A method and system for evaluating the deformation stability of a rock-fill dam based on multi-parameter analysis

Abstract

The application discloses a kind of rockfill dam deformation stability evaluation method based on multi-parameter analysis, it is related to water conservancy engineering technical field, and evaluation step is:S1, based on the data of monitoring point to rockfill dam is collected, data includes dam body deformation data and dam body contour data, data set is formed after data pre-processing;S2, dam body deformation data is converted into deformation time series according to time sequence, its fractal dimension is calculated by box counting method, and the mapping relationship between fractal dimension and rockfill dam deformation stability is established by combining historical data statistical analysis;S3, dam body contour data is imported into three-dimensional modeling software, and three-dimensional geometric model of rockfill dam is constructed.The application constructs comprehensive evaluation system, carries out correlation degree analysis and verification to fractal dimension, displacement strain index, obtains evaluation value based on weighted average method, intelligently divides stability grade, and provides intuitive decision basis for water conservancy engineering management.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering technology, specifically to a method and system for evaluating the deformation stability of rockfill dams based on multi-parameter analysis. Background Technology

[0002] As a key facility in water conservancy projects, the accurate assessment of the deformation stability of rockfill dams is crucial for ensuring project safety, extending service life, and safeguarding the lives and property of people downstream. With the continuous expansion of water conservancy project construction and technological advancements, the height and complexity of rockfill dams are increasing, placing higher demands on their stability assessment technology. Over the past few decades, rockfill dam stability assessment technology has made significant progress. In the early stages, the stability of the dam was mainly assessed using empirical formulas and simple mechanical calculations. These methods only considered a few key factors, such as the dam's self-weight and water pressure, and could not comprehensively and accurately describe the complex deformation and stress conditions during the actual operation of the dam. With the development of monitoring technology, various sensors have been widely used in the monitoring of rockfill dams, enabling real-time acquisition of displacement, stress, and strain data of the dam. Based on this monitoring data, assessment methods based on single or a few parameters have gradually developed. For example, stability can be judged by monitoring the displacement changes of specific parts of the dam, or simple mechanical analysis can be performed based on stress and strain data.

[0003] However, most existing assessment methods focus only on a single or a few parameters, failing to fully exploit the rich information contained in multi-source data such as dam deformation and profile data. The deformation of rockfill dams is highly complex and nonlinear, making it difficult for traditional methods to accurately capture its inherent patterns. Although 3D modeling technology has been widely used in the design and analysis of rockfill dams, in stability assessment, model construction is often overly idealized, failing to fully consider the complex environmental factors in which rockfill dams operate in reality. For example, during model loading, the simulation of actual environmental loads such as seismic force and water scour force is not accurate enough, leading to deviations between the displacement and strain indices extracted from the simulation analysis and the actual situation, affecting the reliability of the assessment results. Furthermore, there is a lack of a comprehensive and systematic integrated assessment system for the deformation stability of rockfill dams. Existing assessment methods are independent and cannot effectively integrate different types of parameters, making it difficult to make a holistic and accurate judgment on the stability of rockfill dams. Therefore, we propose a multi-parameter analysis-based method and system for assessing the deformation stability of rockfill dams. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides a method and system for evaluating the deformation stability of rockfill dams based on multi-parameter analysis. This technical solution solves the problems of insufficient data utilization, weak integration between 3D model simulation and actual conditions, and an imperfect comprehensive evaluation system.

[0005] To achieve the above objectives, the technical solution adopted by this invention is: a method for evaluating the deformation stability of rockfill dams based on multi-parameter analysis, the evaluation steps of which are as follows:

[0006] S1. Data on the rockfill dam is collected based on monitoring points. The data includes dam deformation data and dam outline data. The data is preprocessed to form a dataset.

[0007] S2. Convert the dam deformation data into a deformation time series in chronological order, calculate its fractal dimension using the box counting method, and establish a mapping relationship between the fractal dimension and the deformation stability of the rockfill dam by combining historical data statistical analysis.

[0008] S3. Import the dam outline data into 3D modeling software to construct a 3D geometric model of the rockfill dam; determine the material mechanical parameters based on the rockfill dam design data, experimental data and particle flow theory inversion method, and apply multiple working conditions to the model for simulation analysis according to the actual environmental load conditions, and extract displacement strain index from the simulation results.

[0009] S4. Perform correlation analysis on fractal dimension and displacement-strain index to verify the consistency of the two data; if the correlation is greater than the preset threshold, determine its weight based on the mapping relationship between fractal dimension and deformation stability of rockfill dam, and calculate the comprehensive stability assessment value by weighted average method in combination with displacement-strain index; if the correlation is less than or equal to the preset threshold, trigger the data verification or model correction process; classify the stability level of rockfill dam according to the assessment value.

[0010] Preferably, step S1 includes:

[0011] The horizontal displacement data, vertical displacement data, and strain data of the dam body are collected using displacement sensors and strain gauges.

[0012] The dam outline parameters and upstream and downstream dam slope shape parameters were obtained through a GIS system and a total station.

[0013] The collected dam deformation and contour data are preprocessed to form a dataset.

[0014] Preferably, the step of converting to a deformed time series in step S2 is as follows:

[0015] The timestamp of each record is extracted from the preprocessed dam deformation data, and sorted in ascending order based on the timestamps to obtain a preliminary time series;

[0016] The sampling frequency is unified by using equal-interval interpolation to generate equal-interval time series;

[0017] Logical verification, statistical verification, continuity verification, and integrity verification are performed on the sequence data. For abnormal data that fails verification, linear interpolation is used to repair or mark and remove it.

[0018] The verified time series is stored, and an index is created to associate the time series with the dam outline data to obtain the deformation time series.

[0019] Preferably, the box counting step in step S2 is as follows:

[0020] The deformation time series is mapped onto a two-dimensional plane and divided into square grids with side length ε. The initial value of ε is 1 / 10 of the maximum deformation of the dam body, and the grid is scaled down to the minimum resolution in a proportional sequence.

[0021] Count the number of non-empty boxes N(ε) corresponding to each ε, and record ln(1 / ε) and lnN(ε);

[0022] Perform a least-squares linear fit on the data points (ln(1 / ε), lnN(ε)), and require a goodness-of-fit R0. 2 If the value is ≥0.95, otherwise adjust the ε iteration range and recalculate;

[0023] Take the absolute value of the slope of the fitted line as the fractal dimension D, which satisfies the relationship lnN(ε)=-Dlnε+C, where C is the intercept;

[0024] Correspondingly, the process of establishing the mapping relationship between fractal dimension and deformation stability of rockfill dams is as follows:

[0025] Based on the stability level classification criteria in historical data, the fractal dimension is grouped according to the corresponding level;

[0026] Calculate the mean, standard deviation, and confidence interval of the fractal dimension at each level;

[0027] The significance of the differences in fractal dimension among different levels was verified by the Kruskal-Wallis test, and a scatter plot of fractal dimension-stability level was plotted to show the mapping relationship between fractal dimension and deformation stability of rockfill dams.

[0028] Preferably, the steps for constructing the three-dimensional geometric model of the rockfill dam in step S3 are as follows:

[0029] Obtain dam outline data from the dataset and convert the data into DXF, OBJ, and CSV formats for software recognition.

[0030] Import the data into the 3D modeling software, set the coordinate system and units consistent with the actual measurement, use the Delaunay triangulation algorithm to generate a triangular mesh terrain model, and optimize the model accuracy through mesh smoothing and redundant vertex removal.

[0031] Based on the contour data, the dam surface was constructed using the NURBS surface tool, and the terrain model and dam structure were merged through Boolean operations to generate a three-dimensional geometric model of the rockfill dam.

[0032] Preferably, determining the material mechanical parameters in step S3 includes the following steps:

[0033] The rockfill material was discretized into ellipsoidal particles. The aspect ratio r of the particles was determined statistically through field sieve analysis. A linear spring-damping model was used for particle contact, and the formula for calculating the normal force was:

[0034] F n =k n Δδ n

[0035] Where k n For normal contact stiffness, Δδ n This represents the normal relative displacement increment;

[0036] The formula for calculating tangential force is:

[0037] F t =min(k) t Δδ t μF n )

[0038] Where k t For tangential contact stiffness, Δδ t The tangential relative displacement increment is μ, where μ is the friction coefficient.

[0039] k was calibrated using uniaxial compression and direct shear tests. n k t And μ, and introduce a particle shape correction coefficient η = 1 + 0.1(r-1) to correct the contact stiffness;

[0040] The granular flow simulation results were fitted using least squares, and the elastic modulus was determined based on the slope of the stress-strain curve, using the following formula:

[0041]

[0042] in These are the mean values ​​of stress and strain.

[0043] Based on the relationship between transverse and longitudinal strain, Poisson's ratio is calculated by combining the bulk density ρ and the particle shape correction factor η.

[0044] Preferably, the step of extracting displacement strain parameters in step S3 is as follows:

[0045] The horizontal X-axis, horizontal Y-axis, and vertical Z-axis displacement data of the rockfill dam under each node loading condition were extracted from the simulation results. Statistical analysis was performed on the node displacements in the dam crest region, and the mean and standard deviation of the displacements were calculated.

[0046] Strain parameters are defined according to different regions:

[0047] For the dam crest region, extract the principal strain v1 and the third principal strain v3 data, and calculate the tensile-compression gradient;

[0048] For the dam slope region, shear strain is extracted from the elements along the dam slope direction, and the maximum shear strain is calculated.

[0049] For the dam foundation area, vertical strain is extracted, and the mean compressive strain is calculated;

[0050] The overall displacement index L is calculated by weighting the data by region and summing the values ​​of tensile-compression gradient, maximum shear strain, and average compressive strain.

[0051] Preferably, the correlation analysis step in step S4 is as follows:

[0052] The fractal dimension and the extracted displacement-strain index are dimensionless to eliminate dimensional differences.

[0053] Based on grey relational analysis, the correlation coefficients between fractal dimension and displacement-strain index at each time point are calculated. The average correlation coefficients are then used to obtain the correlation degree, which is calculated using the following formula:

[0054]

[0055] Where μ i (o) is the correlation coefficient, r i For the calculated correlation degree, o is the index of the time point, and n' is the length of the time series;

[0056] If the correlation degree r i If ≥θ, then the consistency between the fractal dimension and the displacement-strain index is reliable; if the correlation degree r i <θ triggers the data verification or model correction process, re-executing steps S1 to S3; where θ is the correlation threshold.

[0057] Preferably, the weighted average calculation formula in step S4 is:

[0058] M = w1 * D + w2 * L

[0059] Where M is the evaluation value calculated by the weighted average method, D is the fractal dimension, L is the displacement strain index; w1 and w2 are the weights, which are determined based on the mapping relationship between the fractal dimension and the stability level and the analytic hierarchy process.

[0060] In addition, the present invention also provides a multi-parameter analysis-based deformation stability assessment system for rockfill dams, comprising:

[0061] The data acquisition module is configured to collect dam deformation data and dam outline data, and construct them into a dataset;

[0062] The fractal dimension analysis module is configured to analyze and calculate the fractal dimension of the collected dam deformation data.

[0063] The modeling and analysis module is configured to construct a three-dimensional model of the rockfill dam, perform simulation analysis, and extract displacement and strain indices.

[0064] The fusion analysis module is configured to integrate fractal dimension and displacement and strain indices, and perform weighted calculations to comprehensively calculate the stability assessment value.

[0065] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0066] In terms of data utilization, this invention constructs a dataset by collecting dam deformation and contour data from monitoring points, integrates multiple data types, comprehensively considers complex deformation factors, avoids assessment bias, and applies fractal theory to construct a time series of deformation data and calculate the fractal dimension to establish an accurate correspondence with stability. This allows for the keen capture of changes in deformation patterns and improves assessment accuracy. In the construction of the three-dimensional model, mechanical parameters are determined by combining design and experimental data to accurately simulate real-world environmental factors of seismic forces. Displacement and strain indices that fit reality are extracted and integrated with fractal dimension parameters to enhance simulation reliability. A comprehensive evaluation system is constructed, and correlation analysis and verification are performed on fractal dimension, displacement, and strain indices. Evaluation values ​​are obtained based on the weighted average method, and stability levels are intelligently classified, providing intuitive decision-making basis for water conservancy project management and improving the level of intelligent management. Attached Figure Description

[0067] Figure 1 This is a flowchart of the deformation stability assessment steps for rockfill dams according to the present invention;

[0068] Figure 2 This is a framework diagram of the rockfill dam deformation stability assessment system of the present invention. Detailed Implementation

[0069] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0070] Reference Figure 1 As shown, this invention provides a method for evaluating the deformation stability of rockfill dams based on multi-parameter analysis. The evaluation steps are as follows:

[0071] S1. Data on the rockfill dam is collected based on monitoring points. The data includes dam deformation data and dam outline data. The data is preprocessed to form a dataset.

[0072] S2. Convert the dam deformation data into a deformation time series in chronological order, calculate its fractal dimension using the box counting method, and establish a mapping relationship between the fractal dimension and the deformation stability of the rockfill dam by combining historical data statistical analysis.

[0073] S3. Import the dam outline data into 3D modeling software to construct a 3D geometric model of the rockfill dam; determine the material mechanical parameters based on the rockfill dam design data, experimental data and particle flow theory inversion method, and apply multiple working conditions to the model for simulation analysis according to the actual environmental load conditions, and extract displacement strain index from the simulation results.

[0074] S4. Perform correlation analysis on fractal dimension and displacement-strain index to verify the consistency of the two data; if the correlation is greater than the preset threshold, determine its weight based on the mapping relationship between fractal dimension and deformation stability of rockfill dam, and calculate the comprehensive stability assessment value by weighted average method in combination with displacement-strain index; if the correlation is less than or equal to the preset threshold, trigger the data verification or model correction process; classify the stability level of rockfill dam according to the assessment value.

[0075] Specifically, in one embodiment of the present invention, step S1 includes:

[0076] The horizontal displacement data, vertical displacement data, and strain data of the dam body are collected using displacement sensors and strain gauges.

[0077] The dam outline parameters and upstream and downstream dam slope shape parameters were obtained through a GIS system and a total station.

[0078] The collected dam deformation and contour data are preprocessed to form a dataset.

[0079] In this embodiment, a network of monitoring points is set up at key monitoring sections of the rockfill dam, such as the dam crest and slope at different elevations. An automated total station measurement system is used to construct a monitoring network with no fewer than three stable benchmark points based on the principle of free network adjustment. The horizontal displacement vector is extracted through coordinate time series analysis.

[0080] Vertical displacement monitoring employs a multi-level monitoring system. Level 1 monitoring: A circular leveling network is established using a digital precision level, with monitoring points spaced every 25-30 meters along the dam axis and arranged along the dam slope according to elevation gradients. The measurement frequency is determined based on the project stage. Level 2 monitoring: In stress concentration areas, surface settlement gauges or hydraulic settlement meters are installed to achieve continuous automated monitoring.

[0081] High-precision strain monitoring devices are installed in the stress concentration areas inside the rockfill dam, including: vibrating wire strain gauges are installed in an orthogonal arrangement to capture the triaxial strain state; a distributed fiber optic grating sensor network is laid along the key structural surfaces to form a continuous strain field monitoring system; and the strain data acquisition frequency is dynamically adjusted according to the dam deformation rate.

[0082] Data collection method based on GIS system: Geographic Information System (GIS) processes and analyzes satellite remote sensing imagery and aerial photogrammetry data to obtain the approximate outline information of rockfill dams. Using spatial analysis tools in GIS software, the image data is vectorized to extract the boundary lines of the dam body, thereby obtaining the outline parameters of the dam body, such as the length and width of the dam body.

[0083] Total station measurement of dam outline and slope shape parameters: The total station is used to accurately measure the dam outline. After setting up the station at known control points, the characteristic points on the dam outline are measured, including the dam crest edge, dam slope turning points, and dam toe. By measuring the three-dimensional coordinates of these points, the accurate outline of the dam is constructed. For the upstream and downstream dam slope shape parameters, measurement points can be selected at certain intervals on the dam slope, and the coordinates of these points can be measured to calculate the slope and slope length parameters of the dam slope.

[0084] In this embodiment, the preprocessing process includes filtering, noise reduction, and data standardization. Wavelet thresholding can be used for noise reduction, Kalman filtering can be used to optimize the data, spatiotemporal standardization can be performed on data from different measurement points and different time periods, a reference datum and a unified time coordinate system can be established, and a formatted dataset can be generated.

[0085] Specifically, in one embodiment of the present invention, the step of converting to a deformed time series in step S2 is as follows:

[0086] The timestamp of each record is extracted from the preprocessed dam deformation data, and sorted in ascending order based on the timestamps to obtain a preliminary time series;

[0087] The sampling frequency is unified by using equal-interval interpolation to generate equal-interval time series;

[0088] Logical verification, statistical verification, continuity verification, and integrity verification are performed on the sequence data. For abnormal data that fails verification, linear interpolation is used to repair or mark and remove it.

[0089] The verified time series is stored, and an index is created to associate the time series with the dam outline data to obtain the deformation time series.

[0090] This embodiment achieves temporal organization of data through timestamp extraction and ascending sorting, establishing a continuous representation model of deformation evolution. Equal-interval interpolation ensures the uniformity of sampling frequency, eliminating temporal irregularities caused by inconsistent monitoring equipment operating cycles, and providing a standardized data foundation for subsequent fractal dimension calculation. A multi-dimensional verification mechanism constructs a data quality control system, systematically identifying abnormal data that does not conform to physical laws or statistical characteristics, and using linear interpolation algorithms for data repair to ensure the physical continuity of the time series. An indexed storage strategy establishes a correlation mapping between the deformation time series and the spatial geometric features of the dam body, realizing integrated spatiotemporal management of deformation data.

[0091] In this embodiment, the box counting method step in step S2 is as follows:

[0092] The deformation time series is mapped onto a two-dimensional plane and divided into square grids with side length ε. The initial value of ε is 1 / 10 of the maximum deformation of the dam body, and the grid is scaled down to the minimum resolution in a proportional sequence.

[0093] Count the number of non-empty boxes N(ε) corresponding to each ε, and record ln(1 / ε) and lnN(ε);

[0094] Perform a least-squares linear fit on the data points (ln(1 / ε), lnN(ε)), and require a goodness-of-fit R0. 2 If the value is ≥0.95, otherwise adjust the ε iteration range and recalculate;

[0095] Take the absolute value of the slope of the fitted line as the fractal dimension D, which satisfies the relationship lnN(ε)=-Dlnε+C, where C is the intercept;

[0096] Correspondingly, the process of establishing the mapping relationship between fractal dimension and deformation stability of rockfill dams is as follows:

[0097] Based on the stability level classification criteria in historical data, the fractal dimension is grouped according to the corresponding level;

[0098] Calculate the mean, standard deviation, and confidence interval of the fractal dimension at each level;

[0099] The significance of the differences in fractal dimension among different levels was verified by the Kruskal-Wallis test, and a scatter plot of fractal dimension-stability level was plotted to show the mapping relationship between fractal dimension and deformation stability of rockfill dams.

[0100] The box counting method used in this embodiment is a classic method in fractal theory for measuring the complexity of irregular geometries. In the deformation analysis of the rockfill dam, the deformation time series is mapped to a two-dimensional phase space to form a trajectory diagram, realizing the transformation from one-dimensional time series to high-dimensional spatial features and effectively capturing the inherent structural characteristics of the deformation mode. An initial grid side length ε is selected as 1 / 10 of the maximum deformation to establish an upper limit benchmark for scale analysis. Then, a sequence with decreasing common ratio is used to cover multiple scale levels, ensuring the consistency verification of fractal features at different observation scales. The fractal dimension D, as the absolute value of the slope of the fitted curve, quantitatively characterizes the spatial filling degree and complexity of the rockfill dam deformation sequence. Theoretically, the closer the D value is to 2, the closer the deformation mode is to random noise, and the lower the system stability; the closer the D value is to 1, the stronger the deterministic trend of the deformation mode, and the higher the system stability.

[0101] In this embodiment, the nonparametric statistical method Kruskal-Wallis test is used in the process of establishing the mapping relationship, which effectively avoids the problem that the fractal dimension distribution may not satisfy the normality assumption, and statistically ensures that the fractal dimension distributions of different stability levels have significant differences.

[0102] Specifically, in one embodiment of the present invention, the step of constructing the three-dimensional geometric model of the rockfill dam in step S3 is as follows:

[0103] Obtain dam outline data from the dataset and convert the data into DXF, OBJ, and CSV formats for software recognition.

[0104] Import the data into the 3D modeling software, set the coordinate system and units consistent with the actual measurement, use the Delaunay triangulation algorithm to generate a triangular mesh terrain model, and optimize the model accuracy through mesh smoothing and redundant vertex removal.

[0105] Based on the contour data, the dam surface was constructed using the NURBS surface tool, and the terrain model and dam structure were merged through Boolean operations to generate a three-dimensional geometric model of the rockfill dam.

[0106] The 3D modeling process in this embodiment follows the "point-line-surface-volume" topology construction principle. First, the discrete coordinate points acquired through monitoring are converted into common CAD / CAE formats (DXF, OBJ, and CSV) through data format standardization. In the import phase, a unified coordinate reference system (UCS) is used to establish a global coordinate transformation matrix, eliminating geometric errors caused by differences in measurement benchmarks. Terrain modeling employs a constrained Delaunay triangulation algorithm to generate irregular triangular networks (TINs). This algorithm ensures maximum interior angles and optimal side length distribution of triangles. Laplacian mesh smoothing filtering and quadrilateral mesh downsampling techniques optimize the geometric accuracy and computational efficiency of the model. The dam structure is constructed using non-uniform rational B-spline surfaces, with the control point mesh density adaptively adjusted according to curvature changes. Finally, CSG Boolean operations are used to topologically fuse the terrain foundation and dam structure, generating a topologically consistent and geometrically accurate 3D geometric model of the rockfill dam.

[0107] In this embodiment, determining the material mechanical parameters in step S3 includes the following steps:

[0108] The rockfill material was discretized into ellipsoidal particles. The aspect ratio r of the particles was determined statistically through field sieve analysis. A linear spring-damping model was used for particle contact, and the formula for calculating the normal force was:

[0109] F n =k n Δδ n

[0110] Where k n For normal contact stiffness, Δδ n This represents the normal relative displacement increment;

[0111] The formula for calculating tangential force is:

[0112] F t =min(k) t Δδ t μF n )

[0113] Where k t For tangential contact stiffness, Δδ t The tangential relative displacement increment is μ, where μ is the friction coefficient.

[0114] k was calibrated using uniaxial compression and direct shear tests. n k t And μ, and introduce a particle shape correction coefficient η = 1 + 0.1(r-1) to correct the contact stiffness;

[0115] The granular flow simulation results were fitted using least squares, and the elastic modulus was determined based on the slope of the stress-strain curve, using the following formula:

[0116]

[0117] in These are the mean values ​​of stress and strain.

[0118] Based on the relationship between transverse and longitudinal strain, Poisson's ratio is calculated by combining the bulk density ρ and the particle shape correction factor η.

[0119] The formula for calculating Poisson's ratio is:

[0120]

[0121] in ρ is the calculated value of Poisson's ratio, and ρ0 is the reference bulk density value.

[0122] This embodiment employs the discrete element method (DEM) to simulate the micromechanical behavior of rockfill materials, achieving a reasonable approximation of the actual rockfill particle morphology through ellipsoidal particle characterization. The particle aspect ratio *r* is obtained by fitting field sieving data using the Rosin-Rammler distribution function. Particle contact mechanics is constructed based on a linear spring-damping model, and the calculation formulas for normal and tangential contact forces reflect the coupling mechanism between elastic deformation and frictional slip. Contact parameter calibration adopts a two-layer verification strategy: first, the macroscopic stress-strain response is determined through uniaxial compression tests, and then the internal friction angle is verified through direct shear tests. A shape correction coefficient *η* is introduced to correct the contact stiffness of non-spherical particles, overcoming the limitations of traditional spherical particle models. Elastic modulus inversion uses a multi-scale homogenization method, achieving cross-scale mapping from microscopic contact parameters to macroscopic elastic parameters. The Poisson's ratio calculation formula considers the effects of bulk density and particle shape, achieving an accurate description of the lateral deformation characteristics of the actual material. This method establishes a quantitative relationship between the microstructure and macroscopic mechanical response of rockfill materials, improving the physical rationality of the material constitutive parameters compared to traditional empirical parameter methods.

[0123] In this embodiment, the step of extracting the displacement strain index in step S3 is as follows:

[0124] The horizontal X-axis, horizontal Y-axis, and vertical Z-axis displacement data of the rockfill dam under each node loading condition were extracted from the simulation results. Statistical analysis was performed on the node displacements in the dam crest region, and the mean and standard deviation of the displacements were calculated.

[0125] Strain parameters are defined according to different regions:

[0126] For the dam crest region, extract the principal strain v1 and the third principal strain v3 data, and calculate the tensile-compression gradient;

[0127] For the dam slope region, shear strain is extracted from the elements along the dam slope direction, and the maximum shear strain is calculated.

[0128] For the dam foundation area, vertical strain is extracted, and the mean compressive strain is calculated;

[0129] The formula for calculating the stretch-compression gradient is:

[0130] L top =|v1-v3|

[0131] Where L top Represented as a stretch-compression gradient;

[0132] The formula for maximum shear strain is:

[0133]

[0134] Where γ max This represents the maximum shear strain value.

[0135] The formula for calculating the mean compressive strain is:

[0136] L base =(1 / n)∑(ε') z )

[0137] Where L base ε' is the mean compressive strain. z This is the data point for the z'th compressive strain;

[0138] Weights are allocated by region, and a weighted sum is performed based on the tensile-compression gradient, maximum shear strain, and mean compressive strain to calculate the overall displacement index L. The calculation formula is as follows:

[0139] L = w3L top +w4γ max +w5L base

[0140] Where w3, w4 and w5 are the weight values ​​of the dam crest region, dam slope region and dam foundation region, respectively, and L is the overall displacement index value.

[0141] This embodiment achieves a comprehensive stability assessment of a rockfill dam through a multi-regional hierarchical parameter extraction strategy. Three-dimensional coordinate displacement field analysis provides a macroscopic characterization of the overall dam deformation, particularly the statistical analysis of the displacement at the dam crest nodes, which effectively measures the central tendency and dispersion of deformation. The regionally differentiated strain index design fully considers the stress characteristics of each structural component of the rockfill dam: the tensile-compression gradient in the dam crest region quantifies the extreme differences in the principal strain space, reflecting the non-uniformity of tensile and compressive deformation at the dam crest; the maximum shear strain in the dam slope region captures the concentrated shear deformation area of ​​the potential sliding surface; and the mean compressive strain in the dam foundation region assesses the overall trend of foundation settlement through vertical strain field integration. The overall displacement-strain index L integrates multi-source strain information through weighted fusion, with weighting coefficients determined based on sensitivity analysis, reflecting the relative contribution of each region to overall stability. This multi-scale, multi-regional strain analysis method overcomes the limitations of traditional single-displacement monitoring and establishes a quantitative assessment pathway from local strain to overall stability.

[0142] Specifically, in one embodiment of the present invention, the correlation analysis step in step S4 is as follows:

[0143] The fractal dimension and the extracted displacement-strain index are dimensionless to eliminate dimensional differences.

[0144] Based on grey relational analysis, the correlation coefficients between fractal dimension and displacement-strain index at each time point are calculated. The average correlation coefficients are then used to obtain the correlation degree, which is calculated using the following formula:

[0145]

[0146] Where μ i (o) is the correlation coefficient, r i For the calculated correlation degree, o is the index of the time point, and n' is the length of the time series;

[0147] If the correlation degree r i If ≥θ, then the consistency between the fractal dimension and the displacement-strain index is reliable; if the correlation degree r i <θ triggers the data verification or model correction process, re-executing steps S1 to S3; where θ is the correlation threshold.

[0148] The specific steps for re-executing S1 to S3 are as follows:

[0149] Verify the integrity of the raw data collection: First, formulate a detailed list based on the collection plan, clarify the requirements of each data field and observation point, check the data volume, investigate the reasons for the difference between the actual and planned collection volume, check the key data fields, and eliminate null values, outliers and values ​​that do not match the type. If the data has time attributes, the time series must be continuous, and the problems of jumps and repetitions must be handled.

[0150] Verification of fractal dimension convergence conditions: During calculation, monitor convergence indices, including the stability of fractal dimension as the box size changes in the box dimension method, and set a reasonable convergence threshold. When the change in fractal dimension is less than this value, it is considered converged. If it does not converge, increase the sample size and adjust the calculation parameters.

[0151] The influence of contact stiffness parameters on simulation results was analyzed to determine the initial adjustment range. Multiple sets of parameters were selected and substituted into the 3D model for simulation. The simulation results under different parameters were compared with the actual data to select the parameter values ​​that fit the reality. The existing loading conditions, including the method, magnitude, direction and time history, were reviewed. The loading method was optimized according to the actual problem, and the loading magnitude and direction were adjusted to fit the actual force. When dynamic loading was performed, the time parameters were precisely set. After the adjustment was completed, the simulation was repeated, and the new results were compared with the data before the adjustment and the actual data. If the results were not ideal, the adjustment was repeated.

[0152] In this embodiment, the weighted average calculation formula in step S4 is as follows:

[0153] M = w1 * D + w2 * L

[0154] Where M is the evaluation value calculated by the weighted average method, D is the fractal dimension, and L is the displacement and strain index; w1 and w2 are the weights, which are determined based on the mapping relationship between the fractal dimension and the stability level and the analytic hierarchy process.

[0155] The initial weight range of the fractal dimension is determined based on the mapping relationship between the fractal dimension and the stability level constructed in S2. Different stability levels correspond to different fractal dimension ranges. Based on this, the initial weight range of the fractal dimension is set to reflect the approximate importance of the fractal dimension to the overall result under different stability levels.

[0156] The weights are adjusted using the analytic hierarchy process (AHP) combined with expert experience:

[0157] A hierarchical model is constructed, with the target layer determining the weighting coefficients w1 and w2, the criterion layer containing the fractal dimension and other related factors, and the layer below the fractal dimension showing the fractal dimension corresponding to different stability levels.

[0158] Construct a judgment matrix and invite experts to compare the relative importance of each factor in the criterion layer pairwise to form the judgment matrix;

[0159] Calculate the weight vector and perform a consistency test. The weight vector is calculated using the eigenvalue method. The consistency test ensures that the expert judgment is reasonable. If the consistency ratio CR < 0.1, it is valid; otherwise, it is adjusted.

[0160] By combining expert experience with practical engineering considerations, the weights calculated based on AHP were adjusted to overcome the limitations of the mapping relationship between fractal dimension and stability level.

[0161] To determine the final weighting coefficients, follow the steps outlined above to determine w1 and w2, both of which take values ​​between 0 and 1 and sum to 1.

[0162] In this application, the evaluation thresholds are determined based on multiple factors, including the specific engineering characteristics of the rockfill dam, design standards, statistical analysis of monitoring data, and engineering experience, to establish a reasonable evaluation value range and stability level classification. Furthermore, the thresholds are dynamically adjusted in conjunction with expert opinions and actual conditions. The defined evaluation value thresholds include:

[0163] Stable 85-100: The evaluation value is within this range, indicating that the rockfill dam has very good stability. The fractal dimension and displacement strain index data are highly consistent, and both reflect that the deformation of the rockfill dam is under control. All indicators show that the dam structure is stable and there are basically no safety hazards under the current working conditions. It can operate normally and only requires routine monitoring and maintenance.

[0164] Relatively Stable 70-84: The rockfill dam is in a relatively stable state. Although there is a certain correlation between the fractal dimension and the displacement-strain index, there may be some subtle changes that need to be paid attention to. The deformation of the dam body is within the allowable range, but there may be some potential factors that need to be further observed. For example, the displacement-strain in local areas may show a slight increasing trend. At this time, the monitoring frequency should be increased and the condition of the dam body should be closely monitored in order to detect any potential problems in a timely manner.

[0165] A value of 55-69 indicates that the rockfill dam is in a critical state of stability. The correlation between the fractal dimension and the displacement-strain index may not be ideal, or some indicators may show that the dam deformation is trending out of the normal range. This may mean that there are some weak links in the dam or that it has been affected by external factors to a certain extent. It is necessary to conduct a detailed inspection and analysis of the dam, assess the potential risks, and take corresponding measures, such as reinforcement and adjustment of the operation mode, to prevent the stability from declining further.

[0166] Unstable 40-54: The stability of the rockfill dam has shown obvious problems. The assessment values ​​indicate poor consistency between the fractal dimension and the displacement strain index, and the deformation of the dam body may have exceeded the safe range, posing a significant safety hazard. At this time, emergency measures must be taken immediately, such as restricting the use of the dam body, carrying out emergency reinforcement, etc., and organizing professional personnel to conduct in-depth research and formulate a comprehensive repair and improvement plan.

[0167] Danger 0-39: This indicates that the rockfill dam is in an extremely dangerous state and may be damaged or unstable at any time. This may be due to serious anomalies in the fractal dimension and displacement strain indicators, obvious signs of damage such as cracks and landslides in the dam body, or deformation of key parts exceeding the design limits. In this case, personnel should be evacuated immediately, related operations should be stopped, and all possible measures should be taken for emergency rescue and disaster relief to avoid a major accident.

[0168] This invention constructs a deformation time series from dam deformation data and calculates the fractal dimension using the box counting method. Fractal theory can effectively describe the high complexity and nonlinear characteristics of rockfill dam deformation. Through in-depth analysis of historical data, an accurate and reliable correspondence between fractal dimension and rockfill dam deformation stability is established. This innovative application overcomes the shortcomings of traditional methods in accurately capturing the inherent laws of rockfill dam deformation. Under complex geological conditions and operating conditions, it can more accurately assess dam stability and effectively improve the accuracy of assessment results.

[0169] In terms of 3D model construction, this invention inputs the dam outline data into 3D modeling software and combines it with the design data and experimental data of the rockfill dam to determine the material mechanical parameters. It fully considers the actual environmental conditions of the rockfill dam, such as seismic force and water scouring force, and performs accurate loading simulation in the 3D model. Unlike the overly idealized model construction in the prior art, this invention can more realistically simulate the stress and deformation of the dam body in the actual environment, making the displacement and strain indicators extracted from the simulation analysis more in line with reality, and greatly enhancing the reliability of the simulation analysis results.

[0170] This invention integrates the calculated fractal dimension with the extracted displacement and strain indices and performs correlation analysis. Through this analysis, the intrinsic relationship between the parameters is studied in depth, and the accuracy of the evaluation results is verified. This process can effectively integrate different types of parameters, avoid the one-sidedness caused by the independent analysis of parameters in existing evaluation methods, and make the evaluation results more reliable and convincing.

[0171] In addition, such as Figure 2 As shown, the present invention also provides a deformation stability assessment system for rockfill dams based on multi-parameter analysis, comprising:

[0172] The data acquisition module is configured to collect dam deformation data and dam outline data, and construct them into a dataset;

[0173] The fractal dimension analysis module is configured to analyze and calculate the fractal dimension of the collected dam deformation data.

[0174] The modeling and analysis module is configured to construct a three-dimensional model of the rockfill dam, perform simulation analysis, and extract displacement and strain indices.

[0175] The fusion analysis module is configured to integrate fractal dimension and displacement and strain indices, and perform weighted calculations to comprehensively calculate the stability assessment value.

[0176] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for evaluating the deformation stability of rockfill dams based on multi-parameter analysis, characterized in that, The evaluation steps are as follows: S1. Data on the rockfill dam is collected based on monitoring points. The data includes dam deformation data and dam outline data. The data is preprocessed to form a dataset. S2. Convert the dam deformation data into a deformation time series in chronological order, calculate its fractal dimension using the box counting method, and establish a mapping relationship between the fractal dimension and the deformation stability of the rockfill dam by combining historical data statistical analysis. S3. Import the dam outline data into 3D modeling software to construct a 3D geometric model of the rockfill dam; determine the material mechanical parameters based on the rockfill dam design data, experimental data and particle flow theory inversion method, and apply multiple working conditions to the model for simulation analysis according to the actual environmental load conditions, and extract displacement strain index from the simulation results. S4. Conduct a correlation analysis between fractal dimension and displacement strain index to verify the consistency of the two data. If the correlation degree is greater than the preset threshold, its weight is determined based on the mapping relationship between fractal dimension and deformation stability of rockfill dam. Combined with displacement strain index, the comprehensive stability assessment value is calculated by weighted average method. If the correlation degree is less than or equal to the preset threshold, the data verification or model correction process will be triggered. The stability level of rockfill dams is determined based on the assessment values.

2. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, Step S1 includes: The horizontal displacement data, vertical displacement data, and strain data of the dam body are collected using displacement sensors and strain gauges. The dam outline parameters and upstream and downstream dam slope shape parameters were obtained through a GIS system and a total station. The collected dam deformation and contour data are preprocessed to form a dataset.

3. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, The step in step S2 to transform the time series is as follows: The timestamp of each record is extracted from the preprocessed dam deformation data, and sorted in ascending order based on the timestamps to obtain a preliminary time series; The sampling frequency is unified by using equal-interval interpolation to generate equal-interval time series; Logical verification, statistical verification, continuity verification, and integrity verification are performed on the sequence data. For abnormal data that fails verification, linear interpolation is used to repair or mark and remove it. The verified time series is stored, and an index is created to associate the time series with the dam outline data to obtain the deformation time series.

4. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, The steps of the box counting method in step S2 are as follows: The deformed time series is mapped to a two-dimensional plane, and the partitions are of side length [missing information]. Square grid, The initial value is 1 / 10 of the maximum deformation of the dam body, and it is reduced to the minimum resolution in a proportional sequence; Statistics for each The corresponding number of non-empty boxes ,Record and ; For data points Perform least squares linear fitting, and require goodness of fit. Otherwise adjust The iteration range is recalculated. Taking the absolute value of the slope of the fitted line as the fractal dimension D, satisfying the relationship ,in The intercept; Correspondingly, the process of establishing the mapping relationship between fractal dimension and deformation stability of rockfill dams is as follows: Based on the stability level classification criteria in historical data, the fractal dimension is grouped according to the corresponding level; Calculate the mean, standard deviation, and confidence interval of the fractal dimension at each level; The significance of the differences in fractal dimension among different levels was verified by the Kruskal-Wallis test, and a scatter plot of fractal dimension-stability level was plotted to show the mapping relationship between fractal dimension and deformation stability of rockfill dams.

5. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, The steps for constructing the 3D geometric model of the rockfill dam in step S3 are as follows: Obtain dam outline data from the dataset and convert the data into DXF, OBJ, and CSV formats for software recognition. Import the data into the 3D modeling software, set the coordinate system and units consistent with the actual measurement, use the Delaunay triangulation algorithm to generate a triangular mesh terrain model, and optimize the model accuracy through mesh smoothing and redundant vertex removal. Based on the contour data, the dam surface was constructed using the NURBS surface tool, and the terrain model and dam structure were merged through Boolean operations to generate a three-dimensional geometric model of the rockfill dam.

6. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, Determining the material's mechanical parameters in step S3 includes the following steps: The rockfill material was discretized into ellipsoidal particles. The aspect ratio r of the particles was determined statistically through field sieve analysis. A linear spring-damping model was used for particle contact, and the formula for calculating the normal force was: ； in For normal contact stiffness, This represents the normal relative displacement increment; The formula for calculating tangential force is: ； in For tangential contact stiffness, This represents the tangential relative displacement increment. The coefficient of friction; Calibration was achieved through uniaxial compression and direct shear tests. , and And introduce a particle shape correction factor. Correct contact stiffness; The granular flow simulation results were fitted using least squares, and the elastic modulus was determined based on the slope of the stress-strain curve, using the following formula: ； in , These are the mean values ​​of stress and strain. Based on the relationship between transverse and longitudinal strain, combined with bulk density ρ and particle shape correction factor Calculate Poisson's ratio.

7. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, The steps for extracting displacement strain parameters in step S3 are as follows: The horizontal X-axis, horizontal Y-axis, and vertical Z-axis displacement data of the rockfill dam under each node loading condition were extracted from the simulation results. Statistical analysis was performed on the node displacements in the dam crest region, and the mean and standard deviation of the displacements were calculated. Strain parameters are defined according to different regions: For the dam crest region, extract the principal strain v1 and the third principal strain v3 data, and calculate the tensile-compression gradient; For the dam slope region, shear strain is extracted from the elements along the dam slope direction, and the maximum shear strain is calculated. For the dam foundation area, vertical strain is extracted, and the mean compressive strain is calculated; Weights are allocated by region, and the overall displacement index is calculated by weighted summation based on the tensile-compression gradient, maximum shear strain, and mean compressive strain. .

8. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, The correlation analysis steps in step S4 are as follows: The fractal dimension and the extracted displacement-strain index are dimensionless to eliminate dimensional differences. Based on grey relational analysis, the correlation coefficients between fractal dimension and displacement-strain index at each time point are calculated. The average correlation coefficients are then used to obtain the correlation degree, which is calculated using the following formula: ； in The correlation coefficient, r i To calculate the correlation degree, For indexes of time points, The length of the time series; If the correlation If the fractal dimension and displacement strain index are consistent, then the correlation is reliable; This triggers a data verification or model correction process, re-executing steps S1 to S3; among which, This is the correlation threshold.

9. The method for evaluating the deformation stability of a rockfill dam based on multi-parameter analysis according to claim 1, characterized in that, The formula for calculating the weighted average in step S4 is: M=w1 D+w2 50 Where M is the evaluation value calculated by the weighted average method, D is the fractal dimension, L is the displacement strain index; w1 and w2 are the weights, which are determined based on the mapping relationship between the fractal dimension and the stability level and the analytic hierarchy process.

10. A multi-parameter analysis-based deformation stability assessment system for rockfill dams, applied to the multi-parameter analysis-based deformation stability assessment method for rockfill dams according to claim 1, characterized in that, include: The data acquisition module is configured to collect dam deformation data and dam outline data, and construct them into a dataset; The fractal dimension analysis module is configured to analyze and calculate the fractal dimension of the collected dam deformation data. The modeling and analysis module is configured to construct a three-dimensional model of the rockfill dam, perform simulation analysis, and extract displacement and strain indices. The fusion analysis module is configured to integrate fractal dimension and displacement and strain indices, and perform weighted calculations to comprehensively calculate the stability assessment value.

Images