Deep overburden earth-rock dam construction material parameter real-time feedback analysis method and system

Abstract

The present application relates to the technical field of dam material parameter analysis, in particular to a deep overburden earth-rock dam material parameter real-time feedback analysis method and system, the method comprising: obtaining the basic engineering information and dam material initial parameter setting value of the deep overburden earth-rock dam; constructing a three-dimensional data acquisition grid system according to the basic engineering information; based on the constructed three-dimensional data acquisition grid system, respectively collecting stress-strain, pore water pressure and geological structure data, and integrating the collected data to obtain a dam body comprehensive monitoring data set; performing multi-scale feature analysis on the dam body comprehensive monitoring data set and the dam material initial parameter setting value to obtain a key feature sample data set; constructing a dam material parameter dynamic inversion model according to the key feature sample data set; inputting the dam body comprehensive monitoring data set into the dam material parameter dynamic inversion model, and the dam material parameter dynamic inversion model outputs a parameter adjustment strategy.

Description

Technical Field

[0001] This invention relates to the technical field of dam material parameter analysis, and in particular to a real-time feedback analysis method and system for dam material parameters of deep overburden earth-rock dams. Background Technology

[0002] In the field of water conservancy and hydropower engineering, earth-rock dams, as a common dam type, are widely used in water conservancy facility construction around the world due to their advantages such as the availability of local materials, ease of construction, and strong adaptability to foundation deformation. As water conservancy projects advance into areas with more complex geological conditions, an increasing number of earth-rock dams need to be built on foundations with deep overburden layers. These deep overburden layers are typically composed of loose deposits of various origins and properties, characterized by their large thickness, complex structure, and heterogeneous mechanical properties, which presents numerous challenges to the design, construction, and operation management of earth-rock dams.

[0003] Accurately determining the parameters of dam materials is crucial for ensuring the safety and stability of earth-rock dams with deep overburden. These parameters directly affect the stress-strain distribution, seepage characteristics, and overall stability of the dam body. Traditional methods for determining dam material parameters mainly rely on laboratory tests and empirical values; however, these methods have limitations. Laboratory tests cannot fully simulate the complex geological conditions and construction processes on-site, leading to discrepancies between test results and actual conditions. Empirical values, on the other hand, lack specificity and fail to fully consider the unique characteristics of specific projects. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method and system for real-time feedback analysis of dam material parameters in deep overburden soil-rock dams, which can improve the safety and stability of the dam body and reduce errors and uncertainties caused by human intervention.

[0005] In a first aspect, the present invention provides a method for real-time feedback analysis of dam material parameters for earth-rock dams with deep overburden layers, the method comprising:

[0006] Obtain foundation engineering information and initial parameter settings for dam construction materials for earth-rock dams with deep overburden layers; construct a three-dimensional data acquisition grid system based on the foundation engineering information;

[0007] Based on the established three-dimensional data acquisition grid system, stress and strain, pore water pressure and geological structure data were collected respectively, and the collected data were integrated to obtain a comprehensive monitoring dataset of the dam body.

[0008] Multi-scale feature analysis was performed on the comprehensive monitoring dataset of the dam body and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data.

[0009] Based on the key feature sample data set, a dynamic inversion model for dam material parameters is constructed;

[0010] The comprehensive monitoring dataset of the dam body is input into the dynamic inversion model of dam construction material parameters, and the dynamic inversion model of dam construction material parameters outputs parameter adjustment strategies.

[0011] The initial parameter settings of the dam material are preprocessed according to the parameter adjustment strategy to obtain the optimized data set and the real-time parameters of the dam material.

[0012] Furthermore, the basic engineering information includes geological profiles, soil and rock mechanical properties, groundwater conditions, dam design parameters, and construction methods.

[0013] Furthermore, the initial parameter settings for the dam construction material include physical property parameters and mechanical property parameters.

[0014] Furthermore, stress-strain, pore water pressure, and geological structure data are collected from the established three-dimensional data acquisition grid system, and the collected data are integrated to obtain a comprehensive monitoring dataset for the dam body, including:

[0015] At each node of the constructed three-dimensional data acquisition grid system, stress and strain data of various parts of the dam body are acquired through strain sensors.

[0016] Pore ​​water pressure data inside the dam were collected at different depths and locations using pore water pressure gauges.

[0017] Geological structure data were collected using ground-penetrating radar;

[0018] The collected data is transmitted in real time through a data transmission network;

[0019] By integrating stress-strain data, pore water pressure data, and geological structure data, a comprehensive monitoring dataset for the dam body is obtained.

[0020] Furthermore, multi-scale feature analysis was performed on the comprehensive monitoring dataset of the dam body and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data, including:

[0021] The comprehensive monitoring dataset of the dam body is integrated with the initial parameter settings of the dam construction materials to obtain a comprehensive dataset;

[0022] Perform data cleaning and standardization on the comprehensive dataset;

[0023] Analyze the variation trends of stress-strain, pore water pressure, and geological structure parameters of the dam body at different time points, and extract time series features;

[0024] Analyze the spatial characteristics of stress-strain distribution, pore water pressure gradient, and geological structure variation at different locations inside the dam body, and extract the spatial distribution patterns;

[0025] Spectral analysis was performed on the dynamic response of the dam body to vibration and wave, and the characteristics of different frequency components were extracted to obtain the dynamic characteristics of the dam body.

[0026] By integrating time series characteristics, spatial distribution patterns, and dynamic properties of the dam body, a set of key feature sample data is constructed.

[0027] Furthermore, the method for constructing the dynamic inversion model of dam construction material parameters includes:

[0028] Machine learning models were selected as the basic architecture for the dynamic inversion model of dam material parameters; the machine learning models included random forest, support vector machine, neural network, multiple linear regression and nonlinear regression;

[0029] The key feature sample data set is divided into a training set, a validation set, and a test set;

[0030] The model is trained using training set data, enabling it to learn the mapping relationship between key features and dam construction material parameters.

[0031] Cross-validation is performed using a validation set to optimize the model;

[0032] The trained model is tested using test set data to evaluate its performance; the performance of the model is then evaluated based on the test results.

[0033] The trained model is deployed to a real-time feedback analysis system, which receives the comprehensive monitoring dataset of the dam body as input and outputs adjustment strategies for the dam construction material parameters.

[0034] Furthermore, the method for generating the parameter adjustment strategy includes:

[0035] Format conversion of the dam body comprehensive monitoring dataset;

[0036] Load the pre-trained and validated dynamic inversion model of dam material parameters;

[0037] The converted dam body comprehensive monitoring dataset is input into the trained machine learning model, and the model calculates the corresponding predicted values ​​of dam construction material parameters.

[0038] The predicted values ​​of dam construction material parameters are compared with the preset standard values ​​of dam construction material to determine the direction and magnitude of parameter adjustment;

[0039] By summarizing the adjustment directions and magnitudes of various dam construction material parameters, a complete parameter adjustment strategy can be obtained.

[0040] On the other hand, this application also provides a real-time feedback analysis system for dam construction material parameters of deep overburden earth-rock dams, the system comprising:

[0041] The information acquisition module acquires foundation engineering information and initial parameter settings for dam construction materials for deep overburden earth-rock dams; based on the foundation engineering information, it constructs a three-dimensional data acquisition grid system.

[0042] The data acquisition and integration module, based on the constructed three-dimensional data acquisition grid system, collects stress and strain, pore water pressure and geological structure data respectively, and integrates the collected data to obtain a comprehensive monitoring dataset of the dam body;

[0043] The feature parsing module performs multi-scale feature parsing on the dam body comprehensive monitoring dataset and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data.

[0044] The model building module constructs a dynamic inversion model of dam material parameters based on the key feature sample data set.

[0045] The strategy generation module inputs the dam body comprehensive monitoring dataset into the dynamic inversion model of dam material parameters, and the dynamic inversion model of dam material parameters outputs parameter adjustment strategies.

[0046] The parameter optimization module preprocesses the initial parameter settings of the dam material according to the parameter adjustment strategy to obtain the optimized data set and obtain the real-time parameters of the dam material.

[0047] Thirdly, this application provides an electronic device including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any of the methods described above.

[0048] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0049] Compared with existing technologies, the beneficial effects of this invention are as follows: Firstly, the method acquires the foundation engineering information and initial parameter settings of the dam material for a deep overburden earth-rock dam, laying the foundation for subsequent analysis. Based on this, a three-dimensional data acquisition grid system can accurately locate each monitoring point on the dam body, comprehensively collecting stress-strain, pore water pressure, and geological structure data. This overcomes the shortcomings of traditional monitoring methods that only acquire data from the dam surface or local areas, fully reflecting the true internal state of the dam body. The acquired data is integrated, and multi-scale feature analysis technology, combined with wavelet transform, principal component analysis, and other methods, is used to deeply analyze the comprehensive monitoring dataset of the dam body and the initial parameter settings of the dam material, extracting key features and obtaining a key feature sample data set. This allows for deeper data mining and provides strong support for subsequent model construction. Based on the key feature sample data set, appropriate methods such as the finite element method or machine learning algorithms can be flexibly selected to construct a dynamic inversion model of the dam material parameters. Furthermore, historical data can be used to verify the model's accuracy, enabling the model to better adapt to the complexities of different projects and overcoming the lack of specificity in traditional methods.

[0050] By acquiring basic engineering information to construct a three-dimensional data acquisition grid system, it is possible to collect and integrate stress-strain, pore water pressure, and geological structure data from all directions, achieving comprehensive and accurate real-time monitoring of the dam's condition. Multi-scale feature analysis, based on the comprehensive monitoring dataset and initial parameter settings, mines deep features and obtains a set of key feature sample data. The comprehensive monitoring dataset is input into a dynamic inversion model to obtain parameter adjustment strategies, and then the initial parameter settings are preprocessed to obtain real-time parameters, forming a dynamic closed loop from data acquisition, analysis, model calculation to parameter optimization. As time goes by and the project progresses, each link continuously interacts, and parameters can be adjusted in real time according to the actual situation, ensuring that the dam material parameters always match the actual project, effectively improving the safety and stability of earth-rock dams in the design, construction, and operation management process, overcoming the shortcomings of traditional static and lagging methods.

[0051] In summary, the real-time feedback analysis method for dam material parameters in deep overburden soil-rock dams can improve the safety and stability of the dam body and reduce errors and uncertainties caused by human intervention. Attached Figure Description

[0052] Figure 1 This is a flowchart of the present invention;

[0053] Figure 2 This is a flowchart illustrating the construction method of a dynamic inversion model for dam construction material parameters;

[0054] Figure 3 This is a structural diagram of a method and system for real-time feedback analysis of dam material parameters in deep overburden soil-rock dams. Detailed Implementation

[0055] As will be apparent to those skilled in the art from the description of this application, this application can be implemented as a method, apparatus, electronic device, and computer-readable storage medium. Therefore, this application can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software. Furthermore, in some embodiments, this application can also be implemented as a computer program product contained in one or more computer-readable storage media, which includes computer program code.

[0056] The aforementioned computer-readable storage medium may be any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, flash memory, optical fiber, optical disc read-only memory, optical storage devices, magnetic storage devices, or any combination thereof. In this application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0057] The acquisition, storage, use, and processing of data in this application all comply with relevant national laws and regulations.

[0058] This application describes the provided methods, apparatus, and electronic devices using flowcharts and / or block diagrams.

[0059] It should be understood that each block of a flowchart and / or block diagram, as well as combinations of blocks in a flowchart and / or block diagram, can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine that, when executed by a computer or other programmable data processing apparatus, creates means for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0060] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to function in a particular manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction apparatus product that includes the functions / operations specified in the blocks of a flowchart and / or block diagram.

[0061] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus or other device to produce a computer-implemented process, such that the instructions that execute on the computer or other programmable data processing apparatus provide a process for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0062] This application will now be described with reference to the accompanying drawings.

[0063] Example 1: As Figures 1 to 2 As shown, the real-time feedback analysis method for dam material parameters of deep overburden soil-rock dams of the present invention specifically includes the following steps:

[0064] S1. Obtain the foundation engineering information and initial parameter settings of the dam material for the deep overburden earth-rock dam; construct a three-dimensional data acquisition grid system based on the foundation engineering information;

[0065] The basic engineering information includes:

[0066] Geological profile: A detailed depiction of the geological structure of the dam site area, including the layering of the thick overburden, the material composition of each layer, and its thickness;

[0067] Soil and rock mechanical properties: Provides mechanical parameters such as density, compressive strength, shear strength, and permeability coefficient of each soil layer, as well as the mechanical properties of the rock;

[0068] Groundwater hydrological conditions: describe the distribution, water level changes, flow velocity, and flow direction of groundwater, as well as the impact of groundwater on the stability of the dam body;

[0069] Dam design parameters: geometric dimensions such as dam height, dam crest width, and dam base width; selection and proportioning of dam materials, including the type, particle size distribution, and moisture content of the dam material; structural layout of the dam, such as drainage system and seepage prevention system;

[0070] Construction methods: dam filling methods, compaction techniques, number of compaction passes, etc.; contact treatment measures between the dam body and the foundation, such as foundation reinforcement and seepage prevention; quality control standards and testing methods during construction.

[0071] The initial parameter settings for the dam construction material include:

[0072] Physical property parameters: Particle size distribution, determined by sieve analysis, reflects the fineness and particle composition distribution of the dam material, affecting the compactness and permeability of the dam body; Natural density, refers to the mass per unit volume of the dam material in its natural state, used to calculate the self-weight stress of the dam body; Moisture content, the ratio of the mass of water in the dam material to the mass of dry soil, has a significant impact on the compaction performance and strength of the dam material.

[0073] Mechanical property parameters: Shear strength indices, including internal friction angle and cohesion, determined through triaxial compression tests and direct shear tests, determine the dam material's ability to resist shear failure and are key parameters for dam stability analysis; Compression modulus reflects the compressive deformation characteristics of the dam material under pressure and is used to calculate the dam's settlement under its own weight and external loads; Permeability coefficient, determined through permeability tests, characterizes the dam material's ability to allow water to pass through.

[0074] The method for constructing the three-dimensional data acquisition grid system includes:

[0075] Based on the structural characteristics, geological conditions, and potential risk areas of the dam, determine the physical quantities that need to be monitored, such as stress and strain, pore water pressure, and displacement; and determine the requirements for monitoring accuracy, frequency, and duration.

[0076] Inside and around the dam, a monitoring point layout plan should be designed based on the monitoring objectives and needs; the monitoring points should cover key parts and potential risk areas of the dam, such as the dam foundation, dam shoulders, and the interior of the dam.

[0077] Select monitoring instruments and equipment based on the physical quantities and accuracy requirements being monitored; ensure that the monitoring instruments and equipment can operate stably for a long period of time and that the data is accurate and reliable.

[0078] The monitoring instruments and equipment are connected to the data acquisition system via wired or wireless means to form a three-dimensional data acquisition grid system. The data acquisition system should have the ability to acquire data in real time, continuously and remotely, and be able to automatically transmit the data to the data center for storage and analysis.

[0079] In this step, detailed collection of basic engineering information, including geological profiles, soil and rock mechanical properties, groundwater conditions, dam design parameters, and construction methods, as well as the physical and mechanical properties of the dam materials, provides a solid foundation for the subsequent construction of a three-dimensional data acquisition grid system and dam stability analysis. This helps to more accurately assess the safety and stability of the dam, ensuring the quality of earth-rock dam construction. Based on the dam's structural characteristics, geological conditions, and potential risk areas, a reasonable monitoring point layout scheme is designed, and appropriate monitoring instruments and equipment are selected. Real-time monitoring of key physical quantities such as stress, strain, pore water pressure, and displacement of the dam provides reliable data support for dam safety monitoring and early warning. Connecting monitoring instruments and equipment to the data acquisition system via wired or wireless means enables real-time, continuous, and remote data acquisition, improving monitoring efficiency and ensuring data accuracy and reliability. The constructed three-dimensional data acquisition grid system allows for real-time monitoring of changes in various physical quantities of the dam, enabling timely detection and handling of potential risks and hazards. This helps to prevent dam instability and other safety accidents, ensuring the safe operation of the earth-rock dam.

[0080] S2. Based on the established three-dimensional data acquisition grid system, stress and strain, pore water pressure and geological structure data are collected respectively, and the collected data are integrated to obtain a comprehensive monitoring dataset of the dam body.

[0081] Stress and strain data acquisition: High-precision vibrating wire strain gauges or fiber optic strain sensors are installed at each node and key part of the constructed three-dimensional data acquisition grid system. Vibrating wire strain gauges determine the strain value by measuring the change in the vibration frequency of the steel wire, and have the characteristics of high accuracy and good stability, making them suitable for long-term monitoring. Fiber optic strain sensors utilize the optical transmission characteristics of optical fibers, are sensitive to strain, and have strong resistance to electromagnetic interference. The sensors are arranged in a certain direction and angle to comprehensively capture the stress and strain of the dam body under different stress states, thereby obtaining stress and strain data of various parts of the dam body under the action of self-weight, water pressure, construction loads, etc.

[0082] Pore ​​water pressure data acquisition: Data acquisition is carried out using pore water pressure gauges; pore water pressure gauges are installed at different depths and locations within the dam body, especially in areas where abnormal seepage may occur; changes in pore water pressure inside the dam body are monitored in real time.

[0083] Geological structure data acquisition: Geological structure data are collected using equipment such as ground-penetrating radar and acoustic wave detectors; ground-penetrating radar detects the geological structure inside the dam body by emitting high-frequency electromagnetic waves and based on the reflection characteristics of electromagnetic waves in different media; acoustic wave detectors analyze the density, uniformity and integrity of the dam body materials and geological structure by utilizing the propagation speed and attenuation characteristics of sound waves in different media.

[0084] During the data acquisition process, a unified data transmission network is established to transmit data collected by various sensors to the data processing center in real time. A data fusion algorithm is used to integrate the collected stress-strain, pore water pressure, and geological structure data. Weighted average fusion assigns corresponding weights to different data based on their reliability and importance, and then performs weighted average calculation to obtain comprehensive data.

[0085] Stress-strain data, pore water pressure data, and geological structure data are integrated to obtain a comprehensive monitoring dataset for the dam body. The integrated monitoring dataset is then stored in a dedicated database and managed effectively.

[0086] In this step, by installing high-precision strain sensors at various nodes and key parts of the dam body, the stress and strain of the dam body under different stress states can be comprehensively captured. Simultaneously, equipment such as pore water pressure gauges and acoustic detectors are used to monitor changes in pore water pressure and geological structural characteristics within the dam body in real time, ensuring the comprehensiveness and accuracy of the data. This helps to gain a deeper understanding of the dam body's working status and potential risks. A unified data transmission network is established to transmit data collected by various sensors to the data processing center in real time, enabling real-time data updates and dynamic monitoring. This helps to promptly detect anomalies in the dam body and take corresponding measures to address them, thereby improving the dam body's safety and stability. Data fusion algorithms are used to integrate the collected data. Through methods such as weighted average fusion, appropriate weights are assigned according to the reliability and importance of different data points, and then a weighted average calculation is performed to obtain comprehensive data. This improves the efficiency and accuracy of data processing and helps to achieve intelligent data analysis and early warning. This step, by comprehensively, accurately, and in real-time collecting and integrating stress-strain, pore water pressure, and geological structural data of the dam body, provides strong technical support for the safety monitoring and early warning of the dam body, improving its safety and stability.

[0087] S3. Perform multi-scale feature analysis on the comprehensive monitoring dataset of the dam body and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data.

[0088] The comprehensive monitoring dataset of the dam body is integrated with the initial parameter settings of the dam construction materials to form a comprehensive dataset containing multiple types and dimensions of data;

[0089] Check the dataset for outliers, missing values, or inconsistencies. Clean the data using methods such as data interpolation, smoothing, or outlier removal to ensure accuracy and completeness. To eliminate differences in scale between different data points, standardize the data to bring all feature values ​​to the same order of magnitude for easier subsequent analysis.

[0090] The study analyzes the changing trends of parameters such as stress, strain, and pore water pressure of the dam body at different time points, and extracts time series characteristics to reflect the characteristics of the dam body changing over time.

[0091] Using spatial interpolation, geostatistics and other methods, we analyzed the spatial characteristics of stress and strain distribution, pore water pressure gradient and geological structure variation at different locations inside the dam body, and extracted the spatial distribution patterns.

[0092] Spectral analysis is performed on the dynamic response of the dam body, such as vibration and wave, to extract the characteristics of different frequency components and reflect the dynamic properties of the dam body;

[0093] Correlation analysis and principal component analysis (PCA) were used to evaluate the correlation between each feature and the dam material parameters, and to screen out features that are sensitive to changes in dam material parameters and have a high correlation.

[0094] For datasets with high feature dimensionality, feature dimensionality reduction techniques can be used to reduce the number of features and improve computational efficiency.

[0095] The contribution of each feature to the prediction of dam construction material parameters is evaluated using the feature importance assessment method, and the feature set is further optimized.

[0096] Based on the features extracted and optimized in the above steps, a key feature sample data set is constructed. The key feature sample data set contains key features that can comprehensively reflect the stress and strain, seepage characteristics, geological structure, and changes in dam construction material parameters of the dam body, providing basic data for the subsequent construction of a dynamic inversion model of dam construction material parameters.

[0097] In this step, by integrating the comprehensive monitoring dataset of the dam body with the initial parameter settings of the dam construction materials, a comprehensive dataset containing multi-type and multi-dimensional data was formed, providing a comprehensive data foundation for subsequent analysis. Simultaneously, the data cleaning process ensured the accuracy and completeness of the data, eliminating the potential impact of outliers, missing values, or inconsistencies on the analysis results. During the multi-scale feature analysis, not only were time-series features extracted to reflect the characteristics of the dam body changing over time, but spatial distribution patterns were also extracted using spatial interpolation and geostatistics methods, and the dynamic characteristics of the dam body were obtained through spectral analysis. This contributes to a more comprehensive understanding of the dam body's behavioral characteristics. Correlation analysis and principal component analysis (PCA) were used to evaluate the correlation between various features and the dam construction material parameters. Features that are sensitive to and highly correlated with changes in dam material parameters were selected. Furthermore, for datasets with high feature dimensionality, feature dimensionality reduction techniques were employed to effectively reduce the number of features and improve computational efficiency, contributing to the construction of a simpler and more efficient model. The contribution of each feature to the prediction of dam material parameters was scientifically evaluated using feature importance assessment methods in machine learning algorithms, aiding in further optimization of the feature set and improving the model's prediction accuracy. Based on the features extracted and optimized in the above steps, a key feature sample dataset was constructed, containing key features that comprehensively reflect the dam's stress-strain, seepage characteristics, geological structure, and changes in dam material parameters, providing high-quality foundational data for subsequent dynamic inversion models of dam material parameters, and contributing to improved model prediction performance.

[0098] S4. Based on the key feature sample data set, construct a dynamic inversion model for dam material parameters;

[0099] The method for constructing the dynamic inversion model of dam construction material parameters includes:

[0100] Based on the characteristics of deep overburden earth-rock dams and the features of key characteristic sample data sets, a machine learning model was selected as the basic architecture for the dynamic inversion model of dam material parameters. The machine learning model includes random forest, support vector machine, neural network, multiple linear regression, and nonlinear regression. The specific structure and parameters of the model were designed according to the selected model type.

[0101] The key feature sample data set is divided into a training set, a validation set, and a test set; the training set is used for training and learning the model; the validation set is used to adjust the model parameters and select the optimal model; and the test set is used to evaluate the model's performance and generalization ability.

[0102] The model is trained using training set data, enabling it to learn the mapping relationship between key features and dam construction material parameters.

[0103] The model is optimized by adjusting its parameters and algorithms, as well as by using methods such as cross-validation and grid search with a validation set, in order to improve its prediction accuracy and generalization ability.

[0104] The trained model is tested using test set data to evaluate its performance; the model's performance is evaluated based on the test results; if the model's performance is poor, the model structure or parameters need to be readjusted, or even the model type needs to be changed.

[0105] The trained model is deployed to a real-time feedback analysis system so that it can receive the comprehensive monitoring dataset of the dam body as input and output adjustment strategies for the dam material parameters in practical applications.

[0106] In this step, by selecting a suitable machine learning model as the basic architecture of the dynamic inversion model for dam material parameters, and designing the model structure and parameters based on the characteristics of deep overburden earth-rock dams and the features of the key feature sample data set, the accuracy and reliability of the model can be significantly improved. By dividing the key feature sample data set into training, validation, and test sets, and using methods such as cross-validation and grid search to optimize the model, it is ensured that the model will not overfit or underfit during training, while improving the model's prediction accuracy and generalization ability, and reducing errors and uncertainties. Deploying the trained model to a real-time feedback analysis system enables online monitoring and dynamic adjustment of dam material parameters. When the comprehensive monitoring dataset of the dam body is used as input, the system can quickly output adjustment strategies for dam material parameters, thereby responding promptly to changes in the dam body and ensuring the safety and stability of the dam body. By constructing a dynamic inversion model for dam material parameters, rapid prediction and evaluation of dam material performance can be achieved, thereby guiding engineering design and construction. This not only improves engineering efficiency and reduces trial-and-error costs, but also allows for the timely detection of potential safety hazards, enabling corresponding preventive and remedial measures to ensure the safety of earth-rock dam projects.

[0107] S5. Input the comprehensive monitoring dataset of the dam body into the dynamic inversion model of dam construction material parameters, and the dynamic inversion model of dam construction material parameters outputs the parameter adjustment strategy.

[0108] The method for generating the parameter adjustment strategy includes:

[0109] The format of the dam body comprehensive monitoring dataset is converted; for machine learning inversion models, the data is normalized or standardized to meet the numerical range requirements of the model input; in addition, the integrity and accuracy of the data are checked, and any outliers or missing values ​​are handled, with interpolation methods used to supplement missing values ​​and statistical methods used to identify and correct outliers.

[0110] Load the pre-trained and validated dynamic inversion model of dam material parameters; for machine learning models, ensure that the programming language environment required to run the model is running normally, and that the model's weights, parameters, and other files are loaded correctly.

[0111] The preprocessed dam body comprehensive monitoring dataset is input into the trained machine learning model; the model uses the forward propagation algorithm to calculate the corresponding predicted values ​​of dam construction material parameters based on the input data and the learned weight and threshold relationships.

[0112] Based on relevant standards and specifications for earth-rock dam engineering and combined with practical engineering experience, the dam material parameter results output by the model are analyzed and judged. If the permeability coefficient value output by the model exceeds the safe range specified in the specifications, it is necessary to consider adjusting the relevant parameters of the dam material to ensure the seepage stability of the dam body.

[0113] Based on the differences between the model output results and engineering standards, determine the direction and magnitude of parameter adjustments; if the shear strength of a certain area of ​​the dam body calculated by the model is lower than the design requirements, consider appropriately increasing the internal friction angle or cohesion parameter of the dam material; the determination of the adjustment magnitude can refer to engineering experience, sensitivity analysis results, and model error conditions.

[0114] Summarize the adjustment direction and range of various dam construction material parameters to form a complete parameter adjustment strategy; the strategy should clearly indicate the name of the parameter to be adjusted, the value or range of change before and after the adjustment, and the order of adjustment.

[0115] In this step, the quality of the input data is ensured by performing format conversion, normalization, or standardization on the dam body comprehensive monitoring dataset, as well as checking data integrity and accuracy. This helps improve the prediction accuracy of the dynamic inversion model of dam material parameters, as high-quality data is the foundation for effective model learning and accurate prediction. Loading the pre-trained and validated model and ensuring its normal operating environment avoids prediction errors caused by model or environmental issues. This guarantees the reliability and practicality of the parameter adjustment strategy, allowing engineers to make decisions based on the model output. The model calculates the predicted values ​​of dam material parameters through the forward propagation algorithm and combines them with soil and rock analysis. The analysis and judgment based on relevant standards and specifications for dam engineering and practical engineering experience ensured the scientific and rational nature of the parameter adjustment strategy. The direction and magnitude of parameter adjustment were determined based on the differences between the model output and engineering standards. This helps engineers quickly and accurately identify the parameters that need adjustment and rationally determine the adjustment range, thereby ensuring the safety and stability of the dam body. The resulting parameter adjustment strategy clearly defines the names of the parameters to be adjusted, their values ​​or ranges of change before and after adjustment, and the order of adjustment. This makes the strategy highly complete and operable, allowing engineers to operate directly based on the strategy without additional analysis and judgment.

[0116] S6. Preprocess the initial parameter settings of the dam material according to the parameter adjustment strategy to obtain the optimized data set and obtain the real-time parameters of the dam material.

[0117] Based on the parameter adjustment strategy, the initial parameter settings of the dam material are initially adjusted, including adjusting the physical properties, mechanical properties, and other relevant parameters of the dam material.

[0118] Using a three-dimensional data acquisition grid system, key indicators such as stress and strain and pore water pressure of the adjusted dam body are monitored in real time.

[0119] Compare the data before and after the adjustment to assess whether the adjustment achieved the expected results and whether it had a positive impact on the overall stability of the dam.

[0120] Based on the evaluation results, the parameters of the dam material are fine-tuned to further optimize the performance of the dam body; the process of monitoring, evaluating and adjusting is repeated until the optimal parameter configuration is achieved.

[0121] During parameter adjustment, a three-dimensional data acquisition grid system is continuously used to collect stress-strain, pore water pressure, and geological structure data of the dam body; ensuring the real-time nature and accuracy of the data so as to reflect the actual state of the dam body in a timely manner;

[0122] Based on real-time monitoring data, the real-time parameters of the dam material are calculated using a dynamic inversion model of dam material parameters. The calculated real-time parameters are compared with the initial parameter settings to update the dam material parameter database, ensuring the timeliness and accuracy of the parameters.

[0123] Real-time parameters are fed back to engineering management personnel so that they can understand the actual condition of the dam body in a timely manner; based on the feedback results of real-time parameters, the parameters of the dam construction materials are further adjusted and optimized.

[0124] In this step, by continuously utilizing a three-dimensional data acquisition grid system, key indicators such as stress and strain, and pore water pressure of the dam body are monitored in real time, ensuring the real-time nature and accuracy of the data. This allows engineering managers to understand the actual state of the dam body in a timely manner, providing reliable data support for subsequent parameter adjustments. After initial adjustments to the initial parameter settings of the dam material according to the parameter adjustment strategy, fine-tuning of the parameters is carried out through real-time monitoring and evaluation to further optimize the performance of the dam body. This ensures the accuracy and rationality of the dam material parameters, thereby improving the overall stability and safety of the dam body. During the parameter adjustment process, the real-time parameters of the dam material are calculated in real time using a dynamic inversion model of dam material parameters and compared with the initial parameter settings, updating the dam material parameter database in a timely manner. This makes the parameter adjustment process more flexible and efficient, enabling continuous improvement based on the actual situation of the dam body. The acquisition and updating of real-time parameters provide important decision support for engineering managers. Based on the changing trends and anomalies of real-time parameters, timely measures can be taken for risk management and response, thereby avoiding potential safety hazards and accidents. By precisely controlling the dam material parameters, the design and construction of the dam body can be optimized, reducing unnecessary material waste and engineering costs.

[0125] Example 2: Figure 3 As shown, the real-time feedback analysis system for dam construction material parameters of deep overburden soil-rock dams of the present invention specifically includes the following modules;

[0126] The information acquisition module acquires foundation engineering information and initial parameter settings for dam construction materials for deep overburden earth-rock dams; based on the foundation engineering information, it constructs a three-dimensional data acquisition grid system.

[0127] The data acquisition and integration module, based on the constructed three-dimensional data acquisition grid system, collects stress and strain, pore water pressure and geological structure data respectively, and integrates the collected data to obtain a comprehensive monitoring dataset of the dam body;

[0128] The feature parsing module performs multi-scale feature parsing on the dam body comprehensive monitoring dataset and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data.

[0129] The model building module constructs a dynamic inversion model of dam material parameters based on the key feature sample data set.

[0130] The strategy generation module inputs the dam body comprehensive monitoring dataset into the dynamic inversion model of dam material parameters, and the dynamic inversion model of dam material parameters outputs parameter adjustment strategies.

[0131] The parameter optimization module preprocesses the initial parameter settings of the dam material according to the parameter adjustment strategy to obtain the optimized data set and obtain the real-time parameters of the dam material.

[0132] By collecting real-time data on the dam's stress and strain, pore water pressure, and geological structure, the system can quickly reflect the actual working state of the dam. Combined with multi-scale feature analysis and dynamic inversion models, the system can more accurately evaluate the performance of dam materials and output real-time parameters, providing strong support for the safety and stability of the dam.

[0133] The system covers the entire process from information acquisition, data collection and integration, feature analysis, model building to strategy generation and parameter optimization, forming a complete and systematic solution. This enables the system to comprehensively consider multiple factors, including geological conditions, construction environment and dam design parameters, thereby more comprehensively evaluating the performance of dam materials.

[0134] The three-dimensional data acquisition grid system constructed by the system can be flexibly adjusted according to different basic engineering information to adapt to the needs of different geological conditions and dam types; the dynamic inversion model can also be updated and optimized in real time based on real-time monitoring data, thereby ensuring that the system can always accurately reflect the actual working status of the dam.

[0135] The system uses automated and intelligent technologies to achieve real-time feedback and adjustment of dam material parameters, greatly improving work efficiency. At the same time, the system can also automatically preprocess the initial parameters of the dam material according to the parameter adjustment strategy, thereby avoiding the tediousness and uncertainty of manual intervention.

[0136] The system employs advanced technologies and data analysis methods to ensure the accuracy and reliability of the data; through real-time monitoring and dynamic adjustment, the system can promptly detect and address potential safety hazards, thereby ensuring the stability and safety of the dam.

[0137] In summary, the real-time feedback analysis system for dam construction material parameters in deep overburden earth-rock dams offers advantages such as real-time performance, accuracy, comprehensiveness, systematic approach, adaptability, flexibility, efficiency, intelligence, reliability, and stability when addressing problems in the field of water conservancy and hydropower engineering. This enables the system to better adapt to the construction needs of earth-rock dams under complex geological conditions, providing strong guarantees for the safety and stability of the dam body.

[0138] In addition, this application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via the bus. When the computer program is executed by the processor, it implements the various processes of the above-described method embodiment for controlling output data and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0139] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for real-time feedback analysis of dam material parameters for earth-rock dams with deep overburden layers, characterized in that, The method includes: Obtain foundation engineering information and initial parameter settings for dam construction materials for earth-rock dams with deep overburden layers; construct a three-dimensional data acquisition grid system based on the foundation engineering information; Based on the established three-dimensional data acquisition grid system, stress and strain, pore water pressure and geological structure data were collected respectively, and the collected data were integrated to obtain a comprehensive monitoring dataset of the dam body. Multi-scale feature analysis was performed on the comprehensive monitoring dataset of the dam body and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data. Based on the key feature sample data set, a dynamic inversion model for dam material parameters is constructed; The comprehensive monitoring dataset of the dam body is input into the dynamic inversion model of dam construction material parameters, and the dynamic inversion model of dam construction material parameters outputs parameter adjustment strategies. The initial parameter settings of the dam material are preprocessed according to the parameter adjustment strategy to obtain the optimized data set and the real-time parameters of the dam material.

2. The real-time feedback analysis method for dam material parameters of deep overburden earth-rock dams as described in claim 1, characterized in that, The basic engineering information includes geological profiles, soil and rock mechanical properties, groundwater conditions, dam design parameters, and construction methods.

3. The real-time feedback analysis method for dam material parameters of deep overburden earth-rock dams as described in claim 1, characterized in that, The initial parameter settings for the dam construction material include physical property parameters and mechanical property parameters.

4. The method for real-time feedback analysis of dam material parameters for deep overburden earth-rock dams as described in claim 1, characterized in that, Based on the established three-dimensional data acquisition grid system, stress-strain, pore water pressure, and geological structure data were collected separately. The collected data were then integrated to obtain a comprehensive monitoring dataset for the dam body, including: At each node of the constructed three-dimensional data acquisition grid system, stress and strain data of various parts of the dam body are acquired through strain sensors. Pore ​​water pressure data inside the dam were collected at different depths and locations using pore water pressure gauges. Geological structure data were collected using ground-penetrating radar; The collected data is transmitted in real time through a data transmission network; By integrating stress-strain data, pore water pressure data, and geological structure data, a comprehensive monitoring dataset for the dam body is obtained.

5. The method for real-time feedback analysis of dam material parameters for deep overburden earth-rock dams as described in claim 1, characterized in that, Multi-scale feature analysis was performed on the comprehensive monitoring dataset of the dam body and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data, including: The comprehensive monitoring dataset of the dam body is integrated with the initial parameter settings of the dam construction materials to obtain a comprehensive dataset; Perform data cleaning and standardization on the comprehensive dataset; Analyze the variation trends of stress-strain, pore water pressure, and geological structure parameters of the dam body at different time points, and extract time series features; Analyze the spatial characteristics of stress-strain distribution, pore water pressure gradient, and geological structure variation at different locations inside the dam body, and extract the spatial distribution patterns; Spectral analysis was performed on the dynamic response of the dam body to vibration and wave, and the characteristics of different frequency components were extracted to obtain the dynamic characteristics of the dam body. By integrating time series characteristics, spatial distribution patterns, and dynamic properties of the dam body, a set of key feature sample data is constructed.

6. The method for real-time feedback analysis of dam material parameters for deep overburden earth-rock dams as described in claim 1, characterized in that, The method for constructing the dynamic inversion model of dam construction material parameters includes: Machine learning models were selected as the basic architecture for the dynamic inversion model of dam material parameters; the machine learning models included random forest, support vector machine, neural network, multiple linear regression and nonlinear regression; The key feature sample data set is divided into a training set, a validation set, and a test set; The model is trained using training set data, enabling it to learn the mapping relationship between key features and dam construction material parameters. Cross-validation is performed using a validation set to optimize the model; The trained model is tested using test set data to evaluate its performance; the performance of the model is then evaluated based on the test results. The trained model is deployed to a real-time feedback analysis system, which receives the comprehensive monitoring dataset of the dam body as input and outputs adjustment strategies for the dam construction material parameters.

7. The method for real-time feedback analysis of dam material parameters for deep overburden earth-rock dams as described in claim 1, characterized in that, The method for generating the parameter adjustment strategy includes: Format conversion of the dam body comprehensive monitoring dataset; Load the pre-trained and validated dynamic inversion model of dam material parameters; The converted dam body comprehensive monitoring dataset is input into the trained machine learning model, and the model calculates the corresponding predicted values ​​of dam construction material parameters. The predicted values ​​of dam construction material parameters are compared with the preset standard values ​​of dam construction material to determine the direction and magnitude of parameter adjustment; By summarizing the adjustment directions and magnitudes of various dam construction material parameters, a complete parameter adjustment strategy can be obtained.

8. A real-time feedback analysis system for dam construction material parameters of earth-rock dams with deep overburden layers, characterized in that, The system includes: The information acquisition module acquires foundation engineering information and initial parameter settings for dam construction materials for deep overburden earth-rock dams; based on the foundation engineering information, it constructs a three-dimensional data acquisition grid system. The data acquisition and integration module, based on the constructed three-dimensional data acquisition grid system, collects stress and strain, pore water pressure and geological structure data respectively, and integrates the collected data to obtain a comprehensive monitoring dataset of the dam body; The feature parsing module performs multi-scale feature parsing on the dam body comprehensive monitoring dataset and the initial parameter settings of the dam construction materials to obtain a set of key feature sample data. The model building module constructs a dynamic inversion model of dam material parameters based on the key feature sample data set. The strategy generation module inputs the dam body comprehensive monitoring dataset into the dynamic inversion model of dam material parameters, and the dynamic inversion model of dam material parameters outputs parameter adjustment strategies. The parameter optimization module preprocesses the initial parameter settings of the dam material according to the parameter adjustment strategy to obtain the optimized data set and obtain the real-time parameters of the dam material.

9. An electronic device for real-time feedback analysis of dam material parameters for earth-rock dams with deep overburden layers, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that, When the computer program is executed by the processor, it implements the steps of the method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-7.

Images