A water allocation method in arid regions based on river loss simulation

Abstract

The present application belongs to the technical field of water allocation in arid regions, and discloses a water allocation method in arid regions based on river loss simulation, comprising: calculating the measured river loss of a basin according to the measured runoff data of the basin in the arid region; constructing a river loss prediction model of the basin according to the hydro-meteorological characteristics of the basin and the corresponding measured river loss; predicting the river loss of the basin according to the river loss prediction model to obtain river loss prediction information; setting a constraint condition according to the river loss prediction information, constructing a water resources optimal allocation model of the arid region based on the constraint condition, and generating a water allocation scheme of the arid region according to the water resources optimal allocation model. Through the river loss prediction model of machine learning, the present application realizes high-precision prediction of river loss, embeds it into the water resources optimal allocation model, and constructs a water allocation method in arid regions based on river loss process simulation suitable for high river loss basins, so that the water resources allocation in arid regions is more reasonable.

Description

TECHNICAL FIELD

[0001] The application discloses a water quantity allocation method based on river loss simulation in a dry area and belongs to the technical field of water quantity allocation in a dry area. BACKGROUND

[0002] River loss refers to the water loss caused by various natural factors during the runoff evolution along the river channel. The water loss after these losses no longer directly participates in the subsequent runoff process of the river channel, which has an important influence on water resource management and allocation. In particular, in arid and semi-arid areas, due to the strong permeability of the river channel underlying surface and the strong evaporation, the river loss phenomenon is particularly significant. Therefore, reasonable and accurate river loss calculation is particularly important in water resource allocation.

[0003] The main influencing factors of river loss include the geometric characteristics of the river channel (such as river length, width, longitudinal slope, etc.), the geological conditions of the riverbed and the two banks (such as soil type, permeability coefficient, etc.), the climate conditions (such as precipitation, evaporation, etc.), and the hydrological conditions (such as upstream inflow, flow velocity, etc.). The traditional river loss calculation methods mainly include the river loss rate method, the unit river length river loss method, and the fitting nonlinear equation method. Although these methods are widely used in practical applications, they all have their own shortcomings.

[0004] River loss rate method: river loss is calculated by setting a fixed river loss rate. This method is simple and easy to implement, but it ignores the differences in river loss under different river sections and different hydrological conditions, resulting in water quantity remaining problems in the calculation results. Unit river length river loss method: the loss amount per unit river length is used to represent river loss. Although the influence of river length is considered, the influence of upstream inflow, flow velocity and other key factors on river loss is not fully reflected, so the calculation result is not accurate enough. Fitting nonlinear equation method: a nonlinear equation is fitted to describe the relationship between river loss and various influencing factors. This method is more flexible in theory, but due to the complexity of actual hydrological conditions, it is often difficult to find a high-precision fitting equation, resulting in low accuracy of the calculation result. SUMMARY

[0005] The purpose of the present application is to provide a water quantity allocation method based on river loss simulation in a dry area to solve the technical problem of low calculation accuracy caused by the river loss calculation method in the prior art. To achieve the above purpose, the present application provides a water quantity allocation method based on river loss simulation in a dry area, and the specific scheme is as follows:

[0006] A water quantity allocation method based on river loss simulation in a dry area, comprising:

[0007] S1, calculating the measured river loss amount of the basin according to the measured runoff data of the basin in the dry area;

[0008] S2, constructing a river loss prediction model of the basin according to the hydro-meteorological characteristics of the basin and the corresponding measured river loss amount of the basin;

[0009] S3, performing river loss prediction on the basin according to the river loss prediction model to obtain river loss prediction information;

[0010] S4, setting a constraint condition according to the river loss prediction information, constructing a water resource optimal allocation model of the arid region based on the constraint condition, and generating a water quantity allocation scheme of the arid region according to the water resource optimal allocation model.

[0011] Preferably, the S1 specifically comprises:

[0012] Obtaining the upstream section inflow, downstream section outflow and interval water diversion quantity of the basin in the arid region;

[0013] Calculating the measured river loss amount of the basin based on the water balance principle.

[0014] Preferably, S2 specifically comprises:

[0015] S21, constructing a feature sliding window and extracting a feature vector of the time series hydro-meteorological characteristics of the basin;

[0016] S22, taking the feature vector as input and the corresponding measured river loss amount as label to construct the river loss prediction model of the basin.

[0017] Preferably, after S21 and before S22, it further comprises:

[0018] Using a multi-dimensional feature clustering hierarchical sampling strategy to perform multi-dimensional clustering on the feature vector;

[0019] Extracting a preset proportion of samples from each layer of the multi-dimensional clustering as a test set, and the remaining samples as a training set;

[0020] Correspondingly, in S22, taking the feature vector in the training set as input and the corresponding measured river loss amount as label to construct the river loss prediction model of the basin.

[0021] Preferably, the S22 further comprises:

[0022] S221, dynamically updating the sampling distribution of the hyperparameters of the to-be-constructed river loss prediction model based on the historical training result set, and generating a hyperparameter candidate set;

[0023] S222, generating a dynamic parameter tuning instruction set of the to-be-constructed river loss prediction model according to the hyperparameter candidate set;

[0024] S223, optimizing the hyperparameters of the river loss prediction model according to the dynamic parameter tuning instruction set.

[0025] Preferably, S221 specifically comprises:

[0026] Adjusting hyperparameters of the river loss prediction model to be constructed automatically through a Bayesian optimization framework;

[0027] Updating a sampling distribution of the hyperparameters according to historical parameter test results;

[0028] Generating a hyperparameter candidate set of the river loss prediction model to be constructed according to the updated sampling distribution.

[0029] Preferably, S222 specifically comprises:

[0030] According to a preset loss convergence threshold, eliminating parameter groups in the hyperparameter candidate set that do not meet the standard, and recording a set of parameter groups that meet the standard as a first candidate parameter group;

[0031] Based on a parameter performance verification matrix, filtering parameter groups that meet a preset standard in precision from the first candidate parameter group;

[0032] Generating a dynamic parameter adjustment instruction set of the river loss prediction model to be constructed according to the filtered parameter groups.

[0033] Preferably, a constraint condition is set according to the river loss prediction information, and a water resource optimization allocation model of the arid region is constructed based on the constraint condition, specifically comprising:

[0034] S41, constructing a water resource allocation model according to the multi-user water demand of the arid region;

[0035] S42, constructing a reservoir scheduling model according to the reservoir operation rules of the arid region;

[0036] S43, coupling the water resource allocation model and the reservoir scheduling model, and establishing a water loss constraint based on the river loss prediction information to construct a water resource optimization allocation model.

[0037] Preferably, the hydro-meteorological characteristics include:

[0038] River section historical inflow, river section current inflow, current water diversion amount, river section current groundwater depth, and current evapotranspiration.

[0039] Preferably, coupling the water resource allocation model and the reservoir scheduling model specifically comprises:

[0040] Taking the reservoir discharge and node water supply of the arid region as decision variables, and taking the minimum multi-user water shortage of the arid region as the optimization objective to couple the model.

[0041] Beneficial effects: the river loss prediction model of the application realizes high-precision prediction of river loss, and embeds the river loss prediction model into a water resource optimization allocation model, constructs a water quantity allocation method suitable for a high river loss basin and based on river loss process simulation in an arid region, so that the water resource allocation in the arid region is more reasonable. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The river loss calculation result of the Alar-Xinjimana section in the embodiment of the application;

[0043] Figure 2 The river loss calculation result of the Xinjimana-Yingbazha section in the embodiment of the application;

[0044] Figure 3 The calculation result of the Yingbazha-Wusiman section in the embodiment of the application;

[0045] Figure 4 The river loss calculation result of the Wusiman-Qiara section in the embodiment of the application;

[0046] Figure 5 The deep neural network structure diagram in the embodiment of the application;

[0047] Figure 6 The comparison diagram of the measured result and the predicted result of the river loss of the Alar-Xinjimana section in the embodiment of the application;

[0048] Figure 7 The comparison diagram of the measured result and the predicted result of the river loss of the Xinjimana-Yingbazha section in the embodiment of the application;

[0049] Figure 8 The comparison diagram of the measured result and the predicted result of the river loss of the Yingbazha-Wusiman section in the embodiment of the application;

[0050] Figure 9 The comparison diagram of the measured result and the predicted result of the river loss of the Wusiman-Qiara section in the embodiment of the application;

[0051] Figure 10 The water resource allocation node schematic diagram in the embodiment of the application;

[0052] Figure 11 The operation process schematic diagram of the aggregate reservoir in each typical year in the embodiment of the application, wherein (i) represents the operation process in a wet year, (j) represents the operation process in an average year, and (k) represents the operation process in a dry year;

[0053] Figure 12 The operation process schematic diagram of the Pamann reservoir in each typical year in the embodiment of the application, wherein (i) represents the operation process in a wet year, (j) represents the operation process in an average year, and (k) represents the operation process in a dry year;

[0054] Figure 13 A flowchart of the present application is shown. DETAILED DESCRIPTION

[0055] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the protection scope of the present application.

[0056] A water allocation method in an arid region based on river loss simulation, as shown in Figure 13 , the method comprises the following steps.

[0057] S1, calculating the measured river loss of the basin according to the measured runoff data of the basin in the arid region;

[0058] Further, the upstream section inflow, downstream section outflow and interval water diversion of the basin reach in the arid region are obtained; and the measured river loss of the basin reach is calculated based on the water balance principle.

[0059] Specifically, the basin in the arid region comprises a plurality of reaches, and the runoff of each reach is directly or indirectly connected. In the study of the upper and middle reaches of the Tarim River mainstream, the basin nodes are as shown in Figure 10 .

[0060] Specifically, the measured runoff data can include the inflow of the upstream section of the basin reach, the outflow of the downstream section, and the interval water diversion of the interval between the upstream section and the downstream section, so the measured river loss of the basin can be calculated based on the water balance principle. Specifically, in the study of the upper and middle reaches of the Tarim River mainstream, the calculated river loss of the Alar-Xinjimana reach, Xinjimana-Yingbazha reach, Yingbazha-Wusimana reach and Wusimana-Qiela reach are as shown in Figure 1 、 Figure 2 、 Figure 3 and Figure 4 .

[0061] Specifically, the calculation formula of the measured river loss is as follows:

[0062]

[0063] In the formula,

[0064] is the river loss, i.e. the measured river loss (10 8 m 3 );

[0065] and are the runoff of the upstream and downstream sections (10 8 m 3 );

[0066] for the water diversion amount (10 8 m 3 ).

[0067] S2, constructing a river loss prediction model of the basin according to the hydro-meteorological characteristics of the basin and the corresponding measured river loss amount;

[0068] Further, S2 can be specifically:

[0069] S21, constructing a feature sliding window, and extracting a feature vector of the time-series hydro-meteorological characteristics of the basin;

[0070] S22, constructing a river loss prediction model of the basin by taking the feature vector as input and the corresponding measured river loss amount as label.

[0071] Specifically, the input features are obtained, i.e., the time-series hydro-meteorological characteristics, which can include river section historical inflow, river section current inflow, current diversion amount, river section current groundwater depth, and current evapotranspiration. It can also include meteorological data (precipitation, air temperature, humidity, and wind speed), vegetation coverage index along the coast, river channel form coefficient, etc.

[0072] Specifically, the river section historical inflow represents the total water inflow into the basin in a certain historical period, which reflects the influence of the previous inflow on the current river section water storage state. The river section current inflow represents the total water inflow into the target basin in the current period, which directly affects the water balance of the river section and is the core input of river loss calculation. The current diversion amount represents the water artificially diverted in the target basin in the current period, including irrigation, industrial water intake, etc., which reflects that human intervention will directly reduce the remaining water amount of the river section. The river section current groundwater depth represents the average depth of the groundwater level around the river section in the current period, which reflects the interaction between groundwater and river water (such as recharge or discharge). Generally, when the river section current groundwater depth is shallow, the river section is more prone to seepage to groundwater. The current evapotranspiration represents the sum of river surface evaporation and vegetation transpiration in the current period, which directly affects the water loss of the river section, especially in arid regions.

[0073] The construction of the feature sliding window can include:

[0074] Specifically, a time window with a preset step size is constructed, and the sliding is performed according to the preset step size. In this embodiment, in order not to damage the natural correlation between the input features, no further feature selection is performed.

[0075] Standardize and normalize the temporal hydrological and meteorological features to unify the dimensions of different features. It is also necessary to align all features to the same river section and the same time period to avoid spatial or temporal misalignment. In this embodiment, the water inflow of the river section last month (runoff of the previous month), the water inflow of the river section this month (runoff of the current month), the water diversion this month (water diversion), the groundwater depth this month (groundwater), and the evapotranspiration this month (evaporation) are used as inputs in the input layer, and the river loss obtained by the above calculation is used as a label. The river loss prediction model is trained based on the machine learning model, and its formula is as follows:

[0076]

[0077] Where:

[0078] This is the water from the river section last month;

[0079] This is the water inflow of the river section this month;

[0080] The amount of water diverted this month;

[0081] The groundwater depth this month;

[0082] is the evaporation volume for this month;

[0083] This is the river loss for this month;

[0084] is the previous step size of the feature sliding window.

[0085] It should be noted that the above inputs and their corresponding labels are divided into training sets and test sets.

[0086] In this embodiment, the machine learning model specifically uses a deep neural network. Specifically, the input layer, hidden layer, and output layer of the deep neural network are set, the previous step size, learning rate, number of training epochs, number of hidden layers, activation function, output features and other hyperparameters of the deep neural network are initialized, and the weights and biases of the deep neural network are optimized. The structural diagram of the deep neural network is shown in FIG. Figure 5 shown.

[0087] To avoid model deviation due to insufficient samples in sparse areas in the multidimensional feature space, this embodiment further includes the following steps after S21:

[0088] Performing multidimensional clustering on the feature vectors using a stratified sampling strategy of multidimensional feature clustering;

[0089] Extracting a preset proportion of samples from each layer of the multidimensional clustering as a test set, and the remaining samples as a training set;

[0090] Correspondingly, in S22, a river loss prediction model of the river basin is constructed by taking the feature vectors in the training set as input and taking the corresponding measured river loss as label.

[0091] Specifically, the K-Means clustering algorithm is used to divide all samples into 10 layers with similar feature spaces. In each layer, if the number of samples is greater than 1, test samples are randomly selected at a proportion of 20% (at least one is guaranteed), and the remaining samples are assigned to the training set. If there is only one sample in a layer, it is directly assigned to the training set. This method can significantly improve the generalization ability and prediction stability of the model.

[0092] In this embodiment, the specific implementation steps of the K-Means clustering process are as follows: first, standardize the input data for preprocessing; then randomly initialize 10 center points (i.e., randomly select 10 samples as initial cluster centers). Then enter the iterative optimization phase, which includes the following steps:

[0093] Sample assignment: assign each sample to the nearest cluster center to form a temporary clustering result;

[0094] Center point update: recalculate the center points of each cluster based on the current clustering result.

[0095] The iteration termination condition is any of the following conditions: the cluster centers no longer change significantly (convergence); the maximum number of iterations is reached.

[0096] If the termination condition is not met, return to continue optimization; if the condition is met, divide the data into different levels according to the final clustering result.

[0097] After the layering is completed, the training set and the test set are divided in each layer according to the proportion. Finally, the training sets of all layers are combined as the global training set, and the test sets of all layers are combined as the global test set. If the clustering result does not meet the expectation (e.g., does not converge), the center points need to be randomly initialized again and the above process needs to be repeated until the termination condition is met before layering and data set division.

[0098] To efficiently search for the hyperparameters of the model, S22 of this embodiment further includes:

[0099] S221, based on the historical training result set, dynamically update the sampling distribution of the hyperparameters of the river loss prediction model to be constructed, and generate a hyperparameter candidate set;

[0100] Specifically, the hyperparameters of the river loss prediction model to be constructed are automatically adjusted through a Bayesian optimization framework;

[0101] The sampling distribution of the hyperparameters is updated according to the historical parameter test results;

[0102] Based on the updated sampling distribution, a candidate set of hyperparameters for the river loss prediction model to be constructed is generated.

[0103] S222: Generate a dynamic parameter adjustment instruction set for the river loss prediction model to be constructed based on the candidate set of hyperparameters;

[0104] Specifically, according to a preset loss convergence threshold, the parameter groups that do not meet the criteria in the hyperparameter candidate set are eliminated, and the set of parameter groups that meet the criteria is recorded as the first candidate parameter group;

[0105] Based on the parameter performance verification matrix, screening a parameter group whose accuracy meets a preset standard from the first candidate parameter group;

[0106] Based on the screened parameter group, a dynamic parameter adjustment instruction set of the river loss prediction model to be constructed is generated.

[0107] S223: Optimizing the hyperparameters of the river loss prediction model according to the dynamic parameter adjustment instruction set. This method improves the prediction performance of the model.

[0108] In this example, the Optuna hyperparameter optimization process is used. Specifically, the objective function and hyperparameter search space are defined, an optimization algorithm is selected, and an iterative optimization loop is initiated. In each iteration, Optuna generates a hyperparameter combination based on the current sampling strategy, evaluates model performance based on the objective function, and updates the parameter search direction based on the evaluation results. During the loop, the number of iterations is continuously monitored. If the number reaches a preset maximum (e.g., 100), the search is terminated and the globally optimal hyperparameter combination and its corresponding objective function value are output. If the upper limit is not reached, the parameter generation, performance evaluation, and model update process are repeated until the termination condition is met. The final output contains the optimal parameter combination and the corresponding optimal objective function value.

[0109] After training the river loss prediction model, the process also includes loading the trained model and performing a single-step rolling simulation on the test set. After testing the model, the process also includes calculating the simulation error and evaluating the simulation results. In this embodiment, the evaluation results are shown in Table 1. As can be seen from Table 1, the simulation accuracy of this embodiment is higher than that of the traditional polynomial model.

[0110] Table 1: Accuracy results of two models

[0111]

[0112] S3. Predicting river damage in the basin according to the river damage prediction model to obtain river damage prediction information;

[0113] Specifically, in the study of the upper and middle reaches of the Tarim River, the measured river losses in each river section were compared with the predicted river losses, such asFigure 6 、 Figure 7 、 Figure 8 and Figure 9 .

[0114] S4, establishing a constraint condition according to the river loss prediction information, constructing a water resource optimal allocation model of the arid region based on the constraint condition, and generating a water allocation scheme of the arid region according to the water resource optimal allocation model.

[0115] S41, constructing a water resource allocation model according to the multi-user water demand of the arid region.

[0116] Specifically, S41 can include:

[0117] S411, collecting the information of water users in the basin, determining the demand of water users at different time periods, and determining agricultural and ecological water demand.

[0118] S412, according to the characteristics and distribution of water conservancy projects in the basin, the reservoirs, different water users and rivers in the basin are unified into a topological node graph, and specifically, in the study of the upper and middle reaches of the Tarim River mainstream, the node graph is as shown in Figure 10 .

[0119] S413, analyzing the water resource allocation situation of different typical years, and specifically, the water resource allocation situation of different typical years is as shown in Table 2.

[0120] Table 2 Water resource allocation situation of each typical year

[0121]

[0122] S42, constructing a reservoir scheduling model according to the reservoir operation rules of the arid region.

[0123] Specifically, S42 can include:

[0124] S421, collecting the characteristic values of the plain reservoirs in the basin, and clarifying the functions and operation rules of the plain reservoirs.

[0125] S422, clarifying the operation constraint conditions of the plain reservoirs in the scheduling period, and the constraints include water balance constraints, reservoir capacity constraints, reservoir discharge constraints, etc.

[0126] S423, establishing a plain reservoir scheduling model, and analyzing the reservoir operation process under different typical years, and specifically, in the study of the upper and middle reaches of the Tarim River mainstream, the operation processes of different reservoirs (aggregated reservoirs and Pamun reservoirs) under different typical years are as shown in Figure 11 、 Figure 12 .

[0127] S43, coupling the water resource allocation model and the reservoir operation model, and establishing a water loss constraint with the river loss prediction information to construct a water resource optimization allocation model.

[0128] Specifically, the reservoir discharge and the node water supply in the arid region are taken as decision variables, and the minimum multi-user water shortage in the arid region is taken as an optimization target for model coupling.

[0129] The objective function of the water resource optimization allocation model is as follows:

[0130]

[0131] In the formula:

[0132] is the minimum water shortage of the user;

[0133] is the agricultural water shortage of the i-th region in the j-th period;

[0134] is the ecological water shortage of the i-th region in the j-th period. In the water resource optimization allocation period, the reservoir operation constraint condition and the water resource allocation constraint condition are as follows:

[0135] The reservoir water balance constraint is as follows:

[0136]

[0137]

[0138] In the formula:

[0139] is the reservoir capacity at the end of the i-th period in the j-th year; is the reservoir inflow in the j-th period in the j-th year;

[0140] is the discharge in the j-th period in the j-th year.

[0141] is the reservoir capacity constraint as follows:

[0142]

[0143]

[0144] In the formula:

[0145] ​​​​​​​​​​ For the Reservoir capacity during the period, billion m 3 ;

[0146] is the minimum storage capacity of the reservoir, 10,000 m 3 ;

[0147] is the maximum storage capacity of the reservoir, 10,000 m 3 .

[0148] The downstream flow constraint is as follows:

[0149]

[0150] Where:

[0151] yes Minimum downstream flow requirement within the time period;

[0152] It is years The outflow from the reservoir at the end of the period;

[0153] yes The maximum downstream flow demand during the time period.

[0154] The node water balance constraint is as follows:

[0155]

[0156] Where:

[0157] For the period The amount of water in the main stream after the node has supplied water, 10,000 m 3 ;

[0158] For the period Node inflow flow, m 3 / s;

[0159] For the period Node water supply, 10,000 m 3 ;

[0160] For the period The river loss amount calculated by the node through the river loss prediction model is the river loss prediction information.

[0161] It should be noted that in actual application, the river loss prediction model corresponding to each runoff in the arid region is different, and the river loss amount corresponding to each node in the condition constraint is obtained according to the river loss prediction model corresponding to the node.

[0162] The node water supply capacity constraint is as follows:

[0163]

[0164] In the formula:

[0165] is the time period Node water supply amount, 10,000 m 3 ;

[0166] is the time period Flow capacity of the water diversion hub, m 2 / s.

[0167] Non-negative constraint, specifically, all variables in the water resource optimization allocation model are non-negative values.

[0168] The present application realizes high-precision prediction of river loss through the river loss prediction model of machine learning, and embeds it into the water resource optimization allocation model, constructs an arid region water allocation method based on river loss process simulation suitable for high river loss basins, so that the water resource allocation in the arid region is more reasonable, and the water supply scheme of the water diversion hub is optimized, so that the reservoir can efficiently release the stored water, assist the water diversion hub to solve the problem of insufficient water supply in the basin, so as to realize reasonable water use of each water use department in the basin, and promote high-quality and sustainable economic and ecological development of the basin.

[0169] The above is only a few embodiments of the present application, and does not limit the present application in any form. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any skilled person in the art can make some changes or modifications to the above disclosed technical content without departing from the scope of the technical solution of the present application, which are equivalent to equivalent embodiments, and belong to the scope of the technical solution.

Claims

1. A water allocation method for arid areas based on river loss simulation, characterized in that: include: S1. Calculating the measured river loss in a river basin in the arid area based on the measured runoff data of the river basin; S2. constructing a river loss prediction model for the river basin based on the hydrological and meteorological characteristics of the river basin and the corresponding measured river loss amount; S2 specifically includes: S21, constructing a feature sliding window to extract the feature vector of the temporal hydrological and meteorological characteristics of the basin; S22. Using the characteristic vector as input and the corresponding measured river loss amount as a label, construct a river loss prediction model for the basin; S3. Predicting river damage in the basin according to the river damage prediction model to obtain river damage prediction information; S4. Establishing constraint conditions according to the river loss prediction information, and constructing a water resource optimization allocation model for the arid area based on the constraint conditions, specifically including: S41, constructing a water resource allocation model according to the water demand of multiple users in the arid area; S42, constructing a reservoir operation model according to the reservoir operation rules in the arid area; S43, coupling the water resource allocation model and the reservoir scheduling model, establishing water loss constraints based on the river loss prediction information, and constructing a water resource optimization allocation model; A water allocation plan for the arid area is generated according to the water resource optimization allocation model.

2. The method for water allocation in arid areas according to claim 1, characterized in that: Said S1 specifically includes: Obtain the upstream section inflow, downstream section outflow and interval water diversion of river sections in arid areas; The measured river loss of the river section in the basin is calculated based on the water balance principle.

3. The method for water allocation in arid areas according to claim 1, characterized in that: After S21 and before S22, the following further includes: Performing multidimensional clustering on the feature vectors using a stratified sampling strategy of multidimensional feature clustering; Extracting a preset proportion of samples from each layer of the multidimensional clustering as a test set, and the remaining samples as a training set; Correspondingly, in S22, a river loss prediction model for the basin is constructed using the feature vector in the training set as input and the corresponding measured river loss amount as a label.

4. The method for water allocation in arid areas according to claim 1, characterized in that: The S22 further includes: S221. Based on the historical training result set, dynamically update the sampling distribution of the hyperparameters of the river loss prediction model to be constructed, and generate a candidate set of hyperparameters; S222: Generate a dynamic parameter adjustment instruction set for the river loss prediction model to be constructed based on the candidate set of hyperparameters; S223. Optimize the hyperparameters of the river loss prediction model according to the dynamic parameter adjustment instruction set.

5. The method for water allocation in arid areas according to claim 4, characterized in that: The S221 specifically includes: Automatically adjusting hyperparameters of the river loss prediction model to be constructed through a Bayesian optimization framework; Updating the sampling distribution of the hyperparameters based on historical parameter test results; Based on the updated sampling distribution, a candidate set of hyperparameters for the river loss prediction model to be constructed is generated.

6. The method for water allocation in arid areas according to claim 4, characterized in that: S222 specifically includes: Eliminate the parameter groups that do not meet the criteria from the hyperparameter candidate set according to a preset loss convergence threshold, and record the set of parameter groups that meet the criteria as the first candidate parameter set; Based on the parameter performance verification matrix, screening a parameter group whose accuracy meets a preset standard from the first candidate parameter group; Based on the screened parameter group, a dynamic parameter adjustment instruction set of the river loss prediction model to be constructed is generated.

7. The method for water allocation in arid areas according to claim 1, characterized in that: The hydrometeorological characteristics include: Historical water inflow of the river section, current water inflow of the river section, current water diversion volume, current groundwater depth of the river section and current evapotranspiration.

8. The method for water allocation in arid areas according to claim 1, characterized in that: Coupling the water resource allocation model and the reservoir operation model specifically includes: The reservoir discharge and node water supply in the arid area are used as decision variables, and the model coupling is performed with minimizing the water shortage of multiple users in the arid area as the optimization goal.

Images