Water power erosion accumulation body risk prevention and control method and system based on data driving

Abstract

The invention relates to the technical field of hydrodynamic erosion and geological safety monitoring, and discloses a hydrodynamic erosion accumulation body risk prevention and control method and system based on data driving, and the method comprises the steps: collecting hydraulic monitoring data and displacement monitoring data; based on a machine learning algorithm and a global sensitivity analysis method, an analysis framework of the influence of hydrodynamic erosion on the stability of the near-dam-area landslide accumulation body is established; establishing a risk analysis method for the accumulation body under the hydrodynamic erosion action according to the risk factors influencing the stability of the accumulation body under the hydrodynamic erosion action; catastrophe evolution characteristics of the accumulation body in the near-dam area of the hydropower station are analyzed, deformation prediction and long-term safety evaluation are carried out, and risk prevention and control measures are put forward. According to the method, a systematic analysis framework is established, the influence of hydrodynamics on a landslide accumulation body is comprehensively understood, a real-time monitoring and dynamic prediction technology is applied, potential stability risks are rapidly identified, comprehensive data analysis and model prediction are provided, and decision makers are assisted to formulate effective emergency response measures.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrodynamic erosion and geological safety monitoring, and in particular to a data-driven method and system for risk prevention and control of hydrodynamic erosion deposits. Background Art

[0002] The long-term effect of hydrodynamic erosion on the stability of landslide deposits cannot be ignored. Over time, it may cause large displacement of the deposits, leading to engineering safety hazards. Due to the heterogeneity, size effect, and discreteness of rocks, the parameters measured by conventional indoor tests are not suitable for numerical studies on the long-term stability of deposits under hydrodynamic erosion. Therefore, it is urgent to propose an effective parameter determination method to evaluate the stability of landslide deposits and predict deformation under hydrodynamic erosion. Combining monitoring data can effectively improve the accuracy of prediction of seepage deformation of deposits, and it is necessary to select a suitable parameter inversion method to solve this problem. Currently, the commonly used parameter inversion methods are divided into two categories: deterministic methods and uncertain methods. Deterministic methods have gradually developed from the earliest local optimization (Newton method, gradient descent method) to global optimization methods (genetic algorithm, particle swarm algorithm); the commonly used method in the uncertain method is the Bayesian method, which can reveal the uncertainty of parameters and evaluate them, and provide corresponding probability distribution intervals. Probabilistic inversion analysis provides a reasonable and feasible method to make full use of monitoring data, quantify and reduce parameter uncertainty, and improve prediction accuracy.

[0003] Clarify the hydrodynamic mechanical characteristics of typical accumulation bodies in the reservoir and dam area of ​​a hydropower station, establish a hydrodynamic erosion model for landslide accumulation bodies, construct a numerical calculation method for hydrodynamic erosion, re-evaluate the stability of typical accumulation bodies, carry out long-term safety research on accumulation bodies under the influence of multiple risk factors considering hydrodynamic erosion and spatial variability of rock and soil materials, and propose risk prevention and control measures for landslides of accumulation bodies caused by hydrodynamic erosion. These measures have important scientific theoretical and engineering practical value for ensuring the safe operation of hydropower projects. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a data-driven risk control method for hydrodynamic erosion deposits, which can clarify the risk factors of hydrodynamic erosion affecting the stability of deposits based on the data-driven method. The long-term safety of hydrodynamic erosion deposits under the influence of multiple risk factors considering different hydrodynamic conditions and spatial variability of rock and soil materials is studied, and risk control measures for hydrodynamic erosion deposits are proposed.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions, which are data-driven methods for risk control of hydrodynamic erosion deposits, including: collecting hydraulic monitoring data and displacement monitoring data; establishing an analysis framework for the impact of hydrodynamic erosion on the stability of landslide deposits in near-dam areas based on machine learning algorithms and global sensitivity analysis methods; establishing a risk analysis method for deposits under hydrodynamic erosion according to the risk factors that affect the stability of deposits under hydrodynamic erosion; analyzing the catastrophic evolution characteristics of deposits in near-dam areas of hydropower stations, conducting deformation prediction and long-term safety evaluation, and proposing risk control measures.

[0007] As a preferred scheme of the data-driven hydrodynamic erosion deposit risk prevention and control method described in the present invention, the collection of hydraulic monitoring data and displacement monitoring data includes obtaining rainfall, reservoir water level, flow velocity, deposit displacement, and sliding surface deformation under different rainfall conditions, and performing data preprocessing on the collected data.

[0008] As a preferred solution of the data-driven hydrodynamic erosion accumulation risk prevention and control method described in the present invention, the machine learning algorithm and the global sensitivity analysis method include selecting key features of rainfall intensity R, water level H, and displacement D according to physical meaning and practical application, and constructing a feature vector X:

[0009] X=[R,H,D,...,N]

[0010] Where N represents other relevant features;

[0011] The improved random forest model and support vector machine are used for training using the weighted loss function L:

[0012]

[0013] Among them, w i is the weight of the i-th sample, y i is the true value, f(X i ) is the model output value;

[0014] The influence of each risk factor on the stability of the accumulation is identified through global sensitivity analysis, the accumulation stability index S is defined, and the sensitivity analysis is performed through the Sobol method to calculate the contribution of each input parameter to the output S.

[0015] As a preferred solution of the data-driven hydrodynamic erosion accumulation risk prevention and control method of the present invention, the analysis framework includes: constructing an accumulation stability assessment model, simulating the accumulation response under different hydrodynamic conditions, and determining the interaction and combination effect of various risk factors;

[0016] Integrate monitoring data, obtain geological exploration data on the accumulation body, collect relevant landslide and collapse event data, analyze different risk factors, select the feature set that most affects the stability of the accumulation body, use the above-mentioned machine learning algorithm to predict stability, and capture complex nonlinear relationships; generate the spatiotemporal characteristics of the hydrodynamic environment through interpolation, use numerical simulation methods to analyze the response of the accumulation body under different conditions, combine short-term and long-term data to adapt the model to dynamic hydrological changes, and evaluate stability trends; use multivariate sensitivity analysis to evaluate the influence of various risk factors and interactions on the stability of the accumulation body.

[0017] As a preferred solution of the data-driven method for risk control of hydrodynamic erosion accumulation bodies described in the present invention, the risk factors include rainfall intensity, reservoir water level fluctuation, accumulation body sliding surface shape and sliding resistance, rock and soil body physical and mechanical properties, and historical deformation record factors;

[0018] Generate interaction terms for each pair of key risk factors and add them to the prediction model to identify the marginal effect on the stability of the pile when both factors change at the same time; use the variance analysis method to evaluate the contribution of different interaction terms to the model output and create different scenarios to observe the specific impact of the interaction effect on the stability of the pile;

[0019] Establish a network relationship diagram between risk factors, analyze how changes in one or more key factors trigger changes in other factors within the system, explore causal relationships, identify core driving factors and estimate the comprehensive impact on the stability of the accumulation.

[0020] As a preferred embodiment of the data-driven risk prevention and control method for hydrodynamic erosion accumulation bodies described in the present invention, the method for establishing a risk analysis method for hydrodynamic erosion accumulation bodies includes formulating a comprehensive evaluation index system, combining quantitative analysis with qualitative evaluation, and grading the impact of different risk factors to determine the potential risk level;

[0021] A quantitative scoring model is established, the weight of each indicator is combined with the indicator value, and the risk scoring matrix RSM is constructed using the scores of different indicators and the corresponding weights:

[0022]

[0023] Each row represents a different monitoring point, and each column is the score of a different risk factor. On the basis of calculating the risk level, the risk of each monitoring point is integrated to obtain the overall risk level L t :

[0024] L t = argmax(R i )

[0025] Among them, the risk level corresponding to the maximum risk score is selected as the final conclusion.

[0026] As a preferred solution of the data-driven risk prevention and control method for hydrodynamic erosion deposits described in the present invention, the deformation prediction and long-term safety evaluation include using numerical simulation methods to perform timeliness analysis on the deposits under different hydrodynamic conditions, and combining historical data and trend analysis to evaluate the stability changes in the future period;

[0027] When the displacement of the sliding surface of the accumulation body exceeds the set threshold of 10mm / month, the emergency assessment mechanism needs to be activated; when the displacement monitored continuously reaches 20mm, a detailed geological survey is carried out; if the displacement exceeds 30mm, relevant engineering activities need to be suspended and on-site reinforcement is carried out;

[0028] Set a threshold combination of rainfall and reservoir water level. When rainfall reaches 50 mm / h and the reservoir water level is above 80% of the impoundment, a high-risk warning is activated. If this condition continues for 48 hours, a risk reassessment is conducted to check the stability of the accumulation. If rainfall reaches 75 mm / h and the reservoir water level is above the critical value, an evacuation order is immediately issued.

[0029] Monitor changes in the shear strength of the rock and soil. When the shear strength is lower than 12 kPa, initiate stability assessment and increase monitoring frequency to once every 12 hours. If the average shear strength below the slope decreases by more than 5%, automatically initiate a prediction model based on machine learning to predict stability risks within the next 48 hours.

[0030] The long-term safety evaluation collects historical risk score data based on past change trends and uses a model to predict risk scores at K future time points:

[0031]

[0032] in, is the predicted future risk score, μ is the constant term, Φ is the autoregressive parameter, θ is the moving average parameter, is the risk score at the previous time point, is the prediction error at the previous time point;

[0033] For each future time point t+k, the risk score and prediction error of the previous moment are used for calculation.

[0034] Another object of the present invention is to provide a data-driven hydrodynamic erosion accumulation body risk prevention and control system, as a preferred solution of the data-driven hydrodynamic erosion accumulation body risk prevention and control system of the present invention, which includes: a data collection and processing module, a machine learning and analysis module, a risk assessment and early warning module, a simulation and prediction module, and a decision support and feedback module;

[0035] The data collection and processing module is responsible for collecting hydraulic monitoring data and displacement monitoring data in real time, and cleaning, normalizing and preprocessing the collected data to ensure data quality and consistency;

[0036] The machine learning and analysis module will use machine learning algorithms to analyze the processed data, screen key factors through feature engineering, build an analysis framework, and identify the main risk factors and interactions that affect the stability of the pile;

[0037] The risk assessment and early warning module is responsible for conducting risk assessment according to the established risk analysis model, generating a risk scoring matrix, identifying potential risk levels, and providing real-time early warnings in combination with set thresholds;

[0038] The simulation and prediction module uses numerical simulation methods to perform dynamic time-effectiveness analysis of the accumulation body under different hydrodynamic conditions, including long-term trend analysis and deformation prediction;

[0039] The decision support and feedback module integrates the outputs of various analysis modules to generate a comprehensive management report, providing a scientific basis for decision makers and guiding on-site emergency response and long-term management strategies. At the same time, a dynamic feedback mechanism is established to review the recorded data and subsequent events in order to continuously optimize risk prevention and control methods.

[0040] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of a data-driven method for preventing and controlling the risk of hydrodynamic erosion deposits are implemented.

[0041] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of a data-driven method for preventing and controlling the risk of hydrodynamic erosion deposits are implemented.

[0042] The beneficial effects of the present invention are as follows: risk factors and interaction effects are quantified by using machine learning algorithms and sensitivity analysis methods, significantly improving the accuracy of stability prediction. A real-time monitoring and early warning system is established to ensure rapid response under dangerous conditions and reduce potential losses. Multiple information is integrated to support decision-making with scientific data and models, promoting long-term safety management of piles. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing 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. Among them:

[0044] Figure 1 A schematic flow chart of a data-driven method for preventing and controlling the risk of hydrodynamic erosion deposits provided in accordance with an embodiment of the present invention.

[0045] Figure 2 An embodiment of the present invention provides a method for preventing and controlling the risk of hydrodynamic erosion deposits based on a data-driven method for predicting deformation and damage of landslide deposits and for evaluating their long-term safety.

[0046] Figure 3 A schematic diagram of the working modules of a data-driven hydrodynamic erosion deposit risk prevention and control system provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0050] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0051] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0052] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0053] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a data-driven method for preventing and controlling the risk of hydrodynamic erosion deposits, including:

[0054] S1: Collect hydraulic monitoring data and displacement monitoring data.

[0055] Furthermore, the rainfall, reservoir water level, flow velocity, displacement of the accumulation body, and deformation of the sliding belt surface under different rainfall conditions are obtained, and the collected data are preprocessed.

[0056] S2: Based on machine learning algorithms and global sensitivity analysis methods, an analytical framework for the impact of hydrodynamic erosion on the stability of landslide deposits near the dam is established.

[0057] Furthermore, according to the physical meaning and practical application, the key features rainfall intensity R, water level H, and displacement D are selected to construct the feature vector X:

[0058] X=[R,H,D,...,N]

[0059] Where N represents other relevant features;

[0060] The improved random forest model and support vector machine are used for training using the weighted loss function L:

[0061]

[0062] Among them, w i is the weight of the i-th sample, y i is the true value, f(X i ) is the model output value;

[0063] Through global sensitivity analysis, the influence of each risk factor on the stability of the deposit is identified, and the deposit stability index S is defined:

[0064] S=g(X)=aR+bH+cD+...+mN

[0065] Among them, a, b, c, and m are parameters corresponding to risk factors;

[0066] The Sobol method is used to perform sensitivity analysis and calculate the contribution of each input parameter to the output S. The main parts include:

[0067]

[0068] Among them, V Y is the variance of the output S, X i and X j are the input parameters, and p represents the total number of input parameters used for sensitivity analysis.

[0069] It should be noted that by constructing a model for assessing the stability of the accumulation, the accumulation response under different hydrodynamic conditions is simulated, and the interaction and combined effects of various risk factors are determined;

[0070] Integrate monitoring data, obtain geological exploration data on the accumulation body, collect relevant landslide and collapse event data, analyze different risk factors, select the feature set that most affects the stability of the accumulation body, use the above-mentioned machine learning algorithm to predict stability, and capture complex nonlinear relationships; generate the spatiotemporal characteristics of the hydrodynamic environment through interpolation, use numerical simulation methods to analyze the response of the accumulation body under different conditions, combine short-term and long-term data to adapt the model to dynamic hydrological changes, and evaluate stability trends; use multivariate sensitivity analysis to evaluate the influence of various risk factors and interactions on the stability of the accumulation body.

[0071] S3: Based on the risk factors that affect the stability of the deposit under hydrodynamic erosion, a risk analysis method for hydrodynamic erosion deposits is established.

[0072] Furthermore, the risk factors include rainfall intensity, reservoir water level fluctuation, the shape and sliding resistance of the sliding belt of the accumulation body, the physical and mechanical properties of the rock and soil body, and historical deformation record factors;

[0073] Generate interaction terms for each pair of key risk factors and add them to the prediction model to identify the marginal effect on the stability of the pile when both factors change at the same time; use the variance analysis method to evaluate the contribution of different interaction terms to the model output and create different scenarios to observe the specific impact of the interaction effect on the stability of the pile;

[0074] Establish a network relationship diagram between risk factors, analyze how changes in one or more key factors trigger changes in other factors within the system, explore causal relationships, identify core driving factors and estimate the comprehensive impact on the stability of the accumulation.

[0075] It should be noted that a comprehensive evaluation indicator system is developed, combining quantitative analysis with qualitative assessment to grade the impact of different risk factors to determine the potential risk level;

[0076] A quantitative scoring model is established, the weight of each indicator is combined with the indicator value, and the risk scoring matrix RSM is constructed using the scores of different indicators and the corresponding weights:

[0077]

[0078] Each row represents a different monitoring point, and each column is the score of a different risk factor. On the basis of calculating the risk level, the risk of each monitoring point is integrated to obtain the overall risk level L t :

[0079] L t = argmax(R i )

[0080] Among them, the risk level corresponding to the maximum risk score is selected as the final conclusion.

[0081] S4: Analyze the catastrophic evolution characteristics of the accumulation body near the dam area of ​​the hydropower station, conduct deformation prediction and long-term safety evaluation, and propose risk prevention and control measures.

[0082] Furthermore, numerical simulation methods are used to conduct a time-effectiveness analysis of the accumulation body under different hydrodynamic conditions, and combined with historical data and trend analysis, its stability changes in the future period are evaluated;

[0083] When the displacement of the sliding surface of the accumulation body exceeds the set threshold of 10mm / month, the emergency assessment mechanism needs to be activated; when the displacement monitored continuously reaches 20mm, a detailed geological survey is carried out; if the displacement exceeds 30mm, relevant engineering activities need to be suspended and on-site reinforcement is carried out;

[0084] Set a threshold combination of rainfall and reservoir water level. When rainfall reaches 50 mm / h and the reservoir water level is above 80% of the impoundment, a high-risk warning is activated. If this condition continues for 48 hours, a risk reassessment is conducted to check the stability of the accumulation. If rainfall reaches 75 mm / h and the reservoir water level is above the critical value, an evacuation order is immediately issued.

[0085] Monitor changes in the shear strength of the rock and soil. When the shear strength is lower than 12 kPa, initiate stability assessment and increase monitoring frequency to once every 12 hours. If the average shear strength below the slope decreases by more than 5%, automatically initiate a prediction model based on machine learning to predict stability risks within the next 48 hours.

[0086] The long-term safety evaluation collects historical risk score data based on past change trends and uses a model to predict risk scores at K future time points:

[0087]

[0088] in, is the predicted future risk score, μ is the constant term, Φ is the autoregressive parameter, θ is the moving average parameter, is the risk score at the previous time point, is the prediction error at the previous time point;

[0089] For each future time point t+k, we use the risk score and prediction error of the previous moment to calculate and first predict Y t+1 :

[0090]

[0091] Then predict Y t+2 :

[0092]

[0093] Repeat the above process until the predicted Y t+k

[0094] Furthermore, a regular assessment procedure should be developed every quarter or every six months to conduct a comprehensive inspection and assessment of monitoring data, risk status and effectiveness of prevention and control measures. In special circumstances (such as after extreme weather events), special assessments should be conducted in a timely manner. Assessment reports should be generated regularly to analyze the stability of the accumulation and its changing trends, and feedback should be provided to management in a timely manner.

[0095] In the identified high-risk areas, the surface of the accumulation body is reinforced by using anchoring, shotcrete or soil reinforcement techniques to enhance its ability to resist scour; a drainage system is established to reduce the accumulation of water inside the accumulation body and reduce the impact of hydrodynamic erosion; deep-rooted plants (such as herbs or shrubs) are planted to fix the soil through the plant root system to reduce soil loss and erosion.

[0096] Introduce ecological restoration technologies, such as filter materials and slope protection lawns, to restore and protect the ecological balance.

[0097] Furthermore, we will regularly review and update the risk assessment model, and adjust management measures in response to new data and new situations. We will establish a feedback mechanism to promptly feedback monitoring results to the decision-making level to quickly adjust strategies and responses, strengthen coordination and cooperation with relevant units, form a collaborative working mechanism for information sharing and resource integration, establish a cross-departmental joint inspection team, regularly survey the accumulation body and surrounding environment, and evaluate the implementation effect.

[0098] Example 2, reference Figure 2 , which is an embodiment of the present invention, provides a data-driven method for risk control of hydrodynamic erosion deposits. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0099] Carry out data monitoring and analysis of the accumulation body near the dam of Suofengying Hydropower Station, and clarify the risk factors of hydrodynamic erosion affecting the stability of the accumulation body based on data-driven methods. Carry out long-term safety research on the accumulation body under the influence of multiple risk factors of hydrodynamic erosion considering different hydrodynamic conditions and spatial variability of rock and soil materials, and propose risk prevention and control measures for the accumulation body under hydrodynamic erosion.

[0100] (1) Data-driven landslide accumulation risk factors and risk analysis methods

[0101] Through in-depth analysis of hydraulic monitoring data and accumulation body displacement monitoring data, an analytical framework for the impact of hydrodynamic erosion on the stability of landslide accumulation bodies near the dam area was established based on machine learning algorithms and global sensitivity analysis methods, revealing the degree of influence of various factors on the stability of the accumulation bodies.

[0102] According to the risk factors that affect the stability of deposits under hydrodynamic erosion, such as different rainfall conditions, reservoir water level, deposit sliding surface and rock and soil strength, a study on multiple risk factors under different hydrodynamic conditions and spatial variability of rock and soil materials is carried out, and a risk analysis method for deposits under hydrodynamic erosion is established.

[0103] (2) Long-term safety assessment and risk control of deposits near dams under hydrodynamic erosion

[0104] Carry out long-term safety research on the accumulation bodies under hydrodynamic erosion, study and evaluate the possible instability and damage forms of the accumulation bodies caused by hydrodynamic erosion and the disaster consequences, analyze the catastrophic evolution characteristics of the accumulation bodies in the vicinity of the Suofengying Hydropower Station, carry out deformation prediction and long-term safety evaluation, and comprehensively propose risk prevention and control measures for the accumulation bodies in the vicinity of the dam under hydrodynamic erosion. Figure 2As shown in the figure, different sliding surfaces (primary slip surface, secondary slip surface and tertiary slip surface), monitoring piles (Monitoringpile) and multiple marks related to heavy rainfall (Heavy Rain) are marked. The horizontal axis is distance and the vertical axis is altitude.

[0105] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0106] Embodiment 3, the third embodiment of the present invention, is different from the first two embodiments in that:

[0107] If the functions are implemented in the form of 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 part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0108] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0109] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0111] Example 4, reference Figure 3 , which is an embodiment of the present invention, provides a data-driven hydrodynamic erosion accumulation risk prevention and control system, which is characterized by: comprising a data collection and processing module 1, a machine learning and analysis module 2, a risk assessment and early warning module 3, a simulation and prediction module 4, and a decision support and feedback module 5;

[0112] Data collection and processing module 1 is responsible for collecting hydraulic monitoring data and displacement monitoring data in real time, and cleaning, normalizing and preprocessing the collected data to ensure data quality and consistency;

[0113] Machine Learning and Analysis Module 2 will use machine learning algorithms to analyze the processed data, screen key factors through feature engineering, build an analysis framework, and identify the main risk factors and interactions that affect the stability of the pile;

[0114] Risk assessment and early warning module 3 is responsible for conducting risk assessment based on the established risk analysis model, generating a risk scoring matrix, identifying potential risk levels, and providing real-time early warnings based on set thresholds;

[0115] Simulation and prediction module 4 uses numerical simulation methods to conduct dynamic and timely analysis of the accumulation body under different hydrodynamic conditions, including long-term trend analysis and deformation prediction;

[0116] Decision support and feedback module 5 integrates the output of each analysis module to generate a comprehensive management report, providing scientific basis for decision makers, guiding on-site emergency response and long-term management strategies. At the same time, a dynamic feedback mechanism is established to review the recorded data and subsequent events in order to continuously optimize risk prevention and control methods.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A data-driven method for risk prevention and control of hydrodynamic erosion deposits, characterized by: include, Collect hydraulic monitoring data and displacement monitoring data; Based on machine learning algorithms and global sensitivity analysis methods, an analytical framework for the impact of hydrodynamic erosion on the stability of landslide deposits near dam areas is established. According to the risk factors that affect the stability of the accumulation body under the action of hydrodynamic erosion, a risk analysis method for the accumulation body under the action of hydrodynamic erosion is established; Analyze the catastrophic evolution characteristics of the accumulation body near the dam area of ​​the hydropower station, conduct deformation prediction and long-term safety evaluation, and propose risk prevention and control measures.

2. The data-driven method for risk control of hydrodynamic erosion deposits according to claim 1, characterized in that: The collection of hydraulic monitoring data and displacement monitoring data includes obtaining rainfall, reservoir water level, flow velocity, displacement of accumulation body, and deformation of sliding belt surface under different rainfall conditions, and performing data preprocessing on the collected data.

3. The data-driven method for risk control of hydrodynamic erosion deposits according to claim 2, characterized in that: The machine learning algorithm and global sensitivity analysis method include selecting key features of rainfall intensity R, water level H, and displacement D according to physical meaning and practical application, and constructing a feature vector X: X=[R,H,D,...,N] Where N represents other relevant features; The improved random forest model and support vector machine are used for training using the weighted loss function L: Among them, w i is the weight of the i-th sample, y i is the true value, f(X i ) is the model output value; The influence of each risk factor on the stability of the accumulation is identified through global sensitivity analysis, the accumulation stability index S is defined, and the sensitivity analysis is performed through the Sobol method to calculate the contribution of each input parameter to the output S.

4. The data-driven method for risk control of hydrodynamic erosion deposits according to claim 3, characterized in that: The analysis framework includes constructing a deposit stability assessment model, simulating the response of the deposit under different hydrodynamic conditions, and determining the interaction and combined effects of various risk factors; Integrate monitoring data, obtain geological exploration data on the accumulation body, collect relevant landslide and collapse event data, analyze different risk factors, select the feature set that most affects the stability of the accumulation body, use the above-mentioned machine learning algorithm to predict stability, and capture complex nonlinear relationships; generate the spatiotemporal characteristics of the hydrodynamic environment through interpolation, use numerical simulation methods to analyze the response of the accumulation body under different conditions, combine short-term and long-term data, adapt the model to dynamic hydrological changes, and evaluate stability trends; Multivariate sensitivity analysis was used to evaluate the impact of various risk factors and their interactions on the stability of the deposit.

5. The data-driven method for risk control of hydrodynamic erosion deposits according to claim 4, characterized in that: The risk factors include rainfall intensity, reservoir water level fluctuation, the shape of the sliding surface of the accumulation body and the sliding resistance, the physical and mechanical properties of the rock and soil body, and the historical deformation record factors; For each pair of key risk factors, interaction terms are generated and added to the prediction model to identify the marginal effect on the stability of the pile when both factors change simultaneously; The variance analysis method was used to evaluate the contribution of different interaction terms to the model output, and different scenarios were created to observe the specific impact of the interaction effect on the stability of the pile; Establish a network relationship diagram between risk factors, analyze how changes in one or more key factors trigger changes in other factors within the system, explore causal relationships, identify core driving factors and estimate the comprehensive impact on the stability of the accumulation.

6. The data-driven method for preventing and controlling the risk of hydrodynamic erosion deposits according to claim 5, characterized in that: The method for establishing a risk analysis method for hydrodynamic erosion deposits includes formulating a comprehensive evaluation index system, combining quantitative analysis with qualitative evaluation, and grading the impact of different risk factors to determine the potential risk level; A quantitative scoring model is established, the weight of each indicator is combined with the indicator value, and the risk scoring matrix RSM is constructed using the scores of different indicators and the corresponding weights: Each row represents a different monitoring point, and each column is the score of a different risk factor. On the basis of calculating the risk level, the risk of each monitoring point is integrated to obtain the overall risk level L t : L t =argmax(R i ) Among them, the risk level corresponding to the maximum risk score is selected as the final conclusion.

7. The data-driven method for preventing and controlling the risk of hydrodynamic erosion deposits according to claim 6, characterized in that: The deformation prediction and long-term safety evaluation include using numerical simulation methods to conduct time-effectiveness analysis of the accumulation body under different hydrodynamic conditions, and combining historical data and trend analysis to evaluate its stability changes in the future period; When the displacement of the sliding surface of the accumulation body exceeds the set threshold of 10mm / month, the emergency assessment mechanism needs to be activated; when the displacement monitored continuously reaches 20mm, a detailed geological survey is carried out; if the displacement exceeds 30mm, relevant engineering activities need to be suspended and on-site reinforcement is carried out; Set a threshold combination of rainfall and reservoir water level. When rainfall reaches 50 mm / h and the reservoir water level is above 80% of the impoundment, a high-risk warning is activated; If this condition continues for 48 hours, a risk reassessment will be conducted to check the stability of the accumulation; if the rainfall reaches 75 mm / h and the reservoir water level is above the critical value, an evacuation order will be issued immediately; Monitor changes in the shear strength of the rock and soil. When the shear strength is lower than 12 kPa, initiate stability assessment and increase monitoring frequency to once every 12 hours. If the average shear strength below the slope decreases by more than 5%, automatically initiate a prediction model based on machine learning to predict stability risks within the next 48 hours. The long-term safety evaluation collects historical risk score data based on past change trends and uses a model to predict risk scores at K future time points: in, is the predicted future risk score, μ is the constant term, φ is the autoregressive parameter, θ is the moving average parameter, is the risk score at the previous time point, is the prediction error at the previous time point; For each future time point t+k, the risk score and prediction error of the previous moment are used for calculation.

8. A system using the data-driven hydrodynamic erosion accumulation risk prevention and control method according to any one of claims 1 to 7, characterized in that: It includes data collection and processing module, machine learning and analysis module, risk assessment and early warning module, simulation and prediction module, and decision support and feedback module; The data collection and processing module is responsible for collecting hydraulic monitoring data and displacement monitoring data in real time, and cleaning, normalizing and preprocessing the collected data to ensure data quality and consistency; The machine learning and analysis module will use machine learning algorithms to analyze the processed data, screen key factors through feature engineering, build an analysis framework, and identify the main risk factors and interactions that affect the stability of the pile; The risk assessment and early warning module is responsible for conducting risk assessment according to the established risk analysis model, generating a risk scoring matrix, identifying potential risk levels, and providing real-time early warnings in combination with set thresholds; The simulation and prediction module uses numerical simulation methods to perform dynamic time-effectiveness analysis of the accumulation body under different hydrodynamic conditions, including long-term trend analysis and deformation prediction; The decision support and feedback module integrates the outputs of various analysis modules to generate a comprehensive management report, providing a scientific basis for decision makers and guiding on-site emergency response and long-term management strategies. At the same time, a dynamic feedback mechanism is established to review the recorded data and subsequent events in order to continuously optimize risk prevention and control methods.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images