment hidden danger identification, and effectively solve the problems in the background art.

[0005] In order to achieve the above purpose, in a first aspect, the present application provides a method for identifying embankment hidden dangers based on multi-source data fusion, comprising:

[0006] Dividing the embankment structure into multiple monitoring control sections according to hydrogeological units, and arranging multiple levels of sensing arrays in each control section;

[0007] Real-time acquisition of the spatio-temporal sequence data of each sensing array, and construction of a field parameter functional containing displacement field, strain field, seepage pressure field and temperature field;

[0008] Extracting the strain energy spectrum entropy value of the field parameter functional through spatio-temporal singular value decomposition, reflecting the energy distribution disorder degree of the embankment; and calculating the coupling degree index based on the stress seepage coupling constitutive equation, representing the strength of soil-water interaction;

[0009] Inputting the strain energy spectrum entropy value, the coupling degree index and the pre-determined water dynamic load correction factor into the embankment structure analysis model to obtain the embankment safety reliability;

[0010] According to the comparison result of the embankment safety reliability and the preset health threshold, triggering a graded early warning response.

[0011] With reference to the first aspect, in a possible design, the method that divides the embankment structure into a plurality of monitoring control sections according to hydrogeological units comprises:

[0012] The zones are divided based on differences in engineering characteristic parameters along the embankment, the engineering characteristic parameters including embankment height gradient change rate, embankment compaction degree distribution coefficient, embankment foundation permeability stability index, and bank slope erosion coefficient;

[0013] Each parameter is weighted and scored by establishing an evaluation model; when a difference between comprehensive scores of adjacent sections exceeds a preset threshold, a monitoring control section boundary is divided, and the length of a single monitoring control section is determined according to embankment structure uniformity, and each control section is ensured to contain a complete microtopography unit.

[0014] With reference to the first aspect, in a possible design, the multi-stage sensing array includes a ground fissure gauge and a soil moisture content monitor arranged on the embankment top, a distributed optical fiber strain gage and a matrix type osmotic pressure gauge embedded in the embankment body, and a pore water pressure sensor array arranged at a set depth below the groundwater level.

[0015] With reference to the first aspect, in a possible design, when the field parameter functional is constructed, the method further includes preprocessing time-space sequence data of each sensing array.

[0016] With reference to the first aspect, in a possible design, the method that extracts the strain energy spectrum entropy value of the field parameter functional by time-space singular value decomposition comprises:

[0017] The time-space matrix M of the field parameter functional is decomposed into M ∈ R T×S , and the characteristic decomposition M = UΣV is performed T , where T is the number of time sampling points, S is the number of spatial monitoring points, U is a left singular matrix, Σ is a diagonal singular value matrix, and V is a right singular matrix;

[0018] The strain energy spectrum entropy value is calculated, and a calculation formula of the strain energy spectrum entropy value is:

[0019]

[0020] , where is a main diagonal element of the singular value matrix, and k is a number of singular values whose cumulative contribution rate exceeds a preset value.

[0021] With reference to the first aspect, in a possible design, the stress-seepage coupling constitutive equation is:

[0022]

[0023] , where CSC represents a coupling degree index; δ str represents a strain field change rate; δ pre represents a seepage pressure gradient field; and ks Ks represents the saturated permeability coefficient of soil; t represents a time variable; H represents the Hamiltonian operator.

[0024] With reference to the first aspect, in a possible design, the water dynamic load correction factor is determined according to a water level fluctuation range and a rainstorm intensity index, and a calculation formula of the water dynamic load correction factor is:

[0025] Φ(W) = 1 + λ1·|ΔH river + λ2·I rain ;

[0026] wherein Φ(W) represents the water dynamic load correction factor; ΔH river represents the water level fluctuation range; I rain represents the rainstorm intensity index; λ1 represents a water level sensitivity coefficient; and λ2 represents a rainfall sensitivity coefficient.

[0027] With reference to the first aspect, in a possible design, a calculation formula of the embankment structure analysis model is:

[0028]

[0029] wherein H SD represents the embankment safety reliability; SES represents a strain energy spectrum entropy value; CSC represents a coupling degree index; and Φ(W) represents a water dynamic load correction operator.

[0030] With reference to the first aspect, in a possible design, the water level sensitivity coefficient λ1 and the rainfall sensitivity coefficient λ2 are determined according to the following method:

[0031] Based on a historical embankment collapse case data set, a coefficient optimization model is established

[0032]

[0033] The constraint condition is λ1 + λ2 = 1, and 0 < λ1, λ2 < 1, wherein N is the number of historical cases, H SDm is a measured health degree, is a predicted health degree.

[0034] Secondly, the present application further provides an embankment hidden danger identification system based on multi-source data fusion, comprising:

[0035] A monitoring control section division module is configured to divide the embankment structure into a plurality of monitoring control sections according to hydrogeological units, and to arrange a multi-stage sensing array in each control section.

[0036] A time-space sequence data acquisition and field parameter construction module is configured to acquire time-space sequence data of each sensing array in real time, and to construct a field parameter functional set including a displacement field, a strain field, a seepage pressure field and a temperature field.

[0037] a characteristic index calculation module configured to extract a strain energy spectrum entropy value of the field parameter functional by spatiotemporal singular value decomposition, and calculate a coupling degree index based on a stress seepage coupling constitutive equation;

[0038] a safety reliability evaluation module configured to input the strain energy spectrum entropy value, the coupling degree index and a predetermined hydrodynamic load correction factor into a embankment body structure analysis model to obtain embankment body safety reliability;

[0039] a hierarchical early warning response triggering module configured to trigger a hierarchical early warning response according to a comparison result of the embankment body safety reliability and a preset health threshold.

[0040] The technical scheme of the present application can achieve the following technical effects:

[0041] By forming a closed loop link of embankment partitioning, multi-stage sensing data acquisition, field parameter functional construction, characteristic index extraction and health degree evaluation, single physical field data such as displacement, strain and seepage pressure can be converted into comprehensive characteristics reflecting the overall structure state of the embankment body through spatiotemporal coupling analysis, which can reduce the problem of hidden danger misjudgment caused by one-sided data. By using strain energy spectrum entropy value to represent energy disorder degree and coupling degree index to quantify soil-water interaction strength, and combining with dynamic correction of hydrodynamic load, initial weak hidden dangers under extreme conditions can be captured, tracked and evaluated accurately in a timely manner, thereby improving the forward prevention and control capability of the flood control system for complex disasters. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 FIG. 1 is a logic flow chart of an embankment hidden danger identification method based on multi-source data fusion in the present application;

[0043] Figure 2 FIG. 2 is a structural block diagram of an embankment hidden danger identification system based on multi-source data fusion in the present application; DETAILED DESCRIPTION

[0044] The present application will be described below in conjunction with the drawings in the present application.

[0045] As shown in FIG. 1, an embankment hidden danger identification method based on multi-source data fusion in the present application specifically includes the following steps: Figure 1 Step S1, divide the embankment structure into multiple monitoring control sections according to hydrogeological units, and arrange multi-stage sensing arrays in each control section;

[0046] Step S2, real-time collect spatiotemporal sequence data of each sensing array, and construct a field parameter functional containing displacement field, strain field, seepage pressure field and temperature field;

[0047]

[0048] ​Step S3, extracting a strain energy spectrum entropy value of the field parameter functional by spatiotemporal singular value decomposition, reflecting an energy distribution disorder degree of the embankment body; and calculating a coupling degree index based on a stress seepage coupling constitutive equation, representing a soil-water interaction strength;

[0049] Step S4, inputting the strain energy spectrum entropy value, the coupling degree index and a predetermined water dynamic load correction factor into an embankment body structure analysis model to obtain an embankment body safety reliability;

[0050] Step S5, triggering a hierarchical early warning response according to a comparison result of the embankment body safety reliability and a preset health threshold.

[0051] In the embodiment, by dividing the monitoring control section according to the hydrogeological unit and arranging the multi-stage sensing array, the synchronous acquisition of the multi-physical field data such as the embankment displacement field, the strain field, the seepage pressure field and the temperature field can be realized. Compared with single physical field monitoring, the multi-source field parameter functional can completely depict the multi-field coupling effect of the embankment body in a complex environment, and avoid the hidden danger misjudgment or omission caused by the one-sidedness of the data. The strain energy spectrum entropy value extracted by the spatiotemporal singular value decomposition can effectively capture the subtle disorder change of the energy distribution of the embankment body, and reflect the decrease of the structural stability caused by the early weak hidden danger. The coupling degree index calculated based on the stress seepage coupling constitutive equation can quantify the abnormal trend of the soil-water interaction strength. The combination of the two and the introduction of the water dynamic load correction factor can make the embankment body safety reliability evaluation more in line with the actual working condition, and improve the sensitivity to the early hidden danger under extreme conditions. At the same time, the embankment body structure analysis model integrates the multi-source characteristic parameters, and the output health quantitative result can be directly compared with the preset threshold to realize the automatic analysis. The hierarchical early warning response mechanism can trigger the corresponding disposal strategy according to the health level, avoid the problems of early warning lag or excessive response, and improve the efficiency and pertinence of the embankment safety management.

[0052] In some embodiments of the present application, the engineering characteristics of different positions along the embankment are different, and unified monitoring cannot accurately identify the hidden danger. The monitoring and analysis can be more targeted by dividing the monitoring control section according to the engineering characteristic parameters, specifically including:

[0053] Step S11, dividing according to the differences of the engineering characteristic parameters along the embankment, the engineering characteristic parameters including embankment height gradient change rate, embankment body compaction degree distribution coefficient, embankment foundation permeability stability index and bank slope erosion coefficient;

[0054] In this step, the embankment height gradient change rate reflects the change of embankment height in the direction of the line. A larger embankment height gradient change indicates that the region may be subjected to different water pressure and soil self-weight stress, which increases the complexity and instability of the embankment structure. The parameter acquisition needs to be combined with topographic survey and data calculation; the embankment compaction degree distribution coefficient reflects the uniformity of embankment soil compaction. Insufficient or uneven compaction will cause differences in the density and strength of embankment soil, and when subjected to external forces such as flood, the weak parts are prone to deformation or even damage. The parameter acquisition needs to be combined with field detection and statistical analysis; the embankment foundation permeability stability index is used to evaluate the ability of the embankment foundation to resist seepage failure. When the embankment foundation permeability stability index is low, under the action of high water level, the embankment foundation is prone to seepage failure phenomena such as piping and soil flow, which threatens the overall stability of the embankment. The parameter acquisition needs to be combined with geological survey and seepage calculation; the bank slope erosion coefficient reflects the degree of bank slope erosion by water flow. In areas with serious bank slope erosion, embankment soil is gradually eroded, the embankment slope stability decreases, and landslides and other hazards may occur. The parameter acquisition needs to be combined with hydrological monitoring and topographic comparison.

[0055] In step S12, each parameter is weighted and scored by establishing an evaluation model. When the difference between the comprehensive scores of adjacent sections exceeds the preset threshold, the monitoring control section boundary is divided. The length of a single monitoring control section is determined according to the uniformity of the embankment structure, and each control section must contain a complete micro-geomorphic unit.

[0056] In this step, the determination of the preset threshold needs to consider historical data, engineering experience and numerical simulation results; the length of a single monitoring control section is determined according to the uniformity of the embankment structure. For sections with relatively uniform embankment structure and gentle parameter changes, the monitoring control section length can be appropriately increased, while for sections with complex embankment structure and large parameter changes, the monitoring control section length should be shortened accordingly, and it must be ensured that each control section contains a complete micro-geomorphic unit, such as a complete river bend or a complete beach, to ensure that the embankment conditions in each monitoring control section are relatively consistent, thereby improving the pertinence and accuracy of monitoring.

[0057] In some embodiments of the present application, the hidden dangers in different parts of the embankment have different manifestations, and multiple types of sensors are needed to monitor from different angles to obtain comprehensive information. The multi-stage sensor array includes:

[0058] a) Surface crack meter: deployed on the top of the embankment to monitor the changes of cracks on the surface of the embankment top; cracks are an intuitive manifestation of embankment structure damage, and the development of early micro cracks may indicate the existence of internal problems; the surface crack meter measures the width, length and depth changes of the crack in real time through high-precision displacement sensors; when installed, the two ends of the crack meter need to be fixed on the embankment on both sides of the crack to ensure that the sensor can accurately perceive the opening degree changes of the crack; at the same time, in order to comprehensively monitor the crack distribution on the embankment top, the surface crack meter should be uniformly laid according to a certain interval, and the interval size is determined according to the width of the embankment top and the historical crack occurrence;

[0059] b) Soil moisture monitor: deployed at different elevations on the embankment top and slope to monitor the soil moisture content and water change rate of the embankment slope in real time; when there is a seepage problem inside the embankment, seepage water will change the water content distribution of the surrounding soil, causing abnormal fluctuations in local area soil moisture data, such as sudden increase or persistent high water content; the soil moisture monitor can capture abnormal changes in soil water content by collecting parameters such as soil volume water content and soil water potential at high frequency; when laid, the monitoring points are laid along the embankment slope strike according to the preset interval, and each monitoring point is buried with a certain number of soil moisture sensors at different depths along the vertical direction of the embankment slope, forming a three-dimensional monitoring network; the sensor uses time domain reflection technology or frequency domain reflection technology to ensure stable operation in different soil textures; the monitoring frequency is adjusted according to the season;

[0060] c) Distributed optical fiber strain grid: embedded in the embankment body to monitor the strain changes inside the embankment body; the embankment will produce strain during stress process, and abnormal changes in strain may reflect stress concentration or damage of the embankment structure; the distributed optical fiber strain grid uses optical time domain reflection technology to calculate the strain value by measuring the scattering characteristic changes of light in the optical fiber; when buried, the optical fiber strain grid should be laid in layers along the depth direction of the embankment, and each layer of optical fiber strain grid should be laid horizontally and uniformly distributed on the embankment cross section to comprehensively obtain the strain information at different positions inside the embankment. The burial depth of the optical fiber strain grid is determined according to the height of the embankment;

[0061] d) Matrix type osmometer: embedded in the embankment body to monitor the seepage pressure distribution inside the embankment body; seepage pressure is an important factor leading to seepage problems such as piping and flow soil of embankment; the matrix type osmometer is composed of a matrix arrangement of multiple osmotic sensors; when installed, the matrix type osmometer should be buried in key parts of the embankment where seepage may occur, such as the joint between the embankment and the embankment foundation, the water-facing surface of the embankment, etc., and the spacing between the sensors is determined according to the size of the embankment and the seepage characteristics;

[0062] e) Pore water pressure sensor array: laid at a set depth below the groundwater level, used to monitor the change of pore water pressure in the lower part of the embankment foundation and embankment body below the groundwater level; the change of pore water pressure directly affects the effective stress of the soil body, and further affects the stability of the embankment body; each sensor in the pore water pressure sensor array adopts a vibrating string or piezoresistive principle, can measure the pore water pressure value in real time, and the determination of the set depth needs to comprehensively consider the groundwater level change range, embankment foundation depth and engineering experience; the sensor array should be uniformly laid on the plane, and the interval is determined according to the embankment foundation width and geological conditions.

[0063] In some embodiments of the present application, the original spatio-temporal sequence data collected by each sensing array may have problems such as noise, error or incompleteness, which will affect the subsequent analysis and calculation results. If the original data without processing is directly used for analysis, it will lead to distortion of the field parameter functional, and further affect the accuracy of feature index extraction, and finally cause deviation of the hidden danger identification result. Therefore, preprocessing is needed when constructing the field parameter functional, which specifically includes:

[0064] In step S21, the noise and abnormal values in the original data are removed through data cleaning. For continuous data collected by the ground crack meter and the distributed optical fiber strain grid, a wavelet threshold denoising algorithm can be used to filter the noise. The algorithm decomposes the data into wavelet coefficients of different frequencies by wavelet decomposition, sets a threshold for the coefficients belonging to high-frequency noise to suppress them, and then restores the true signal by wavelet reconstruction. For time series soil moisture data collected by the soil moisture monitor, such as soil volume water content and soil water potential, a moving average filtering algorithm can be used to eliminate random noise and drift error in the data, while retaining the mutation points and trend characteristics of abnormal water changes. The length of the filtering window is reasonably set according to the monitoring frequency and the soil moisture change rate. In the abnormal value processing, a detection method based on statistical criteria is used. When the deviation of a data point from the mean value exceeds the normal range, it is determined as an abnormal value, and the interpolation data of the adjacent time is used to replace it to ensure the continuity of the data sequence.

[0065] In step S22, the time and space dimensions of different sensor data are synchronized through spatio-temporal alignment. In the time dimension, the collection time stamps of each sensing array are uniformly calibrated based on a high-precision clock, the time resolution of all data is adjusted to a unified standard, the sensor data with a collection frequency higher than the resolution is down-sampled, and the data lower than the resolution is supplemented by linear interpolation to ensure the consistency of the time series. In the spatial dimension, based on the geographic coordinate system of the monitoring control section, a spatial index matrix is established to map the discrete measurement points of the distributed optical fiber strain grid and the array data of the matrix osmometer to a unified three-dimensional spatial grid. The data gaps in the grid are filled by a spatial interpolation algorithm to realize the spatial coordinate alignment of different physical field data.

[0066] Step S23, eliminate the dimensional difference of different physical field data by feature normalization; that is, the normalized processing is carried out on the processed space-time sequence data by using the standardization method, and the data is compressed into a specific interval;

[0067] Step S24, repair the missing data caused by sensor failure or transmission interruption; when the data missing condition of a single time period is light, the sliding window mean method can be used for filling, and the window size is set according to the data change frequency; when the missing condition is moderate, the spatial correlation of adjacent monitoring points can be combined to repair the missing data by using the spatial collaborative interpolation method; when the missing condition is more serious, the sensor failure warning is triggered, the trend characteristics of the historical data of the same period are called to supplement, and the data validity of the region is checked in the subsequent data acquisition process;

[0068] Through the above steps, the preprocessed space-time sequence data can more truly reflect the physical field distribution characteristics of the embankment body; at the same time, the dimensional interference can be eliminated by normalization processing, and the feature weight imbalance problem caused by the data magnitude difference can be avoided; in the extreme working condition, the preprocessed data can more clearly present the characteristic signal of the initial weak hidden danger.

[0069] In some embodiments of the application, the strain energy spectrum entropy value can reflect the disorder degree of the energy distribution of the embankment body, and the value is extracted by time-space singular value decomposition of the field parameter functional, which is helpful for analyzing the state change of the embankment body. The strain energy spectrum entropy value of the field parameter functional is extracted by time-space singular value decomposition, and specifically includes:

[0070] The time-space matrix M of the field parameter functional is decomposed into UΣV. T×S T Wherein, T is the number of time sampling points, S is the number of spatial monitoring points, U is the left singular matrix, Σ is the diagonal singular value matrix, and V is the right singular matrix.

[0071] The strain energy spectrum entropy value is calculated, and the calculation formula of the strain energy spectrum entropy value is:

[0072]

[0073] Wherein, The main diagonal elements of the singular value matrix are k, and k is the number of singular values whose cumulative contribution rate exceeds the preset value.

[0074] ​In the embodiment, the time sampling point number T is determined according to the acquisition frequency of the sensor and the monitoring time length, the higher the acquisition frequency and the longer the monitoring time length, the more detailed the data change in the time dimension can be reflected; the space monitoring point number S depends on the number of the sensing array arranged in the monitoring control section and the monitoring point distribution of each sensing array, the more intensive the monitoring points, the more comprehensive the space coverage; the left singular matrix U reflects the characteristic mode in the time dimension, the singular value matrix Sigma diagonal element indicates the importance of the corresponding singular vector, and the right singular matrix V embodies the characteristic mode in the space dimension; k is usually determined according to engineering experience and data characteristics; through the spatiotemporal singular value decomposition, the dimension can be effectively reduced, the data structure can be simplified under the premise of retaining the main characteristics of the field parameter functional, the strain energy spectrum entropy value is taken as a quantitative index, and the disorder degree of the embankment energy distribution can be directly reflected, when the initial weak hidden danger occurs in the embankment, the energy distribution will abnormally change, and the entropy value will also change, so that early perception of the hidden danger is realized.

[0075] In some embodiments of the application, during the operation of the embankment, there is a complex interaction between the soil skeleton and the pore water in the embankment, and the soil-water interaction is a core factor affecting the stability of the embankment; when the embankment is subjected to external loads such as water level change, rainfall infiltration and the like, the stress field and the seepage field will interact with each other, and the coupling degree index can be calculated through a stress-seepage coupling constitutive equation to represent the strength of the interaction, and the stress-seepage coupling constitutive equation is:

[0076]

[0077] Wherein, CSC represents the coupling degree index; δ str represents the strain field change rate; δ pre represents the seepage pressure gradient field; k s represents the saturated permeability coefficient of the soil; t represents a time variable; represents a Hamiltonian operator.

[0078] In the embodiment, δ str reflects the deformation rate of the embankment structure under the action of stress, and can be calculated through the strain data collected by the distributed optical fiber strain gage arranged in the embankment; δ pre characterizes the spatial variation rate of the seepage pressure in the embankment, and the seepage pressure data collected by the matrix seepage pressure gauge and the pore water pressure sensor array are processed through a spatial gradient algorithm; k s is a key parameter reflecting the water permeability of the soil, which can be determined according to the soil test data in the embankment survey stage, and dynamically corrected combined with the seepage data monitored on site; is used to describe the convergence or divergence state of the seepage pressure field; The time rate of change of the osmotic pressure gradient field is represented, and the dynamic evolution characteristics of the osmotic pressure field over time are embodied; through the stress-seepage coupling constitutive equation, the complex state inside the embankment can be more comprehensively reflected; when the embankment has initial weak hidden dangers, the strength of soil-water interaction will change abnormally, and the coupling degree index can be sensitively captured to improve the sensitivity to initial hidden dangers, and in the scenes of water level sudden change, continuous heavy rain and the like, the hidden danger risks in the embankment caused by the intensification of the coupling effect can be timely reflected.

[0079] In some embodiments of the present application, the stability of the embankment is affected by the hydrodynamic load, and the water level amplitude and the rainstorm intensity are two key factors in the hydrodynamic load; wherein, the large change of the water level will cause the water pressure borne by the embankment to change significantly, thereby affecting the stress distribution and seepage condition inside the embankment; continuous heavy rain will increase the water content of the embankment, reduce the strength of the soil body, and the rainwater infiltration will also change the osmotic pressure field of the embankment; therefore, it is necessary to determine the hydrodynamic load correction factor to correct the evaluation result of the embankment safety reliability, so as to improve the accuracy of hidden danger identification; specifically, the calculation formula of the hydrodynamic load correction factor is:

[0080] Φ(W) = 1 + λ1·|ΔH river + λ2·I rain ;

[0081] Wherein, Φ(W) represents the hydrodynamic load correction factor; ΔH river represents the water level amplitude; I rain represents the rainstorm intensity index; λ1 represents the water level sensitivity coefficient; λ2 represents the rainfall sensitivity coefficient.

[0082] In this embodiment, ΔH river is the difference value of the water level change of the river along the embankment within a certain time, which can be calculated by the data collected in real time by the water level monitoring station deployed near the embankment, and the collection frequency is determined according to the water level change, which can be increased when the water level changes sharply; I rain is a parameter reflecting the intensity of the rainstorm, which can be calculated according to the rainfall data provided by the meteorological department combined with the specific topographic and geomorphic features of the area where the embankment is located, and is usually represented by the rainfall amount per unit time; λ1 is used to measure the influence weight of the water level change on the safety reliability of the embankment, and λ2 is used to measure the influence weight of the rainstorm intensity on the safety reliability of the embankment; the influence of the water level amplitude and the rainstorm intensity is weighted through the water level sensitivity coefficient and the rainfall sensitivity coefficient, which can reasonably allocate the influence weight of the two according to the actual situation and historical data of different embankment projects, so that the corrected safety reliability of the embankment is more in line with the actual embankment state, and in the extreme working condition, for the extreme working conditions such as water level sudden change and continuous heavy rain, the correction factor can sensitively reflect the sharp change of the hydrodynamic load.

[0083] The health condition of the embankment body can be more comprehensively evaluated by establishing an embankment body structure analysis model based on the strain energy spectrum entropy value, the coupling degree index and the water dynamic load correction factor.

[0084]

[0085] wherein H SD represents the embankment body safety reliability; SES represents the strain energy spectrum entropy value; CSC represents the coupling degree index; and Φ(W) represents the water dynamic load correction operator.

[0086] In the embodiment, H SD The value reflects the stability of the embankment body structure, and the smaller the value, the greater the possibility of hidden dangers in the embankment body and the poorer the health condition.

[0087] In some embodiments of the present application, the accuracy of the water dynamic load correction factor directly affects the evaluation result of the embankment body safety reliability, and the water level sensitivity coefficient λ1 and the rainfall sensitivity coefficient λ2 are key parameters for calculating the water dynamic load correction factor. The hydrogeological environment and the embankment body structure characteristics of different embankment projects are different, and the influence of water level and rainstorm on the health of the embankment body is also different. If a fixed sensitivity coefficient value is used, it cannot adapt to the actual situation of different embankments, which will lead to inaccurate calculation of the water dynamic load correction factor, and further affect the evaluation accuracy of the embankment body safety reliability. Therefore, it is necessary to determine the water level sensitivity coefficient λ1 and the rainfall sensitivity coefficient λ2 based on historical embankment failure case data sets through scientific methods to improve the rationality and applicability of the water dynamic load correction factor. The method for determining the water level sensitivity coefficient λ1 and the rainfall sensitivity coefficient λ2 comprises:

[0088] Establishing a coefficient optimization model based on historical embankment failure case data sets

[0089]

[0090] The constraint condition is λ1+λ2=1, 0<λ1, λ2<1, wherein N is the number of historical cases, H SDm is the measured health degree, is the predicted health degree.

[0091] The historical embankment failure case data sets should cover various conditions such as different hydro-meteorological conditions, embankment structure types and geological environments to ensure that the data sets have wide representativeness and rich information amount. By using historical experience data, the determined sensitivity coefficients can better fit the actual engineering situation, improve the calculation accuracy of the water dynamic load correction factor, and further enhance the reliability of the evaluation result of the embankment body safety reliability. In the solving process, various optimization algorithms such as gradient descent method and genetic algorithm can be used to iteratively optimize the sensitivity coefficients until the optimal solution that minimizes the objective function value is found.

[0092] In some embodiments of the present application, according to the comparison result of the embankment safety reliability and the preset health threshold, the triggered hierarchical early warning response is as follows:

[0093] The preset health threshold is divided into three levels, namely, a first threshold, a second threshold and a third threshold, corresponding to green, yellow and red three early warning levels respectively, and the early warning response mechanism of each level is as follows:

[0094] Green early warning: when the embankment safety reliability reaches or exceeds the first threshold, it is determined that the embankment is currently in a stable state without significant hidden risk; at this time, the monitoring platform can push a routine inspection prompt to the embankment management department, including the health degree value of each monitoring control section, the stable trend of the main field parameter functional and the recent maintenance suggestion; the management department carries out work according to the daily inspection plan, and updates the monitoring data summary table once every 72 hours without the need to start the emergency disposal process;

[0095] Yellow early warning: if the health degree is in the second threshold interval, it indicates that there is a potential weak hidden danger in the embankment, the intensity of soil-water interaction appears abnormal fluctuation, and the energy distribution disorder degree rises slightly; at this time, the yellow early warning signal is triggered immediately, in addition to pushing a detailed analysis report containing the strain energy spectrum entropy mutation point and the abnormal period of coupling degree index to the management department, the multi-stage sensing array of the monitoring control section is adjusted to high-frequency acquisition mode, and the local embankment management and protection unit is notified to carry out targeted investigation, focusing on checking the seepage outlet, embankment deformation area and the environment around the sensing device layout point of the control section, and uploading the investigation results to the monitoring platform in real time, if no obvious hidden danger is found in the investigation, the health degree change trend is reviewed once every 48 hours;

[0096] Red early warning: when the health degree is lower than the third threshold, there is a serious structural hidden danger in the embankment, and the initial signs of local piping, crack expansion or slope instability may have appeared; at this time, the red early warning response is started immediately, and the sound and light alarm signal and the emergency report containing the hidden danger position coordinates, real-time field parameter data curve and coupling degree index peak value are sent to the management department, emergency command center and local government, at the same time, the emergency monitoring equipment around the control section is activated to intensively collect displacement field and seepage pressure field data, and the video monitoring equipment is started to continuously shoot the hidden danger area; the emergency command center needs to dispatch professional technical team to the scene within 1 hour to carry out fine detection such as drilling sampling and ground penetrating radar scanning, and according to the detection result, temporary disposal measures such as sandbag pressure seepage and slope cutting and load reduction are taken immediately, and a warning area is delineated to prohibit irrelevant personnel from entering; the health degree evaluation result is updated once every 15 minutes until the hidden danger is effectively controlled or the health degree rises to above the second threshold.

[0097] In the embodiment, the determination of the preset health threshold needs to be made in combination with the dike engineering grade, the historical hidden danger database and the safety redundancy requirement, and is comprehensively formulated and is individually adjusted in combination with the dike engineering characteristics; for important dikes such as a first-class dike for protecting a core area of a city, the first-class threshold can be floated to improve the safety redundancy; and for secondary dikes such as farmland protection dikes, the first-class threshold can be lowered to balance the monitoring cost; the threshold calibration needs to be performed once every certain period of time, by comparing the correlation between the measured value of the health degree in the period of time and the historical hidden danger occurrence probability, the least square method can be used to optimize the threshold boundary, so as to ensure the accuracy of the threshold division; meanwhile, before the arrival of extreme weather such as typhoon season and flood season, a dynamic threshold correction coefficient needs to be temporarily used to temporarily increase the lower limit of the second-class threshold, so as to enhance the sensitivity to initial hidden dangers.

[0098] As shown in Figure 2 The application further provides a dike hidden danger identification system based on multi-source data fusion, which specifically comprises the following modules.

[0099] A monitoring control section division module is used to divide the dike structure into a plurality of monitoring control sections according to hydrogeological units, and a multi-stage sensing array is arranged in each control section.

[0100] A space-time sequence data acquisition and field parameter construction module is used to acquire space-time sequence data of each sensing array in real time, and construct a field parameter functional relation containing a displacement field, a strain field, a seepage pressure field and a temperature field.

[0101] A feature index calculation module is used to extract a strain energy spectrum entropy value of the field parameter functional relation through space-time singular value decomposition; and a coupling degree index is calculated based on a stress seepage coupling constitutive equation.

[0102] A safety reliability evaluation module is used to input the strain energy spectrum entropy value, the coupling degree index and a pre-determined hydrodynamic load correction factor into a dike body structure analysis model, so as to obtain a dike body safety reliability.

[0103] A graded early warning response triggering module is used to trigger a graded early warning response according to the comparison result of the dike body safety reliability and a preset health threshold.

[0104] In the embodiment, by dividing the monitoring control section according to the hydrogeological unit and arranging the multi-stage sensing array, the comprehensive collection of multi-physical field data such as displacement, strain, seepage pressure and temperature can be realized, and the limitation of single sensor monitoring can be overcome; by extracting the strain energy spectrum entropy value through the time-space singular value decomposition and combining the stress seepage coupling constitutive equation to calculate the coupling degree index, the hidden danger characteristics under the multi-field coupling effect can be captured, and the sensitivity to the initial weak hidden danger can be improved; by fusing the multi-feature index and the water dynamic load correction factor to evaluate the safety reliability and trigger the graded early warning, the false alarm and missed judgment problems under the extreme working condition can be reduced, the timely and accurate identification and graded response of the embankment hidden danger can be realized, and the comprehensiveness, timeliness and reliability of the hidden danger identification can be improved.

[0105] The basic principles, main features and advantages of the present application are shown and described above. It should be understood by those skilled in the art that the present application is not limited by the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for identifying potential risks of embankments based on multi-source data fusion, characterized in that, The method comprises the following steps: The embankment structure is divided into multiple monitoring control sections according to hydrogeological units, and multiple-stage sensing arrays are arranged in each control section; Real-time acquisition of the spatio-temporal sequence data of each sensing array, construction of a field parameter functional including displacement field, strain field, seepage pressure field and temperature field; Extracting the strain energy spectrum entropy value of the field parameter functional through spatio-temporal singular value decomposition, reflecting the energy distribution disorder degree of the embankment body; and calculating the coupling degree index based on the stress-seepage coupling constitutive equation, representing the strength of soil-water interaction; Inputting the strain energy spectrum entropy value, the coupling degree index and the pre-determined water dynamic load correction factor into the embankment body structure analysis model to obtain the embankment body safety reliability; According to the comparison result of the embankment body safety reliability and the preset health threshold, triggering a hierarchical early warning response.

2. The method according to claim 1, wherein, The method for dividing the embankment structure into multiple monitoring control sections according to hydrogeological units comprises the following steps: Based on the differences of engineering characteristic parameters along the embankment, the embankment is divided into zones, and the engineering characteristic parameters include embankment height gradient change rate, embankment compaction degree distribution coefficient, embankment foundation permeability stability index and bank slope erosion coefficient; Through the establishment of an evaluation model, each engineering characteristic parameter is weighted and scored, and when the difference between the comprehensive scores of adjacent sections exceeds the preset threshold, the monitoring control section boundary is divided; the length of a single monitoring control section is determined according to the uniformity of the embankment structure, and each control section contains a complete micro-geomorphic unit.

3. The method according to claim 1, wherein, The multiple-stage sensing array comprises a ground fissure meter and a soil moisture content monitor arranged on the embankment top, a distributed optical fiber strain gage and a matrix type seepage pressure gauge embedded in the embankment body, and a pore water pressure sensor array arranged at a set depth below the groundwater level.

4. The method according to claim 1, wherein, When constructing the field parameter functional, the spatio-temporal sequence data of each sensing array is also pre-processed, and the processing method comprises the following steps: Through data cleaning, removing noise and outliers in the original data; Through spatio-temporal alignment, synchronizing different sensor data in time and space dimensions; Through feature normalization, eliminating the dimension difference of different physical field data; Repairing missing data caused by sensor failure or transmission interruption.

5. The method for embankment hidden danger identification based on multi-source data fusion according to claim 1, characterized in that, The method for extracting the strain energy spectrum entropy value of the field parameter functional through spatio-temporal singular value decomposition comprises the following steps: The space-time matrix M of the field parameter functional is obtained T×S Eigen decomposition is performed on M=U∑V T Wherein, T is the number of time sampling points, S is the number of spatial monitoring points, U is the left singular matrix, ∑ is the diagonal singular value matrix, and V is the right singular matrix. Calculating the strain energy spectrum entropy value, and the calculation formula of the strain energy spectrum entropy value is: wherein is the main diagonal element of the singular value matrix, and k is the number of singular values whose cumulative contribution rate exceeds a preset value.

6. The method according to claim 5, wherein, The stress-seepage coupling constitutive equation is: wherein CSC represents a coupling degree index; δ str represents a strain field change rate; δ pre represents an osmotic pressure gradient field; k s represents a saturated permeability coefficient of the soil body; t represents a time variable; represents a Hamiltonian operator.

7. The method according to claim 6, wherein, The water dynamic load correction factor is determined by the water level amplitude and the rainstorm intensity index, and the calculation formula of the water dynamic load correction factor is: Φ(W) = 1 + λ1 • |ΔH river + λ2 • I rain ; wherein Φ(W) represents a water dynamic load correction factor; ΔH river represents a water level amplitude; I rain represents a storm intensity index; λ1represents a water level sensitivity coefficient; and λ2represents a rainfall sensitivity coefficient.

8. The method according to claim 7, wherein, The calculation formula of the embankment body structure analysis model is: where H SD represents the levee body safety reliability; SES represents the strain energy spectrum entropy value; CSC represents the coupling degree index; and Φ(W) represents a water dynamic load correction operator.

9. The method according to claim 8, wherein, The determination method of the water level sensitive coefficient λ1 and the rainfall sensitive coefficient λ2 comprises the following steps: Based on historical embankment breach case data sets, a coefficient optimization model is established The constraint condition is λ1+λ2=1, 0<λ1, λ2<1, where N is the number of historical cases, H SDm is the measured health, is the predicted health.

10. A dike hidden danger identification system based on multi-source data fusion, characterized in that, The method comprises the following steps: A monitoring control section division module is configured to divide the embankment structure into multiple monitoring control sections according to hydrogeological units, and to arrange multiple-stage sensing arrays in each control section; A spatio-temporal sequence data acquisition and field parameter construction module is configured to acquire the spatio-temporal sequence data of each sensing array in real time, and to construct a field parameter functional including displacement field, strain field, seepage pressure field and temperature field; A feature index calculation module is configured to extract the strain energy spectrum entropy value of the field parameter functional through spatio-temporal singular value decomposition; and to calculate the coupling degree index based on the stress-seepage coupling constitutive equation. A safety reliability evaluation module is configured to input the strain energy spectrum entropy value, the coupling degree index, and a predetermined hydrodynamic load correction factor into a embankment structure analysis model to obtain embankment safety reliability; A hierarchical early warning response triggering module is configured to trigger a hierarchical early warning response according to a comparison result of the embankment safety reliability and a preset health threshold.

Images