Oasis eco-hydrological simulation method, device, equipment and storage medium

Abstract

The application provides an oasis ecological hydrology simulation method, device, equipment and storage medium. It relates to the technical field of ecological hydrology under geophysics. The method comprises the following steps: based on ecological hydrology basic data, using ecological hydrology equation to calculate oasis evapotranspiration, oasis area carrying capacity of groundwater and vegetation coverage carrying capacity of groundwater, as input features of machine learning modeling, and based on machine learning model suitable for regression modeling, respectively constructing lake area module, groundwater depth module, oasis area module and vegetation coverage module; integrating each module to construct an oasis ecological hydrology model, and realizing oasis ecological hydrology simulation based on the oasis ecological hydrology model. The method can provide an effective tool for the total water delivery amount and water delivery process optimization design of ecological water delivery engineering, and help to improve the ecological benefit of oasis recovery.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ecological hydrology under geophysics, and particularly relates to an oasis ecological hydrology simulation method, device, equipment and storage medium. BACKGROUND

[0002] Current oasis ecological hydrology simulation has developed physical mechanism models, conceptual models, etc., such as a method for simulating and optimizing evaluation of water delivery in a desert oasis based on ecological hydrology (application / patent number: CN202010372702.8), an oasis ecological water delivery process optimization method based on monthly scale ecological hydrology simulation (application / patent number: CN202410454651.1), etc. With the rapid development of artificial intelligence technology, various machine learning algorithms have also been gradually applied to oasis ecological hydrology simulation research, such as an ecological hydrology simulation method based on a machine learning model (application / patent number: CN202510055611.4), a groundwater ecological burial depth evaluation method fusing CatBoost regression and SHAP attribution (application / patent number: CN202411452589.9), a groundwater critical burial depth calculation method based on geographic big data and residual analysis (application / patent number: CN202510198101.2), etc.

[0003] Specifically, the invention patent application with the application / patent number CN202510055611.4 discloses an ecological hydrology simulation method based on a machine learning model, which constructs an ecological hydrology model based on an improved ConvLSTM network structure, trains the ecological hydrology model based on a data set, obtains the trained ecological hydrology model, inputs the to-be-tested data into the trained ecological hydrology model, and performs prediction of evapotranspiration. The groundwater ecological burial depth evaluation method fusing CatBoost regression and SHAP attribution uses a CatBoost machine learning algorithm to respectively construct an EVI regression model and an ET regression model, respectively performs SHAP attribution analysis on the EVI and ET regression models, uses a Boltzmann function to establish a correlation between the SHAP value of the groundwater burial depth in the EVI and ET regression models and the groundwater burial depth, and evaluates the groundwater ecological burial depth by using inflection point analysis. The groundwater critical burial depth calculation method based on geographic big data and residual analysis evaluates the groundwater critical burial depth through steps of groundwater burial depth data calculation, environmental factor neighboring covariate data calculation, grid data space matching, background data set construction by random sampling, groundwater burial depth influence quantity calculation, and empirical model fitting.

[0004] Current research on eco-hydrological simulation using machine learning algorithms still has certain shortcomings in the following three aspects: (1) It only considers the impact of meteorological and hydrological elements such as precipitation and groundwater depth on ecological elements such as EVI and vegetation cover, or only considers the impact of vegetation elements on evapotranspiration, realizing the simulation of hydrological process-driven ecological process or ecological process-driven hydrological process, but has not yet realized eco-hydrological feedback simulation; (2) It fails to organically combine machine learning algorithms with eco-hydrological equations, making it difficult to ensure that machine learning algorithms grasp the correct eco-hydrological mechanisms when data is scarce; (3) It does not design models for oases for ecological water conveyance recovery, making it difficult to apply to simulating oase eco-hydrological processes driven by ecological water conveyance. Therefore, it is necessary to develop an oase eco-hydrological simulation method that integrates machine learning and eco-hydrological equations, providing an effective tool for optimizing the total water conveyance and water conveyance process design of ecological water conveyance projects, helping to improve the ecological benefits of oase recovery, and serving the ecological civilization construction and high-quality sustainable development of ecologically fragile areas in inland river basins. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, and storage medium for simulating oasis eco-hydrology, which overcomes the shortcomings of existing oasis eco-hydrology simulation studies, such as the failure to utilize machine learning to achieve eco-hydrological feedback simulation, the failure to organically combine machine learning algorithms with eco-hydrological equations, and the lack of design models for oasis ecological water conveyance projects. It improves the methodological system for simulating oasis eco-hydrology in arid inland river basins, enhances the utilization efficiency of meteorological, hydrological, and ecological observation data such as precipitation, potential evapotranspiration, ecological water conveyance, groundwater depth, oasis area, and vegetation cover, and provides an effective tool for optimizing the total water conveyance and water conveyance process design of ecological water conveyance projects. This helps to improve the ecological benefits of oasis recovery and serves the ecological civilization construction and high-quality sustainable development of ecologically fragile areas in inland river basins.

[0006] Firstly, this application provides a method for simulating oasis eco-hydrology, including:

[0007] Acquire basic eco-hydrological data; wherein, the basic eco-hydrological data includes monthly meteorological data, hydrological data and ecological data after standardization processing, the meteorological data includes temperature, precipitation and potential evapotranspiration, the hydrological data includes ecological water transport and groundwater depth, and the ecological data includes oasis area, normalized vegetation index and vegetation cover.

[0008] Based on the aforementioned eco-hydrological basic data, the eco-hydrological equations are used to calculate the oasis groundwater evapotranspiration, the oasis area carrying capacity of groundwater, and the vegetation cover carrying capacity of groundwater. These are used as input features for machine learning modeling. Based on a machine learning model suitable for regression modeling, modules for lake area, groundwater depth, oasis area, and vegetation cover are constructed respectively.

[0009] The lake area module, groundwater depth module, oasis area module, and vegetation coverage module are integrated to construct an oasis eco-hydrological model, and oasis eco-hydrological simulation is realized based on the oasis eco-hydrological model.

[0010] In one possible design, the machine learning model includes one of the following: MLP model, SVR model, RF model, XGBoost model, LightGBM model, CatBoost model, DDPG model, A2C model, PPO model, and TD3 model.

[0011] In one possible design, based on a machine learning model suitable for regression modeling, modules for lake area, groundwater depth, oasis area, and vegetation cover are constructed, including:

[0012] Based on a machine learning model, a lake area module is constructed:

[0013] ;

[0014] In the formula, This indicates a machine learning algorithm suitable for regression modeling; express Lake area at any given time, in km² 2 ; express Lake area at any given time, in km² 2 ; express Ecological water transfer volume at any given time, in millions of cubic meters. 3 ; express Ecological water transfer volume at any given time, in millions of cubic meters. 3 ; express Potential evapotranspiration at any given time, in mm; express Rainfall at any given time, in mm; This indicates time; the simulation time step is 1 month.

[0015] Based on machine learning models and combined with the oasis groundwater evapotranspiration equation, a groundwater depth module is constructed:

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] In the formula, express The depth of groundwater at any given time, in meters (m). express The depth of groundwater at any given time, in meters (m). express Total water consumption by evapotranspiration at Oasis, in millions of cubic meters. 3 ; express The rate of groundwater evaporation at any given time, expressed in mm. express Oasis groundwater evapotranspiration at any given time, in mm; express Oasis vegetation cover at any given time, in % express Oasis area at any given time, in km² 2 ; , , These are empirical parameters of the submerged evaporation equation. These are empirical parameters of the oasis groundwater evapotranspiration equation, determined based on hydrogeological characteristics;

[0021] Based on a machine learning model and combined with the groundwater oasis area carrying capacity equation, an oasis area module is constructed:

[0022] ;

[0023] ;

[0024] ;

[0025] In the formula, express Oasis area at any given time, in km² 2 ; express Groundwater oasis area carrying capacity at any given time, in km² 2 ; express Time factors that constantly reflect seasonal changes in vegetation growth under the influence of temperature and phenology; express The month in which the time occurs, with a value ranging from 1 to 12; , , These are empirical parameters of the groundwater oasis area bearing capacity equation. , , These are empirical parameters of the time factor equation, determined based on the eco-hydrological characteristics of the study area.

[0026] Based on a machine learning model and combined with the groundwater-vegetation cover carrying capacity equation, a vegetation cover module is constructed:

[0027] ;

[0028] ;

[0029] ;

[0030] In the formula, express Oasis vegetation cover at any given time, in % express Groundwater vegetation cover carrying capacity at any given time, in percentages (%) The vegetation cover equation represents the vegetation cover calculated from the normalized vegetation index. , , These are empirical parameters in the groundwater-vegetation-coverage-bearing capacity equation. , , It is an empirical parameter of the vegetation cover equation, determined based on eco-hydrological characteristics;

[0031] In the process of constructing the lake area module, groundwater depth module, oasis area module, and vegetation coverage module, meteorological, hydrological, and ecological measured data are acquired. The meteorological, hydrological, and ecological measured data are divided into training and testing sets according to the proportion. The 5-fold cross-validation method is used to train each module, and the root mean square error, correlation coefficient, and / or Nash efficiency coefficient are used to evaluate the simulation accuracy.

[0032] In one possible design, the input to the groundwater depth module in the oasis eco-hydrological model includes the results of the lake area module, the oasis area module, and the vegetation cover module, and the inputs of the oasis area module and the vegetation cover module include the results of the groundwater depth module.

[0033] In one possible design, oasis eco-hydrological simulation is achieved based on the oasis eco-hydrological model, including: solving the groundwater depth by enumeration and using a recursive method to achieve continuous oasis eco-hydrological simulation.

[0034] The groundwater depth is determined by enumeration, including:

[0035] The fluctuation range of groundwater depth is determined based on measured groundwater depth data. And determine the range of groundwater burial depths enumerated as follows: , for Time of the first The depth of the groundwater is [missing information]. Based on the groundwater depth at the previous moment, set the search step size and determine the number of search steps. ;

[0036] Will Input the oasis area module and output the corresponding simulated oasis area value. ;

[0037] Will Input the vegetation cover module and output the corresponding simulated vegetation cover value. ;

[0038] Will , , Input the groundwater depth module and output the corresponding simulated groundwater depth value. ;

[0039] Pick and The smallest difference corresponds to As Simulated values ​​of groundwater depth at any time The simulated value of the groundwater depth Input oasis area module and vegetation coverage module, output Simulated values ​​of oasis area and vegetation coverage at any time , .

[0040] In one possible design, a recursive method is used to achieve continuous oasis eco-hydrological simulation, including:

[0041] Will Simulated lake area at time t is used as Input for the lake area module at any given time;

[0042] Will Simulated values ​​of groundwater depth at time t are used as Input for the groundwater depth module at all times;

[0043] Will Simulated oasis area at time t is used as Input for the Oasis area module;

[0044] Will Simulated vegetation cover value at time t as Input to the vegetation cover module at any time;

[0045] In one possible design, the method further includes:

[0046] Based on simulation accuracy evaluation and SHAP attribution analysis, a model was selected from multiple machine learning models to construct modules for lake area, groundwater depth, oasis area, and vegetation cover. These modules, along with the optimized model, were then integrated into an oasis eco-hydrological model. The methods for selecting the optimal model from multiple machine learning models based on simulation accuracy evaluation and SHAP attribution analysis included:

[0047] The simulation accuracy of the lake area module, groundwater depth module, oasis area module, and vegetation coverage module after integration during the testing period is obtained. Machine learning models with simulation accuracy exceeding a set accuracy threshold are selected as the basis for constructing the corresponding modules. The simulation accuracy is characterized by one or more of the following: root mean square error, correlation coefficient, and Nash efficiency coefficient.

[0048] SHAP attribution analysis was performed on the lake area module, and scatter plots were used to identify... SHAP value and Relationships SHAP value and Relationships SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that makes the SHAP values ​​of each input variable in the lake area module conform to the following correlation with each input variable is selected as the preferred model:

[0049] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0050] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0051] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0052] The SHAP value varies As the value increases, it tends to decrease, and the two are basically negatively correlated.

[0053] SHAP attribution analysis was conducted on the groundwater depth module, and scatter plots were used to identify [factors / items]. SHAP value and Relationships SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that makes the SHAP values ​​of each input variable of the groundwater depth module conform to the following correlation with each input variable is selected as the preferred model:

[0054] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0055] The SHAP value varies As the value increases, it tends to decrease, and the two are basically negatively correlated.

[0056] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0057] SHAP attribution analysis was performed on the oasis area module, and scatter plots were used to identify... SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that ensures the SHAP values ​​of each input variable in the oasis area module conform to the following correlation with each input variable is selected as the preferred model:

[0058] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0059] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0060] SHAP attribution analysis was performed on the vegetation cover module, and scatter plots were used to identify... SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that makes the SHAP values ​​of each input variable of the vegetation cover module conform to the following correlation with each input variable is selected as the preferred model:

[0061] The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated.

[0062] The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0063] Secondly, this application provides an oasis eco-hydrological simulation device, the device comprising:

[0064] The data construction unit is configured to acquire basic eco-hydrological data; wherein, the basic eco-hydrological data includes monthly meteorological data, hydrological data and ecological data after standardization processing, the meteorological data includes temperature, precipitation and potential evapotranspiration, the hydrological data includes ecological water transport and groundwater depth, and the ecological data includes oasis area, normalized vegetation index and vegetation coverage.

[0065] The module construction unit is configured to calculate the oasis groundwater evapotranspiration, the oasis area carrying capacity of groundwater, and the vegetation cover carrying capacity of groundwater using the eco-hydrological basic data and eco-hydrological equations, as input features for machine learning modeling, and to construct the lake area module, groundwater depth module, oasis area module, and vegetation cover module respectively based on the machine learning model suitable for regression modeling.

[0066] The hydrological model integration unit is configured to integrate the lake area module, groundwater depth module, oasis area module, and vegetation coverage module to construct an oasis eco-hydrological model, and to realize oasis eco-hydrological simulation based on the oasis eco-hydrological model.

[0067] Thirdly, embodiments of this application provide an electronic device, including: at least one processor and a memory; the memory stores computer execution instructions; the at least one processor executes the computer execution instructions stored in the memory, causing the at least one processor to perform the oasis eco-hydrological simulation method as described in the first aspect and various possible designs of the first aspect.

[0068] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the oasis eco-hydrological simulation method described in the first aspect and various possible designs of the first aspect.

[0069] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the oasis eco-hydrological simulation method described in the first aspect and various possible designs of the first aspect.

[0070] The oasis eco-hydrological simulation method, apparatus, equipment, and storage medium provided in this application have at least the following beneficial effects:

[0071] This application addresses the shortcomings of existing oasis eco-hydrological simulation studies, such as the failure to utilize machine learning for eco-hydrological feedback simulation, the failure to organically combine machine learning algorithms with eco-hydrological equations, and the lack of design models for oasis ecological water conveyance projects. It improves the methodological system for oasis eco-hydrological simulation in arid inland river basins, enhances the utilization efficiency of meteorological, hydrological, and ecological observation data such as precipitation, potential evapotranspiration, ecological water conveyance, groundwater depth, oasis area, and vegetation cover, and provides an effective tool for optimizing the total water conveyance and process design of ecological water conveyance projects. This will help improve the ecological benefits of oasis recovery and serve the ecological civilization construction and high-quality sustainable development of ecologically fragile areas in inland river basins. Attached Figure Description

[0072] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0073] Figure 1 A flowchart of an oasis eco-hydrological simulation method provided in this application embodiment.

[0074] Figure 2 The following are comparison charts showing the accuracy of lake area simulation during the test period of the Qingtuhu Oasis Eco-hydrological Model provided in this application embodiment: (a) comparison chart of root mean square error of lake area simulation by different machine learning algorithms; (b) comparison chart of correlation coefficient of lake area simulation by different machine learning algorithms; and (c) comparison chart of Nash efficiency coefficient of lake area simulation by different machine learning algorithms.

[0075] Figure 3 Comparison of groundwater depth simulation accuracy during the testing period of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment; wherein, (a) comparison of root mean square error of groundwater depth simulation by different machine learning algorithms; (b) comparison of correlation coefficient of groundwater depth simulation by different machine learning algorithms; (c) comparison of Nash efficiency coefficient of groundwater depth simulation by different machine learning algorithms.

[0076] Figure 4 The following are comparison charts showing the simulation accuracy of the oasis area during the test period of the Qingtuhu oasis eco-hydrological model provided in this application embodiment: (a) comparison chart of the root mean square error of the oasis area simulated by different machine learning algorithms; (b) comparison chart of the correlation coefficient of the oasis area simulated by different machine learning algorithms; and (c) comparison chart of the Nash efficiency coefficient of the oasis area simulated by different machine learning algorithms.

[0077] Figure 5Comparison of vegetation cover simulation accuracy during the test period of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment; wherein, (a) comparison of root mean square error of vegetation cover simulation by different machine learning algorithms; (b) comparison of correlation coefficient of vegetation cover simulation by different machine learning algorithms; (c) comparison of Nash efficiency coefficient of vegetation cover simulation by different machine learning algorithms.

[0078] Figure 6 SHAP attribution analysis diagram of the lake area module of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment; wherein, (a) in the lake area module SHAP value and (a) Scatter plot of correlations; (b) Lake area module SHAP value and (c) Scatter plot of correlation relationships in the lake area module SHAP value and Scatter plot of correlations; (d) Lake area module SHAP value and The correlation scatter plot.

[0079] Figure 7 SHAP attribution analysis diagram of the groundwater depth module of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment; wherein, (a) in the groundwater depth module SHAP value and (a) Scatter plot of correlations; (b) Groundwater depth module SHAP value and (c) Scatter plot of correlation relationships in the groundwater depth module SHAP value and The correlation scatter plot.

[0080] Figure 8 SHAP attribution analysis diagram of the oasis area module of the Qingtuhu oasis eco-hydrological model provided in this application embodiment; wherein, (a) in the oasis area module SHAP value and (a) Scatter plot of correlations; (b) Oasis area module SHAP value and The correlation scatter plot.

[0081] Figure 9 The SHAP attribution analysis diagram of the vegetation cover module of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment; wherein, (a) in the vegetation cover module SHAP value and (a) Scatter plot of correlations; (b) Vegetation cover module SHAP value and The correlation scatter plot.

[0082] Figure 10 This is a structural diagram of the oasis eco-hydrological simulation device provided in the embodiments of this application.

[0083] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0084] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0085] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0086] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0087] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0088] This application provides an oasis eco-hydrological simulation method, the overall implementation process of which is as follows: collecting and organizing oasis meteorological, hydrological, and ecological data as the basic data for constructing an oasis eco-hydrological model; using eco-hydrological equations to calculate oasis groundwater evapotranspiration, oasis area carrying capacity of groundwater, and vegetation cover carrying capacity of groundwater as input for machine learning modeling; selecting machine learning algorithms suitable for regression modeling to construct lake area modules, groundwater depth modules, oasis area modules, and vegetation cover modules respectively; using an enumeration method to solve for groundwater depth, so that it basically satisfies the groundwater depth module, oasis area module, and vegetation cover module, and using a recursive method to achieve continuous eco-hydrological simulation; comprehensively using simulation accuracy evaluation and SHAP attribution analysis to select the best machine learning algorithm, and establishing an oasis eco-hydrological model integrating machine learning and eco-hydrological equations. Figure 1 This is a flowchart of the oasis eco-hydrological simulation method provided in this application embodiment, which specifically includes the following steps S10-S40.

[0089] S10: Collect and organize monthly meteorological data such as temperature, precipitation, and potential evapotranspiration in the study area; collect and organize monthly hydrological data such as ecological water transport and groundwater depth in the study area; and collect and organize monthly ecological data such as oasis area, normalized difference vegetation index, and vegetation cover in the study area. The original meteorological, hydrological, and ecological data are standardized using the z-score method and then used to construct an oasis eco-hydrological model.

[0090] S20: Based on the basic eco-hydrological data in step S10, calculate the oasis groundwater evapotranspiration, oasis area carrying capacity of groundwater, and vegetation cover carrying capacity of groundwater using eco-hydrological equations, and use them as inputs for machine learning modeling; select machine learning algorithms / models suitable for regression modeling, and construct lake area module, groundwater depth module, oasis area module, and vegetation cover module respectively.

[0091] In some embodiments, step S20 specifically includes the following steps S201-S206.

[0092] S201: Select a machine learning algorithm or model suitable for regression modeling, including Multi-Layer Perceptron (MLP model), Support Vector Regression (SVR model), Random Forest (RF model), eXtreme Gradient Boosting (XGBoost model), Light Gradient Boosting Machine (LightGBM model), Categorical Boosting (CatBoost model), Deep Deterministic Policy Gradient (DDPG model), Advantage Actor-Critic (A2C model), Proximal Policy Optimization (PPO model), Twin Delayed Deep Deterministic Policy Gradient (TD3 model), etc.

[0093] S202: Using machine learning algorithms to construct a lake area module:

[0094] ;

[0095] In the formula, This indicates a machine learning algorithm suitable for regression modeling; express Lake area at any given time, in km² 2 ; express Lake area at any given time, in km² 2 ; express Ecological water transfer volume at any given time, in millions of cubic meters. 3 ; express Ecological water transfer volume at any given time, in millions of cubic meters. 3 ; express Potential evapotranspiration at any given time, in mm; express Rainfall at any given time, in mm; This indicates the time, with a simulation time step of 1 month.

[0096] S203: Using machine learning algorithms and combining the oasis groundwater evapotranspiration equation, a groundwater depth module is constructed.

[0097] ;

[0098] ;

[0099] ;

[0100] ;

[0101] In the formula, express The depth of groundwater at any given time, in meters (m). express The depth of groundwater at any given time, in meters (m). express Total water consumption by evapotranspiration at Oasis, in millions of cubic meters. 3 ; express The rate of groundwater evaporation at any given time, expressed in mm. express Oasis groundwater evapotranspiration at any given time, in mm; express Oasis vegetation cover at any given time, in % express Oasis area at any given time, in km² 2 ; , , These are empirical parameters of the submerged evaporation equation. These are empirical parameters of the oasis groundwater evapotranspiration equation, determined based on the hydrogeological characteristics of the study area.

[0102] S204: Using machine learning algorithms and combining the groundwater oasis area carrying capacity equation, an oasis area module is constructed.

[0103] ;

[0104] ;

[0105] ;

[0106] In the formula, express Oasis area at any given time, in km² 2 ; express Groundwater oasis area carrying capacity at any given time, in km² 2 ; express Time factors that constantly reflect seasonal changes in vegetation growth under the influence of temperature and phenology; express The month in which the time occurs, with a value ranging from 1 to 12; , , These are empirical parameters of the groundwater oasis area bearing capacity equation. , , These are empirical parameters of the time factor equation, determined based on the eco-hydrological characteristics of the study area.

[0107] S205: Using machine learning algorithms and combining the groundwater-vegetation cover carrying capacity equation, a vegetation cover module is constructed.

[0108] ;

[0109] ;

[0110] ;

[0111] In the formula, express Oasis vegetation cover at any given time, in % express Groundwater vegetation cover carrying capacity at any given time, in percentages (%) This represents the vegetation cover equation used to calculate fractional vegetation cover (FVC) from the Normalized Difference Vegetation Index (NDVI). , , These are empirical parameters in the groundwater-vegetation-coverage-bearing capacity equation. , , These are empirical parameters of the vegetation cover equation, determined based on the eco-hydrological characteristics of the study area.

[0112] S206: In constructing the lake area module, groundwater depth module, oasis area module, and vegetation cover module, 85% of the measured meteorological, hydrological, and ecological data were used to train the modules, denoted as the training set, and the corresponding period was denoted as the training period; the remaining 15% of the measured meteorological, hydrological, and ecological data were used to test the modules, denoted as the test set, and the corresponding period was denoted as the test period. A 5-fold cross-validation method was used to train each module, and the root mean square error, correlation coefficient, and Nash efficiency coefficient were used to evaluate the simulation accuracy.

[0113] S30: Based on the lake area module, groundwater depth module, oasis area module, and vegetation coverage module constructed in step S20, integrate the modules to construct an oasis eco-hydrological model: use the enumeration method to solve the groundwater depth so that it basically meets the requirements of the groundwater depth module, oasis area module, and vegetation coverage module, and use the recursive method to achieve continuous eco-hydrological simulation.

[0114] In some embodiments, step S30 specifically includes the following steps S301-S303.

[0115] S301: During the integration of various modules, the lake area module does not require results from other modules as input; the groundwater depth module requires results from the lake area module, oasis area module, and vegetation cover module as input, while the oasis area module and vegetation cover module also require results from the groundwater depth module as input. An enumeration method is used to solve for the groundwater depth to ensure that it basically meets the needs of the groundwater depth module, oasis area module, and vegetation cover module.

[0116] Specifically, step S301 includes the following steps S3011-S3015.

[0117] S3011: Analysis of its fluctuation amplitude based on measured groundwater depth data Based on this, the range of enumerated groundwater depths is determined. The search step size is 0.01 m. .

[0118] S3012: Will Input the oasis area module and output its corresponding simulated oasis area value. .

[0119] S3013: Will Input the vegetation cover module and output its corresponding simulated vegetation cover value. .

[0120] S3014: Will , , Input the groundwater depth module and output its corresponding simulated groundwater depth value. .

[0121] S3015: Take and The difference is minimal, that is The corresponding As Simulated values ​​of groundwater depth at any given time The simulated value of the groundwater depth Input oasis area module and vegetation coverage module, output Simulated values ​​of oasis area and vegetation coverage at any time , .

[0122] S302: In the process of integrating various modules, a recursive method is used to achieve continuous oasis eco-hydrological simulation.

[0123] Specifically, step S302 includes the following steps S3021-S3025.

[0124] S3021: Will Simulated lake area at time t is used as Input for the lake area module at any given time;

[0125] S3022: Will Simulated values ​​of groundwater depth at time t are used as Input for the groundwater depth module at all times;

[0126] S3023: Will Simulated oasis area at time t is used as Input for the Oasis area module;

[0127] S3024: Will Simulated vegetation cover value at time t as Input to the vegetation cover module at any time;

[0128] S3025: Based on the inputs of each module obtained in steps S3021 to S3024, the simulated values ​​of each module at continuous time points are recursively calculated to ensure the continuity and accuracy of the oasis eco-hydrological simulation.

[0129] During the recursive process of steps S3021-S3025 above, the simulated values ​​of each module are verified and adjusted in real time to improve the reliability and accuracy of the simulation results. Through continuous iteration and optimization, a complete oasis eco-hydrological simulation result is finally obtained.

[0130] S303: The root mean square error, correlation coefficient, and Nash efficiency coefficient are used to evaluate the simulation accuracy of lake area, groundwater depth, oasis area, and vegetation coverage after module integration during the training and testing periods, respectively.

[0131] S40: By comprehensively utilizing simulation accuracy evaluation and SHAP attribution analysis, an optimal algorithm is selected from machine learning algorithms such as MLP, SVR, RF, XGBoost, LightGBM, CatBoost, DDPG, A2C, PPO, and TD3 to establish an oasis eco-hydrological model that integrates machine learning and eco-hydrological equations.

[0132] In some embodiments, step S40 specifically includes the following steps S401-S406.

[0133] S401: The focus is on evaluating the simulation accuracy during the testing period after integrating the lake area module, groundwater depth module, oasis area module, and vegetation coverage module. Machine learning algorithms that achieve high simulation accuracy during the testing period after integrating each module will be selected. Among them, machine learning algorithms that achieve high simulation accuracy are those with relatively small root mean square error, relatively large correlation coefficient, and relatively large Nash efficiency coefficient.

[0134] S402: For the lake area module, conduct SHAP attribution analysis and use scatter plots to identify... SHAP value and Relationships SHAP value and Relationships SHAP value and Relationships SHAP value and The correlation between the input variables of the lake area module and the input variables should, in theory, conform to the following correlation:

[0135] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0136] (b) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0137] (c) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0138] (d) The SHAP value varies As the value increases, the value tends to decrease, and the two are basically negatively correlated.

[0139] S403: For the groundwater depth module, conduct SHAP attribution analysis and use scatter plots to identify... SHAP value and Relationships SHAP value and Relationships SHAP value and The correlation between the input variables of the groundwater depth module and the actual input variables should, in theory, conform to the following correlation:

[0140] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0141] (b) The SHAP value varies As the value increases, the value tends to decrease, and the two are basically negatively correlated.

[0142] (c) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0143] S404: For the oasis area module, perform SHAP attribution analysis and use scatter plots to identify... SHAP value and Relationships SHAP value and The correlation between the input variables of the oasis area module and the actual input variables should, in theory, conform to the following correlation:

[0144] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0145] (b) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0146] S405: For the vegetation cover module, conduct SHAP attribution analysis and use scatter plots to identify... SHAP value and Relationships The correlation between the SHAP value and the input variables of the vegetation cover module. Theoretically, the SHAP value of each input variable in the vegetation cover module should conform to the following correlation with each input variable:

[0147] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0148] (b) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0149] S406: By comprehensively selecting the lake area module, groundwater depth module, oasis area module, and vegetation coverage module, and integrating the simulation with a high accuracy and reasonable SHAP attribution interpretation during the testing period, an oasis eco-hydrological model is constructed to realize the oasis eco-hydrological simulation integrating machine learning and eco-hydrological equations.

[0150] Example 2:

[0151] Based on the oasis eco-hydrological simulation method provided in Example 1, this example further illustrates the feasibility and advancement of the method proposed in this application by combining the following specific implementation cases. This example takes the Qingtuhu oasis in the Shiyang River basin of the Hexi Corridor in Northwest my country as the application research area. The specific implementation method of applying the oasis eco-hydrological simulation method integrating machine learning and eco-hydrological equations to the Qingtuhu oasis is as follows:

[0152] Step 1: Collect and organize monthly meteorological data (temperature, precipitation, potential evapotranspiration, etc.) for Qingtuhu Oasis from 2010 to 2024; collect and organize monthly hydrological data (ecological water transport, groundwater depth, etc.) for Qingtuhu Oasis from 2010 to 2024; and collect and organize monthly ecological data (oasis area, normalized difference vegetation index, vegetation cover, etc.) for Qingtuhu Oasis from 2010 to 2024. The original meteorological, hydrological, and ecological data were standardized using the z-score method and then used to construct the Qingtuhu Oasis eco-hydrological model.

[0153] Step 2: Based on the basic eco-hydrological data of Qingtu Lake oasis from Step 1, calculate the oasis groundwater evapotranspiration, oasis area carrying capacity of groundwater, and vegetation cover carrying capacity of groundwater using eco-hydrological equations. These will serve as inputs for machine learning modeling. Select machine learning algorithms suitable for regression modeling to construct the lake area module, groundwater depth module, oasis area module, and vegetation cover module of the Qingtu Lake oasis eco-hydrological model, respectively. Specifically:

[0154] ① Select machine learning algorithms suitable for regression modeling, including MLP, SVR, RF, XGBoost, LightGBM, CatBoost, DDPG, A2C, PPO, TD3, etc.

[0155] ② Using machine learning algorithms, construct the lake area module of the Qingtuhu oasis eco-hydrological model:

[0156] ;

[0157] ③ Using machine learning algorithms and combining them with the oasis groundwater evapotranspiration equation, a groundwater depth module for the Qingtu Lake oasis eco-hydrological model was constructed:

[0158] ;

[0159] ;

[0160] ;

[0161] ;

[0162] ④ Using machine learning algorithms and combining them with the groundwater oasis area carrying capacity equation, the oasis area module of the Qingtu Lake oasis eco-hydrological model is constructed:

[0163] ;

[0164] ;

[0165] ;

[0166] ⑤ Using machine learning algorithms and combining the groundwater vegetation cover carrying capacity equation, a vegetation cover module for the Qingtuhu oasis eco-hydrological model was constructed:

[0167] ;

[0168] ;

[0169] ;

[0170] ⑥ In constructing the lake area module, groundwater depth module, oasis area module, and vegetation cover module of the Qingtuhu oasis eco-hydrological model, measured meteorological, hydrological, and ecological data from 2010 to 2022 were used for training the modules, and measured meteorological, hydrological, and ecological data from 2023 to 2024 were used for testing the modules. A 5-fold cross-validation method was used to train each module, and root mean square error, correlation coefficient, and Nash efficiency coefficient were used to evaluate the simulation accuracy.

[0171] Step 3: Based on the lake area module, groundwater depth module, oasis area module, and vegetation cover module of the Qingtu Lake oasis eco-hydrological model constructed in Step 2, integrate these modules to construct the Qingtu Lake oasis eco-hydrological model: use an enumeration method to solve for the groundwater depth, ensuring it basically satisfies the requirements of the groundwater depth module, oasis area module, and vegetation cover module; and use a recursive method to achieve continuous Qingtu Lake oasis eco-hydrological simulation. Specifically:

[0172] ① An enumeration method is used to solve for the groundwater depth, ensuring it basically satisfies the requirements of the groundwater depth module, oasis area module, and vegetation cover module. Specifically:

[0173] (a) Analyze the fluctuation range of groundwater depth based on the measured groundwater depth data of Qingtuhu Oasis, and determine the enumerated range of groundwater depth based on this. The search step size is 0.01 m. .

[0174] (b) will Input the oasis area module and output its corresponding simulated oasis area value. .

[0175] (c) will Input the vegetation cover module and output its corresponding simulated vegetation cover value. .

[0176] (d) will , , Input the groundwater depth module and output its corresponding simulated groundwater depth value. .

[0177] (e) Take and The difference is minimal, that is The corresponding As Simulated values ​​of groundwater depth at any given time The simulated value of the groundwater depth Input oasis area module and vegetation coverage module, output Simulated values ​​of oasis area and vegetation coverage at any time , .

[0178] ② In the process of integrating the various modules of the Qingtuhu Oasis Eco-hydrological Model, a recursive method is used to achieve continuous oasis eco-hydrological simulation. Specifically:

[0179] (a) will Simulated lake area at time t is used as Input for the lake area module at any given time.

[0180] (b) will Simulated values ​​of groundwater depth at time t are used as Input for the groundwater depth module at all times.

[0181] (c) will Simulated oasis area at time t is used as Input for the Oasis area module.

[0182] (d) will Simulated vegetation cover value at time t as Input to the vegetation coverage module at any time.

[0183] ③ The root mean square error, correlation coefficient, and Nash efficiency coefficient were used to evaluate the simulation accuracy of lake area, groundwater depth, oasis area, and vegetation coverage after module integration during the training and testing periods, respectively.

[0184] (4) By comprehensively utilizing simulation accuracy evaluation and SHAP attribution analysis, an optimal algorithm is selected from machine learning algorithms such as MLP, SVR, RF, XGBoost, LightGBM, CatBoost, DDPG, A2C, PPO, and TD3 to establish an integrated machine learning and eco-hydrological equation eco-hydrological model of the Qingtu Lake oasis. Specifically:

[0185] ① The focus is on examining the simulation accuracy of the integrated lake area module, groundwater depth module, oasis area module, and vegetation cover module of the Qingtuhu oasis eco-hydrological model during the testing period. Machine learning algorithms that achieve high simulation accuracy after integrating each module are selected. Among them, machine learning algorithms that achieve high simulation accuracy are those with relatively small root mean square error, relatively large correlation coefficient, and relatively large Nash efficiency coefficient. Figure 2 This is a comparison chart of the accuracy of lake area simulation during the testing period of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment. Figure 3 This is a comparison chart of the groundwater depth simulation accuracy during the testing period of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment. Figure 4 This is a comparison chart of the simulation accuracy of the oasis area during the testing period of the Qingtuhu oasis eco-hydrological model provided in this application embodiment. Figure 5 This is a comparison chart of the simulation accuracy of vegetation cover during the test period of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment. The comprehensive evaluation results show that the DDPG algorithm has high accuracy in simulating the lake area, groundwater depth, oasis area, and vegetation cover of the Qingtuhu Oasis during the test period.

[0186] ② For the lake area module of the Qingtuhu oasis eco-hydrological model constructed using the DDPG algorithm, SHAP attribution analysis was conducted, and scatter plots were used to identify... SHAP value and Relationships SHAP value and Relationships SHAP value and Relationships SHAP value and The relevant relationship. Figure 6 This is a SHAP attribution analysis diagram of the lake area module of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment. The SHAP values ​​of each input variable of the lake area module of the Qingtuhu Oasis eco-hydrological model constructed based on the DDPG algorithm conform to the following correlation with each input variable, and the SHAP attribution explanation is reasonable:

[0187] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0188] (b) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0189] (c) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0190] (d) The SHAP value varies As the value increases, the value tends to decrease, and the two are basically negatively correlated.

[0191] Step 4: Conduct SHAP attribution analysis on the groundwater depth module of the Qingtuhu Oasis eco-hydrological model constructed using the DDPG algorithm, and use scatter plots to identify... SHAP value and Relationships SHAP value and Relationships SHAP value and The relevant relationship. Figure 7 This is a SHAP attribution analysis diagram of the groundwater depth module of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment. The SHAP values ​​of each input variable of the groundwater depth module of the Qingtuhu Oasis eco-hydrological model constructed based on the DDPG algorithm conform to the following correlation with each input variable, and the SHAP attribution explanation is reasonable:

[0192] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0193] (b) The SHAP value varies As the value increases, the value tends to decrease, and the two are basically negatively correlated.

[0194] (c) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0195] ④ For the oasis area module of the Qingtuhu oasis eco-hydrological model constructed using the DDPG algorithm, SHAP attribution analysis was conducted, and scatter plots were used to identify the oasis area. SHAP value and Relationships SHAP value and The relevant relationship. Figure 8This is a SHAP attribution analysis diagram of the oasis area module of the Qingtuhu oasis eco-hydrological model provided in this application embodiment. The SHAP values ​​of each input variable of the oasis area module of the Qingtuhu oasis eco-hydrological model constructed based on the DDPG algorithm conform to the following correlation with each input variable, and the SHAP attribution explanation is reasonable:

[0196] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0197] (b) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0198] ⑤ For the vegetation cover module of the Qingtuhu Oasis eco-hydrological model constructed using the DDPG algorithm, SHAP attribution analysis was conducted, and scatter plots were used to identify... SHAP value and Relationships SHAP value and The relevant relationship. Figure 9 This is the SHAP attribution analysis diagram of the vegetation cover module of the Qingtuhu Oasis eco-hydrological model provided in this application embodiment. The SHAP values ​​of each input variable of the vegetation cover module of the Qingtuhu Oasis eco-hydrological model constructed based on the DDPG algorithm conform to the following correlation with each input variable, and the SHAP attribution explanation is reasonable:

[0199] (a) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0200] (b) The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

[0201] ⑥ By comprehensively selecting the lake area module, groundwater depth module, oasis area module, and vegetation coverage module, and integrating the simulation with the DDPG algorithm which has high accuracy and reasonable SHAP attribution interpretation during the testing period, the Qingtu Lake oasis eco-hydrological model is established, realizing the Qingtu Lake oasis eco-hydrological simulation integrating machine learning and eco-hydrological equations.

[0202] The oasis eco-hydrological simulation method integrating machine learning and eco-hydrological equations provided in this embodiment can be applied to constructing oasis eco-hydrological models driven by ecological water conveyance. Therefore, this application can overcome the shortcomings of existing oasis eco-hydrological simulation studies, such as the failure to utilize machine learning for eco-hydrological feedback simulation, the failure to organically combine machine learning algorithms with eco-hydrological equations, and the lack of design models for oasis ecological water conveyance projects. It improves the methodological system for oasis eco-hydrological simulation in arid inland river basins, enhances the utilization efficiency of meteorological, hydrological, and ecological observation data such as precipitation, potential evapotranspiration, ecological water conveyance, groundwater depth, oasis area, and vegetation cover, and provides an effective tool for optimizing the total water conveyance and water conveyance process design of ecological water conveyance projects. This helps improve the ecological benefits of oasis recovery and serves the ecological civilization construction and high-quality sustainable development of ecologically fragile areas in inland river basins.

[0203] Example 3:

[0204] This application also provides an oasis eco-hydrological simulation device, such as... Figure 10 As shown, the oasis eco-hydrological simulation device includes:

[0205] Data construction unit 1001 is configured to acquire basic eco-hydrological data; wherein, the basic eco-hydrological data includes monthly meteorological data, hydrological data and ecological data after standardization processing, the meteorological data includes temperature, precipitation and potential evapotranspiration, the hydrological data includes ecological water transport and groundwater depth, and the ecological data includes oasis area, normalized vegetation index and vegetation coverage.

[0206] The module construction unit 1002 is configured to calculate the oasis groundwater evapotranspiration, the oasis area carrying capacity of groundwater, and the vegetation coverage carrying capacity of groundwater using the eco-hydrological basic data and eco-hydrological equations, as input features for machine learning modeling, and to construct the lake area module, groundwater depth module, oasis area module, and vegetation coverage module respectively based on the machine learning model suitable for regression modeling.

[0207] The hydrological model integration unit 1003 is configured to integrate the lake area module, groundwater depth module, oasis area module and vegetation coverage module to construct an oasis eco-hydrological model, and to realize oasis eco-hydrological simulation based on the oasis eco-hydrological model.

[0208] In some embodiments, the device further includes a model optimization unit configured to select a model from multiple machine learning models based on simulation accuracy evaluation and SHAP attribution analysis, and to construct a lake area module, a groundwater depth module, an oasis area module, and a vegetation coverage module, respectively, and to integrate the oasis eco-hydrological model based on the lake area module, groundwater depth module, oasis area module, and vegetation coverage module constructed by the optimized model.

[0209] This application provides an electronic device. The electronic device may include a processor and a memory, wherein the processor and the memory can communicate; exemplarily, the processor and the memory communicate via a communication bus.

[0210] The processor executes computer execution instructions stored in memory, causing the processor to perform the scheme in the above embodiments. The processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0211] The communication bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. Transceivers are used to enable communication between database access devices and other computers (e.g., clients, read-write libraries, and read-only libraries). Memory may include random access memory (RAM) and may also include non-volatile memory.

[0212] The electronic device provided in this application embodiment can be the terminal device described in the above embodiments.

[0213] This application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer performs the technical solution of the oasis eco-hydrological simulation method described in the above embodiments.

[0214] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, it can implement the technical solution of the oasis eco-hydrological simulation method in the above embodiments.

[0215] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0216] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to implement the solution of this embodiment according to actual needs.

[0217] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The unit composed of the above modules can be implemented in hardware or in the form of hardware plus software functional units.

[0218] The integrated modules described above, implemented as software functional modules, can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods of the various embodiments of this application.

[0219] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0220] The memory may include high-speed RAM, and may also include non-volatile storage (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0221] Buses can be Industry Standard Architecture (ISA) buses, Peripheral Component Interconnect (PCI) buses, or Extended Industry Standard Architecture (EISA) buses, etc. Buses can be categorized into address buses, data buses, control buses, etc.

[0222] The aforementioned storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0223] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Alternatively, the storage medium can be an integral part of the processor. The processor and storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and storage medium can exist as discrete components in an electronic control unit or main control device.

[0224] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0225] Finally, it should be noted that the above 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 or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for simulating oasis eco-hydrology, characterized in that, The method includes: Acquire basic eco-hydrological data; wherein, the basic eco-hydrological data includes monthly meteorological data, hydrological data and ecological data after standardization processing, the meteorological data includes temperature, precipitation and potential evapotranspiration, the hydrological data includes ecological water transport and groundwater depth, and the ecological data includes oasis area, normalized vegetation index and vegetation cover. Based on the aforementioned eco-hydrological basic data, the eco-hydrological equations are used to calculate the oasis groundwater evapotranspiration, the oasis area carrying capacity of groundwater, and the vegetation cover carrying capacity of groundwater. These are used as input features for machine learning modeling. Based on a machine learning model suitable for regression modeling, modules for lake area, groundwater depth, oasis area, and vegetation cover are constructed respectively. The lake area module, groundwater depth module, oasis area module, and vegetation coverage module are integrated to construct an oasis eco-hydrological model, and oasis eco-hydrological simulation is realized based on the oasis eco-hydrological model. Based on a machine learning model, a lake area module is constructed: ； In the formula, This indicates a machine learning algorithm suitable for regression modeling; express Lake area at any given time, in km² 2 ; express Lake area at any given time, in km² 2 ; express Ecological water transfer volume at any given time, in millions of cubic meters. 3 ; express Ecological water transfer volume at any given time, in millions of cubic meters. 3 ; express Potential evapotranspiration at any given time, in mm; express Rainfall at any given time, in mm; express The difference between potential evapotranspiration and precipitation at any given time, in mm; This indicates time; the simulation time step is 1 month. The method further includes: Based on simulation accuracy evaluation and SHAP attribution analysis, a model was selected from multiple machine learning models to construct modules for lake area, groundwater depth, oasis area, and vegetation cover. These modules, along with the optimized model, were then integrated into an oasis eco-hydrological model. The methods for selecting the optimal model from multiple machine learning models based on simulation accuracy evaluation and SHAP attribution analysis included: The simulation accuracy of the lake area module, groundwater depth module, oasis area module, and vegetation coverage module after integration during the testing period is obtained. Machine learning models with simulation accuracy exceeding a set accuracy threshold are selected as the basis for constructing the corresponding modules. The simulation accuracy is characterized by one or more of the following: root mean square error, correlation coefficient, and Nash efficiency coefficient. SHAP attribution analysis was performed on the lake area module, and scatter plots were used to identify... SHAP value and Relationships SHAP value and Relationships SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that makes the SHAP values ​​of each input variable in the lake area module conform to the following correlation with each input variable is selected as the preferred model: The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. The SHAP value varies As the value increases, the value tends to decrease, and the two are basically negatively correlated.

2. The oasis eco-hydrological simulation method according to claim 1, characterized in that, The machine learning model includes one of the following: MLP model, SVR model, RF model, XGBoost model, LightGBM model, CatBoost model, DDPG model, A2C model, PPO model, and TD3 model.

3. The oasis eco-hydrological simulation method according to claim 1, characterized in that, Based on machine learning models suitable for regression modeling, modules for groundwater depth, oasis area, and vegetation cover are constructed, including: Based on machine learning models and combined with the oasis groundwater evapotranspiration equation, a groundwater depth module is constructed: ； ； ； ； In the formula, express The depth of groundwater at any given time, in meters (m). express The depth of groundwater at any given time, in meters (m). express Total water consumption by evapotranspiration at Oasis, in millions of cubic meters. 3 ; express The rate of groundwater evaporation at any given time, expressed in mm; express Oasis groundwater evapotranspiration at any given time, in mm; express Oasis vegetation cover at any given time, in % express Oasis area at any given time, in km² 2 ; , , These are empirical parameters of the submerged evaporation equation. These are empirical parameters of the oasis groundwater evapotranspiration equation, determined based on hydrogeological characteristics; Based on a machine learning model and combined with the groundwater oasis area carrying capacity equation, an oasis area module is constructed: ； ； ； In the formula, express Oasis area at any given time, in km² 2 ; express Groundwater oasis area carrying capacity at any given time, in km² 2 ; express Time factors that constantly reflect seasonal changes in vegetation growth under the influence of temperature and phenology; express The month in which the time occurs, with a value ranging from 1 to 12; , , These are empirical parameters of the groundwater oasis area bearing capacity equation. , , These are empirical parameters of the time factor equation, determined based on the eco-hydrological characteristics of the study area. Based on a machine learning model and combined with the groundwater-vegetation cover carrying capacity equation, a vegetation cover module is constructed: ； ； ； In the formula, express Oasis vegetation cover at any given time, in % express Groundwater vegetation cover carrying capacity at any given time, in percentages (%) The vegetation cover equation represents the vegetation cover calculated from the normalized vegetation index. , , These are empirical parameters in the groundwater-vegetation-coverage-bearing capacity equation. , , It is an empirical parameter of the vegetation cover equation, determined based on eco-hydrological characteristics; In the process of constructing the lake area module, groundwater depth module, oasis area module, and vegetation coverage module, meteorological, hydrological, and ecological measured data are acquired. The meteorological, hydrological, and ecological measured data are divided into training and testing sets according to the proportion. The 5-fold cross-validation method is used to train each module, and the root mean square error, correlation coefficient, and / or Nash efficiency coefficient are used to evaluate the simulation accuracy.

4. The oasis eco-hydrological simulation method according to claim 1, characterized in that, In the oasis eco-hydrological model, the inputs to the groundwater depth module include the results of the lake area module, the oasis area module, and the vegetation coverage module. The inputs to the oasis area module and the vegetation coverage module include the results of the groundwater depth module.

5. The oasis eco-hydrological simulation method according to claim 4, characterized in that, Based on the oasis eco-hydrological model, oasis eco-hydrological simulation is realized, including: solving the groundwater depth by enumeration method and realizing continuous oasis eco-hydrological simulation by recursion method. The groundwater depth is determined by enumeration, including: The fluctuation range of groundwater depth is determined based on measured groundwater depth data. And determine the range of groundwater burial depths enumerated as follows: , for Time of the first The depth of the groundwater is [missing information]. Based on the groundwater depth at the previous moment, set the search step size and determine the number of search steps. ; Will Input the oasis area module and output the corresponding simulated oasis area value. ; Will Input the vegetation cover module and output the corresponding simulated vegetation cover value. ; Will , , Input the groundwater depth module and output the corresponding simulated groundwater depth value. ; Pick and The smallest difference corresponds to As Simulated values ​​of groundwater depth at any time The simulated value of the groundwater depth Input oasis area module and vegetation coverage module, output Simulated values ​​of oasis area and vegetation coverage at any time , .

6. The oasis eco-hydrological simulation method according to claim 5, characterized in that, A recursive method is used to achieve continuous oasis eco-hydrological simulation, including: Will Simulated lake area at time t is used as Input for the lake area module at any given time; Will Simulated values ​​of groundwater depth at time t are used as Input for the groundwater depth module at all times; Will Simulated oasis area at time t is used as Input for the Oasis area module; Will Simulated vegetation cover value at time t as Input to the vegetation coverage module at any time.

7. The oasis eco-hydrological simulation method according to claim 3, characterized in that, SHAP attribution analysis was conducted on the groundwater depth module, and scatter plots were used to identify [factors / items]. SHAP value and Relationships SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that makes the SHAP values ​​of each input variable of the groundwater depth module conform to the following correlation with each input variable is selected as the preferred model: The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. The SHAP value varies As the value increases, it tends to decrease, and the two are basically negatively correlated. The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. SHAP attribution analysis was performed on the oasis area module, and scatter plots were used to identify... SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that ensures the SHAP values ​​of each input variable in the oasis area module conform to the following correlation with each input variable is selected as the preferred model: The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. SHAP attribution analysis was performed on the vegetation cover module, and scatter plots were used to identify... SHAP value and Relationships SHAP value and Based on the correlation, the machine learning model that makes the SHAP values ​​of each input variable of the vegetation cover module conform to the following correlation with each input variable is selected as the preferred model: The SHAP value varies The value tends to increase as it increases, and the two are basically positively correlated. The SHAP value varies The value tends to increase as the value increases, and the two are basically positively correlated.

8. An oasis eco-hydrological simulation device, used to implement the oasis eco-hydrological simulation method as described in any one of claims 1-7, characterized in that, The device includes: The data construction unit is configured to acquire basic eco-hydrological data; wherein, the basic eco-hydrological data includes monthly meteorological data, hydrological data and ecological data after standardization processing, the meteorological data includes temperature, precipitation and potential evapotranspiration, the hydrological data includes ecological water transport and groundwater depth, and the ecological data includes oasis area, normalized vegetation index and vegetation coverage. The module construction unit is configured to calculate the oasis groundwater evapotranspiration, the oasis area carrying capacity of groundwater, and the vegetation cover carrying capacity of groundwater using the eco-hydrological basic data and eco-hydrological equations, as input features for machine learning modeling, and to construct the lake area module, groundwater depth module, oasis area module, and vegetation cover module respectively based on the machine learning model suitable for regression modeling. The hydrological model integration unit is configured to integrate the lake area module, groundwater depth module, oasis area module, and vegetation coverage module to construct an oasis eco-hydrological model, and to realize oasis eco-hydrological simulation based on the oasis eco-hydrological model.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes the computer execution instructions stored in the memory to implement the oasis eco-hydrological simulation method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the oasis eco-hydrological simulation method as described in any one of claims 1-7.

Images