Double-layer multi-target-based irrigation district water resource allocation method

Abstract

The invention discloses an irrigation district water resource allocation method based on double layers and multiple targets, which comprises the following steps of: 1, collecting related data of an irrigation district, and preprocessing all the data to obtain a data set; 2, according to the data set, various parameters in the model are determined, and an upper-layer model and a lower-layer model are constructed; 3, processing a multi-target problem by adopting a fuzzy mathematical programming method, and converting interval parameters into deterministic parameters by applying an optimal-worst method; step 4, adopting a genetic algorithm, setting a proper population scale and genetic operation probability parameters, and on the basis of satisfying a target function and constraint conditions, realizing optimal configuration of water resources in the whole irrigation area through data interaction and iterative optimization of the upper-layer model and the lower-layer model; and 5, analyzing the final scheme, and evaluating economic benefits and social effects of water resource allocation in the irrigated area. The invention relates to the technical field of water resource optimal configuration, the water resource configuration is considered thoroughly, and the adjusting efficiency is high.

Description

Technical Field

[0001] The invention relates to the technical field of water resource optimization configuration, and in particular to an irrigation area water resource configuration method based on double layers and multiple objectives. Background Art

[0002] In today's irrigation area water resources management practice, the rational allocation of water resources is a complex and critical task. As an important infrastructure for agricultural production, the effective use of water resources in irrigation areas is directly related to the sustainable development of agriculture, social stability and the balance of the ecological environment. However, the traditional irrigation area water resources allocation method has many limitations. On the one hand, in the pursuit of economic benefits, it often only focuses on the irrigation benefits of local areas or single crops, lacks the overall consideration of the comprehensive benefits of irrigation in different channels of the entire irrigation area, resulting in low macroeconomic efficiency of resource allocation; on the other hand, in terms of social equity, previous methods have failed to fully measure the degree of balance in the allocation of water resources among different channels, which is easy to cause water resources redundancy in some channels and water resources shortage in some channels, affecting the fair development of agricultural production in different regions. In addition, for the water loss of irrigation area water resources in the process of transporting water resources to downstream water users and the actual water shortage of water users, the traditional method lacks accurate models and effective control means, which further aggravates the irrationality of water resources utilization.

[0003] Therefore, an innovative water resources allocation method is urgently needed to improve the current situation and ensure that irrigation area water resources are scientifically allocated among different channels and the lower-level water users served by the channels to meet the challenges of limited water resources and diverse water demand. Summary of the invention

[0004] The purpose of the present invention is to provide a water resource allocation method for irrigation areas based on a double-layer multi-objective approach, which solves the problem that the prior art does not start from a macroscopic and microscopic perspective, the water resource allocation is not well considered, and the regulation efficiency is low.

[0005] The technical solution adopted by the present invention is a method for allocating water resources in irrigation areas based on a double-layer multi-objective approach, which is implemented according to the following steps: Step 1, collect relevant data of the irrigation area, pre-process all data to obtain a data set; Step 2: According to the data set obtained in step 1, determine the parameters in the model and construct the upper model and the lower model; Step 3, using fuzzy mathematical programming method to deal with multi-objective problems, using the best-worst method to convert interval parameters into deterministic parameters; Step 4: Using genetic algorithm, setting appropriate population size and genetic operation probability parameters, on the basis of satisfying the objective function and constraint conditions, the upper model and the lower model interact with each other through data, and then optimize iteratively to achieve the optimal allocation of water resources in the entire irrigation area; Step 5: Analyze the final plan and evaluate the economic benefits and social effects of water resources allocation in the irrigation area.

[0006] The beneficial effects of the present invention are as follows: 1) The upper model takes into account the two key elements of economic benefits and social equity from a macro perspective; the lower model focuses on the micro level, focusing on reducing the water shortage caused by unreasonable water resource allocation of lower-level water users, and controlling the water loss in the process of channel water delivery, so as to achieve a comprehensive scientific allocation of irrigation area water resources from macro to micro. 2) The optimal allocation of irrigation area water resources at the irrigation area level and the actual canal level is achieved to improve the utilization efficiency of water resources, including economic and social benefits, and to ensure the sustainable utilization and supply-demand balance of irrigation area water resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a schematic diagram of the flow of the irrigation area water resources allocation method of the present invention; Figure 2 It is a structural relationship diagram of a double-layer multi-objective model constructed by the method of the present invention; Figure 3 It is a schematic diagram of the physical structure of the water resource allocation electronic device of the present invention. DETAILED DESCRIPTION

[0008] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0009] Reference Figure 1 The present invention is based on a double-layer multi-objective irrigation area water resources allocation method, which is specifically implemented according to the following steps: Step 1: Collect meteorological, hydrological, soil and crop growth data of the irrigation area and pre-process all data; 1.1) Collect meteorological data, hydrological data, soil data, crop planting structure and area data, irrigation facility data, and farmers’ water use behavior data in the irrigation area; 1.2) The collected data were preprocessed, including data cleaning, denoising, missing value filling and other operations to ensure the accuracy and completeness of the data; in addition, the mean interpolation method was used to supplement the missing meteorological data, and the sliding average method was used to smooth the fluctuations in the hydrological data to obtain the data set.

[0010] Step 2: Based on the data set obtained in step 1, determine the parameters in the model and construct the upper and lower models (collectively referred to as the irrigation area-channel two-layer model). 2.1) The construction process of the upper model is: The upper model comprehensively considers the economic and social effect goals. Based on the function related to the irrigation benefit of the main canal of the irrigation area and the irrigation water consumption in the lower model, in order to achieve the effective constraint of the irrigation area-channel double-layer model, the surface water diversion and groundwater extraction of different main canals are optimized in intervals, considering factors such as water supply and canal project scale, so as to achieve the optimal allocation of limited water supply on different channels, so that the economic output and social effect of the irrigation area can be coordinated and optimized. 2.1.1) The economic benefit target is to maximize the total irrigation benefit of the irrigation area as a whole. The calculation method is the sum of the irrigation benefits of each channel minus the irrigation area management cost. The irrigation benefit of each channel is related to the irrigation water volume. The calculation formula is: , in, is the surface water diversion volume of the channel, in m 3 ; is the groundwater extraction volume of the channel, in m 3 ; is the function related to irrigation benefit and irrigation water volume of each channel system; is the surface water management cost of the irrigation area, in yuan / m 3 ; is the groundwater management cost in the irrigation area, in yuan / m 3 ; 2.1.2) Social impact target using the Gini coefficient Gini To reflect the fairness of water resources distribution between different channels. The Gini coefficient ranges from 0 to 1. The larger the value, the more unfair the distribution and the greater the difference in water resources distribution. The calculation of the Gini coefficient is based on the Lorenz curve. For water resources distribution, its calculation formula is: , in: m is the number of channels; i The water distribution data is arranged from small to large. W (i) After sorting i data, Its mean value, unit is m 3 ; 2.1.3) The constraints of the upper model are: Surface water availability: , Groundwater availability: , Constraints on the scale of main canal projects: , in, is the total amount of surface water available for agriculture in the irrigation area, in m 3 ; is the total amount of groundwater available for agriculture, in m 3 ; For the upper channel system i The upper limit of water transfer capacity of the inner channel, in m 3 , increase traffic through channels; For the upper channel system i The lower limit of water delivery capacity of the inner channel, in m 3 , deduced through the minimum channel flow.

[0011] 2.2) The specific construction process of the lower-level model is: 2.2.1) The lower model is based on N Channels in T Headwater flow rate for each period q it As the decision variable, the objective function is set to minimize the water shortage of lower-level water users and the water loss of channel water supply, and the expression is: , in, For lower-level water users j Time i The water shortage coefficient of the channel; s ij for j Time i The water shortage of the channel, in m 3 ; For lower-level water users j Time i Weight coefficient of channel water loss; l ij for j Time i The water loss of the channel, in m 3 ; 2.2.2) The constraints of the lower model are: Channel water delivery capacity constraint: In order to ensure safe and stable operation of the channel, the channel water delivery flow must not exceed the design flow, expressed as: , Water quantity constraint: Due to the water supply restriction, the water diversion volume of the upper channel head shall not exceed the water distribution volume of the main channel. The product of the water distribution flow and the water diversion time of each lower channel is equal to the water distribution volume of the channel, which is expressed as follows: , , Channel section water balance constraints: i The flow at the end of the channel section should be equal to the flow at the head of each lower channel connected to the end of the channel section and the flow at the head of the upper channel. i+1The sum of the flow at the head of the canal section is expressed as: , Cycle constraint: The start and end time of water distribution of each lower-level channel should be within the cycle, and the start time should be greater than zero. The expression is: , in, is the design flow of the lower channel, in m 3 / s; for j The duration of the time period, in seconds; For the upper channel j The total water volume allocated to the lower channel during the period, in m 3 ; for j Time i The water distribution volume of the channel, in m 3 ; For the upper channel k The collection of subordinate channels connected at the end of a channel section; For the upper channel k +1 channel section in j The headwater flow rate during the period, in m 3 .

[0012] Step 3, using fuzzy mathematical programming method to deal with multi-objective problems, using the best-worst method to convert interval parameters into deterministic parameters; 3.1) Fuzzy mathematical programming is a mathematical method widely used to deal with problems with uncertainty and ambiguity. Using fuzzy mathematical programming, the economic benefit objective function and the social effect objective function are transformed into a single objective function, which is expressed as follows: ; Among them, for the economic benefit objective function , according to its value range and expected satisfaction level, define the membership function For the social effect objective function Gini , also based on the reasonable value range of the Gini coefficient and the expected degree of fairness, define the membership function The smaller the Gini coefficient (the higher the fairness), The closer it is to 1, the larger the Gini coefficient is. The closer to 0; Determine the weight of the decision makers' emphasis on economic and social benefits through expert consultation, analytic hierarchy process or other decision analysis methods. and For example, if the irrigation district focuses more on economic development, It may take a larger value; if social equity is emphasized more, The sum of the weights is equal to 1; according to Construct a comprehensive fuzzy objective function to transform the multi-objective optimization problem of the upper model into a single-objective optimization problem; 3.2) For interval parameters in the model, for example, for the interval of surface water supply ,Will Determine the optimal value. Determine the worst value; for other interval parameters, similarly determine their optimal and worst values; For decision variables in the model (such as and ) and calculate the distance between it and the best and worst values ​​of each interval parameter; for example, the Euclidean distance formula is used to calculate the distance in terms of surface water availability. ,in For the optimal surface water supply, is the water supply of the worst surface water. Then the relative proximity is calculated based on the distance ,in for The distance to the optimal value, for The distance to the worst value; Similar calculations are performed for other interval parameter and decision variable combinations.

[0013] Step 4: Genetic algorithm is used to set appropriate parameters such as population size and genetic operation probability. On the basis of satisfying the objective function and constraint conditions, the upper model and the lower model interact with each other through data and then optimize iteratively to achieve the optimal allocation of water resources in the entire irrigation area. 4.1) Using the principle of large system decomposition and coordination, the upper model is associated and coordinated with the lower model, and the comprehensive fuzzy objective function is solved by the optimization algorithm to obtain the surface water diversion volume of the backbone channel. x i and groundwater extraction y i The initial plan; 4.2) According to the upper model x i and y i , determine the relevant parameters in the lower model (such as Then, the genetic algorithm is used to solve the lower model to obtain the headwater flow of the lower channel at each time period. q ij The initial plan; The constraints of the lower model must be met when solving the problem, such as channel water delivery capacity constraints, water volume constraints, channel section water volume balance constraints, and rotation constraints; 4.3) The lower model feeds back the calculated water shortage of lower-level water users, water loss in channel water supply, and water distribution flow to the upper model; the upper model re-evaluates the rationality of the current water resource allocation plan based on this feedback information and adjusts the configuration plan of surface water diversion and groundwater extraction; Repeat the above steps and solve the upper model and the lower model again until the convergence conditions are met and the final solution is obtained.

[0014] Step 5: Analyze the final plan and evaluate the economic benefits (total irrigation benefits, management costs) and social effects (fairness of water resource allocation reflected by the Gini coefficient) of water resource allocation in the irrigation area.

[0015] 5.1) Calculate the economic benefits (total irrigation benefits, management costs) and social effects (equity of water resource allocation as reflected by the Gini coefficient) of water resource allocation in irrigation areas; 5.2) Using actual monitoring data and simulation results, evaluate the final solution obtained in step 4 to determine whether it meets the requirements of irrigation district water resources management.

[0016] Figure 3 It is a schematic diagram of the physical structure of the water resource configuration electronic device provided by the present invention. The water resource configuration electronic device of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned irrigation area water resource configuration method based on double-layer multi-objective is implemented. The specific description is as follows: The memory is used to store a large amount of irrigation area related data, including but not limited to channel information, crop planting data, water user water demand data, etc., and also stores computer programs and model parameters. The processor, as the core computing unit of the device, is responsible for executing instructions in the computer program to realize the calculation and processing of the irrigation area water resource configuration method. During operation, the processor reads data and programs from the memory, and performs calculations according to the steps of the irrigation area water resource configuration method based on multi-level intervals.

[0017] The present invention also provides a storage medium, in which a computer program of the aforementioned method of the present invention is stored. When the computer program is executed by a processor, each step of the above-mentioned irrigation area water resource allocation method based on multi-level intervals can be implemented. The computer program exists in the form of binary code in the storage medium, which contains all instructions and algorithm logic for realizing functions such as data collection, model construction, solution and output. The independence of this storage medium enables the water resource allocation method of the present invention to be run on different electronic devices, as long as the device has a corresponding processor and operating environment, thereby facilitating the wide application of irrigation area water resource allocation technology.

[0018] The present invention also provides a computer program product, including a computer program, which implements the above-mentioned irrigation area water resource allocation method based on double-layer multi-objectives when executed by a processor.

[0019] Example 1 In this embodiment 1, the water resource allocation of a certain irrigation area in Shaanxi Province is taken as an example. The model structure relationship is as follows Figure 2 As shown, the specific steps are as follows: Step 1: Meteorological station data include latitude and longitude coordinates, altitude, monthly average temperature, and monthly precipitation data of meteorological stations around the irrigation area; meteorological station data are obtained from the National Meteorological Science Data Center; hydrological data include Linjia Village, Xianyang, and Lintong stations upstream of the irrigation area, which are derived from the Yellow River Basin Hydrological Yearbook (Weihe River System); Irrigation channel information, including the geographical location, length, cross-sectional dimensions, material and other physical parameters of the channel, can be obtained through geographic information system technology combined with field measurements; The water demand data of water users is initially estimated based on the irrigation plans and irrigation areas reported by water users. On the other hand, it is dynamically adjusted based on the historical irrigation data of water users and real-time soil moisture sensor data (installed in the irrigation area of ​​water users) to ensure the accuracy of the water demand data.

[0020] Step 2: According to the structure and function of the irrigation district water resources system, it is clearly divided into the upper model and the lower model.

[0021] The upper model is mainly responsible for water resource allocation decisions at the macro level, involving the optimal configuration of surface water diversion and groundwater extraction in different backbone channels; the lower model focuses on water resource scheduling and execution at the micro level, with the water distribution flow of each lower-level channel in different time periods as the decision variable.

[0022] When constructing the upper model, the various parameters in the economic benefit objective function are accurately set according to the collected channel and crop data, and the decision variable is the surface water diversion of the backbone channel. and groundwater extraction , parameters include surface water supply range , Groundwater supply range , Main canal project scale constraint parameters and , surface water and groundwater management costs in irrigation areas and At the same time, the Gini coefficient is calculated based on the size of the irrigation area and the number of channels. m The water allocation fairness constraint is incorporated into the model, and the expression is: , , When constructing the lower-level model, the objective function and constraints are accurately set based on the measured data of the channel water loss coefficient and the estimated water demand of water users. The decision variable is the headwater distribution flow of the lower channel in each period. The parameters include the total water volume interval allocated from the upper channel to the lower channel in each period , Design flow of lower-level channels , duration of each period and water shortage and water transmission loss weight coefficients and Etc. Ensure that the model can accurately reflect the actual situation of water resources in the irrigation area during the distribution process to the downstream, and provide a solid model foundation for subsequent solutions and optimization. The expression is: , Step 3, using fuzzy mathematical programming method to deal with multi-objective problems, using the best-worst method to convert interval parameters into deterministic parameters; For the economic benefit objective function , according to its value range and expected satisfaction level, define the membership function For the social effect objective function Gini , also based on the reasonable value range of the Gini coefficient and the expected degree of fairness, define the membership function The smaller the Gini coefficient (the higher the fairness), The closer it is to 1, the larger the Gini coefficient is. The closer to 0.

[0023] Through expert consultation, determine the weight of economic and social impact goals that decision makers attach importance to and For example, if the irrigation district focuses more on economic development, It may take a larger value; if social equity is emphasized more, The sum of the weights should be equal to 1.

[0024] according to Construct a comprehensive fuzzy objective function to transform the multi-objective optimization problem of the upper model into a single-objective optimization problem, that is, to solve F .

[0025] For interval parameters in the upper model, such as the surface water supply interval ,Will Determine the optimal value. For other interval parameters, their optimal and worst values ​​are also determined.

[0026] For the decision variables in the upper model (such as and ) and calculate the distance between it and the best and worst values ​​of each interval parameter. For example, the Euclidean distance formula is used to calculate the distance in terms of surface water availability. ,in For the optimal surface water supply, is the water supply of the worst surface water. Then the relative proximity is calculated based on the distance ,in for The distance to the optimal value, for The distance to the worst value. Similar calculations are performed for other interval parameters and decision variable combinations.

[0027] Based on the relative closeness, the deterministic equivalent value of the interval parameter is determined by weighted average or other appropriate methods. For example, for surface water availability, the deterministic equivalent value (in and These deterministic equivalent values ​​are used to replace the interval parameters in the upper model, and the upper model is converted into a deterministic model.

[0028] For the interval parameters in the lower model, such as the total water volume interval allocated from the upper channel to the lower channel in each period , using the same best-worst method steps to determine its deterministic equivalent value , and replace the original interval parameters to transform the lower model into a deterministic model.

[0029] Step 4: Use genetic algorithm to solve the transformed upper model and obtain the surface water diversion volume of the backbone channel x i and groundwater extraction y i In the solution process, the constraints of the upper model must be met, including the constraints on the available surface water and groundwater, the scale of the main canal project, etc.

[0030] According to the upper model x i and y i , determine the relevant parameters in the lower model (such as Then the genetic algorithm is used to solve the lower model to obtain the headwater flow of the lower channel at each time period. q it The initial solution is to satisfy the constraints of the lower model, such as the channel water delivery capacity constraint, water volume constraint, channel section water volume balance constraint and rotation period constraint.

[0031] The lower model feeds back the calculated water shortage of lower-level water users, water loss in channel water delivery, and water distribution flow to the upper model. Based on this feedback information, the upper model re-evaluates the rationality of the current water resource allocation plan and adjusts the configuration plan of surface water diversion and groundwater extraction.

[0032] Repeat the above steps and solve the upper and lower layer models again until the convergence conditions are met.

[0033] Step 5: Analyze the optimization plan and evaluate the economic benefits and social effects of the irrigation area.

[0034] According to the actual situation and possible changing factors (such as climate conditions, crop planting structure adjustment, etc.), the model is further verified and improved to ensure its practicality and adaptability.

[0035] Example 2 This embodiment 2 takes the water resource allocation of a certain irrigation area in Shaanxi Province as an example and is implemented according to the following steps: Step 1: Collect relevant data of the irrigation area and pre-process all data.

[0036] Step 2: Determine the parameters in the model and build the upper and lower models: The upper model comprehensively considers economic benefits and social effect goals; Economic benefit target: There are 4 main channels in the irrigation area. The irrigation benefit and irrigation water volume function of each channel (the coefficient is obtained by fitting historical data) is: Channel 1 、Channel 2 、Channel 3 、Channel 4 , the surface water management cost of the irrigation area is 0.1 yuan / m³, and the groundwater management cost is 0.2 yuan / m³. Then the economic benefit objective function is: ; Social impact goal: Using the Gini coefficient Gini To reflect the fairness of water resources distribution among different channels, the calculation of the Gini coefficient is based on the Lorenz curve, and its calculation formula is: ; Upper model constraints: The total amount of surface water available for agriculture in the irrigation area is in the range of [1.07×10 8 , 3.05×10 8 ]m 3 The total amount of groundwater available for agriculture is in the range of [1.30×10 8 , 1.68×10 8 ]m 3 The upper limit of water transfer capacity in the upper channel system 1 is 3.0×10 7 m3 , the lower limit is 1.0×10 7 m 3 ; The upper limit of water transfer capacity in the upper channel system 2 is 2.5×10 7 m 3 , the lower limit is 1.0×10 7 m 3 ; The upper limit of water delivery in the upper channel system 3 is 2.5×10 7 m 3 , the lower limit is 0.8×10 7 m 3 ; The upper limit of water delivery capacity in the upper channel system 4 is 2.0×10 7 m 3 , the lower limit is 0.5×10 7 m 3 ; Lower model: headwater flow of 4 channels in 12 periods (each period is 1 month) q it is the decision variable, and the objective function is to minimize the water shortage of the lower-level water users and the water loss of the channel. The water shortage coefficient of the lower-level water users in channel 1 in period 1 is 0.1, and the water shortage of channel 1 in period 1 is 500m 3 The weight coefficient of the water loss of channel 1 in one period of time for the lower-level water user is 0.3, and the water loss of channel 1 in one period of time is 300m 3 etc. (data for other time periods and channels are similar); Lower model constraints: The design flow rate of the lower channel 1 is 10m 3 / s, the duration of one period is 2592000s, and the total water volume allocated from the upper channel to the lower channel 1 in one period is in the range of [8000, 10000]m 3 , the water supply of channel 1 in period 1 is 9000m 3 Etc. (data for other time periods and channels are similar).

[0037] Step 3: Use fuzzy mathematical programming to deal with multi-objective problems and use the best-worst method to convert interval parameters into deterministic parameters: For the economic benefit objective function, its value range is [ E min , E max ]=[3.71×10 8 , 3.79×10 8 ], the expected satisfaction level is between [0, 1]. Define the membership function of the economic benefit objective function for: ; For the social effect objective function Gini coefficient, the reasonable value range of Gini coefficient is [ Gini min , Gini max ]=[0, 1], the expected fairness level is between [0, 1]. Define the Gini coefficient membership function for: ; Through the analytic hierarchy process, the decision makers' emphasis on economic benefit goals =0.6, weight of the importance attached to social effect goals =0.4. Then the comprehensive fuzzy objective function F is constructed as: ; The range of surface water supply is known to be [1.07×10 8 , 3.05×10 8 ]m 3 , determine the worst value W worst =1.07×10 8 m 3 , the optimal value W opt =3.05×10 8 m 3 , the decision variable is the surface water diversion of channel 1 x 1=1.5×10 7 m 3 ; Calculate the distance of the decision variable from the optimal value d opt (Using the Euclidean distance formula, here it is the absolute value of the difference in one dimension): ; Calculate the distance of a decision variable from its worst value: ; Calculate relative closeness: .

[0038] Step 4: Use the principle of large system decomposition and coordination to associate the upper model with the lower model. Solve the comprehensive fuzzy objective function through genetic algorithm F , and the initial plan for the surface water diversion and groundwater extraction of the backbone channel is obtained: the surface water diversion of channel 1 x 1=1.5×10 7 m 3 , Groundwater extraction y 1=1.3×10 7 m 3 ; Surface water diversion of channel 2 x 2=1.2×10 7 m3 , Groundwater extraction y 2=1.2×10 7 m 3 ; Surface water diversion of channel 3 x 3=1.0×10 7 m 3 , Groundwater extraction y 3=1.1×10 7 m 3 ; Surface water diversion of channel 4 x 4=0.8×10 7 m 3 , Groundwater extraction y 4=1.0×10 7 m 3 ; The relevant parameters in the lower model are determined based on the results obtained from the upper model, for example, the constraints related to the water diversion of the channel or the parameters in the objective function in the lower model.

[0039] Solve the lower model and get the initial scheme of the headwater flow of the lower channel in each time period: the headwater flow of channel 1 in time period 1 q 11 =5m 3 / s, headwater flow rate in 2 periods q 21 =6m 3 / s,…, headwater flow rate in 12 periods q 121 =7m 3 / s; Channel 2 headwater flow in period 1 q 12 =4.5m 3 / s, headwater flow rate in 2 periods q 22 =5.2m 3 / s,…, headwater flow rate in 12 periods q 122 =6.5m 3 / s; headwater flow of channel 3 in period 1 q 13 =4m 3 / s, headwater flow rate in 2 periods q 23 =4.8m 3 / s,…, headwater flow rate in 12 periods q 123 =6m 3 / s; Channel 4 headwater flow in period 1 q 14 =3.5m 3 / s, headwater flow rate in 2 periods q 24 =6m 3 / s,…, headwater flow rate in 12 periods q 124 =5.5m 3 / s; The lower model feeds back the calculated water shortage of lower-level water users, water loss in channel water delivery, and water distribution flow to the upper model. Based on this feedback information, the upper model re-evaluates the rationality of the current water resource allocation plan and adjusts the configuration plan of surface water diversion and groundwater extraction; Repeat the above process of solving the upper and lower models until the convergence condition is met and the final solution is obtained: the surface water diversion volume of channel 1 x 1=1.45×10 7 m 3 , Groundwater extraction y 1=1.28×10 7 m 3 ; Surface water diversion of channel 2 x 2 = 1.15 × 10 7 m 3 , Groundwater extraction y 2 = 1.18 × 10 7 m 3 ; Surface water diversion of channel 3 x 3 = 0.95 × 10 7 m 3 , Groundwater extraction y 3=1.08×10 7 m 3 ; Surface water diversion of channel 4 x 4=0.75×10 7 m 3 , Groundwater extraction y 4=0.98×10 7 m 3 The headwater flow of the lower channel at each time period is shown in Table 1.

[0040] Table 1. Headwater flow of the lower channel at different time periods

[0041] Step 5: Analyze the final plan and evaluate the economic and social effects of the irrigation area.

[0042] According to the surface water diversion volume of each backbone channel in the final plan x i and groundwater extraction y i Substituting into the economic benefit objective function, the total irrigation benefit is (unit: yuan): According to the actual surface water management cost and groundwater management cost, the management cost is (unit: yuan): , According to the calculation of water allocation for each channel in the final plan, the water allocation for each channel in the final plan is: W 1= x 1 +y 1=1.28×10 7 +1.15×10 7 =2.73×10 7 m 3 , W 2= x 2 +y 2 = 1.15 × 10 7 +1.18×10 7 =2.33×10 7 m 3 , W 3= x 3 +y 3 = 0.95 × 10 7 +1.08×10 7 =2.03×10 7 m 3 , W 4= x 4 +y 4=0.75×10 7 +0.98×10 7 =1.73×10 7 m 3 , the Gini coefficient is: , Comparing the total irrigation benefit with the expected benefit value, the expected total irrigation benefit E exp = 3.75×10 8 Yuan, E exp < E , meet the requirements in terms of benefits; Gini coefficient Gini =0.09,0< Gini< 1. It shows that the fairness of water resources allocation meets the requirements.

[0043] Example 3 Based on Example 1, this Example 3 is specifically implemented according to the following steps: Step 1: Collect relevant data of the irrigation area and pre-process all data.

[0044] Step 2: Determine the parameters in the model and build the upper and lower models: The upper model comprehensively considers economic benefits and social effect goals; Economic benefit target: There are three main canals in the irrigation area. The correlation function between irrigation benefit and irrigation water volume of each canal is: 、Channel 2 、Channel 3 , the surface water management cost of the irrigation area is 0.1 yuan / m³, and the groundwater management cost is 0.2 yuan / m³. Then the economic benefit objective function is:

[0045] Social impact goal: Using the Gini coefficient Gini To reflect the fairness of water resources distribution among different channels, the calculation of the Gini coefficient is based on the Lorenz curve, and its calculation formula is: ; Upper model constraints: The total amount of surface water available for agriculture in the irrigation area is between [50000, 60000]m 3 , the total amount of groundwater available for agriculture is between [30000, 40000]m 3 ; Lower model: headwater flow of three channels in four time periods q it is the decision variable, and the objective function is to minimize the water shortage of lower-level water users and the water loss of channel water delivery; Lower model constraints: The design flow rate of the lower channel 1 is 10m 3 / s, the duration of one period is 2592000s, and the total water volume allocated from the upper channel to the lower channel 1 in one period is in the range of [8000, 10000]m 3 , the water supply of channel 1 in period 1 is 9000m 3 Etc. (data for other time periods and channels are similar).

[0046] Step 3: Use fuzzy mathematical programming to deal with multi-objective problems and use the best-worst method to convert interval parameters into deterministic parameters: For the economic benefit objective function, its value range is [ E min , E max ]=[100000, 300000], the expected satisfaction level is between [0, 1]. Define the membership function of the economic benefit objective function for: , For the social effect objective function Gini coefficient, the reasonable value range of Gini coefficient is [ Gini min , Gini max ]=[0, 1], the expected fairness level is between [0, 1]. Define the Gini coefficient membership function for: , Through expert consultation, the decision makers attach importance to the economic benefit target =0.6, weight of the importance attached to social effect goals =0.4. Then the comprehensive fuzzy objective function F is constructed as: ; The surface water supply range is known to be [50000, 60000]m 3 , determine the worst value W worst =50000m 3 , the optimal value W opt =60000m 3 , the decision variable is the surface water diversion of channel 1 x 1=55000m 3 , calculate the distance between the decision variable and the optimal value d opt (Using the Euclidean distance formula, here it is the absolute value of the difference in one dimension): ; Calculate the distance between the decision variable and the worst value: ; Calculate relative closeness: .

[0047] Step 4: Use the principle of large system decomposition and coordination to associate the upper model with the lower model. Solve the comprehensive fuzzy objective function through genetic algorithm F , and the initial plan for the surface water diversion and groundwater extraction of the backbone channel is obtained: the surface water diversion of channel 1 x 1=15000m 3 , Groundwater extraction y 1=8000m 3 ; Surface water diversion of channel 2 x 2=12000m 3 , Groundwater extraction y 2=6000m 3 ; Surface water diversion of channel 3 x 3=10000m 3 , Groundwater extraction y 3=5000m 3 ; The relevant parameters in the lower model are determined based on the results obtained from the upper model, for example, the constraints related to the water diversion of the channel or the parameters in the objective function in the lower model.

[0048] Solve the lower model and get the initial scheme of the headwater flow of the lower channel in each time period. q 11 =8m 3 / s, headwater flow of 2 channels in 1 period q 12 =6m 3 / s, headwater flow of 3 channels in 1 period q 13 =5m 3 / s; water distribution flow at the head of channel 1 in period 2 q 21 =7m 3 / s, headwater flow of 2 channels in 2 periods q 22 =5.5m 3 / s, headwater flow of 3 channels in 2 periods q 23 =4.5m 3 / s; water distribution flow at the head of channel 1 in 3 periods q 31 =6m 3 / s, headwater flow of 2 channels in 3 periods q 32 =5m 3 / s, headwater flow of 3 channels in 3 periods q 33 =4m 3 / s; water distribution flow at the head of channel 1 in 4 periods q 41 =5m 3 / s, headwater flow of 2 channels in 4 periods q 42 =4.5m 3 / s, headwater flow of 3 channels in 4 periods q 43 =3.5m 3 / s; The lower model feeds back the calculated water shortage of lower-level water users, water loss in channel water delivery, and water distribution flow to the upper model. Based on this feedback information, the upper model re-evaluates the rationality of the current water resource allocation plan and adjusts the configuration plan of surface water diversion and groundwater extraction; Repeat the above process of solving the upper and lower models until the convergence condition is met and the final solution is obtained: the surface water diversion volume of channel 1x 1=14500m 3 , Groundwater extraction y 1=7800m 3 ; Surface water diversion of channel 2 x 2 = 11500m 3 , Groundwater extraction y 2=5800m 3 ; Surface water diversion of channel 3 x 3=9500m 3 , Groundwater extraction y 3=4800m 3 At the same time, the final headwater flow of the lower channel in each period is as follows: The headwater flow of channel 1 in period 1 q 11 =7.8m 3 / s, headwater flow of 2 channels in 1 period q 12 =5.8m 3 / s, headwater flow of 3 channels in 1 period q 13 =4.8m 3 / s; water distribution flow at the head of channel 1 in period 2 q 21 =6.8m 3 / s, headwater flow of 2 channels in 2 periods q 22 =5.3m 3 / s, headwater flow of 3 channels in 2 periods q 23 =4.3m 3 / s; water distribution flow at the head of channel 1 in 3 periods q 31 =5.8m 3 / s, headwater flow of 2 channels in 3 periods q 32 =4.8m 3 / s, headwater flow of 3 channels in 3 periods q 33 =3.8m 3 / s; water distribution flow at the head of channel 1 in 4 periods q 41 =4.8m 3 / s, headwater flow of 2 channels in 4 periods q 42 =4.3m 3 / s, headwater flow of 3 channels in 4 periods q 43 =3.3m 3 / s.

[0049] Step 5: Analyze the final plan and evaluate the economic and social effects of the irrigation area: According to the surface water diversion volume of each backbone channel in the final plan x i and groundwater extraction y i Substituting into the economic benefit objective function, the total irrigation benefit is (unit: yuan):

[0050] Calculated based on the actual surface water management cost and groundwater management cost, the management cost unit is RMB: , According to the calculation of water allocation for each channel in the final plan, the water allocation for each channel in the final plan W 1= x 1 +y 1=14500+7800=22300m 3 , W 2= x 2 +y 2=11500+5800=17300m 3 , W 3= x 3 +y 3=9500+4800=14300m 3 , the Gini coefficient is:

[0051] Comparing the total irrigation benefit with the expected benefit value, the expected total irrigation benefit E exp = 13000, E exp < E =14530, considered to meet the requirements in terms of benefits; Gini coefficient Gini =0.10,0< Gini< 1. It shows that the fairness of water resources allocation meets the requirements.

[0052] Example 4 Based on Example 1, this Example 4 is specifically implemented according to the following steps: Step 1: Collect relevant data of the irrigation area and pre-process all data.

[0053] Step 2: Determine the parameters in the model and build the upper and lower models: The upper model comprehensively considers economic benefits and social effect goals; Economic benefit target: There are 4 main canals in the irrigation area, the surface water management cost of the irrigation area is 0.13 yuan / m³, and the groundwater management cost is 0.23 yuan / m³. The economic benefit target function is: , Social impact goal: Using the Gini coefficient Gini To reflect the fairness of water resources distribution among different channels, the calculation of the Gini coefficient is based on the Lorenz curve, and its calculation formula is: , Upper model constraints: The total amount of surface water available for agriculture in the irrigation area is between [35000, 45000]m 3 , the total amount of groundwater available for agriculture is between [18000, 28000]m 3 , the upper and lower limits of the water delivery capacity of each superior channel system shall be determined according to the actual conditions of the channel; Lower model: headwater flow of 4 channels in 3 time periods q it The objective function is to minimize the water shortage of lower-level water users and the water loss of channel water supply. The objective function is to minimize the water shortage and water supply loss. The water shortage coefficient, water shortage, weight coefficient and water supply loss are set. Set the constraints of the lower model, including channel water delivery capacity, water volume constraints, channel section water volume balance constraints and rotation constraints.

[0054] Step 3: Use fuzzy mathematical programming to deal with multi-objective problems and use the best-worst method to convert interval parameters into deterministic parameters: The comprehensive fuzzy objective function F is constructed as: ; Determine the optimal and worst values ​​of interval parameters, and calculate the distance and relative closeness between decision variables and interval parameters.

[0055] Step 4: Use the principle of large system decomposition and coordination to associate the upper and lower models, and solve the comprehensive fuzzy objective function through genetic algorithm F , get the initial plan of surface water diversion and groundwater extraction of the backbone channel; solve the lower model to get the initial plan of the headwater distribution flow of the lower channel in each period; the lower layer is fed back to the upper layer, and the solution is repeated until convergence to get the final plan: the surface water diversion of channel 1 x 1=9500m 3 , Groundwater extraction y 1=5800m 3 ; Surface water diversion of channel 2 x 2=8000m 3 , Groundwater extraction y 2=5000m 3; Surface water diversion of channel 3 x 3=6500m 3 , Groundwater extraction y 3=4000m 3 ; Surface water diversion of channel 4 x 4=5000m 3 , Groundwater extraction y 4=3000m 3 .

[0056] Step 5: Analyze the final plan and evaluate the economic and social effects of the irrigation area.

[0057] According to the surface water diversion volume of each backbone channel in the final plan x i and groundwater extraction y i Substituting into the economic benefit objective function, the total irrigation benefit is (unit: yuan): , According to the actual surface water management cost and groundwater management cost, the management cost is (unit: yuan): , According to the calculation of water allocation for each channel in the final plan, the water allocation for each channel in the final plan W 1= x 1 +y 1=15300m 3 , W 2= x 2 +y 2=13000m 3 , W 3= x 3 +y 3=10500m 3 , W 4= x 4 +y 4=8000m 3 , the Gini coefficient is: ; Comparing the total irrigation benefit with the expected benefit value, the expected total irrigation benefit E exp = 10000, E exp < E , and believed that the requirements were met in terms of benefits; Gini coefficient Gini =0.13,0< Gini< 1. It shows that the fairness of water resources allocation meets the requirements.

[0058] Example 5 Based on Example 1, this Example 5 is specifically implemented according to the following steps: Step 1: Collect relevant data of the irrigation area and pre-process all data.

[0059] Step 2: Determine the parameters in the model and build the upper and lower models: The upper model comprehensively considers economic benefits and social effect goals; Economic benefit target: There are three main canals in the irrigation area, the surface water management cost of the irrigation area is 0.3 yuan / m³, and the groundwater management cost is 0.5 yuan / m³. The economic benefit target function is: ; Social impact goal: Using the Gini coefficient Gini To reflect the fairness of water resources distribution among different channels, the calculation of the Gini coefficient is based on the Lorenz curve, and its calculation formula is: , Upper model constraints: The total amount of surface water available for agriculture in the irrigation area is in the range of [8×10 5 , 10×10 5 ]m 3 The total amount of groundwater available for agriculture is in the range of [4×10 5 , 6×10 5 ]m 3 , the upper and lower limits of the water delivery capacity of each superior channel system shall be determined according to the actual conditions of the channel; Lower model: headwater flow of three channels in three time periods q it The objective function is to minimize the water shortage of lower-level water users and the water loss of channel water supply. The objective function is to minimize the water shortage and water supply loss. The water shortage coefficient, water shortage, weight coefficient and water supply loss are set. Set the constraints of the lower model, including channel water delivery capacity, water volume constraints, channel section water volume balance constraints and rotation constraints.

[0060] Step 3: Use fuzzy mathematical programming to deal with multi-objective problems and use the best-worst method to convert interval parameters into deterministic parameters: The comprehensive fuzzy objective function F is constructed as: ; Determine the optimal and worst values ​​of interval parameters, and calculate the distance and relative closeness between decision variables and interval parameters.

[0061] Step 4: Use the principle of large system decomposition and coordination to associate the upper and lower models, and solve the comprehensive fuzzy objective function through genetic algorithm F, get the initial plan of surface water diversion and groundwater extraction of the backbone channel; solve the lower model to get the initial plan of the headwater distribution flow of the lower channel in each period; the lower layer is fed back to the upper layer, and the solution is repeated until convergence to get the final plan: the surface water diversion of channel 1 x 1=2.5×10 5 m 3 , Groundwater extraction y 1=1.8×10 5 m 3 ; Surface water diversion of channel 2 x 2=2.0×10 5 m 3 , Groundwater extraction y 2 = 1.5 × 10 5 m 3 ; Surface water diversion of channel 3 x 3=1.5×10 5 m 3 , Groundwater extraction y 3=1.2×10 5 m 3 .

[0062] Step 5: Analyze the final plan and evaluate its economic and social effects.

[0063] According to the surface water diversion volume of each backbone channel in the final plan x i and groundwater extraction y i Substituting into the economic benefit objective function, the total irrigation benefit is (unit: yuan): ; According to the actual surface water management cost and groundwater management cost, the management cost unit is: ; According to the calculation of water allocation for each channel in the final plan, the water allocation for each channel in the final plan W 1= x 1 +y 1=4.1×10 5 m 3 , W 2= x 2 +y 2=3.3×10 5 m 3 , W 3= x 3 +y 3=2.5×10 5 m 3 , the Gini coefficient is: ; Comparing the total irrigation benefit with the expected benefit value, the expected total irrigation benefit E exp = 300000, E exp < E , and believed that the requirements were met in terms of benefits; Gini coefficient Gini =0.05,0< Gini< 1. It shows that the fairness of water resources allocation meets the requirements.

[0064] Example 6 Based on Example 1, this Example 6 is specifically implemented according to the following steps: Step 1: Collect relevant data of the irrigation area and pre-process all data.

[0065] Step 2: Determine the parameters in the model and build the upper and lower models: Analyze the supply and demand of water resources in the irrigation area and establish a two-layer model of irrigation area and channel. The upper model comprehensively considers economic benefits and social effects; Economic benefit target: There are 4 main canals in the irrigation area, the surface water management cost of the irrigation area is 0.2 yuan / m³, and the groundwater management cost is 0.35 yuan / m³. The economic benefit target function is: , Social impact goal: Using the Gini coefficient Gini To reflect the fairness of water resources distribution among different channels, the calculation of the Gini coefficient is based on the Lorenz curve, and its calculation formula is: , Upper model constraints: The total amount of surface water available for agriculture in the irrigation area is in the range of [8×10 5 , 10×10 5 ]m 3 The total amount of groundwater available for agriculture is in the range of [4×10 5 , 6×10 5 ]m 3 , the upper and lower limits of the water delivery capacity of each superior channel system shall be determined according to the actual conditions of the channel; Lower model: headwater flow of 4 channels in 5 time periods q it The objective function is to minimize the water shortage of lower-level water users and the water loss of channel water supply. The objective function is to minimize the water shortage and water supply loss. The water shortage coefficient, water shortage, weight coefficient and water supply loss are set. Set the constraints of the lower model, including channel water delivery capacity, water volume constraints, channel section water volume balance constraints and rotation constraints.

[0066] Step 3: Use fuzzy mathematical programming to deal with multi-objective problems and use the best-worst method to convert interval parameters into deterministic parameters: The comprehensive fuzzy objective function F is constructed as: ; Determine the optimal and worst values ​​of interval parameters, and calculate the distance and relative closeness between decision variables and interval parameters.

[0067] Step 4: Use the principle of large system decomposition and coordination to associate the upper and lower models, and solve the comprehensive fuzzy objective function through genetic algorithm F , get the initial plan of surface water diversion and groundwater extraction of the backbone channel; solve the lower model to get the initial plan of the headwater distribution flow of the lower channel in each period; the lower layer feeds back to the upper layer, and the solution is repeated until convergence to get the final plan: Surface water diversion of channel 1 x 1=2.15×10 5 m 3 , Groundwater extraction y 1=1.55×10 5 m 3 ; Surface water diversion of channel 2 x 2 = 1.75 × 10 5 m 3 , Groundwater extraction y 2 = 1.25 × 10 5 m 3 ; Surface water diversion of channel 3 x 3 = 1.35 × 10 5 m 3 , Groundwater extraction y 3 = 0.95 × 10 5 m 3 ; Surface water diversion of channel 4 x 4=0.95×10 5 m 3 , Groundwater extraction y 4=0.65×10 5 m 3 .

[0068] Step 5: Analyze the final plan and evaluate its economic and social effects.

[0069] According to the surface water diversion volume of each backbone channel in the final plan x i and groundwater extraction y i Substituting into the economic benefit objective function, the total irrigation benefit is (unit: yuan): , According to the actual surface water management cost and groundwater management cost, the management cost is: , unit is yuan According to the calculation of water allocation for each channel in the final plan, the water allocation for each channel in the final plan W 1= x 1 +y 1=3.7×10 5 m 3 , W 2= x 2 +y 2=3.0×10 5 m 3 , W 3= x 3 +y 3=2.3×10 5 m 3 , W 4= x 4 +y 4=1.6×10 5 m 3 , the Gini coefficient is: ; Comparing the total irrigation benefit with the expected benefit value, the expected total irrigation benefit E exp = 300000, E exp < E , and believed that the requirements were met in terms of benefits; Gini coefficient Gini =0.17,0< Gini< 1. It shows that the fairness of water resources allocation meets the requirements.

Claims

1. The irrigation area water resources allocation method based on double-layer multi-objective is characterized by: Follow these steps to implement: Step 1, collect relevant data of the irrigation area, pre-process all data to obtain a data set; Step 2, according to the data set obtained in step 1, determine the parameters in the model and construct the irrigation area-channel two-layer model, namely the upper model and the lower model; Step 3, using fuzzy mathematical programming method to deal with multi-objective problems, using the best-worst method to convert interval parameters into deterministic parameters; Step 4: Using genetic algorithm, setting population size and genetic operation probability parameters, on the basis of satisfying the objective function and constraint conditions, the upper model and the lower model interact with each other through data, and then optimize iteratively to achieve the optimal allocation of water resources in the entire irrigation area; Step 5: Analyze the final plan and evaluate the economic benefits and social effects of water resources allocation in the irrigation area.

2. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 1 is characterized in that: Step 1 specifically includes: 1.1) Collect meteorological data, hydrological data, soil data, crop planting structure and area data, irrigation facility data, and farmers’ water use behavior data in the irrigation area; 1.2) The collected data are preprocessed, including data cleaning, denoising, and missing value filling operations; in addition, the mean interpolation method is used to fill in the missing meteorological data, and the sliding average method is used to smooth the fluctuations in the hydrological data to obtain the data set.

3. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 1 is characterized in that: In step 2, the specific construction process of the upper model is: The upper model comprehensively considers the economic and social effect objectives, optimizes the interval configuration of surface water diversion and groundwater extraction in different trunk channels, and realizes the optimal allocation of limited water supply in different channels, so that the economic output and social effect of the irrigation area can be coordinated and optimized; 2.1.1) The economic benefit target is to maximize the total irrigation benefit of the irrigation area as a whole. The calculation method is the sum of the irrigation benefits of each channel minus the irrigation area management cost. The irrigation benefit of each channel is related to the irrigation water volume. The calculation formula is: ， in, is the surface water diversion volume of the channel, in m 3 ; is the groundwater extraction volume of the channel, in m 3 ; is the function related to irrigation benefit and irrigation water volume of each channel system; is the surface water management cost of the irrigation area, in yuan / m 3 ; is the groundwater management cost in the irrigation area, in yuan / m 3 ; 2.1.2) Social impact target using the Gini coefficient Gini To reflect the fairness of water resources distribution among different channels, the calculation of the Gini coefficient is based on the Lorenz curve. For water resources distribution, its calculation method is: ， in: m is the number of channels; i The water distribution data is arranged from small to large. W (i) After sorting i data, Its mean value, unit is m 3 ; 2.1.3) The constraints of the upper model are: Surface water availability: , Groundwater availability: , Constraints on the scale of main canal projects: , in, is the total amount of surface water available for agriculture in the irrigation area, in m 3 ; is the total amount of groundwater available for agriculture, in m 3 ; For the upper channel system i The upper limit of water transfer capacity of the inner channel, in m 3 , increase traffic through channels; For the upper channel system i The lower limit of water delivery capacity of the inner channel, in m 3 , deduced through the minimum channel flow.

4. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 1 is characterized in that: In step 2, the specific construction process of the lower model is: 2.2.1) The lower model is based on N Channels in T Headwater flow rate for each period q it As the decision variable, the objective function is set to minimize the water shortage of lower-level water users and the water loss of channel water supply, and the expression is: ， in, For lower-level water users j Time i The water shortage coefficient of the channel; s ij for j Time i The water shortage of the channel, in m 3 ; For lower-level water users j Time i Weight coefficient of channel water loss; l ij for j Time i The water loss of the channel, in m 3 ; 2.2.2) The constraints of the lower model are: Channel water delivery capacity constraint: The channel water delivery flow rate must not exceed the design flow rate, expressed as: , Water quantity constraint: Due to the water supply restriction, the water diversion volume of the upper channel head shall not exceed the water distribution volume of the main channel. The product of the water distribution flow and the water diversion time of each lower channel is equal to the water distribution volume of the channel, which is expressed as follows: ， ， Channel section water balance constraints: i The flow at the end of the channel section should be equal to the flow at the head of each lower channel connected to the end of the channel section and the flow at the head of the upper channel. i+1 The sum of the flow at the head of the canal section is expressed as: , Cycle constraint: The start and end time of water distribution of each lower-level channel should be within the cycle, and the start time should be greater than zero. The expression is: ， in, is the design flow rate of the lower channel, in m 3 / s; for j The duration of the time period, in seconds; For the upper channel j The total water volume allocated to the lower channel during the period, in m 3 ; for j Time i The water distribution volume of the channel, in m 3 ; For the upper channel k The collection of subordinate channels connected at the end of a channel section; For the upper channel k +1 channel section in j The headwater flow rate during the period, in m 3 .

5. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 1 is characterized in that: Step 3 specifically includes: 3.1) Using fuzzy mathematical programming method, the economic benefit objective function and social effect objective function are transformed into a single objective function, which is expressed as: ； Among them, for the economic benefit objective function , according to its value range and expected satisfaction level, define the membership function ; For the social effect objective function Gini , also based on the reasonable value range of the Gini coefficient and the expected degree of fairness, define the membership function ; according to Construct a comprehensive fuzzy objective function to transform the multi-objective optimization problem of the upper model into a single-objective optimization problem; 3.2) For the interval parameters in the model, for the interval of surface water supply ,Will Determine the optimal value. Determine the worst value; for other interval parameters, similarly determine their optimal and worst values; For each possible set of values ​​of the decision variables in the model, the distance between it and the optimal and worst values ​​of each interval parameter is calculated, and then the relative closeness is calculated based on the distance; similar calculations are performed for other combinations of interval parameters and decision variables.

6. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 5 is characterized in that: In step 3.1), determine the weights of the economic and social benefits that decision makers place on through expert consultation, analytic hierarchy process or other decision analysis methods. and , the sum of the weights is equal to 1.

7. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 5 is characterized in that: In step 3.2), the Euclidean distance formula is used to calculate the distance in terms of surface water availability: ,in For the optimal surface water supply, The water supply of the worst surface water; The relative closeness is calculated based on the distance: ， in for The distance to the optimal value, for The distance to the worst value.

8. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 1 is characterized in that: In step 4, the specific process is: 4.1) Using the principle of large system decomposition and coordination, the upper model is associated and coordinated with the lower model, and the comprehensive fuzzy objective function is solved by the optimization algorithm to obtain the surface water diversion volume of the backbone channel. x i and groundwater extraction y i The initial plan; 4.2) According to the upper model x i and y i , determine the relevant parameters in the lower model, use genetic algorithm to solve the lower model, and obtain the headwater flow of the lower channel in each period q ij The initial plan; The constraints of the underlying model must be met when solving; 4.3) The lower model feeds back the calculated water shortage of lower-level water users, water loss of channel water transmission and water distribution flow to the upper model; Based on these feedback information, the upper model re-evaluates the rationality of the current water resource allocation plan and adjusts the configuration plan of surface water diversion and groundwater extraction; Repeat the above steps and solve the upper model and the lower model again until the convergence conditions are met and the final solution is obtained.

9. The irrigation area water resources allocation method based on double layers and multiple objectives according to claim 1 is characterized in that: Step 5 specifically includes: 5.1) Calculate the economic and social effects of water resources allocation in irrigation areas; 5.2) Using actual monitoring data and simulation results, evaluate the final solution obtained in step 4 to determine whether it meets the requirements of irrigation district water resources management.

Images