Physical deposition simulation experiment design planning method based on artificial intelligence

Abstract

The invention discloses a physical deposition simulation experiment design planning method based on artificial intelligence, and particularly relates to the field of physical experiments, and the method comprises the following steps: S100, collecting a large amount of known actual data of ancient lake basins and modern lake basins; step S200, classifying and labeling the collected data, and establishing a lake basin parameter database; s300, establishing a lake basin parameter prediction model by taking the geographical environment and climate conditions of the lake basin as input characteristics and taking various parameters of the lake basin as output targets; step S400, training the lake basin parameter prediction model by using the collected lake basin parameter database; step S500, in combination with a similarity principle, predicting an initial value of lake basin data suitable for laboratory setting; and step S600, carrying out experiment design, and carrying out a simulation experiment. According to the method, the reasonable range of all parameters can be automatically calculated, the time for manually setting step by step is shortened, and the accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of physical experiments, and more specifically, to a method for designing and planning a physical deposition simulation experiment based on artificial intelligence. Background Art

[0002] Sedimentation simulation is an important experimental means and technical method in the theoretical research of sedimentology, which can be divided into numerical simulation and physical simulation. Physical simulation is an indoor simulation of the physical process of sediment, and by simulating the sedimentation conditions at that time, the sedimentation process of natural sediments is restored in the laboratory. The initial physical simulation experiments were mostly applied to the research of hydrology and fluvial geomorphology, and only in the past 20 years has the focus been on simulating the formation process and evolution law of lake basin sedimentary sand bodies.

[0003] The sediment physical simulation technology is divided into 4 development stages: ① The primary stage mainly focuses on phenomenon observation and description. The representative achievements of this stage include the systematic research reports on flume experiments by Simmons et al.; ② The rapid development stage mainly focuses on the study of bed forms. In this stage, the graphs of velocity - grain size - water depth, the physical methods reflected by the research of fluid mechanics and loose boundary hydraulics, and the study of fluid dynamics at river junctions and its control over sediment transport and bed forms have laid a solid foundation for sedimentation simulation; ③ From the 1980s to the 1990s of the 20th century, the lake basin sand body simulation stage mainly focuses on the study of the sand body formation process. Several large - scale laboratories suitable for sand body simulation were established in this stage; ④ From the 21st century to the present, the semi - quantitative - quantitative simulation stage. With the rapid development and wide application of computer technology, the characteristics of this stage are mainly reflected in the close combination of sediment physical simulation and numerical simulation, and the close combination of sediment physical simulation and high - precision photography and measurement technologies, enabling the detailed recording of the sedimentation process and the fine description and characterization of the temporal - spatial distribution characteristics of sediment bodies to be realized.

[0004] However, when a paleo - lake basin or a modern lake basin is set up in the laboratory according to a certain scale, parameters such as the range area, sediment grain size, lake level change range, river width, and depth are set according to the similarity principle. This process often requires manual calculation and multiple experiments, which is rather cumbersome and has poor accuracy.

[0005] To solve the above problems, a technical solution is provided now. Summary of the Invention

[0006] To overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method for establishing a lake basin parameter database. Taking the geographical environment and climate conditions of the lake basin as input features and various parameters of the lake basin as output targets, a lake basin parameter prediction model is established. The collected lake basin parameter database is used to train the lake basin parameter prediction model. Combining the similarity principle, the initial value of the lake basin data suitable for laboratory settings is predicted, and the reasonable range of all parameters is automatically calculated, shortening the time for manual step-by-step setting and increasing the accuracy, making it more in line with similarity to solve the problems raised in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for designing and planning a physical sedimentation simulation experiment based on artificial intelligence, comprising the following steps:

[0009] Step S100, collecting a large amount of actual data of known ancient and modern lake basins;

[0010] Step S200, classifying and labeling the collected data to establish a lake basin parameter database;

[0011] Step S300, taking the geographical environment and climate conditions of the lake basin as input features and various parameters of the lake basin as output targets, and establishing a lake basin parameter prediction model;

[0012] Step S400, using the collected lake basin parameter database to train the lake basin parameter prediction model;

[0013] Step S500, combining the similarity principle to predict the initial value of the lake basin data suitable for laboratory settings;

[0014] Step S600, conducting experimental design and carrying out simulation experiments.

[0015] In a preferred embodiment, step S100 specifically includes the following content:

[0016] The actual data of ancient and modern lake basins includes the lake basin range area, sediment grain size distribution, lake level change range, river width and depth parameters, as well as the corresponding geographical environment and climate condition related information.

[0017] In a preferred embodiment, step S200 specifically includes the following content:

[0018] Create a lake basin information table, including lake basin identifier, name, geographical location, and type fields;

[0019] Spatially annotate each data point using a geographic coordinate system to clarify the specific location corresponding to the data, add a time tag to the data to record the collection time of the data or the time range it represents, and indicate the source channel of each data;

[0020] According to the classified parameter categories and annotation information, design the architecture of the lake basin parameter database, and input the classified and annotated lake basin data into the database according to the requirements of the database architecture;

[0021] Select the relational database management system MySQL to manage the lake basin parameter database.

[0022] In a preferred embodiment, step S300 specifically includes the following contents:

[0023] Sort out the data in the lake basin database, use the data related to the geographical environment and climate conditions as the input feature set, and use the various parameters of the lake basin as the output target set;

[0024] Check for outliers and missing values in the input features and output target data and process them. For outliers, delete or replace them with reasonable values; for missing values, fill them using mean filling, median filling, or interpolation methods;

[0025] Calculate the correlation coefficient between the input features and the output target, and select the features with higher correlation;

[0026] For numerical features, avoid the influence of the dimensional differences of different features on the model, and perform feature scaling and standardization processing;

[0027] Establish a lake basin parameter prediction model, which includes a machine learning model consideration and a physical-mathematical model combination;

[0028] According to the selected model type and input-output features, design the data structure of the model. For the machine learning model, determine the dimension and sample format of the data. For the physical-mathematical model, clarify the variables, parameters, and boundary conditions in the equation;

[0029] Establish the connection relationship between the input features and the output target in the model. For the machine learning model, determine the weights of the input features and the parameters of the model through the training algorithm. For the physical-mathematical model, establish this connection by deriving and solving the equation.

[0030] In a preferred embodiment, step S400 specifically includes the following contents:

[0031] Extract the required data from the lake basin parameter database, sort out the structure and field information of the lake basin parameter database, clarify the specific data tables containing geographical environment and climate condition characteristic data and the corresponding field names therein. For each lake basin record, according to the established field screening rules, accurately extract the above-mentioned geographical environment and climate condition characteristic data as input data, and at the same time extract all the corresponding lake basin parameters completely as output target data. Organize all the extracted data into a structured dataset format, ensuring that each row of data represents the relevant information of a lake basin, and each column corresponds to different input features and output targets;

[0032] Divide the data into a training set, a validation set and a test set;

[0033] Use the training set to train the selected model. Input the geographical environment and climate condition characteristic data in the training set into the model, and continuously adjust the parameters of the model to make the output of the model close to the actual lake basin parameters;

[0034] Use the validation set to validate the trained model, and adjust the parameters of the model according to the performance on the validation set;

[0035] Use the test set to evaluate the finally determined model, calculate the performance metrics of the model on the test set, compare with the performance on the training set and the validation set, and evaluate the generalization ability of the model;

[0036] For machine learning models, use the training set data to train the model. Take the input features and the corresponding output targets as training samples, and adjust the parameters of the model through the gradient descent method to make the predicted output of the model close to the actual output. During the training process, use the validation set to monitor whether the model has overfitting phenomenon, and adjust the hyperparameters of the model according to the validation results;

[0037] For physical-mathematical models, use the training set data to estimate the parameters in the model, and determine the values of the mass exchange coefficient and the hydrodynamic coefficient by fitting the actual data with the least squares method, so that the model can better describe the change process of the lake basin parameters;

[0038] After completing the model training and validation and determining the final model parameters, input the geographical environment and climate condition characteristic data in the test set into the model to obtain the predicted output of the model on the test set. Then calculate the corresponding performance metrics according to the specific task type, and evaluate the performance of the model on the brand-new data that has not participated in the training and validation;

[0039] Adjust and optimize the model according to the results of the evaluation metrics.

[0040] In a preferred embodiment, step S500 specifically includes the following contents:

[0041] Compare the similarity in geographical environment between the target experimental lake basin and known lake basins, obtain the precise geographical location information of the target experimental lake basin, including longitude and latitude coordinates, and set circular search areas with different radius ranges centered on the target lake basin. Then, in the lake basin parameter database, screen out the known lake basins located within these search areas;

[0042] For the screened neighboring lake basins, further investigate the position of the tectonic plates in their regions, the mountain ranges' orientations, and their relative relationships with water systems such as the ocean and rivers, and judge their similarity degrees with the target experimental lake basin in the macroscopic geographical pattern, and assign different weight scores to quantify the contribution of geographical location proximity to the overall similarity;

[0043] Analyze the similarity in climate conditions between the target experimental lake basin and other lake basins, compare the annual average temperature, precipitation patterns, and evaporation intensity, and find the lake basins with similar climate conditions as reference cases;

[0044] Integrate the similarities in geographical environment and climate conditions to determine the most similar known lake basin cases. Based on the lake basin parameters of these similar cases, combined with appropriate proportional relationships or correction factors, predict the initial values of the lake basin data suitable for laboratory settings;

[0045] When predicting the initial values, fully consider the purpose of the experiment and the actual limiting conditions of the laboratory. At the same time, consider the range that the laboratory equipment can simulate, including the water temperature control range and the water volume simulation accuracy, and reasonably adjust the predicted initial values.

[0046] In a preferred embodiment, step S600 specifically includes the following content:

[0047] According to the purpose of the experiment and the predicted initial values of the lake basin data, determine the experimental variables and control variables. The experimental variables include changing the water inflow into the lake and adjusting the water temperature, and the control variables include keeping the lake basin shape unchanged and fixing the lighting conditions. Clearly define the value range and change gradient of each variable and design a reasonable experimental plan;

[0048] Formulate a detailed experimental plan, including the time period of the experiment, the number of repetitions of the experiment, the time intervals and locations for sample collection;

[0049] According to the experimental plan, prepare the equipment and materials required for the experiment, including the container for simulating the lake basin, the water circulation device, the temperature control equipment, and the water quality monitoring instruments. At the same time, prepare the chemical reagents for adjusting salinity and density and the samples for simulating the sediment content in the water.

[0050] In a preferred embodiment, step S600 specifically further includes the following content:

[0051] Conduct simulation experiments according to the experimental plan, precisely control the experimental variables, and record various data during the experiment, including changes in hydrological, water quality, ecological and other parameters of the lake basin;

[0052] Establish a standardized data recording system to ensure the accuracy and integrity of experimental data. The recorded data includes experimental conditions, set values of experimental variables, actual changes, and measured values of lake basin parameters, etc. At the same time, back up and organize the data in a timely manner.

[0053] The technical effects and advantages of a physical deposition simulation experiment design and planning method based on artificial intelligence of the present invention:

[0054] Construct a complete process from data collection to simulation experiment, organically combine actual lake basin data with artificial intelligence modeling and experimental design, provide a scientific, systematic and operable method for in-depth study of the physical deposition process of the lake basin, help improve the accuracy of understanding the sedimentation law of the lake basin, and provide strong theoretical and practical basis for related fields (such as lake ecological protection, water resource management, geological research, etc.). Brief Description of the Drawings

[0055] Figure 1 It is a schematic flow chart of a physical deposition simulation experiment design and planning method based on artificial intelligence of the present invention.

[0056] Figure 2 It is a geomorphic map restored according to the data.

[0057] Figure 3 It is an experimental topographic map designed according to the data.

[0058] Figure 4 It is an experimental parameter map designed according to the data.

[0059] Figure 5 It is an experimental result map.

[0060] Figure 6 It is a structural diagram of the lake basin parameter database of the present invention. Detailed Description of the Invention

[0061] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0062] Embodiment 1

[0063] Figure 1A method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to the present invention is provided, which specifically includes the following steps:

[0064] Step S100: Collect a large amount of actual data of known ancient lake basins and modern lake basins.

[0065] Step S200: Classify and label the collected data to establish a lake basin parameter database.

[0066] Step S300: Establish a lake basin parameter prediction model with the geographical environment and climate conditions of the lake basin as input features and various parameters of the lake basin as output targets.

[0067] Step S400: Use the collected lake basin parameter database to train the lake basin parameter prediction model.

[0068] Step S500: Combine the similarity principle to predict the initial values of lake basin data suitable for laboratory settings.

[0069] Step S600: Conduct experimental design and carry out simulation experiments.

[0070] Step S100 specifically includes the following content:

[0071] The actual data of ancient lake basins and modern lake basins include the lake basin range area, sediment grain size distribution, lake level change range, river width and depth parameters, as well as corresponding geographical environment and climate condition related information.

[0072] It should be added that through a large amount of actual data, the true situations of different types of lake basins under various natural conditions can be understood, including the shape, size, depth, sediment grain size distribution, etc. of the lake basin. This provides an accurate reference for the physical deposition simulation experiment, enabling the experimental design to be closer to the actual lake basin environment, thereby improving the authenticity and reliability of the simulation. The actual data can help determine the reasonable range of various parameters in the experiment. For example, according to the water level change range, river width and depth, etc. of ancient lake basins and modern lake basins, appropriate upper and lower limits can be set for similar parameters in the experiment to avoid overly arbitrary or unreasonable parameter settings.

[0073] The collected actual data can be used to verify the similarity principle adopted in the physical deposition simulation experiment. By comparing the performance of experimental parameters and actual lake basin data in terms of similarity, the experimental design can be continuously adjusted to ensure that the experiment can accurately simulate the physical deposition process in natural lake basins.

[0074] The actual data can reveal the relationships between different lake basin characteristics and the key factors affecting the deposition process. Using this information, the experimental scheme can be optimized, such as adjusting the water flow velocity, sediment source, lake basin topography, etc. in the experiment to better simulate the deposition situation in actual lake basins.

[0075] Step S200 specifically includes the following content:

[0076] Create a lake basin information table, including fields such as lake basin identifier, name, geographical location, and type;

[0077] Use the geographic coordinate system to perform spatial annotation on each data point to clarify the specific location corresponding to the data, add a time tag to the data to record the collection time or the time range represented by the data, and indicate the source channel of each data;

[0078] According to the classified parameter categories and annotation information, design the architecture of the lake basin parameter database, and enter the classified and annotated lake basin data into the database according to the requirements of the database architecture;

[0079] Select the relational database management system MySQL to manage the lake basin parameter database.

[0080] It should be added that using the geographic coordinate system for spatial annotation can accurately determine the specific location corresponding to each data point. This precise spatial positioning is very important for studying the regional characteristics of the lake basin, such as analyzing the ecological environment differences in different areas around the lake basin or the impact of human activities on different positions of the lake basin. At the same time, adding a time tag to record the data collection time or the represented time range makes the data have time series characteristics. This provides support in the time dimension for studying the dynamic change process of the lake basin, such as seasonal changes in water level and long-term evolution of the ecosystem. Indicating the data source channel ensures the traceability of the data, and the source can be easily found when validating the data or further consulting the original materials.

[0081] Design the database architecture according to the classified parameter categories and annotation information, so that the lake basin data can be stored in a scientific and reasonable manner. Different types of parameters (such as geological, hydrological, ecological, human activity parameters, etc.) are stored in corresponding data tables, and the data tables can be linked through key association fields (such as lake basin identifier, time, etc.). This architecture is conducive to the classified management and comprehensive query of data. For example, the geological data table and the ecological data table can be associated through the lake basin identifier to analyze the impact of the geological conditions of a specific lake basin on the ecosystem. And this design method is also convenient for expanding the database in subsequent research. When new parameter categories or data types need to be added, flexible expansion can be carried out on the basis of the existing architecture.

[0082] The classified and labeled lake basin data is entered into the database according to the requirements of the database architecture, ensuring the standardization of data entry. All data is stored in a predefined format and category, avoiding data chaos and inconsistency. At the same time, this also helps to ensure data integrity, as data can be checked and verified during the entry process to ensure that each data point contains necessary information, such as spatial annotations, time tags, and source channels, providing a high-quality data foundation for subsequent data analysis and model building.

[0083] Step S300 specifically includes the following:

[0084] Organize the data in the lake basin database, taking the data related to the geographical environment and climate conditions as the input feature set, and taking the various parameters of the lake basin as the output target set;

[0085] Check for outliers and missing values in the input features and output target data and process them. For outliers, delete or replace them with reasonable values; for missing values, use mean filling, median filling, or interpolation methods to fill them;

[0086] Calculate the correlation coefficient between the input features and the output target, and select the features with higher correlation;

[0087] For numerical features, avoid the impact of different feature dimensions on the model, and perform feature scaling and standardization processing;

[0088] Establish a lake basin parameter prediction model, which includes considering machine learning models and combining physical-mathematical models;

[0089] According to the selected model type and input-output features, design the data structure of the model. For machine learning models, determine the dimension and sample format of the data. For physical-mathematical models, clarify the variables, parameters, and boundary conditions in the equations;

[0090] Establish the connection relationship between the input features and the output target in the model. For machine learning models, determine the weights of the input features and the parameters of the model through training algorithms. For physical-mathematical models, establish this connection through derivation and solution of equations.

[0091] It should be added that by organizing the data and clarifying the input feature set (data related to the geographical environment and climate conditions) and the output target set (various parameters of the lake basin), the data can be made more compliant with the requirements of model training. This targeted data organization method allows the model to directly use effective data for learning and prediction, avoiding interference from irrelevant data.

[0092] Checking and handling outliers and missing values can significantly improve data quality. Outliers may be caused by measurement errors or other special circumstances. If not handled, they can mislead the training of the model and cause the model to learn incorrect patterns. By deleting or reasonably replacing outliers and using appropriate methods to fill in missing values (such as mean filling, median filling, or interpolation), the data can more accurately reflect the true situation of the lake basin and provide a more reliable data basis for the model.

[0093] Considering the combination of machine learning models and physical-mathematical models to establish a lake basin parameter prediction model provides a flexible modeling strategy. Machine learning models can automatically learn complex patterns and relationships from large amounts of data and are suitable for dealing with non-linear and complex data relationships. Physical-mathematical models are based on physical principles and mathematical derivations and can explain the change process of lake basin parameters from a mechanism perspective. Combining the advantages of both, the most suitable model can be selected according to specific research questions and data characteristics, or the two can be used in combination to obtain more accurate lake basin parameter prediction results.

[0094] Establishing the connection relationship between input features and output targets in the model is the key for the model to make predictions. For machine learning models, by determining the weights of input features and the parameters of the model through training algorithms, the model can learn the quantitative relationship between input features and output targets. For physical-mathematical models, by deriving and solving equations to establish this connection, the internal law of the change of lake basin parameters can be revealed from a physical and mathematical perspective. Such a connection relationship enables the model to accurately predict various parameters of the lake basin based on input features such as geographical environment and climate conditions.

[0095] Step S400 specifically includes the following content:

[0096] Extract the required data from the lake basin parameter database, sort out the structure and field information of the lake basin parameter database, clarify the specific data tables containing geographical environment and climate condition feature data and the corresponding field names, and for each lake basin record, accurately extract the above-mentioned geographical environment and climate condition feature data as input data according to the established field screening rules. At the same time, extract all the corresponding lake basin parameters completely as output target data, and organize all the extracted data into a structured dataset format to ensure that each row of data represents the relevant information of a lake basin, and each column corresponds to different input features and output targets;

[0097] Divide the data into a training set, a validation set, and a test set;

[0098] Use the training set to train the selected model, input the geographical environment and climate condition feature data in the training set into the model, and continuously adjust the parameters of the model to make the output of the model close to the actual lake basin parameters;

[0099] Use the validation set to validate the trained model, and adjust the model parameters according to the performance on the validation set;

[0100] Use the test set to evaluate the finally determined model, calculate the performance metrics of the model on the test set, compare with the performance on the training set and the validation set, and evaluate the generalization ability of the model;

[0101] For a machine learning model, use the training set data to train the model. Take the input features and the corresponding output targets as training samples, and adjust the model parameters through the gradient descent method, so that the predicted output of the model is close to the actual output. During the training process, use the validation set to monitor whether the model has overfitting, and adjust the hyperparameters of the model according to the validation results;

[0102] For a physical-mathematical model, use the training set data to estimate the parameters in the model, and determine the values of the mass exchange coefficient and the hydrodynamic coefficient by fitting the actual data through the least squares method, so that the model can better describe the change process of the lake basin parameters;

[0103] After completing the model training and validation and determining the final model parameters, input the geographical environment and climate condition feature data in the test set into the model to obtain the predicted output of the model on the test set. Then calculate the corresponding performance metrics according to the specific task type, and evaluate the performance of the model on the brand-new data that has not participated in training and validation;

[0104] Adjust and optimize the model according to the results of the evaluation metrics.

[0105] It should be added that using the training set data, taking the input features and the corresponding output targets as training samples, and adjusting the model parameters through optimization algorithms such as the gradient descent method can enable the model to gradually learn the complex relationship between the input (geographical environment and climate condition features) and the output (various lake basin parameters). The gradient descent method can continuously update the model parameters along the direction where the loss function decreases fastest, so that the predicted output of the model continuously approaches the actual output, thereby improving the fitting degree of the model to the known data and enhancing the prediction accuracy.

[0106] For example, when predicting the parameter of the lake basin water level change, the model can more accurately simulate the actual water level change situation by continuously adjusting the parameters based on the input climate condition features such as precipitation and evaporation, as well as the geographical environment features such as the topography and landform of the lake basin.

[0107] For physical-mathematical models, training set data is used to estimate the parameters in the models (such as mass exchange coefficients, hydrodynamic coefficients, etc.). By fitting the actual data through methods such as the least squares method, the models can better describe the variation process of lake basin parameters based on physical principles. These physical-mathematical models are constructed starting from physical mechanisms such as mass and energy exchange within the lake basin. After determining the parameters through fitting with actual data, their prediction results have clear physical meanings and relatively reliable theoretical bases, which helps to deeply understand the internal relationships and variation laws among the various parameters of the lake basin system.

[0108] For example, when studying the water quality changes in a lake basin, by accurately estimating parameters such as mass exchange coefficients, physical-mathematical models can simulate the input, output, and transformation of substances such as nutrients within the lake basin under different conditions, providing a scientific basis for the research and management of the lake basin ecological environment.

[0109] Step S500 specifically includes the following content:

[0110] Compare the similarity in geographical environment between the target experimental lake basin and known lake basins, obtain the precise geographical location information of the target experimental lake basin, including longitude and latitude coordinates. Taking the target lake basin as the center, set circular search areas with different radius ranges, and screen out the known lake basins located within these search areas in the lake basin parameter database;

[0111] For the screened neighboring lake basins, further investigate the positions of the tectonic plates in their regions, the directions of mountain ranges, and the relative relationships with water systems such as the ocean and rivers, judge their similarity degrees with the target experimental lake basin in the macroscopic geographical pattern, and assign different weight scores to quantify the contribution of geographical location proximity to the overall similarity;

[0112] Analyze the similarity in climate conditions between the target experimental lake basin and other lake basins, compare the annual average temperature, precipitation patterns, and evaporation intensities, and find the lake basins with similar climate conditions as reference cases;

[0113] Integrate the similarities in geographical environment and climate conditions, determine the most similar known lake basin cases, and based on the lake basin parameters of these similar cases, combined with appropriate proportional relationships or correction coefficients, predict the initial values of the lake basin data suitable for laboratory settings;

[0114] When predicting the initial values, fully consider the purpose of the experiment and the actual limiting conditions of the laboratory. At the same time, consider the range that the laboratory equipment can simulate, including the water temperature control range and the water volume simulation accuracy, and reasonably adjust the predicted initial values.

[0115] It should be added that screening reference cases by comparing the similarities of geographical environments and climatic conditions can make the initial values of predictions more targeted. Considering geographical environment factors such as geographical location proximity, topography and basin characteristics, as well as climatic conditions such as annual average temperature, precipitation pattern and evaporation intensity, the similar cases found are consistent with the target experimental lake basin in multiple key factors. For example, when predicting the initial value of the lake basin water level change, selecting known lake basin cases with similar geographical environments (such as the same closed-basin topography) and similar climatic conditions (similar annual precipitation and evaporation), the water level change law is more likely to provide a reliable reference for the target experimental lake basin, thus improving the accuracy of the initial value prediction.

[0116] Predicting the initial value considering the experimental purpose can ensure that the laboratory simulation experiment serves the research goal more effectively. For example, if the experimental purpose is to study the ecosystem evolution of the lake basin under specific climatic conditions, then when predicting the initial value, focus on the lake basin parameters related to the ecosystem (such as water quality, biodiversity, etc.), and set the initial value according to the ecological parameters of similar cases, so that the experiment can more specifically explore the target problem and enhance the effectiveness of the experimental design.

[0117] Using similar cases of known lake basins to predict the initial value provides practical data references for the experiment. The parameters of these known lake basins are based on actual observations or research and have a certain degree of reliability. By reasonably referring to the parameters of these cases to set the experimental initial value, the experimental results can be linked to the actual lake basin situation, increasing the referenceability of the experimental results in the actual lake basin research. For example, when studying the water quality change of the lake basin, referring to the initial value of the water quality parameters of a similar lake basin to set the experiment, the experimental results can better analogize the possible water quality change situation in the actual lake basin, providing more valuable references for the ecological protection and management of the actual lake basin.

[0118] It should also be added that when combining the similarity principle, the following formula can be used for calculation:

[0119]

[0120]

[0121] Step S600 specifically includes the following content:

[0122] According to the experimental purpose and the predicted initial value of the lake basin data, determine the experimental variables and control variables, where the experimental variables include changing the inflow water volume and adjusting the water temperature, and the control variables include keeping the lake basin shape unchanged and fixing the light conditions. Define the value range and change gradient of each variable and design a reasonable experimental plan;

[0123] Formulate a detailed experimental plan, including the time period of the experiment, the number of repetitions of the experiment, the time intervals and locations for sample collection;

[0124] According to the experimental plan, prepare the equipment and materials required for the experiment, including the container for the simulated lake basin, the water circulation device, the temperature control equipment, and the water quality monitoring instruments. At the same time, prepare the chemical reagents for adjusting salinity and density and the samples for simulating the sediment content in the water.

[0125] Step S600 specifically further includes the following content:

[0126] Conduct the simulation experiment according to the experimental plan, precisely control the experimental variables, and record various data during the experiment, including the changes in the hydrology, water quality, ecology and other parameters of the lake basin;

[0127] Establish a standardized data recording system to ensure the accuracy and integrity of the experimental data. The recorded data includes the experimental conditions, the set values of the experimental variables, the actual changes, and the measured values of the lake basin parameters, etc. At the same time, back up and organize the data in a timely manner.

[0128] It should be added that by determining the experimental variables (such as changing the water inflow into the lake, adjusting the water temperature) and the control variables (such as keeping the shape of the lake basin unchanged, fixing the lighting conditions), the research direction and key points of the experiment can be clarified. This distinction can help researchers focus on the impact of specific factors on the lake basin, avoid the interference of other factors, and thus more accurately reveal the causal relationship between variables. For example, when studying the impact of water temperature on the lake basin ecosystem, by fixing other conditions (such as lighting, the shape of the lake basin, etc.) and only changing the water temperature, the changes in the ecosystem caused by the water temperature change can be clearly observed, such as the changes in the types and quantities of organisms, the chemical changes in the water quality, etc.

[0129] Defining the value range and change gradient of each variable can make the experimental plan more reasonable and detailed. A reasonable value range can cover the actual situation involved in the research problem, and an appropriate change gradient can more accurately capture the influence law of the variable on the lake basin parameters. For example, when studying the impact of the water inflow into the lake on the water level of the lake basin, determine the value range of the water inflow into the lake according to the actual water volume change range of the lake basin and the research purpose, and set an appropriate change gradient, so that the changes in the water level under different water volume inputs can be observed in detail, providing data support for establishing an accurate mathematical model or theory.

[0130] Preparing the corresponding equipment and materials according to the experimental plan can ensure the smooth progress of the experiment. Equipment such as the container for simulating the lake basin, the water circulation device, the temperature control equipment, and the water quality monitoring instrument can accurately simulate and measure the physical and chemical processes of the lake basin. For example, using a high-precision water quality monitoring instrument can obtain the changes in water quality parameters (such as pH value, dissolved oxygen, etc.) in real time and accurately, providing reliable data for studying water quality changes. At the same time, preparing chemical reagents for adjusting salinity and density and samples for simulating the sediment content in water can artificially control and change the ecological environment of the lake basin in the experiment to meet the requirements of different experimental purposes.

[0131] Background of the Embodiment

[0132] Suppose we want to study the physical sedimentation process of lakes in a specific area, aiming to reveal the distribution law of sediments in the lake basin under different conditions and the key factors affecting sedimentation through simulation experiments, providing a scientific basis for the ecological protection and resource development of lakes in this area.

[0133] Specific Steps of the Embodiment

[0134] Step S100: Data Collection

[0135] Determine the data sources

[0136] Obtain detailed geological exploration reports of multiple ancient and modern lake basins in this area and its surrounding areas from the geological survey department. These reports contain parameter information such as the area of the lake basin, the sediment grain size distribution, the range of lake level changes, the width and depth of the river, etc. At the same time, collect multi-year meteorological data of the area where the corresponding lake basin is located from the meteorological department to obtain information related to the geographical environment and climate conditions, such as the annual average temperature, precipitation pattern, evaporation intensity, etc.

[0137] Illustrate the data collection situation

[0138] For example, in the collected geological exploration reports, it is found that the lake level of a certain ancient lake basin has experienced several large-scale changes in the past thousand years. Its maximum area can reach 50 square kilometers, and the minimum is only 20 square kilometers; the sediment grain size distribution shows a law of changing from fine to coarse from the lake center to the lake shore; the corresponding river width is about 10 meters on average at the lake inlet, and the depth is about 2 meters. At the same time, the climate data of this area shows that the annual average temperature is about 15°C, the precipitation is concentrated in summer, the annual precipitation is about 800 mm, and the evaporation intensity reaches its peak in summer, etc. By collecting actual data of many such lake basins, a basic data set for this study is constructed.

[0139] Step S200: Data Classification and Annotation, Establish a Database

[0140] Create a lake basin information table and annotate the data

[0141] Create a lake basin information table. For each lake basin data collected above, assign it a unique lake basin identifier, and record the lake basin name, precise geographical location (latitude and longitude coordinates), and lake basin type (such as tectonic lake, crater lake, etc.). For example, one of the lake basin identifiers is "LB001", the name is "XX Lake", the geographical location is XX degrees north latitude and XX degrees east longitude, and the type is a tectonic lake.

[0142] Use the geographic coordinate system to spatially annotate each data point. For example, for the sediment grain size data of different sampling points within the lake basin, clearly define their corresponding specific coordinate positions. Add time tags to the data. For data such as the range of lake level changes in ancient lake basins, mark the corresponding geological era; for data of modern lake basins, mark the specific collection year. Indicate the source channel of each data. For example, the sediment grain size data of the "LB001" lake basin is sourced from the XX geological exploration project report.

[0143] Design the database architecture and enter the data

[0144] According to the classified parameter categories, design a data architecture that includes a geological parameter table (storing information such as sediment grain size distribution and lake basin stratigraphic structure), a hydrological parameter table (range of lake level changes, river width and depth, etc.), a climate parameter table (annual average temperature, precipitation pattern, evaporation intensity, etc.), and an ecological parameter table (which can be used to store information such as biodiversity if ecological-related research is involved later). Establish connections between these data tables through key association fields such as lake basin identifier and time.

[0145] Enter the classified and annotated lake basin data into the lake basin parameter database managed by MySQL. During the entry process, check the integrity and accuracy of the data to ensure that each data point has complete annotation information.

[0146] Step S300: Establish a lake basin parameter prediction model

[0147] Data sorting and preprocessing

[0148] Extract data from the lake basin parameter database, sort out data related to the geographical environment (such as the topography and land area of the lake basin) and climate conditions (annual precipitation, evaporation, etc.) as the input feature set, and use the various parameters of the lake basin (such as changes in sediment thickness, grain size distribution at different locations, etc.) as the output target set.

[0149] Check for outliers and missing values in the input features and output target data. For example, in the river depth data of a certain lake basin, individual extremely large values that deviate significantly from the normal range are found and judged as outliers caused by measurement errors and are deleted; for the missing annual evaporation data of some lake basins, fill it with the average annual evaporation of other lake basins in the same climate region.

[0150] Calculate the correlation coefficient between the input features and the output targets, and it is found that the terrain slope of the lake basin has a relatively high correlation with the accumulation thickness of sediments near the lake shore. Select such features with relatively high correlations for subsequent modeling. Perform feature scaling and standardization on numerical features. For example, unify data with different dimensions such as lake basin area and precipitation into a specific interval to avoid the impact of dimensional differences on the model.

[0151] Model construction

[0152] Consider establishing a lake basin parameter prediction model by combining the random forest model in machine learning with the physical-mathematical model constructed based on the principles of hydrodynamics and material transport.

[0153] For the random forest model, determine the dimension of the data as the number of input features and the dimension of the output target. The sample format is organized such that each row of data represents a lake basin sample, including the input feature values and the corresponding output target values. For the physical-mathematical model, clarify the variables (such as water flow velocity, sediment concentration, etc.), parameters (such as sediment settling velocity, water flow diffusion coefficient, etc.), and boundary conditions (such as the water flow in and out conditions at the lake basin boundary) in the equations.

[0154] Establish the connection relationship between the input features and the output targets in the model through training algorithms (the random forest uses the decision tree training method based on information gain, and the physical-mathematical model derives and solves the hydrodynamics and mass conservation equations), so that the model can learn the variation laws of lake basin parameters under different geographical environments and climate conditions.

[0155] Step S400: Model training and evaluation

[0156] Data division

[0157] Extract a sufficient amount of data from the lake basin parameter database and divide it into a training set, a validation set, and a test set according to the ratio of 7:2:1. For example, if there are a total of 1000 groups of lake basin data, 700 groups are used as the training set, 200 groups are used as the validation set, and 100 groups are used as the test set.

[0158] Model training (machine learning model part)

[0159] For the random forest model, it is trained using the training set data. The input features and the corresponding output targets are used as training samples. Through optimization algorithms such as the gradient descent method, the node splitting rules of the decision trees and parameters such as the weights of each feature in the model are continuously adjusted to make the predicted output of the model gradually approach the actual output. During the training process, the validation set is used to monitor whether the model has overfitting. When it is found that as the depth of the decision tree increases and the error on the validation set starts to rise, it indicates overfitting. Accordingly, the hyperparameters of the model (such as limiting the maximum depth of the decision tree, increasing the minimum number of samples required for node splitting, etc.) are adjusted to optimize the performance of the model.

[0160] Model Training (Physical - Mathematical Model Part)

[0161] For the physical - mathematical model, the training set data is used to estimate the parameters in the model. For example, by using the least - squares method to fit the observed data such as the actual lake basin water flow velocity and sediment concentration, the values of the mass exchange coefficient and hydrodynamic coefficient are determined, enabling the model to better describe the variation process of parameters such as sediment transport and deposition in the lake basin.

[0162] Model Evaluation

[0163] The finally determined model is evaluated using the test set, and the performance metrics of the model on the test set are calculated, such as the mean squared error (MSE), coefficient of determination (R 2 ) etc. Suppose the MSE of the random forest model on the test set is 0.05 and the R 2 reaches 0.85, and the MSE of the physical - mathematical model is 0.06 and the R 2 is 0.80. By comparing with the performance on the training set and validation set, it is found that the performance of the model is relatively stable on different data sets and has good generalization ability, indicating that the model can better predict the parameters of new lake basin situations.

[0164] Step S500: Predict the initial values of the lake basin data set in the laboratory

[0165] Similarity Analysis

[0166] We plan to simulate a sedimentary environment similar to a small lake (target experimental lake basin) in the study area in the laboratory. By comparing the similarity of the target experimental lake basin and known lake basins in terms of geographical environment, it is found that several lake basins are adjacent in geographical location, with the same mountain - basin - type topography and similar basin characteristics, and the vegetation coverage and soil type within their basins are similar; when analyzing the similarity of climate conditions, by comparing the annual average temperature, precipitation pattern, and evaporation intensity, the lake basins with similar climate conditions are selected as reference cases.

[0167] Initial Value Prediction and Adjustment

[0168] Based on the similarity of the comprehensive geographical environment and climatic conditions, determine the several most similar known lake basin cases. According to the lake basin parameters of these similar cases (such as lake basin area, initial water depth, sediment grain size composition, etc.), combined with appropriate proportional relationships or correction factors for prediction. For example, the average initial water depth of a known similar lake basin is 5 meters. Considering that the area of the target experimental lake basin is slightly smaller, the initial water depth of the target experimental lake basin is predicted to be 4 meters according to the area proportional relationship. At the same time, fully consider the experimental purpose (studying the sedimentation law of sediments under different water temperatures and water volume changes) and the actual limiting conditions of the laboratory (the water temperature control range in the laboratory is 5°C - 30°C, and the water volume simulation accuracy is ±0.1 cubic meters per minute), and reasonably adjust the predicted initial value. Finally, determine the initial water depth to be 3.5 meters to ensure that the experiment can be carried out smoothly under the existing equipment conditions and better meet the research needs.

[0169] Step S600: Experimental design and simulation experiment

[0170] Experimental design

[0171] According to the experimental purpose (studying the sedimentation law of sediments under different conditions) and the predicted initial values of the lake basin data, determine the experimental variables and control variables. The experimental variables include changing the incoming lake water volume (the value range is set to 0.1 - 1 cubic meters per minute, and the change gradient is 0.1 cubic meters per minute) and adjusting the water temperature (the value range is set to 10°C - 25°C, and the change gradient is 5°C). The control variables include keeping the lake basin shape unchanged (using a simulated lake basin container with a fixed shape) and fixing the lighting conditions (simulating a constant natural light intensity).

[0172] Formulate a detailed experimental plan, set the time period of the experiment to 30 days, and collect samples once a day; the number of repetitions of the experiment is 3 times to reduce the influence of accidental factors; the time interval for sample collection is fixed at the same time every day (such as 10 am), and the collection locations include the center of the lake basin and three points at different distances from the lake shore, etc., to ensure that the parameter changes at different positions of the lake basin can be comprehensively obtained.

[0173] According to the experimental plan, prepare the equipment and materials required for the experiment, build the container of the simulated lake basin, the size of which is designed according to the predicted initial values of the lake basin; install a water circulation device to control the incoming lake water volume, equip a temperature control device to accurately adjust the water temperature, and place high-precision water quality monitoring instruments to monitor the water quality changes in real time (such as pH value, dissolved oxygen, suspended solid concentration, etc., these indicators can indirectly reflect the sediment situation); at the same time, prepare chemical reagents for adjusting the water quality (simulating the influence of different water body environments on sediments) and biological samples of the simulated biological community collected from similar natural lake basins (such as plankton, benthic organisms, etc., for studying the influence of biological activities on sedimentation).

[0174] Simulation experiment and data recording

[0175] Conduct simulation experiments according to the experimental plan, precisely control the experimental variables. For example, accurately adjust the water inflow into the lake through a water circulation device to change according to the set value range and variation gradient, and strictly control the water temperature at the set value using temperature control equipment. Record various data during the experiment, including changes in the hydrology of the lake basin (such as water level changes, water flow velocity, etc.), water quality (such as pH value, dissolved oxygen, suspended solid concentration, etc.), and ecology (such as changes in the types and quantities of organisms, etc.).

[0176] Establish a standardized data recording system. The data recorded daily includes experimental conditions (such as external environmental factors like the then room temperature, weather conditions, etc., and set values of experimental variables such as the water inflow into the lake and water temperature), actual changes in experimental variables (such as the fluctuation range of the actually measured water inflow into the lake, real-time changes in water temperature, etc.), and measured values of lake basin parameters (water quality parameters, organism quantities, etc. measured at each sampling location), etc. At the same time, back up and organize the data of each day in a timely manner, store it in a dedicated experimental data folder, and classify it according to the date and number of experiments, etc., for convenient subsequent query and analysis.

[0177] Analysis of the results of the embodiments

[0178] By analyzing the data recorded during the simulation experiment, for example, observing the relationship between the changes in water quality parameters in the lake basin and the stacking thickness and grain size distribution changes of sediments at different positions on the lake bottom under different water inflows into the lake and water temperatures. It can be found that as the water inflow into the lake increases, the water flow velocity speeds up, and the sediment grain size at the center of the lake basin has a gradually coarser trend. And when the water temperature is higher, the biological activities increase, and the disturbance and resuspension effects on the sediments are more obvious, affecting the final distribution law of the sediments.

[0179] Compare these experimental results with some sedimentation phenomena observed in the actual lake basin, and it is found that the laws obtained from the simulation experiment are consistent with the actual situation to a certain extent, verifying the feasibility and effectiveness of this physical sedimentation simulation experiment design and planning method based on artificial intelligence. It can provide a reliable reference basis for further research on the physical sedimentation process of the lake basin, help us better understand the sedimentation mechanism in natural lake basins and the influence of different factors on sedimentation, and thus provide strong support for practical applications such as lake ecological protection and resource development.

[0180] As described above, it is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claimed rights.

[0181] Finally, the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A design and planning method for physical deposition simulation experiments based on artificial intelligence, characterized in that: The steps include: Step S100, collecting a large amount of actual data of known ancient lake basins and modern lake basins; Step S200, classifying and labeling the collected data to establish a lake basin parameter database; Step S300, using the geographical environment and climate conditions of the lake basin as input features and various parameters of the lake basin as output targets, to establish a lake basin parameter prediction model; Step S400, training a lake basin parameter prediction model using the collected lake basin parameter database; Step S500, combining the similarity principle, predicting the initial value of the lake basin data suitable for laboratory setting; Step S600, designing an experiment and conducting a simulation experiment.

2. The method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to claim 1, characterized in that: Step S100 specifically includes the following contents: The actual data of ancient lake basins and modern lake basins include the area of ​​the lake basin, sediment particle size distribution, lake level change range, river width and depth parameters, and corresponding geographical environment and climatic conditions.

3. The method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to claim 2, characterized in that: Step S200 specifically includes the following contents: Create a lake basin information table, including lake basin identifier, name, geographic location, and type fields; Use the geographic coordinate system to spatially annotate each data point, clarify the specific location of the data, add time tags to the data, record the data collection time or the time range it represents, and indicate the source channel of each data; According to the classified parameter categories and annotation information, the architecture of the lake basin parameter database is designed, and the classified and annotated lake basin data is entered into the database according to the requirements of the database architecture; The relational database management system MySQL is selected to manage the lake basin parameter database.

4. The method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to claim 3, characterized in that: Step S300 specifically includes the following contents: Organize the data in the lake basin database, take the data related to the geographical environment and climate conditions as the input feature set, and take the various parameters of the lake basin as the output target set; Check the outliers and missing values ​​in the input features and output target data, and process them. For outliers, delete them or replace them with reasonable values; for missing values, use mean filling, median filling or interpolation to fill them. Calculate the correlation coefficient between the input features and the output target, and select the features with higher correlation; For numerical features, we should avoid the influence of different dimensional differences of different features on the model by performing feature scaling and standardization. Establish a lake basin parameter prediction model, which includes machine learning model considerations and a combination of physical-mathematical models; Design the data structure of the model based on the selected model type and input and output characteristics. For machine learning models, determine the data dimension and sample format. For physical-mathematical models, specify the variables, parameters, and boundary conditions in the equations. Establish a connection between input features and output targets in the model. For machine learning models, the weights of input features and model parameters are determined by training algorithms. For physical-mathematical models, this connection is established by deriving and solving equations.

5. The method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to claim 4, characterized in that: Step S400 specifically includes the following contents: Extract the required data from the lake basin parameter database, sort out the structure and field information of the lake basin parameter database, clarify the specific data table containing the geographical environment and climate condition characteristic data and the corresponding field names, and for each lake basin record, accurately extract the above geographical environment and climate condition characteristic data as input data according to the established field screening rules, and at the same time completely extract the corresponding lake basin parameters as output target data, organize all the extracted data into a structured data set format, ensure that each row of data represents the relevant information of a lake basin, and each column corresponds to different input features and output targets; Divide the data into training, validation and test sets; The selected model is trained using the training set, the geographical environment and climate condition characteristic data in the training set are input into the model, and the parameters of the model are continuously adjusted so that the output of the model is close to the actual lake basin parameters; Use the validation set to validate the trained model and adjust the model parameters based on the performance on the validation set. Use the test set to evaluate the finalized model, calculate the performance indicators of the model on the test set, compare them with the performance on the training set and validation set, and evaluate the generalization ability of the model; For machine learning models, use the training set data to train the model, use the input features and the corresponding output targets as training samples, and adjust the model parameters through the gradient descent method so that the model's predicted output is close to the actual output. During the training process, use the validation set to monitor whether the model is overfitting, and adjust the model's hyperparameters based on the validation results. For the physical-mathematical model, the training set data is used to estimate the parameters in the model, and the values ​​of the material exchange coefficient and the hydrodynamic coefficient are determined by fitting the actual data through the least square method, so that the model can better describe the change process of the lake basin parameters; After completing model training and validation and determining the final model parameters, the geographical environment and climate condition feature data in the test set are input into the model to obtain the model's predicted output on the test set. Then, the corresponding performance indicators are calculated according to the specific task type to evaluate the model's performance on new data that has not participated in training and validation; According to the results of the evaluation indicators, the model is adjusted and optimized.

6. The method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to claim 5, characterized in that: Step S500 specifically includes the following contents: Compare the similarities between the target experimental lake basin and the known lake basins in terms of geographical environment, obtain the precise geographical location information of the target experimental lake basin, including the longitude and latitude coordinates, set circular search areas with different radii with the target lake basin as the center, and select the known lake basins within these search areas in the lake basin parameter database; For the selected adjacent lake basins, further investigate the location of the tectonic plates, the direction of the mountains, and the relative relationship with the ocean, rivers and other water systems in the area where they are located, judge their similarity with the target experimental lake basin in the macro-geographic pattern, and assign different weight scores to quantify the contribution of geographical proximity to the overall similarity; Analyze the similarity of climate conditions between the target experimental lake basin and other lake basins, compare the annual average temperature, precipitation pattern and evaporation intensity, and find lake basins with similar climate conditions as reference cases; Based on the similarity of geographical environment and climate conditions, the most similar known lake basin cases are determined. Based on the lake basin parameters of these similar cases and appropriate proportional relationships or correction coefficients, the initial values ​​of lake basin data suitable for laboratory settings are predicted. When predicting the initial value, fully consider the purpose of the experiment and the actual limitations of the laboratory. At the same time, consider the range that the laboratory equipment can simulate, including the water temperature control range and water volume simulation accuracy, and make reasonable adjustments to the predicted initial value.

7. The method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to claim 6, characterized in that: Step S600 specifically includes the following contents: According to the experimental purpose and the predicted initial values ​​of the lake basin data, determine the experimental variables and control variables. The experimental variables include changing the amount of water entering the lake and adjusting the water temperature, and the control variables include keeping the shape of the lake basin unchanged and fixing the light conditions. Clarify the value range and change gradient of each variable and design a reasonable experimental plan. Develop a detailed experimental plan, including the experimental time period, the number of experimental repetitions, and the time intervals and locations for sample collection; According to the experimental plan, prepare the equipment and materials required for the experiment, including containers simulating lake basins, water circulation devices, temperature control equipment and water quality monitoring instruments. Also prepare chemical reagents for adjusting salinity and density and samples of sediment content in simulated water.

8. The method for designing and planning a physical deposition simulation experiment based on artificial intelligence according to claim 7, characterized in that: Step S600 specifically includes the following contents: Conduct simulation experiments according to the experimental plan, accurately control experimental variables, and record various data during the experiment, including changes in the hydrology, water quality, ecology and other parameters of the lake basin; A standardized data recording system is established to ensure the accuracy and completeness of experimental data. The recorded data include experimental conditions, setting values ​​of experimental variables, actual changes and measured values ​​of lake basin parameters, etc. At the same time, the data is backed up and organized in a timely manner.

Images