Deformation prediction method and system for hydrodynamic landslide under reservoir water level cyclic fluctuation

Abstract

The invention provides a hydrodynamic landslide deformation prediction method and system under reservoir water level cyclic fluctuation, and relates to the technical field of data processing. The method comprises the steps that a landslide numerical model is constructed through ABAQUS software, landslide displacement and inclination angle changes under the condition that the reservoir water level rises and falls circularly at different rates are simulated, and a displacement change curve and an inclination angle change curve are obtained; accumulative displacement and inclination angle data of the landslide are obtained through field monitoring, and the two curves are compared and verified based on the accumulative displacement and the inclination angle data; splitting the displacement and dip angle change curve into a trend term and a periodic term by utilizing a component splitting method, and predicting the trend term by adopting Fourier curve fitting; constructing a periodic term component prediction model based on support vector regression, and respectively predicting a displacement periodic term and a dip angle periodic term; and superposing prediction results of the displacement trend term and the periodic term and prediction results of the dip angle trend term and the periodic term to obtain displacement and dip angle prediction values of the landslide so as to identify a high-risk area of landslide deformation.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level. Background Art

[0002] Landslides are the most common and devastating type of geological disaster, particularly in my country, where they occur frequently due to the complexity of the topography and diversity of geological structures. The construction of large-scale hydropower projects, particularly the operation of high dams and large reservoirs, has had a profound impact on the ecological and geological environment of the reservoir area. Water level fluctuations, in particular, pose a serious threat to the stability of reservoir bank slopes. Landslides not only pose a significant risk to the safety of people and property, but also cause immeasurable damage to the regional economy, transportation, and ecological environment. To effectively address this challenge, landslide monitoring, early warning, and prediction technologies have gradually become a research focus both domestically and internationally.

[0003] Currently, landslide stability analysis methods can be roughly divided into qualitative analysis methods, quantitative analysis methods, and uncertainty analysis methods. Qualitative analysis methods assess slope safety through engineering geological surveys and comparisons with historical data, but are often subject to interference from subjective factors and have certain limitations. Quantitative analysis methods, such as the limit equilibrium method and numerical simulation methods, have become the main technical means for assessing landslide stability. The limit equilibrium method has the advantages of simple operation and intuitive results and is widely used in slope stability analysis. However, this method has shortcomings when considering deformation effects and complex geological conditions. Numerical analysis methods, especially the finite element method combined with the strength reduction method, have demonstrated unique advantages in dealing with complex slope problems. They can more accurately simulate landslide processes and deformation characteristics, making them an important tool for predicting and preventing landslide disasters in reservoir areas.

[0004] However, traditional analysis methods often ignore the small deformations and nonlinear deformation behaviors that may occur in the slope during the landslide process. When faced with complex geological conditions, they are unable to accurately predict the stability of the slope, resulting in deviations in the assessment of landslide hazards. In addition, traditional methods often simplify the complex changes in landslide displacement and inclination when predicting landslide hazards, resulting in low prediction accuracy of deformation trends, which in turn affects the reliability of the prediction results. Summary of the Invention

[0005] In order to solve the technical problems that traditional analysis methods often ignore the small deformation and nonlinear deformation behavior that may occur in the slope during the landslide process, and are unable to accurately predict the stability of the slope when facing complex geological conditions, thereby leading to deviations in the assessment of landslide disasters; in addition, traditional methods often simplify the complex changes in landslide displacement and inclination when predicting landslide disasters, resulting in low prediction accuracy of deformation trends, which in turn affects the reliability of the prediction results, the present invention provides a method and system for predicting hydrodynamic landslide deformation under cyclic fluctuations of reservoir water levels.

[0006] The technical solutions provided by the embodiments of the present invention are as follows: First aspect: An embodiment of the present invention provides a method for predicting hydrodynamic landslide deformation under cyclic fluctuations of reservoir water levels, comprising: S1: Construct a numerical model of the landslide using ABAQUS software; S2: Based on the landslide numerical model, simulate the displacement and inclination changes of the landslide when the reservoir water level rises and falls at different rates, and determine the displacement change curve and inclination change curve of the landslide; S3: Obtaining cumulative displacement data and cumulative inclination data of the landslide through on-site monitoring; S4: Based on the accumulated displacement data and the accumulated inclination angle data, comparing and verifying the displacement change curve and the inclination angle change curve; S5: By using the component decomposition method, the verified displacement change curve and the inclination change curve are respectively decomposed into a displacement trend term component, a displacement period term component, an inclination trend term component, and an inclination period term component; S6: performing curve fitting on the displacement trend item component and the inclination trend item component respectively through Fourier curve fitting to determine a displacement trend item prediction result and an inclination trend item prediction result; S7: Construct a periodic component prediction model based on support vector regression; S8: taking the displacement periodic term component and the inclination periodic term component as input respectively, and outputting a displacement periodic term prediction result and an inclination periodic term prediction result through the periodic term component prediction model; S9: superimposing the displacement trend item prediction result and the displacement period item prediction result, and superimposing the inclination trend item prediction result and the inclination period item prediction result, to determine the displacement prediction value and the inclination prediction value of the landslide; S10: Determine high-risk areas for landslide deformation based on the predicted displacement value and the predicted inclination angle value.

[0007] Second aspect: An embodiment of the present invention provides a hydrodynamic landslide deformation prediction system under cyclic fluctuations of reservoir water levels, comprising: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level as described in the first aspect is implemented.

[0008] The third aspect: An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for predicting hydrodynamic landslide deformation under cyclic fluctuations of reservoir water level as described in the first aspect is implemented.

[0009] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: (1) In the embodiment of the present invention, the influence of the cyclic rise and fall of the reservoir water level on the landslide is accurately considered through the landslide numerical model, which can capture the subtle and nonlinear deformation that may occur in the slope during the landslide process. Using the component decomposition method, the displacement and inclination change curves are decomposed into trend terms and periodic terms, thereby determining the complex deformation behavior during the landslide process and avoiding the simplification and neglect of traditional methods. In addition, Fourier curve fitting further refines the capture of nonlinear deformation and enhances the accurate prediction of the dynamic response of the landslide process.

[0010] (2) In the embodiment of the present invention, the displacement and inclination changes of the landslide are decomposed into trend terms and periodic terms through the component decomposition method, which effectively improves the accuracy of deformation trend prediction and avoids the simplification of traditional methods. The periodic term component prediction model based on support vector regression accurately predicts the periodic term components, thereby enhancing the prediction accuracy of the periodic changes of the landslide. Finally, by superimposing the prediction results of the trend term and the periodic term, the complex changes of the landslide displacement and inclination are comprehensively considered, thereby significantly improving the reliability and accuracy of the overall prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic flow chart of a method for predicting hydrodynamic landslide deformation under cyclic fluctuations of reservoir water levels provided by an embodiment of the present invention; Figure 2 This is a structural schematic diagram of a hydrodynamic landslide deformation prediction system under cyclic fluctuations of reservoir water levels provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0014] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0015] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0016] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0017] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0018] Reference Manual Figure 1 , which shows a flow chart of a method for predicting hydrodynamic landslide deformation under cyclic fluctuations of reservoir water level provided by an embodiment of the present invention.

[0019] An embodiment of the present invention provides a method for predicting hydrodynamic landslide deformation under cyclical fluctuations in reservoir water levels. This method can be implemented by a device for predicting hydrodynamic landslide deformation under cyclical fluctuations in reservoir water levels, which can be a terminal or a server. The process flow of the method for predicting hydrodynamic landslide deformation under cyclical fluctuations in reservoir water levels may include the following steps: S1: Construct a landslide numerical model using ABAQUS software.

[0020] Among them, ABAQUS is a finite element analysis (FEA) software widely used in the engineering field. The software can simulate and analyze complex mechanical behaviors, covering multiple fields such as structural mechanics, thermal analysis, fluid-solid coupling, electromagnetic analysis, etc., and is widely used in aerospace, civil engineering, mechanical manufacturing, biomedicine and other industries.

[0021] In a possible implementation, S1 specifically includes: S101: Obtaining landslide terrain data.

[0022] S102: Based on the topographic data of the landslide, a two-dimensional finite element model of the landslide is created in ABAQUS software to define the landslide structure including the potential slip surface and boundary range.

[0023] S103: Based on the two-dimensional finite element model, a meshing tool is used to perform finite element meshing on the landslide.

[0024] S104: Based on the meshing results, mechanical parameters of the soil and bedrock in the landslide are selected.

[0025] Specifically, based on the meshed structure, the mechanical parameters of the soil and bedrock that match the structure are selected.

[0026] S105: Based on the mechanical parameters, the Mohr-Coulomb constitutive model is used to characterize the mechanical properties of the landslide.

[0027] The Mohr-Coulomb constitutive model is a constitutive relationship model widely used in geotechnical engineering and material mechanics. It is primarily used to describe the yield and failure behavior of brittle materials such as soil, rock, and concrete. Based on the Coulomb shear failure criterion and Mohr's stress circle theory, this model is used to analyze the failure characteristics of materials under shear stress.

[0028] S106: Setting the boundary conditions of the landslide numerical model.

[0029] It should be noted that boundary conditions can generally be divided into two categories: static boundary conditions and fluid boundary conditions. The model constructed in this paper contains four main surfaces. Given that the landslide surface is not subject to any constraints in its natural state, static boundary constraints are not applied to this surface, and only displacement constraints are applied to the other three surfaces. Considering the water level in contact with the surface in front of the landslide, it is set as a permeable boundary, and the other four surfaces are set as water-repellent boundaries.

[0030] S107: Set the initial reservoir water level and apply the reservoir water level fluctuation rate to simulate the deformation of the landslide under different water level fluctuations.

[0031] S108: Based on the boundary conditions, fluid-solid coupling analysis is used to analyze the effect of pore water pressure on landslide stability under different water level fluctuations.

[0032] Fluid-Solid Coupling (FSI) analysis, a computational method involving the interaction between fluids and solids, has broad applications in civil engineering, aerospace, petroleum exploration, and biomedicine. It is used to study the effects of fluids on solid structures, or the effects of solid deformation on fluid flow, to accurately simulate fluid-solid interaction phenomena in real engineering applications.

[0033] S109: Based on the boundary conditions, the strength reduction method is used to gradually reduce the cohesion and friction angle of the landslide to determine the stability and safety factor of the landslide, thereby completing the construction of the landslide numerical model.

[0034] The Strength Reduction Method (SRM) is a landslide stability calculation method based on finite element analysis, primarily used to calculate the slope stability factor (FS). This method determines slope stability by gradually reducing the shear strength parameters (cohesion and internal friction angle) of the soil or rock mass until the landslide fails.

[0035] Specifically, the main section with the most detailed stratigraphic information in the study area was selected as the subject of numerical simulation, and a two-dimensional model was constructed. A quadrilateral mesh was used for the model, with specifications of 1156.64m in length in the x-direction, 50m in width in the y-direction, and a maximum height of 724.62m in the z-direction. A total of 1832 cells and 1938 nodes were defined, and the stratigraphic layers were divided into seven layers. Since the landslide surface is naturally unconstrained, no static boundary constraints were applied to it; only displacement constraints were applied to the other three surfaces. Considering the water level at the front of the landslide in contact with the surface, this surface was designated as a permeable boundary, while the other four surfaces were designated as impermeable boundaries. Through engineering analogies, the physical and mechanical parameters of each layer were determined in detail. The bulk modulus and shear modulus played a crucial role in this parameter determination. Based on the basic principles of elasticity, relevant formulas were used to accurately convert these two key parameters. In combination with the engineering report and on-site practice, the present invention sets three different reservoir water level cycle rise and fall speeds of 0.8m / d, 1.6m / d, and 3.2m / d to study the deformation law and stability of the landslide during the rise and fall of the reservoir water level. Before the reservoir water level is raised or lowered, it is necessary to calculate the initial stress field of the reservoir water level in a stable state (i.e., natural state). The influence of the initial water level should be considered when calculating the initial equilibrium state stress field. The dead water level in the study area is 765.0m (corresponding to a model height of 165m), so the initial water level is set to 165m. The reservoir water level comes into contact with fewer loose deposits, and the seepage effect mainly affects the deposits at the front edge of the slope. After settlement occurs, it will drive a part of the rear edge slope to slide, but the overall amplitude is small. According to the calculation results of the strength reduction method, the current safety factor of the landslide reaches 0.98, which indicates that the landslide is still in a stable state. Based on the strength reduction method and fluid-solid coupling analysis, ABAQUS was secondary developed using the Python language to complete parametric modeling and batch operations, realizing the cyclic fluctuation of reservoir water levels at different rates. The deformation evolution law and stability of the slope were studied during the process.

[0036] In the present invention, the stress distribution, deformation mode and instability process of the landslide under different water level conditions can be simulated through the strength reduction method and finite element analysis, thereby improving the accuracy of the calculation and making the analysis results more consistent with the actual situation.

[0037] Furthermore, by incorporating fluid-structure interaction analysis in ABAQUS, we can simulate the pore water pressure caused by reservoir water level fluctuations. Using the reduction method, we can analyze the stability of the landslide under different water levels and assess its safety factor (FS). This method can more accurately reflect the impact of hydraulic forces on landslide instability and avoid underestimating or overestimating landslide risks.

[0038] S2: Based on the landslide numerical model, simulate the displacement and inclination changes of the landslide when the reservoir water level rises and falls at different rates, and determine the displacement change curve and inclination change curve of the landslide.

[0039] In a possible implementation, S2 specifically includes: S201: Use the landslide numerical model to simulate the rise and fall of reservoir water level at different rates.

[0040] S202: Use the strength reduction method to gradually reduce the cohesion and friction angle of the landslide, and record the displacement trend and inclination trend of the landslide under different water level conditions.

[0041] S203: Determine a displacement curve and an inclination change curve of the landslide according to the displacement trend and inclination trend of the landslide.

[0042] In this study, the cohesion and friction angle of the landslide soil are gradually reduced through the strength reduction method. This method can determine the displacement and inclination trends of the landslide under different water level conditions. These displacement and inclination curves are then generated, allowing the key instability stages of the landslide to be identified and improving the accuracy of landslide prediction. Furthermore, the impact of different water level change rates on the landslide is considered, making the study more comprehensive and enabling analysis of landslide stability under various working conditions.

[0043] S3: Obtain the cumulative displacement data and cumulative inclination data of the landslide through on-site monitoring.

[0044] Specifically, before the reservoir was initially filled to its minimum level, project personnel were responsible for installing GNSS displacement and reservoir level monitoring equipment. To address deformation observed on-site, GNSS monitoring equipment was installed at the leading edge, mid-to-back edge, and top of the slope's central axis, numbered GNSS-1 through GNSS-3. Simultaneously, new situational awareness equipment (for inclination monitoring) was also installed at the same locations to capture cumulative displacement and inclination data.

[0045] In the present invention, field monitoring is an important data source for numerical simulation and prediction in landslide analysis. Obtaining cumulative displacement and inclination data from field monitoring can effectively improve prediction accuracy and enhance landslide early warning capabilities. Furthermore, field monitoring data can directly reflect the actual deformation of the landslide, which can be used to correct and optimize numerical simulation results, ensure the rationality of model parameters, and reduce errors caused by discrepancies between model assumptions and actual conditions.

[0046] In a possible implementation, after S3, the following steps are further included: The accumulated displacement data and accumulated inclination angle data are processed to eliminate outliers and smooth them.

[0047] In a possible implementation, the outlier elimination process specifically includes: The outlier data points in the cumulative displacement data and the cumulative inclination data are determined by the following formula:

[0048]

[0049] in, O represents an outlier, Q 1 represents the first quartile, Q 3 represents the third quartile, IQR represents the interquartile range, x Represents all data points in the dataset.

[0050] Remove outlier data points.

[0051] Specifically, the interquartile range, which is the difference between the third quartile (Q3) and the first quartile (Q1), is called the IQR (IQR = Q3 - Q1). In this context, we define observations as outliers if their values ​​are below the lower whisker of the box plot (i.e., Q1 minus 1.5 times the IQR) or above the upper whisker (i.e., Q3 plus 1.5 times the IQR).

[0052] In this paper, the interquartile range method can effectively eliminate outliers that are too large or too small, ensuring data stability and making the analysis more reliable. At the same time, after eliminating outliers, the data can more realistically reflect the actual deformation trend of the landslide, ensuring the rationality of the prediction.

[0053] S4: Based on the accumulated displacement data and the accumulated inclination angle data, the displacement change curve and the inclination angle change curve are compared and verified.

[0054] Specifically, the displacement and inclination curves from the numerical simulation are displayed in the same graph as the cumulative displacement and inclination data from field monitoring. This visualization allows for an intuitive understanding of the degree of agreement between the two, helping to identify discrepancies between the model and actual observations. Based on these comparisons, the effectiveness of the numerical simulation model can be evaluated.

[0055] In the present invention, the numerical simulation results are compared and verified with the field monitoring data, which can improve the simulation accuracy, optimize the model parameters, identify the instability stage of the landslide, and provide a scientific basis for landslide prediction, early warning and control. This method can ensure the reliability of the landslide prediction model.

[0056] S5: By means of the component decomposition method, the verified displacement change curve and the inclination change curve are respectively decomposed into a displacement trend term component, a displacement period term component, an inclination trend term component and an inclination period term component.

[0057] The component decomposition method (CDM) is a mathematical technique used for time series data analysis. It decomposes the original signal (data) into its components to extract key trends and cyclical variation characteristics. In landslide prediction and monitoring, the CDM is often used to separate landslide displacement and inclination data into trend and cyclic terms, enabling more accurate analysis of landslide deformation patterns.

[0058] It should be noted that the displacement trend component represents the long-term cumulative displacement trend of the landslide, reflecting the long-term stability or instability of the landslide. The displacement period component reflects the cyclical changes in landslide displacement. The inclination trend component reflects the long-term trend of the slope inclination. The inclination period component reflects the cyclical changes in the slope inclination.

[0059] In one possible implementation, the calculation method of the component splitting method is specifically as follows:

[0060] in, C (t) represents the cumulative deformation, T (t) Indicates trend item data, P (t) Represents periodic item data.

[0061] Specifically, based on the theory of time series analysis, the accumulated displacement and inclination data of the landslide were systematically decomposed into three main components: trend term and period term. C (t) represents the cumulative deformation, T (t) The trend item data is mainly affected by its inherent geological conditions.

[0062] In this study, the component decomposition method effectively extracts long-term trends and cyclical changes in landslides, improving the accuracy of landslide predictions, optimizing numerical simulations, and enhancing the predictive capabilities of machine learning models. This method also provides a scientific basis for landslide early warning and management. This method can reduce data noise and improve the reliability of landslide stability analysis, making landslide monitoring and early warning more accurate and efficient.

[0063] S6: Perform curve fitting on the displacement trend term component and the inclination trend term component respectively through Fourier curve fitting to determine the displacement trend term prediction result and the inclination trend term prediction result.

[0064] It should be noted that the displacement trend item prediction result is based on the long-term displacement trend item of the landslide, and is mainly used to analyze the long-term stability or instability trend of the landslide.

[0065] If the displacement trend term increases, it may indicate that the landslide has entered a phase of accelerated deformation. If the rate of increase is too rapid, instability may occur. If the displacement trend term is stable, it indicates that the landslide is in a slow creep phase and will not become unstable in the short term. If the displacement trend term decreases, the deformation of the landslide has slowed down after reinforcement and treatment, or the stability has improved after the reservoir water level has dropped.

[0066] The prediction result of the inclination trend item is based on the long-term inclination change trend of the landslide area, and is mainly used to analyze the overall inclination change and stability of the landslide slope.

[0067] If the inclination trend term increases, it indicates that the slope is gradually tilting, which may be a precursor to landslide instability. If the inclination trend term is stable, it means that the slope will not tilt significantly in the short term and is generally stable. If the inclination trend term decreases, it indicates that the topography in the landslide area has adjusted, resulting in the inclination recovering.

[0068] Among them, Fourier Curve Fitting is a mathematical method based on Fourier series expansion, which is often used to analyze and fit data with periodic changes.

[0069] In a possible implementation, the calculation method of Fourier curve fitting is specifically as follows:

[0070] in, y (t) represents the forecast value of the periodic term, a 0 represents a constant, a s represents the coefficient of the cosine term, b s represents the coefficient of the sine term, ω represents the angular frequency, s represents the harmonic order, t Indicates time, q Indicates the fitting order.

[0071] It's important to note that after separating the landslide's trend terms, displacement and inclination, the data for both sets of trend terms exhibit a nearly linear pattern, with a monotonically increasing trend over time. Given this stable growth, the use of Fourier curve fitting is particularly appropriate for prediction. Fourier curve fitting, as an effective data analysis tool, relies on the statistical analysis of discrete data using orthogonal functions, and then achieves accurate curve fitting through Fourier series approximation.

[0072] In this invention, a set of sine and cosine terms can be used to approximate the long-term trend of a landslide through a Fourier series expansion, improving the accuracy of landslide trend prediction. Meanwhile, traditional linear regression or polynomial fitting can struggle to accurately describe the cyclical fluctuations and nonlinear trends of landslide data, easily leading to large prediction errors. Fourier curve fitting, which considers both trend and cyclic terms, is suitable for data with both cyclical and long-term trends, providing a more stable prediction of future deformation trends.

[0073] S7: Construct a periodic term component prediction model based on support vector regression.

[0074] Landslide displacement, inclination, and reservoir water level change data are collected as input for support vector regression.

[0075] Determine the optimization objectives and constraints of the periodic component prediction model.

[0076] The specific optimization goals are:

[0077] Among them, min means minimization, ω represents the weight vector of the regression function, 、 represents the slack variable, C represents the penalty coefficient, n Indicates the sample size.

[0078] The specific constraints are:

[0079] in, y i represents the actual observed value, represents the mapping function, b represents the bias term, Indicates the allowable error.

[0080] By introducing the Lagrange multiplier, the optimization objective is transformed into a dual optimization problem:

[0081] Among them, max means maximization, 、 represents the Lagrange multiplier, y i represents the actual observed value, represents the kernel function, x i Indicates the i The feature vector of the sample, x j Indicates the j The feature vector of the sample, 、 represents the Lagrange multiplier.

[0082] Introduce kernel function for data transformation to achieve nonlinear regression:

[0083] Among them, exp represents the exponential function, g represents the kernel parameters.

[0084] The penalty coefficient and kernel parameters are optimized by using the improved particle swarm optimization algorithm to determine the optimal Lagrange multiplier.

[0085] According to the optimal Lagrange multiplier, the periodic component prediction model is determined as:

[0086] in, f ( x ) represents the predicted value of the landslide period term, b represents the bias term, 、 represents the Lagrange multiplier, represents the kernel function, x Represents input data, x i represents the support vector, λ represents the regularization parameter, n represents the number of support vectors.

[0087] In this paper, by introducing a kernel function (such as the radial basis function (RBF)), SVR can map input data from the original low-dimensional space to a high-dimensional feature space, thereby improving prediction accuracy. Furthermore, by introducing slack variables, the model can tolerate a certain degree of error, ensuring good adaptability and stability when used with actual landslide data.

[0088] In a possible implementation, parameters of the periodic term component prediction model are optimized using an improved particle swarm optimization algorithm.

[0089] In a possible implementation, the improved particle swarm optimization algorithm specifically includes: S701: Initialize the particle swarm, which contains Togo particles, each of which represents a combination of a penalty coefficient and kernel parameters.

[0090] S702: Determine the fitness function of the improved particle swarm optimization algorithm:

[0091] in, represents the value of the fitness function,n represents the number of samples, y i represents the actual observed value, represents the predicted value of the model, Indicates the allowable error range.

[0092] S703: Calculate the fitness function value of each particle in the particle swarm.

[0093] S704: Update the particle with the highest fitness value to the global optimal position.

[0094] S705: Determine the neighborhood of each particle, and determine the particle with the highest fitness value in the neighborhood as the local optimal position based on the particle information in the neighborhood.

[0095] In this invention, the search capability of particles in a local area can be improved by guiding the local optimal position and sharing neighborhood information. Particles adjust their positions according to the particle with the highest fitness value in the neighborhood, thereby improving the accuracy of local search.

[0096] S706: Update the particle's velocity and position using the following formula:

[0097] in, V u represents the velocity of the particle, ω represents the inertia weight, c 1. c 2 represents the acceleration constant, r 1. r 2 represents a random number, Pbest u Indicates the u The best historical position of a particle, Gbest represents the global optimal position.

[0098] S707: Generate a random number and determine whether the currently generated random number is less than the preset mutation probability. If so, proceed to S708. Otherwise, proceed to S709.

[0099] S708: Using the mutation operation, update the particle position using the following formula and proceed to S712:

[0100] in, T uv Indicates the u The particle in v The particle position after mutation in the dimension, X uv Indicates the u The particle inv Current location information in dimensions, A Represents the amplification factor.

[0101] S709: Apply a random permutation function to the local optimal position to increase the diversity of the solution space:

[0102] in, Lbest represents the optimal solution position found by the particle in its neighborhood, and permuting() represents the random permutation function.

[0103] In the present invention, by introducing mutation operation and random arrangement of local solution space, the diversity of solution space can be maintained during the optimization process, the global search capability for complex landslide period prediction problems can be enhanced, and a better solution can be found.

[0104] S710: Generate the test position of the particle using the following formula:

[0105] in, Mutant represents the position of the particle after mutation, X represents the current position of the particle, Lbest represents the local optimal position, F Represents the amplification factor.

[0106] S711: Use the crossover operator to perform a crossover operation on the current particle to generate a solution for the next generation, and then proceed to 712.

[0107] S712: Calculate the fitness of each particle and update the global optimal position according to the fitness of the particle.

[0108] S713: Determine whether the maximum number of iterations is met, and if so, output the optimal result. Otherwise, return to S705.

[0109] In this paper, an advanced particle swarm optimization algorithm optimizes the parameters of a support vector regression (SVM) model to provide reliable and efficient solutions for predicting landslide periodic components. This provides more accurate periodic predictions and enhances the operability of landslide early warning and prevention. Furthermore, the improved particle swarm optimization algorithm effectively improves optimization efficiency, model stability, and search capabilities by incorporating mutation operations, random permutation of the local solution space, and information sharing between particles. Optimizing the landslide periodic component prediction model can find better solutions to complex nonlinear problems, making landslide prediction, early warning, and prevention more reliable.

[0110] S8: Taking the displacement periodic term component and the inclination periodic term component as input respectively, outputting the displacement periodic term prediction result and the inclination periodic term prediction result through the periodic term component prediction model.

[0111] In the present invention, by predicting the displacement periodic term and the inclination periodic term, the specific impact of different periodic changes on the landslide can be determined, and targeted prevention and control measures can be formulated in advance, thereby improving the effect of landslide control.

[0112] S9: superimposing the displacement trend item prediction result and the displacement period item prediction result, and superimposing the inclination trend item prediction result and the inclination period item prediction result, to determine the displacement prediction value and the inclination prediction value of the landslide.

[0113] It should be noted that by superimposing the trend term predicted by Fourier fitting and the period term predicted by PSO-SVR, the final goodness of fit of the cumulative displacement and inclination angle reached 0.9986 and 0.9965, respectively. The overall prediction effect of the model is good, and it has high reliability for the prediction of step-type landslide deformation.

[0114] In this invention, the displacement trend term and the period term reflect the long-term trend and cyclical fluctuations of a landslide, respectively. By superimposing the prediction results of these two terms, both trend changes and cyclical fluctuations are comprehensively considered, improving the overall prediction accuracy of landslide displacement changes. Furthermore, the superimposed prediction results can form a dynamic monitoring system for long-term landslide deformation. By comparing the actual deformation of the landslide with the predicted results in real time, the prediction model can be dynamically adjusted to improve the accuracy of predicting future deformation trends.

[0115] S10: Determine high-risk areas for landslide deformation based on predicted displacement and inclination values.

[0116] Optionally, a criterion for determining a high-risk area is determined, and when a displacement prediction value or an inclination prediction value of an area exceeds a threshold, the area is identified as a high-risk area.

[0117] For example, if the displacement exceeds a certain value (for example, 0.5 meters), it can be considered a high risk; if the tilt angle changes by more than a certain angle (for example, 5 degrees), it can be considered a high risk.

[0118] In this invention, the clear definition of high-risk areas helps decision-makers and engineers quickly identify key areas where landslides may occur, and quickly formulate appropriate protective measures based on the displacement and inclination changes in these areas. Furthermore, by monitoring landslide areas in real time and judging them based on set thresholds, the sensitivity and response speed of the landslide early warning system can be improved. This timely warning can prevent potential landslide disasters in advance.

[0119] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: (1) In the embodiment of the present invention, the influence of the cyclic rise and fall of the reservoir water level on the landslide is accurately considered through the landslide numerical model, which can capture the subtle and nonlinear deformation that may occur in the slope during the landslide process. Using the component decomposition method, the displacement and inclination change curves are decomposed into trend terms and periodic terms, thereby determining the complex deformation behavior during the landslide process and avoiding the simplification and neglect of traditional methods. In addition, Fourier curve fitting further refines the capture of nonlinear deformation and enhances the accurate prediction of the dynamic response of the landslide process.

[0120] (2) In the embodiment of the present invention, the displacement and inclination changes of the landslide are decomposed into trend terms and periodic terms through the component decomposition method, which effectively improves the accuracy of deformation trend prediction and avoids the simplification of traditional methods. The periodic term component prediction model based on support vector regression accurately predicts the periodic term components, thereby enhancing the prediction accuracy of the periodic changes of the landslide. Finally, by superimposing the prediction results of the trend term and the periodic term, the complex changes of the landslide displacement and inclination are comprehensively considered, thereby significantly improving the reliability and accuracy of the overall prediction.

[0121] Reference Manual Figure 2 , showing a structural schematic diagram of a hydrodynamic landslide deformation prediction system under cyclic fluctuations of reservoir water level provided by the present invention.

[0122] The present invention further provides a hydrodynamic landslide deformation prediction system 20 under cyclic fluctuation of reservoir water level, which is applied to the above-mentioned hydrodynamic landslide deformation prediction method under cyclic fluctuation of reservoir water level, and comprises: Processor 201.

[0123] The memory 202 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 201 , the method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level as described in the method embodiment is implemented.

[0124] The hydrodynamic landslide deformation prediction system 20 under cyclic fluctuation of reservoir water level provided by the present invention can execute the above-mentioned hydrodynamic landslide deformation prediction method under cyclic fluctuation of reservoir water level and achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate on it.

[0125] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: (1) In the embodiment of the present invention, the influence of the cyclic rise and fall of the reservoir water level on the landslide is accurately considered through the landslide numerical model, which can capture the subtle and nonlinear deformation that may occur in the slope during the landslide process. Using the component decomposition method, the displacement and inclination change curves are decomposed into trend terms and periodic terms, thereby determining the complex deformation behavior during the landslide process and avoiding the simplification and neglect of traditional methods. In addition, Fourier curve fitting further refines the capture of nonlinear deformation and enhances the accurate prediction of the dynamic response of the landslide process.

[0126] (2) In the embodiment of the present invention, the displacement and inclination changes of the landslide are decomposed into trend terms and periodic terms through the component decomposition method, which effectively improves the accuracy of deformation trend prediction and avoids the simplification of traditional methods. The periodic term component prediction model based on support vector regression accurately predicts the periodic term components, thereby enhancing the prediction accuracy of the periodic changes of the landslide. Finally, by superimposing the prediction results of the trend term and the periodic term, the complex changes of the landslide displacement and inclination are comprehensively considered, thereby significantly improving the reliability and accuracy of the overall prediction.

[0127] It should be understood that the processor in the embodiments of the present invention may be a central processing unit (CPU), but may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0128] It should also be understood that the memory in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0129] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0130] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0131] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0132] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0133] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0134] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0135] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0136] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0137] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0138] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0139] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for predicting hydrodynamic landslide deformation under cyclic fluctuations of reservoir water level as described in the method embodiment is implemented.

[0140] The computer-readable storage medium provided by the present invention can realize the steps and effects of the hydrodynamic landslide deformation prediction method under cyclic fluctuation of reservoir water level in the above method embodiment. To avoid repetition, the present invention will not elaborate on them.

[0141] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: (1) In the embodiment of the present invention, the influence of the cyclic rise and fall of the reservoir water level on the landslide is accurately considered through the landslide numerical model, which can capture the subtle and nonlinear deformation that may occur in the slope during the landslide process. Using the component decomposition method, the displacement and inclination change curves are decomposed into trend terms and periodic terms, thereby determining the complex deformation behavior during the landslide process and avoiding the simplification and neglect of traditional methods. In addition, Fourier curve fitting further refines the capture of nonlinear deformation and enhances the accurate prediction of the dynamic response of the landslide process.

[0142] (2) In the embodiment of the present invention, the displacement and inclination changes of the landslide are decomposed into trend terms and periodic terms through the component decomposition method, which effectively improves the accuracy of deformation trend prediction and avoids the simplification of traditional methods. The periodic term component prediction model based on support vector regression accurately predicts the periodic term components, thereby enhancing the prediction accuracy of the periodic changes of the landslide. Finally, by superimposing the prediction results of the trend term and the periodic term, the complex changes of the landslide displacement and inclination are comprehensively considered, thereby significantly improving the reliability and accuracy of the overall prediction.

[0143] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0144] There are a few points to note: (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0145] (2) For the sake of clarity, the thickness of layers or regions in the drawings used to describe the embodiments of the present invention are exaggerated or reduced, that is, these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element may be "directly on" or "under" the other element or intervening elements may be present.

[0146] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to form new embodiments.

[0147] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level, characterized by: include: S1: Construct a numerical model of the landslide using ABAQUS software; S2: Based on the landslide numerical model, simulate the displacement and inclination changes of the landslide when the reservoir water level rises and falls at different rates, and determine the displacement change curve and inclination change curve of the landslide; S3: Obtaining cumulative displacement data and cumulative inclination data of the landslide through on-site monitoring; S4: Based on the accumulated displacement data and the accumulated inclination angle data, comparing and verifying the displacement change curve and the inclination angle change curve; S5: By using the component decomposition method, the verified displacement change curve and the inclination change curve are respectively decomposed into a displacement trend term component, a displacement period term component, an inclination trend term component, and an inclination period term component; S6: performing curve fitting on the displacement trend item component and the inclination trend item component respectively through Fourier curve fitting to determine a displacement trend item prediction result and an inclination trend item prediction result; S7: Construct a periodic component prediction model based on support vector regression; S8: taking the displacement periodic term component and the inclination periodic term component as input respectively, and outputting a displacement periodic term prediction result and an inclination periodic term prediction result through the periodic term component prediction model; S9: superimposing the displacement trend item prediction result and the displacement period item prediction result, and superimposing the inclination trend item prediction result and the inclination period item prediction result, to determine the displacement prediction value and the inclination prediction value of the landslide; S10: Determine high-risk areas for landslide deformation based on the predicted displacement value and the predicted inclination angle value.

2. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 1 is characterized in that: Said S1 specifically includes: S101: Acquire landslide terrain data; S102: creating a two-dimensional finite element model of the landslide in the ABAQUS software according to the topographic data of the landslide to define a landslide structure including a potential slip surface and a boundary range; S103: Based on the two-dimensional finite element model, using a meshing tool to perform finite element meshing on the landslide; S104: Selecting mechanical parameters of soil and bedrock in the landslide based on the meshing result; S105: Based on the mechanical parameters, a Mohr-Coulomb constitutive model is used to characterize the mechanical properties of the landslide; S106: setting the boundary conditions of the landslide numerical model; S107: setting the initial reservoir water level and applying the reservoir water level fluctuation rate to simulate the deformation of the landslide under different water level fluctuations; S108: Based on the boundary conditions, fluid-solid coupling analysis is used to analyze the effect of pore water pressure on landslide stability under different water level fluctuations; S109: Based on the boundary conditions, a strength reduction method is used to gradually reduce the cohesion and friction angle of the landslide to determine the stability and safety factor of the landslide, thereby completing the construction of the landslide numerical model.

3. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 1, characterized in that: The S2 specifically includes: S201: simulating the cyclic rise and fall of the reservoir water level at different rates using the landslide numerical model; S202: Use the strength reduction method to gradually reduce the cohesion and friction angle of the landslide, and record the displacement trend and inclination trend of the landslide under different water level conditions; S203: Determine the displacement curve and the inclination change curve of the landslide according to the displacement trend and the inclination trend of the landslide.

4. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 1, characterized in that: After the S3, the following is also included: The accumulated displacement data and the accumulated inclination angle data are subjected to outlier elimination and smoothing processing.

5. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 4 is characterized in that: The outlier elimination process specifically includes: The outlier data points in the cumulative displacement data and the cumulative inclination data are determined by the following formula: in, O represents an outlier, Q 1 represents the first quartile, Q 3 represents the third quartile, IQR represents the interquartile range, x Represents all data points in the dataset; The outlier data points are eliminated.

6. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 1, characterized in that: The calculation method of the component splitting method is specifically as follows: in, C (t) represents the cumulative deformation, T (t) Indicates trend item data, P (t) Represents periodic item data.

7. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 1, characterized in that: The calculation method of the Fourier curve fitting is specifically as follows: in, y (t) represents the forecast value of the periodic term, a 0 represents a constant, a s represents the coefficient of the cosine term, b s represents the coefficient of the sine term, ω represents the angular frequency, s represents the harmonic order, t Indicates time, q Indicates the fitting order.

8. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 1, characterized in that: The periodic term component prediction model based on support vector regression is specifically: in, f ( x ) represents the predicted value of the landslide period term, b represents the bias term, 、 represents the Lagrange multiplier, represents the kernel function, x Represents input data, x i represents the support vector, λ represents the regularization parameter, n represents the number of support vectors.

9. The method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to claim 8, characterized in that: The parameters of the periodic term component prediction model are optimized by using an improved particle swarm optimization algorithm.

10. A hydrodynamic landslide deformation prediction system under cyclic fluctuation of reservoir water level, characterized by: include: processor; A memory storing computer-readable instructions, wherein the computer-readable instructions, when executed by the processor, implement the method for predicting hydrodynamic landslide deformation under cyclic fluctuation of reservoir water level according to any one of claims 1 to 9.

Images