Hydrological engineering geological safety analysis method based on hydrological situation change

Abstract

The invention relates to the technical field of industrial big data processing, and discloses a hydrological engineering geological safety analysis method based on hydrological situation changes, which comprises the following steps of: acquiring environmental data in real time through Internet of Things monitoring equipment, ecological monitoring equipment, an unmanned aerial vehicle and a remote sensing technology, extracting environmental characteristic parameters by utilizing a big data analysis technology, and analyzing the environmental characteristic parameters; the method comprises the following steps: establishing a hydrogeology digital twinborn model based on a digital twinborn technology to obtain hydrologic situation information, constructing a safety evaluation model in combination with hydrologic engineering geology safety indexes, performing risk evaluation and early warning according to the hydrologic situation information, adjusting the model by using an Adam optimization algorithm based on a TensorFlow framework, and performing early warning on the hydrologic engineering geology safety indexes. And finally integrating the data to generate a hydrological engineering geological safety analysis report. According to the method, the comprehensiveness and accuracy of data acquisition, the dynamic adaptability of the model and the timeliness and accuracy of safety evaluation are effectively improved, and a sufficient scientific basis is provided for geological safety management of hydrological engineering.

Description

Technical Field

[0001] This invention relates to the field of industrial big data processing technology, and in particular to a method for hydrological engineering geological safety analysis based on changes in hydrological conditions. Background Technology

[0002] In traditional hydrogeological safety analysis, data acquisition mainly relies on fixed sensor networks. While these sensors can provide some time-series data, they have limitations in spatial coverage and cannot fully reflect the changes in hydrological conditions under complex terrain and variable geological conditions. For example, in mountainous areas or river basins, due to the complex terrain, the limited installation locations of sensors result in blind spots in data acquisition, making it impossible to accurately capture local hydrological changes. In addition, for some hard-to-reach areas or areas with sparse monitoring points, traditional methods often cannot obtain enough data, which limits the accuracy and reliability of the analysis results.

[0003] In terms of data processing, traditional methods typically involve simple integration of data from different sources, lacking effective feature extraction and fusion techniques. This fails to fully leverage the advantages of multi-source data. For example, traditional methods may simply overlay sensor data and remote sensing data without deeply exploring the intrinsic connections and features between the data, resulting in low-quality input data for the model. This, in turn, affects the accuracy of safety analysis and makes it difficult for the model to handle complex hydrogeological and engineering problems.

[0004] In terms of model building and optimization, traditional methods often use static model parameter settings instead of dynamically adjusting the model's operating parameters and boundary conditions based on real-time data. For example, after initialization, the parameters and boundary conditions of traditional hydrogeological models are usually fixed and cannot be dynamically updated based on real-time monitoring data. This results in the model being unable to respond accurately and promptly to rapid changes in hydrological conditions, reducing the timeliness and accuracy of risk warnings.

[0005] In terms of risk warning and report generation, traditional methods provide relatively simple risk warning information, lacking intuitive visualization and timely notification mechanisms. For example, traditional methods may simply output the risk level without providing detailed information such as the specific location of the risk area, the probability of the risk, and the expected time of occurrence. This makes it difficult for relevant personnel to quickly understand the risk situation and take corresponding measures, affecting the efficiency and effectiveness of hydrogeological safety management. Summary of the Invention

[0006] The purpose of this invention is to propose a hydrogeological safety analysis method based on changes in hydrological conditions in order to solve the problems in the prior art.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: a hydrogeological safety analysis method based on changes in hydrological conditions, the method comprising:

[0008] Step S1: Collect environmental data through literature, IoT monitoring equipment, ecological monitoring equipment, drones, and remote sensing technology;

[0009] Step S2 involves denoising, normalizing, and extracting features from the environmental data to obtain environmental feature parameters.

[0010] Step S3: Establish a hydrogeological digital twin model using digital twin technology, and generate hydrological situation information based on environmental characteristic parameters;

[0011] Step S4: Set hydrogeological safety indicators according to hydrogeological safety requirements and establish a safety evaluation model;

[0012] Step S5: Based on the hydrological situation information, obtain the safety assessment results through the safety assessment model;

[0013] Step S6: Compare and analyze the safety assessment results and environmental characteristic parameters, and optimize and adjust the hydrogeological digital twin model and the safety assessment model.

[0014] Step S7: Generate and save a hydrogeological safety analysis report based on environmental data, environmental characteristic parameters, and safety assessment results.

[0015] Furthermore, step S1 also includes the following sub-steps:

[0016] S1-1, Determine the objectives and scope of hydrogeological safety analysis by reviewing literature and obtain regional environmental information;

[0017] S1-2, Deploy renewable energy-driven IoT monitoring equipment to collect hydrological, meteorological and geological data in real time through the IoT monitoring equipment to obtain IoT monitoring data. The IoT monitoring equipment includes smart water level gauges, smart weather stations and smart geological sensors.

[0018] S1-3, Deploy ecological monitoring equipment to monitor biological activity in groundwater and soil and obtain ecological monitoring data. The ecological monitoring equipment includes infrared sensors, bioelectric signal sensors and biochemical signal sensors.

[0019] S1-4, Conduct terrain and environmental monitoring using drones to acquire drone monitoring data, which includes high-resolution images and videos;

[0020] S1-5, According to monitoring needs, remote sensing data is acquired through an online remote sensing data platform;

[0021] S1-6, The environmental data is transmitted to the data processing center and stored in the big data storage system through the wireless communication module. The environmental data includes regional environmental information, Internet of Things monitoring data, ecological monitoring data, UAV monitoring data and remote sensing data.

[0022] Furthermore, step S2 also includes the following sub-steps:

[0023] S2-1, noise and outliers in environmental data are removed by smoothing and interpolation methods to obtain denoised environmental data;

[0024] S2-2 uses a linear normalization method to convert the denoised environmental data into a unified format, including a unified numerical type, a unified data structure, and a unified timestamp, to obtain preprocessed environmental data;

[0025] S2-3, features are extracted from preprocessed environmental data using big data analytics, and the extracted features are fused to obtain environmental feature parameters, including hydrological parameters, geological parameters, meteorological parameters, and ecological parameters.

[0026] Furthermore, step S3 also includes the following sub-steps:

[0027] S3-1, establish a hydrogeological model through a distributed hydrogeological model framework, and initialize the boundary conditions and operating parameters of the hydrogeological model;

[0028] S3-2, combining hydrogeological models with environmental characteristic parameters, establishes a hydrogeological digital twin model through digital twin technology;

[0029] S3-3, Hydrological situation changes are simulated through a hydrogeological digital twin model to obtain hydrological situation information, which includes water level changes, seepage field distribution, groundwater distribution, and flood inundation range;

[0030] S3-4 uses a Geographic Information System (GIS) platform to visualize hydrological information.

[0031] Furthermore, step S4 also includes the following sub-steps:

[0032] S4-1, Hydrogeological safety indicators are set according to the requirements of hydrogeological engineering, including geological stability indicators, seepage stability indicators, and ecological stability indicators.

[0033] S4-2, combining expert scoring and the analytic hierarchy process (AHP) to determine the weights of each hydrogeological safety indicator, obtaining the weighted fusion indicators, and establishing a safety evaluation indicator system based on the weighted fusion indicators;

[0034] S4-3. Based on the safety evaluation index system, a safety evaluation model is established using the regression algorithm of support vector machine.

[0035] Furthermore, step S5 also includes the following sub-steps:

[0036] S5-1, Input the changes in hydrological situation into the safety assessment model;

[0037] S5-2, Run the safety assessment model and generate safety assessment results based on hydrological situation information. The safety assessment results include geological stability, seepage stability and ecological stability.

[0038] S5-3 uses a Geographic Information System (GIS) platform to visualize the security assessment results.

[0039] Furthermore, step S6 also includes the following sub-steps:

[0040] S6-1, Compare and analyze the safety assessment results and environmental characteristic parameters, and extract the deviation between the safety assessment results and environmental characteristic parameters;

[0041] S6-2, Invite experts in the field to set an evaluation fault tolerance threshold based on their experience and expertise, and determine whether the deviation between the safety evaluation result and the environmental characteristic parameter exceeds the evaluation fault tolerance threshold;

[0042] If the error exceeds the limit, the evaluation result will be recorded as negative, and the cause of the deviation will be analyzed. The cause of the deviation includes unreasonable model structure, inaccurate model parameters, and data acquisition error.

[0043] If it does not exceed the limit, the evaluation result is recorded as positive;

[0044] S6-3 If the recorded evaluation result is negative, then based on the cause of the deviation, the hydrogeological digital twin model and the safety evaluation model are optimized and adjusted using the Adam optimization algorithm based on the TensorFlow framework to obtain the optimized hydrogeological digital twin model and the optimized safety evaluation model.

[0045] If the recorded evaluation result is positive, no optimization or adjustment will be made to the hydrogeological digital twin model and the safety evaluation model.

[0046] Furthermore, step S7 also includes the following sub-steps:

[0047] S7-1 involves structurally stitching together environmental data, environmental characteristic parameters, and safety assessment results to obtain hydrogeological safety information.

[0048] S7-2, The hydrogeological safety information is analyzed using a multi-criteria decision analysis method to obtain the analysis results, and recommendations are made based on the analysis results;

[0049] S7-3, The hydrogeological safety information, analysis results and recommended measures are structured and processed to generate a hydrogeological safety analysis report, which is then saved to a big data storage system.

[0050] S7-4 uses a Geographic Information System (GIS) platform to visualize hydrological engineering geological safety analysis reports and sends these reports to hydrological engineering departments and professionals via SMS and email.

[0051] The beneficial effects of the technical solution provided by this invention include at least the following:

[0052] This invention, by introducing IoT monitoring equipment, drone technology, and ecological monitoring technology, enables real-time and comprehensive collection of hydrological, meteorological, geological, and ecological environment data. It significantly improves the spatial coverage and acquisition efficiency of environmental data, can more comprehensively reflect changes in hydrological conditions, enhances the richness of environmental data, and strengthens the accuracy and reliability of environmental data. In particular, by monitoring the biological activity in groundwater and soil through ecological monitoring equipment, it provides a scientific basis for the protection of the ecological environment.

[0053] This invention establishes a hydrogeological digital twin model using digital twin technology, simulates changes in hydrological conditions in real time, and combines it with a safety evaluation model to dynamically assess the safety of hydrogeological engineering, thereby improving the accuracy and timeliness of the simulation results and significantly enhancing the dynamic adaptability, timeliness, and precision of hydrogeological engineering safety analysis methods.

[0054] This invention visualizes risk warning information through a Geographic Information System (GIS) platform and sends it to relevant personnel in real time via email or SMS, enabling them to quickly understand the risk situation and take corresponding measures. This improves the efficiency and effectiveness of hydrological engineering geological safety management and solves the shortcomings of traditional methods in risk warning and report generation.

[0055] This invention provides a systematic hydrogeological safety analysis process, covering the entire process from data acquisition, preprocessing, model building, safety assessment, risk warning to model optimization and report generation. This makes the entire safety analysis work more standardized, efficient and scientific, providing a comprehensive and systematic solution for hydrogeological safety analysis and promoting the standardization and normalization of hydrogeological safety analysis technology. Attached Figure Description

[0056] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a flowchart of a method provided in an embodiment of the present invention. Detailed Implementation

[0058] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a hydrogeological safety analysis method based on hydrological situation changes proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0060] The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0061] The following description, in conjunction with the accompanying drawings, details a specific scheme for a hydrogeological safety analysis method based on changes in hydrological conditions provided by this invention.

[0062] Please see Figure 1 The diagram illustrates a flowchart of a hydrogeological safety analysis method based on changes in hydrological conditions, according to an embodiment of the present invention. The method includes the following steps:

[0063] Step S1: Collect environmental data through literature, IoT monitoring equipment, ecological monitoring equipment, drones, and remote sensing technology;

[0064] Step S1 further includes the following sub-steps:

[0065] S1-1, Determine the objectives and scope of hydrogeological safety analysis by reviewing literature and obtain regional environmental information;

[0066] S1-2, deploy renewable energy-driven IoT monitoring equipment to collect hydrological, meteorological and geological data in real time through IoT monitoring equipment to obtain IoT monitoring data. IoT monitoring equipment includes smart water level gauges, smart weather stations and smart geological sensors;

[0067] S1-3, Deploy ecological monitoring equipment to monitor biological activity in groundwater and soil and obtain ecological monitoring data. The ecological monitoring equipment includes infrared sensors, bioelectric signal sensors and biochemical signal sensors.

[0068] S1-4 uses drones to monitor terrain and environment, and acquires drone monitoring data, including high-resolution images and videos;

[0069] S1-5, According to monitoring needs, remote sensing data is acquired through an online remote sensing data platform;

[0070] S1-6 transmits environmental data to the data processing center and stores it in the big data storage system via a wireless communication module. The environmental data includes regional environmental information, IoT monitoring data, ecological monitoring data, UAV monitoring data, and remote sensing data.

[0071] It should be noted that the purpose of reviewing the literature is to understand the current research hotspots and cutting-edge technologies in the field of hydrogeology and engineering, to determine the specific objectives of this analysis (such as assessing the flood risk, landslide stability, or groundwater pollution in a specific area), to determine the geographical scope of the analysis based on actual needs, including the boundaries of the study area, key monitoring points, and areas of focus, and to collect background information on the study area (such as topography, geological structure, hydrological conditions, meteorological characteristics, etc.) to provide a basis for subsequent sensor installation and data acquisition.

[0072] Intelligent geological sensors include:

[0073] Hydrological sensors, such as rain gauges, water level gauges, flow meters, and flow meters, are used to measure hydrological parameters such as rainfall, water level, flow velocity, and flow rate.

[0074] Geological sensors, such as permeameters, porosimeters, and compressors, are used to measure geological parameters of soil and rock layers, such as permeability, porosity, and compressibility modulus.

[0075] Meteorological sensors: such as temperature sensors, humidity sensors, wind speed sensors, etc., are used to measure meteorological parameters such as air temperature, humidity, and wind speed.

[0076] The installation location of the sensor can be determined using the following method:

[0077] The installation range of the sensors is determined based on the topographic map of the hydrogeological area (e.g., installing flow meters and water level gauges at river cross-sections, water level gauges at groundwater observation wells, and geological sensors at landslide bodies).

[0078] Adjust the specific installation location of the sensors based on the geological survey report to ensure that the sensors can accurately reflect changes in geological conditions (for example, install permeameters and porosity measuring instruments in areas where there are significant changes in soil and rock layers).

[0079] The installation density and distribution of sensors are determined based on monitoring needs (e.g., increasing the installation density of rain gauges and water level gauges in flood-prone areas, and increasing the number of geological sensors in areas with high risk of geological disasters).

[0080] Infrared sensors convert infrared radiation into electrical signals through thermal elements (such as thermocouples and thermistors), and then calculate the temperature of the object being measured. They are mainly used to monitor temperature changes in soil and groundwater, as well as the thermal radiation signals of organisms. Temperature changes can reflect the heat flow of soil and groundwater, while the thermal radiation signals of organisms can be used to detect biological activities.

[0081] Bioelectric signal sensors capture weak electrical signals generated by organisms through electrodes, and are used to monitor the electrophysiological activities of organisms, such as electrocardiogram (ECG), electroencephalogram (EEG), and electromyography (EMG). In soil and groundwater, these sensors can detect the electrophysiological activities of microorganisms and plant roots, providing data support for ecological risk assessment.

[0082] Biochemical signal sensors are used to monitor chemical substances produced by organisms, such as enzyme activity, metabolites, and biomarkers. These sensors can detect biochemical changes in soil and groundwater, reflecting the metabolic activities and health status of organisms.

[0083] Biochemical signal sensors are typically based on biorecognition elements to detect specific chemical substances (such as enzymes and antibodies). These sensors can convert chemical signals into electrical signals, which can then be detected by electrochemical methods (such as potentiometry and electrometry) or optical methods (such as fluorescence and colorimetry).

[0084] Remote sensing technology can cover large areas and provide macro-level environmental information, especially in areas that are difficult to reach or where monitoring points are sparse. Choosing a suitable online remote sensing data platform can acquire high-resolution images and data, which helps to analyze changes in topography, geological structure and hydrological conditions in greater detail. Combined with data from ground sensor networks, it can provide more comprehensive environmental information and enhance the accuracy and reliability of the analysis.

[0085] Online remote sensing data platforms typically include:

[0086] Optical remote sensing satellites, such as Landsat and Sentinel-2, are used to acquire information such as topography, land use, and vegetation cover.

[0087] Meteorological satellites, such as Fengyun satellites and NOAA satellites, are used to acquire meteorological data, including cloud images, temperature distribution, precipitation distribution, etc.

[0088] Radar remote sensing satellites, such as Sentinel-1, are used to acquire information such as surface deformation and soil moisture. Radar data can operate in cloudy and foggy weather and provide continuous monitoring data.

[0089] Remote sensing data acquisition methods include:

[0090] By building or utilizing existing satellite data receiving stations, data can be directly received from satellites. This method can obtain the latest raw data, but it requires a high level of technical and equipment investment.

[0091] Accessing processed and calibrated remote sensing data through online remote sensing data platforms (such as Google Earth Engine, Copernicus Open Access Hub, etc.) provides abundant data resources and convenient data acquisition methods, making it suitable for most users.

[0092] Step S2 involves denoising, normalizing, and extracting features from the environmental data to obtain environmental feature parameters.

[0093] Step S2 further includes the following sub-steps:

[0094] S2-1, noise and outliers in environmental data are removed by smoothing and interpolation methods to obtain denoised environmental data;

[0095] S2-2 uses a linear normalization method to convert the denoised environmental data into a unified format, including a unified numerical type, a unified data structure, and a unified timestamp, to obtain preprocessed environmental data;

[0096] S2-3 uses big data analytics to extract features from preprocessed environmental data and fuses these features to obtain environmental characteristic parameters, including hydrological, geological, meteorological, and ecological parameters.

[0097] It should be noted that noise reduction is a crucial step in environmental data processing to ensure data quality, including:

[0098] Outlier identification: Using machine learning algorithms (such as isolated forests and local outliers) to automatically identify outliers in data. These algorithms can handle complex data distributions and are suitable for large-scale datasets.

[0099] Noise handling: For noise in time series data, smoothing algorithms (such as moving average and median filtering) are used for processing;

[0100] For missing data points or outliers, interpolation methods (such as linear interpolation and spline interpolation) are used to fill in the missing values. Linear interpolation fills in the missing values ​​by calculating the linear relationship between adjacent data points, while spline interpolation fills in the missing values ​​by fitting a polynomial curve.

[0101] For obviously erroneous outliers, delete these data points directly to avoid misleading subsequent analysis.

[0102] The above methods can effectively identify and process outliers and noise in environmental data, resulting in high-quality denoised data.

[0103] In environmental data processing, normalization is a key step in ensuring data consistency and comparability, including:

[0104] Consistent Numerical Data Types: Ensure that all numerical data is formatted consistently (e.g., unify all data as floating-point numbers or integers). For non-numerical data, such as percentages or proportions, convert them to decimal form.

[0105] Timestamp unification: Convert all timestamp data to a unified format (such as the ISO 8601 standard format) to ensure that the time zones of the timestamps are consistent, and perform time zone conversion when necessary;

[0106] Unified data structure:

[0107] The data is organized into a tabular format, with each row representing an observation point and each column representing a variable. This structure facilitates data storage, transmission, and analysis. For data that needs to be transmitted, it is converted to JSON format for easier storage and transmission. JSON format has good readability and compatibility, and is suitable for various programming languages ​​and data processing tools.

[0108] The above methods can convert denoised data into a unified format to obtain preprocessed data, providing a consistent and comparable data foundation for subsequent data analysis and model training.

[0109] In environmental data processing, feature extraction and fusion are key steps in improving model performance, including:

[0110] Feature extraction: Select features related to hydrogeological safety analysis from the preprocessed data, such as rainfall, water level, flow velocity, and flow rate (hydrological parameters), permeability coefficient, porosity, and compression modulus (geological parameters), and temperature, humidity, and wind speed (meteorological parameters). Use big data analysis tools (such as Python's Pandas library and Scikit-learn library) to scale, encode, and directly combine the features. Then, use big data algorithms (such as principal component analysis PCA and feature selection algorithms) to extract key features, reduce data dimensionality, and improve model training efficiency.

[0111] Feature fusion: fusing data from different sources (such as sensor data and remote sensing data) to form a complete dataset (for example, fusing water level data collected by sensors and topographic data from remote sensing images to more comprehensively reflect changes in hydrological conditions).

[0112] Feature fusion methods: The extracted features are fused using feature fusion methods (such as weighted averaging and feature concatenation) to generate comprehensive environmental feature parameters. The weighted averaging method assigns different weights according to the importance of the features, while the feature concatenation method combines multiple features into a high-dimensional feature vector.

[0113] The effectiveness of the fused feature parameters is verified by methods such as cross-validation to ensure that the feature parameters can accurately reflect environmental changes and provide reliable data support for subsequent security analysis and model training.

[0114] Using the above methods, key features can be extracted and fused from preprocessed data to obtain environmental characteristic parameters, providing high-quality data input for hydrogeological safety analysis.

[0115] Step S3: Establish a hydrogeological digital twin model using digital twin technology, and generate hydrological situation information based on environmental characteristic parameters;

[0116] Step S3 further includes the following sub-steps:

[0117] S3-1, establish a hydrogeological model through a distributed hydrogeological model framework, and initialize the boundary conditions and operating parameters of the hydrogeological model;

[0118] S3-2, combining hydrogeological models with environmental characteristic parameters, establishes a hydrogeological digital twin model through digital twin technology;

[0119] S3-3 uses a hydrogeological digital twin model to simulate changes in hydrological conditions and obtain hydrological information, including water level changes, seepage field distribution, groundwater distribution, and flood inundation range.

[0120] S3-4 uses a Geographic Information System (GIS) platform to visualize hydrological information.

[0121] It should be noted that choosing a suitable distributed hydrological model framework (such as SWAT or MIKE SHE) is important. These models can take into account various factors such as topography, soil, and vegetation to simulate the spatial distribution and temporal variation of hydrological processes.

[0122] Distributed hydrological models, by dividing the study area into multiple sub-regions and simulating the hydrological processes in each sub-region separately, can more accurately reflect changes in hydrological conditions and are suitable for areas with complex terrain and variable geological conditions.

[0123] Model initialization methods include:

[0124] The initial conditions for setting up a hydrogeological model include initial water level, initial soil moisture content, initial groundwater level, etc. These initial conditions are usually determined based on historical data or field measurement data.

[0125] Set the boundary conditions for the model, including precipitation boundary, river boundary, groundwater boundary, etc. The precipitation boundary is set based on meteorological data, the river boundary is set based on hydrological data, and the groundwater boundary is set based on geological data.

[0126] Model parameters are set based on environmental characteristic parameters, such as soil permeability coefficient, porosity, evaporation rate, and vegetation cover. These environmental characteristic parameters can be determined through field measurements, literature review, or empirical formulas.

[0127] Cross-validation methods include:

[0128] The collected environmental data is divided into training and validation sets. Common division methods include simple random division and time series division (such as division by year or season) to ensure that the data distribution of the training and validation sets is representative.

[0129] The hydrogeological model was validated using k-fold cross-validation. The training set was further divided into k subsets. Each time, k-1 subsets were used for model training, and the remaining 1 subset was used for validation. This process was repeated k times to ensure that each subset was used as a validation set once.

[0130] Choosing appropriate evaluation metrics to assess model performance is crucial. Commonly used metrics include mean squared error (MSE), root mean squared error (RMSE), and coefficient of determination (R²). 2 ) etc., to quantify the prediction error and goodness of fit of the model.

[0131] Based on the model validation results, appropriate big data machine learning algorithms are selected to train and optimize the hydrogeological model. Commonly used algorithms include support vector machine (SVM), neural network (NN), random forest (RF), etc. The selected machine learning algorithm is used to train the model. By adjusting the algorithm parameters (such as learning rate, regularization parameters, etc.), the performance of the model is optimized, and the prediction accuracy and generalization ability of the model are improved.

[0132] Hydrological situation information includes:

[0133] Water level change: Generates water level change curves to simulate the changes in river, lake, and groundwater levels over time, and is used to analyze the dynamic changes in water levels;

[0134] Seepage field distribution: Generates a seepage field distribution map, showing the direction and velocity of groundwater flow, which is used to assess the impact of groundwater seepage on geological stability;

[0135] Groundwater distribution: Generates a groundwater distribution map, showing the groundwater level and distribution range, which is used to assess groundwater reserves and dynamic changes;

[0136] Flood inundation extent: Generates a flood inundation extent map, showing the areas that may be flooded, for use in flood risk assessment and emergency management.

[0137] GIS platforms, such as ArcGIS and QGIS, provide powerful data processing and visualization capabilities.

[0138] Data import: Convert the simulation results of the hydrogeological model into a format supported by the GIS platform, such as Shapefile, GeoTIFF, etc., and import it into the selected GIS platform.

[0139] Visual presentation:

[0140] Map drawing: Using the drawing tools of the GIS platform, the changes in hydrological conditions are displayed intuitively through visualization methods such as color coding and contour lines;

[0141] Dynamic display: For time series data (such as water level changes), generate dynamic maps to show the process of water level changes over time, enhancing the dynamism and interactivity of the results;

[0142] Interpretation of Results: The simulation results are explained and illustrated using the annotation and commentary functions of the GIS platform, helping users to better understand the meaning and application value of the results.

[0143] Step S4: Set hydrogeological safety indicators according to hydrogeological safety requirements and establish a safety evaluation model;

[0144] Step S4 further includes the following sub-steps:

[0145] S4-1, Hydrogeological safety indicators are set according to the requirements of hydrogeological engineering. Hydrogeological safety indicators include geological stability indicators, seepage stability indicators, and ecological stability indicators.

[0146] S4-2, combining expert scoring and the analytic hierarchy process (AHP) to determine the weights of each hydrogeological safety indicator, obtaining the weighted fusion indicators, and establishing a safety evaluation indicator system based on the weighted fusion indicators;

[0147] S4-3. Based on the safety evaluation index system, a safety evaluation model is established using the regression algorithm of support vector machine.

[0148] It should be noted that geological stability indicators include:

[0149] Slope stability coefficient: The stability of a slope under different working conditions is evaluated by calculating the slope stability coefficient (such as the safety factor FoS).

[0150] Compression deformation of soil and rock layers: By measuring the changes in the compression modulus and porosity of soil and rock layers, the compression deformation of soil and rock layers under hydrological changes (such as changes in groundwater level) is assessed, and potential ground subsidence risks are predicted.

[0151] Historical records of geological disasters: Statistics on the frequency and scale of historical geological disasters (such as landslides and debris flows) in the region, serving as a reference for assessing geological stability.

[0152] Seepage stability indicators include:

[0153] Seepage pressure: By monitoring the groundwater level and seepage velocity, the seepage pressure generated by groundwater seepage on the soil and rock mass is calculated, and the impact of seepage on slope and foundation stability is assessed.

[0154] Seepage stability coefficient: The seepage stability coefficient is calculated by analyzing factors such as seepage path and hydraulic gradient to assess the impact of groundwater seepage on engineering geological structures;

[0155] Permeability coefficient variation: Monitor the changes in the permeability coefficient of soil and rock layers to assess the impact of groundwater flow on geological structures.

[0156] Ecological stability indicators include:

[0157] Groundwater quality changes: By monitoring the chemical composition of groundwater (such as pH value, total dissolved solids (TDS), heavy metal content, etc.), changes in groundwater quality are assessed.

[0158] Soil erosion level: The impact of hydrological activities on soil erosion is assessed by monitoring indicators such as soil loss and vegetation cover.

[0159] Ecological impact assessment coefficient: comprehensively assesses the impact of hydrogeological and engineering activities on the ecological environment, including the impact on animal and plant habitats and aquatic ecosystems.

[0160] The expert scoring method is a weighting method based on expert experience and subjective judgment. It determines the weight of each indicator by inviting experts in the field to score the importance of each model indicator. Its advantages are that it is simple and easy to implement, can make full use of the experience and knowledge of experts, and is suitable for situations where there are few indicators and experts have clear judgments on the indicators.

[0161] The Analytic Hierarchy Process (AHP) is a systems analysis method that decomposes complex problems into multiple levels and factors by establishing a hierarchical model. It determines the relative importance of each factor through pairwise comparisons, thereby determining the weight of each indicator. The implementation steps include:

[0162] The problem of hydrogeological safety assessment is decomposed into three layers: the target layer, the criterion layer, and the indicator layer. The target layer is the overall objective (e.g., hydrogeological safety assessment), the criterion layer is the main evaluation aspects (e.g., geological stability, seepage stability, environmental ecology), and the indicator layer is the specific evaluation indicators (e.g., slope stability coefficient, seepage pressure, groundwater quality changes, etc.).

[0163] By using a pairwise comparison method, a judgment matrix is ​​constructed for the criterion layer and the indicator layer. The consistency of the judgment matrix is ​​checked. The weight of each factor is calculated by the eigenvalue method or the geometric mean method. The weights of the criterion layer and the indicator layer are combined to obtain the final weight of each indicator.

[0164] The advantage of this method is its strong systematic nature, its ability to handle complex problems with multiple criteria and levels, and its high scientific rigor and objectivity.

[0165] Support Vector Machine (SVM) is a machine learning algorithm based on statistical learning theory. It is suitable for classification and regression problems of high-dimensional data and has good generalization ability. SVM is selected as the construction method of safety evaluation model. Safety evaluation indicators and environmental feature parameters are used as input features, and historical risk events or risk levels labeled by experts are used as output labels to establish a safety evaluation model.

[0166] External datasets related to hydrogeological safety are collected. These datasets can come from monitoring data, historical disaster records, literature, etc. in other regions. The collected external data are preprocessed, and the external datasets are merged with the internal datasets. The merged dataset is then used to retrain the safety assessment model to further improve the model's performance.

[0167] Step S5: Based on the hydrological situation information, obtain the safety assessment results through the safety assessment model;

[0168] Step S5 further includes the following sub-steps:

[0169] S5-1, Input the changes in hydrological situation into the safety assessment model;

[0170] S5-2, Run the safety assessment model and generate safety assessment results based on hydrological situation information. The safety assessment results include geological stability, seepage stability and ecological stability.

[0171] S5-3 uses a Geographic Information System (GIS) platform to visualize the security assessment results.

[0172] It should be noted that the safety assessment results can be used to comprehensively evaluate the geological safety status of hydrological engineering projects, guide the design and management of hydrological projects, and visualize the safety assessment results through a geographic information system (GIS) platform. This can intuitively show the spatial distribution of the geological safety status of hydrological engineering projects and provide decision support for the design and management of hydrological projects.

[0173] Step S6: Compare and analyze the safety assessment results and environmental characteristic parameters, and optimize and adjust the hydrogeological digital twin model and the safety assessment model.

[0174] Step S6 further includes the following sub-steps:

[0175] S6-1, Compare and analyze the safety assessment results and environmental characteristic parameters, and extract the deviation between the safety assessment results and environmental characteristic parameters;

[0176] S6-2, invite experts in the field to set evaluation tolerance thresholds based on their experience and expertise, and determine whether the deviation between the safety evaluation results and environmental characteristic parameters exceeds the evaluation tolerance thresholds;

[0177] If the error exceeds the limit, the evaluation result will be recorded as negative, and the reasons for the deviation will be analyzed. The reasons for the deviation include unreasonable model structure, inaccurate model parameters, and data acquisition error.

[0178] If it does not exceed the limit, the evaluation result is recorded as positive;

[0179] S6-3 If the recorded evaluation result is negative, then based on the cause of the deviation, the hydrogeological digital twin model and the safety evaluation model are optimized and adjusted using the Adam optimization algorithm based on the TensorFlow framework to obtain the optimized hydrogeological digital twin model and the optimized safety evaluation model.

[0180] If the recorded evaluation result is positive, no optimization or adjustment will be made to the hydrogeological digital twin model and the safety evaluation model.

[0181] It should be noted that an unreasonable model structure means that the model may not have considered certain important hydrogeological processes, or the model's boundary conditions may be inaccurate. Inaccurate model parameter settings may lead to discrepancies between the model output and actual data. Data acquisition errors may be caused by sensor malfunctions, data transmission errors, or improper data processing. The presence of data acquisition errors can be determined by checking the sensor calibration records, data transmission logs, and data preprocessing steps.

[0182] Methods for analyzing the causes of deviations include: employing causal analysis methods (such as fishbone diagram analysis), combining model operation records, data collection logs, and field investigation results to systematically identify and analyze the causes of deviations. When necessary, consult with domain experts to obtain professional judgments on the causes of deviations. Expert experience and knowledge can help quickly pinpoint the root cause of the problem.

[0183] TensorFlow is an open-source machine learning framework widely used for various machine learning tasks. TensorFlow supports multiple programming languages, such as Python, C++, and Java. It also provides a rich API to facilitate developers in building, training, and deploying machine learning models. TensorFlow can run on multiple platforms, including CPUs, GPUs, and TPUs (Tensor Processing Units), as well as distributed computing environments, which enables it to handle a wide range of tasks, from small-scale experiments to large-scale production deployments.

[0184] Adam optimization (Adaptive Moment Estimation) is a deep learning optimization algorithm with an adaptive learning rate, which is particularly suitable for optimization problems with non-stationary objectives or large-scale data.

[0185] Step S7: Generate and save a hydrogeological safety analysis report based on environmental data, environmental characteristic parameters, and safety assessment results;

[0186] Step S7 further includes the following sub-steps:

[0187] S7-1 involves structurally stitching together environmental data, environmental characteristic parameters, and safety assessment results to obtain hydrogeological safety information.

[0188] S7-2, analyzes hydrogeological safety information through multi-criteria decision analysis, obtains analysis results, and formulates recommended measures based on the analysis results;

[0189] S7-3, performs structured processing on hydrogeological safety information, analysis results and recommended measures, generates a hydrogeological safety analysis report, and saves the hydrogeological safety analysis report to a big data storage system;

[0190] S7-4 uses a Geographic Information System (GIS) platform to visualize hydrological engineering geological safety analysis reports and sends these reports to hydrological engineering departments and professionals via SMS and email.

[0191] It should be noted that the hydrogeological safety analysis report includes:

[0192] Changes in hydrological conditions: Describe in detail the changes in hydrological parameters such as water level, rainfall, flow velocity, and discharge, and provide data charts to enhance intuitiveness;

[0193] Geological stability analysis: Analyze slope stability, soil and rock layer compression deformation, and groundwater level changes to assess the stability of the geological structure;

[0194] Safety assessment results: Display the output of the safety assessment model, including key information such as risk level, risk area, and risk probability;

[0195] Risk warning information: List the specific details of the risk warning, such as the risk level, the risk area, and the expected time of the risk occurrence;

[0196] Recommended measures: Based on the analysis results, propose targeted measures, such as strengthening monitoring, taking engineering measures, and developing emergency response plans;

[0197] Appendix: Includes supplementary information such as data sources, calculation formulas, references, and a table and chart index.

[0198] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for hydrogeological safety analysis based on changes in hydrological conditions, characterized in that, The method includes: Step S1: Collect environmental data through literature, IoT monitoring equipment, ecological monitoring equipment, drones, and remote sensing technology; Step S2 involves denoising, normalizing, and extracting features from the environmental data to obtain environmental feature parameters. Step S3: Establish a hydrogeological digital twin model using digital twin technology, and generate hydrological situation information based on environmental characteristic parameters; Step S4: Set hydrogeological safety indicators according to hydrogeological safety requirements and establish a safety evaluation model; Step S5: Based on the hydrological situation information, obtain the safety assessment results through the safety assessment model; Step S6: Compare and analyze the safety assessment results and environmental characteristic parameters, and optimize and adjust the hydrogeological digital twin model and the safety assessment model. Step S7: Generate and save a hydrogeological safety analysis report based on environmental data, environmental characteristic parameters, and safety assessment results.

2. The hydrogeological safety analysis method based on hydrological situation changes according to claim 1, characterized in that: Step S1 further includes the following sub-steps: S1-1, Determine the objectives and scope of hydrogeological safety analysis by reviewing literature and obtain regional environmental information; S1-2, Deploy renewable energy-driven IoT monitoring equipment to collect hydrological, meteorological and geological data in real time through the IoT monitoring equipment to obtain IoT monitoring data. The IoT monitoring equipment includes smart water level gauges, smart weather stations and smart geological sensors. S1-3, Deploy ecological monitoring equipment to monitor biological activity in groundwater and soil and obtain ecological monitoring data. The ecological monitoring equipment includes infrared sensors, bioelectric signal sensors and biochemical signal sensors. S1-4, Use drones to monitor terrain and environment, and acquire drone monitoring data, which includes high-resolution images and videos; S1-5, Based on monitoring needs, remote sensing data is acquired through an online remote sensing data platform; S1-6, The environmental data is transmitted to the data processing center and stored in the big data storage system through the wireless communication module. The environmental data includes regional environmental information, Internet of Things monitoring data, ecological monitoring data, UAV monitoring data and remote sensing data.

3. The hydrogeological safety analysis method based on hydrological situation changes according to claim 1, characterized in that: Step S2 further includes the following sub-steps: S2-1, noise and outliers in environmental data are removed by smoothing and interpolation methods to obtain denoised environmental data; S2-2 uses a linear normalization method to convert the denoised environmental data into a unified format, including a unified numerical type, a unified data structure, and a unified timestamp, to obtain preprocessed environmental data; S2-3, features are extracted from preprocessed environmental data using big data analytics, and the extracted features are fused to obtain environmental feature parameters, including hydrological parameters, geological parameters, meteorological parameters, and ecological parameters.

4. The hydrogeological safety analysis method based on hydrological situation changes according to claim 1, characterized in that: Step S3 further includes the following sub-steps: S3-1, establish a hydrogeological model through a distributed hydrogeological model framework, and initialize the boundary conditions and operating parameters of the hydrogeological model; S3-2, combining hydrogeological models with environmental characteristic parameters, establishes a hydrogeological digital twin model through digital twin technology; S3-3, Hydrological situation changes are simulated through a hydrogeological digital twin model to obtain hydrological situation information, which includes water level changes, seepage field distribution, groundwater distribution, and flood inundation range; S3-4 uses a Geographic Information System (GIS) platform to visualize hydrological information.

5. The hydrogeological safety analysis method based on hydrological situation changes according to claim 1, characterized in that: Step S4 further includes the following sub-steps: S4-1, Hydrogeological safety indicators are set according to the requirements of hydrogeological engineering, including geological stability indicators, seepage stability indicators, and ecological stability indicators. S4-2, combining expert scoring and the analytic hierarchy process (AHP) to determine the weights of each hydrogeological safety indicator, obtaining the weighted fusion indicators, and establishing a safety evaluation indicator system based on the weighted fusion indicators; S4-3. Based on the safety evaluation index system, a safety evaluation model is established using the regression algorithm of support vector machine.

6. The hydrogeological safety analysis method based on hydrological situation changes according to claim 1, characterized in that: Step S5 further includes the following sub-steps: S5-1, Input the changes in hydrological situation into the safety assessment model; S5-2, Run the safety assessment model and generate safety assessment results based on hydrological situation information. The safety assessment results include geological stability, seepage stability and ecological stability. S5-3 uses a Geographic Information System (GIS) platform to visualize the security assessment results.

7. The hydrogeological safety analysis method based on hydrological situation changes according to claim 1, characterized in that: Step S6 further includes the following sub-steps: S6-1, Compare and analyze the safety assessment results and environmental characteristic parameters, and extract the deviation between the safety assessment results and environmental characteristic parameters; S6-2, Invite experts in the field to set an evaluation fault tolerance threshold based on their experience and expertise, and determine whether the deviation between the safety evaluation result and the environmental characteristic parameter exceeds the evaluation fault tolerance threshold; If the error exceeds the limit, the evaluation result will be recorded as negative, and the cause of the deviation will be analyzed. The cause of the deviation includes unreasonable model structure, inaccurate model parameters, and data acquisition error. If it does not exceed the limit, the evaluation result is recorded as positive; S6-3 If the recorded evaluation result is negative, then based on the cause of the deviation, the hydrogeological digital twin model and the safety evaluation model are optimized and adjusted using the Adam optimization algorithm based on the TensorFlow framework to obtain the optimized hydrogeological digital twin model and the optimized safety evaluation model. If the recorded evaluation result is positive, no optimization or adjustment will be made to the hydrogeological digital twin model and the safety evaluation model.

8. The hydrogeological safety analysis method based on hydrological situation changes according to claim 1, characterized in that: Step S7 further includes the following sub-steps: S7-1 involves structurally stitching together environmental data, environmental characteristic parameters, and safety assessment results to obtain hydrogeological safety information. S7-2, The hydrogeological safety information is analyzed using a multi-criteria decision analysis method to obtain the analysis results, and recommendations are made based on the analysis results; S7-3, The hydrogeological safety information, analysis results and recommended measures are structured and processed to generate a hydrogeological safety analysis report, which is then saved to a big data storage system. S7-4 uses a Geographic Information System (GIS) platform to visualize hydrological engineering geological safety analysis reports and sends these reports to hydrological engineering departments and professionals via SMS and email.

Images