A multi-stage multi-objective based reservoir water resource optimal allocation method

Abstract

The application discloses a kind of reservoir water resource optimal allocation method based on multi-stage multi-objective belonging to water resource management technical field, comprising the following steps: A, predict the water resource demand, water resource available quantity, social economy and other basic data in basin;B, with the maximum unit water resource utilization benefit as target, with reservoir water balance, irrigation and power generation water demand, salinity leaching and so on as constraint condition, construct water resource optimal allocation model;C, coupling fixed combination stochastic programming and fractional programming method is solved to model, according to the water resource shortage obtained under different probability level arrangement production to reach maximum unit allocation water benefit.The application fully considers the necessity of joint management of upstream hydropower generation and downstream agricultural irrigation in reservoir management system, reasonably allocates water resources, land resources and energy in the basin, obtains the maximum unit allocation water benefit, so as to relieve water contradiction and improve the utilization efficiency of resources.

Description

Technical Field

[0001] This invention belongs to the field of water resources management technology, and specifically relates to a method for optimizing the allocation of reservoir water resources based on multi-stage and multi-objective approaches, specifically a fixed-combination stochastic fractional programming method for optimizing the allocation of water resources. Background Technology

[0002] Water resources are an indispensable natural resource for human survival. However, due to continuous population growth and intensifying climate change, a series of problems have arisen, including water shortages, food security, and energy crises. For example, there is a conflict between upstream hydropower generation and downstream agricultural irrigation: upstream countries increase reservoir storage in summer to meet electricity demand for winter hydropower generation, leading to frequent winter floods and insufficient irrigation water in summer for downstream countries. Furthermore, water resource management systems are dynamic and complex systems with uncertainties: the randomness of precipitation and climate change results in significant uncertainty regarding the availability of water resources; the subjectivity of human perception and the incompleteness of regional information lead to inaccuracies in data acquisition and model building. Moreover, water resource allocation involves not only the effective use of water resources but is also influenced by socioeconomic factors (such as water prices and water conveyance channels). Different decision-makers have different objectives; for example, the finance department pursues maximum economic benefits, while the water resource management department aims for minimum water consumption. These conflicting objectives further complicate water resource management systems. How to select a suitable reservoir water resource allocation scheme under the constraints of limited water and land resources, alleviate the contradiction between upstream hydropower generation and downstream agricultural irrigation water use, and thus ensure production requirements is an important issue.

[0003] However, traditional research either only considers the uncertainty in the system, ignoring the interaction and influence between multiple objectives and failing to balance the conflicting objectives among stakeholders from a holistic perspective; or it only considers the multi-objective nature of the system, ignoring the impact of random uncertainty on decision-making, leading to significant irrationality in water resource optimization models. Traditional research rarely considers the multiple uncertainties and multi-objective nature of a system simultaneously, which has become a major bottleneck restricting the research on rational water resource allocation. Therefore, this invention develops a fixed-combination stochastic fractional programming method for reservoir water resource optimization. This method can fully characterize the multiple uncertainties in the system and balance the contradictions between two different objectives, enabling joint management of hydropower generation and agricultural irrigation to achieve efficient water resource utilization. Summary of the Invention

[0004] The purpose of this invention is to provide a reservoir water resource optimization allocation method based on multi-stage and multi-objective methods. The method is characterized by being a fixed-combination stochastic fractional programming method for water resource optimization allocation, comprising the following steps:

[0005] Step A: Predict the available water resources and water demand within the watershed, and use these as input data for the water resources optimization allocation module;

[0006] Step B: With the goal of maximizing the efficiency of unit water resource utilization, and with the constraints of reservoir water balance, irrigation, power generation water demand and salinity leaching, construct a water resource optimization model;

[0007] Step C: Optimize the model using coupled fixed-combination stochastic programming and fractional programming; based on the optimization results, determine the water allocation for each sector, and arrange hydropower generation and crop planting patterns according to the water shortage under different water inflow and power generation levels to achieve the maximum water allocation benefit per unit.

[0008] Step A includes:

[0009] Step A1: Based on the economic development of the study area and the impact of human activities, predict the water resource demand of each user in the basin, and then set the upper and lower limits of the water demand of each user.

[0010] Step A2: Based on the natural geographical conditions within the basin, predict the available water resources within the basin, and use the obtained available water resources as the input boundary of the optimization model.

[0011] Step A3: Use the obtained water resource demand and available water resources of each user as the basic input data for the water resource optimization and allocation module.

[0012] In step B, the objective function Maxf for maximizing marginal benefits in the constructed reservoir water resource optimization allocation model includes (a) the system benefits generated by hydropower and agricultural irrigation activities, including expected benefits and economic penalties (economic losses) when the expected goals cannot be met; and (b) total water consumption, specifically represented as follows:

[0013] Max f = (a) / (b)

[0014] in,

[0015]

[0016]

[0017] In the formula: i represents the specific crop type (i = 1, 2, 3, 4, 5); j represents the specific region (j = 1, 2, 3); t represents the specific period (t = 1, 2, 3, 4, 5); h represents different water inflow levels (h = 1, 2, 3, 4, 5); f is the marginal benefit (in US dollars); NB ij Net benefit of crop i in region j; TA ij Irrigation targets for crop i in region j; TEg tThe target for hydropower generation during period t; NP t The net benefit of hydropower generation during period t; Eg t p represents the reservoir's power generation during period t; th C represents the probability of water level h occurring in period t; ij Economic penalties for failing to meet irrigation targets for crop i in region j; CP t The economic penalty when the hydropower generation target for period t cannot be met; AC ijth Planting area for each crop; SW ijth Irrigation water allocation for various crops; GW th Water allocation for hydropower generation; m jt The efficiency of channel transportation in region j during period t.

[0018] The constraints are B01 to B09.

[0019] B01) Water Balance Constraints

[0020]

[0021] In the formula, S th S(t) represents the reservoir's water storage capacity at the level of the inflow at the end of period t; t-1),h q represents the reservoir's water storage capacity at the initial inflow level h during period t; th WR represents the inflow volume of the reservoir at the inflow level of time t (h). th E represents the reservoir discharge at the inflow level of time t (h). th The evapotranspiration rate of the reservoir is denoted by h, representing the inflow level during period t.

[0022] B02) Water resource availability constraints

[0023]

[0024] In the formula, LW ijth The amount of water allocated for leaching each crop based on its salinity.

[0025] B03) Constraints on available irrigable water resources

[0026]

[0027] In the formula, WPC ijt Let θ be the water requirement per unit area of ​​crop i in region j at period t; jt The field irrigation efficiency in region j during period t.

[0028] B04) Soil salinity leaching constraint

[0029]

[0030] In the formula, LF ijtSalinity leaching fraction of crop i in region j during period t

[0031] B05) Water resource constraints for power generation

[0032]

[0033] In the formula, GW min,t The minimum water demand for hydropower generation during period t; GW max,t Let t be the water demand corresponding to the installed capacity of the hydropower station during period t.

[0034] B06) Power Supply Constraints

[0035]

[0036] In the formula, PED ijt The electricity required per unit of water transport for crop i in region j during period t; EP jh for

[0037] Maximum irrigation power distribution in region j during period t.

[0038] B07) Farmland Area Constraints

[0039]

[0040] B08) Food Security Constraints

[0041]

[0042]

[0043] In the formula, TAC j This represents the maximum permissible arable land area in region j. The maximum area ratio of crop i in region j; Let i be the minimum area ratio of crop i in region j.

[0044] B09) Agricultural Permitted Emissions of Non-point Source Pollution Constraints

[0045]

[0046] In the formula, MF ij Fertilizer application rate for crop i in region j during period t; PP k The content of pollutant K in fertilizer; TP represents the emission rate of pollutant k during period t. kt Let k be the total emissions of pollutant during period t.

[0047] Step C includes:

[0048] Step C1: Based on the basic water resources situation within the basin and the actual water demand of users, a fixed-combination stochastic fractional programming method is developed to optimize the water allocation for each user. Compared with traditional methods, this algorithm can simultaneously handle the uncertainties represented by probability functions in the model and the contradictory relationship between the two objectives;

[0049] Step C2: Select 5 inflow levels (h=1,2,3,4,5) and 5 hydropower generation levels (α=1,0,75,0.6,0.45,0.3), and use Lingo programming software to calculate sub-models under different level combinations to obtain reservoir water resource optimization allocation schemes, thereby alleviating the water use contradiction between upstream hydropower generation and downstream agricultural irrigation.

[0050] The beneficial effects of this invention are:

[0051] Compared with existing water resource optimization methods, this invention has the following advantages:

[0052] Fixed-combination stochastic programming, based on multi-stage stochastic programming, simplifies the solution process by using a fixed probability level. Therefore, it can simplify solutions while reflecting the advantages of multi-stage stochastic programming, making it better suited for large-scale practical problems in the context of long-term planning. Furthermore, traditional allocation methods only consider uncertainties in the model or the contradictory relationships between multiple objectives, rarely addressing the system's multiple uncertainties and multi-objective problems simultaneously. This leads to the acquisition of optimal power production and water resource allocation schemes, alleviating water use conflicts between upstream and downstream areas and achieving rational resource utilization. Attached Figure Description

[0053] Figure 1 A flowchart for the optimal allocation of reservoir water resources based on multiple stages and multiple objectives;

[0054] Figure 2 This is a water resource allocation scheme under uncertain conditions in the embodiments;

[0055] Figure 3 The optimization results of the method of this invention are compared with those of the traditional method; where a is the overall curve, b is a magnified view of a local area, and c is a magnified view of a local area. Detailed Implementation

[0056] This invention provides a multi-stage, multi-objective method for optimizing reservoir water resource allocation. This method is a fixed-combination stochastic fractional programming approach for water resource optimization. The invention will now be described in further detail with reference to the accompanying drawings.

[0057] This invention applies a fixed-combination stochastic fractional programming method to develop a reservoir water resource optimization allocation model based on multi-stage and multi-objectives. This model solves the problems of complexity, uncertainty, and multi-objectives in the water resource allocation system and effectively alleviates the water use conflict between upstream hydropower generation and downstream agricultural irrigation.

[0058] like Figure 1 The diagram shows a multi-stage, multi-objective reservoir water resource optimization allocation flowchart. First, based on the water resource situation within the basin, the upper and lower limits of available water resources and water demand are determined. These are used as input data to construct an optimization model. A fixed-combination stochastic algorithm and a fractional programming algorithm are then introduced to solve the model, thus addressing the stochastic uncertainty and multi-objective nature of the water resource management system. Finally, the water resource allocation scheme for the reservoir under uncertain conditions is output. Specific practical steps include:

[0059] Step A: Based on the natural geographical conditions and historical data within the watershed, predict the available water resources and water demand within the watershed, and use these as input data for the water resources optimization module; the specific implementation steps are as follows:

[0060] Step A1: Based on the economic development of the study area and the impact of human activities, predict the water demand of each user in the basin, and then set the upper and lower limits of the water demand of each user.

[0061] Step A2: Based on the natural geographical conditions within the basin, predict the available water resources within the basin, and use the obtained available water resources as the input boundary of the optimization model.

[0062] Step A3: Load the water resource demand and available water resources for each user, using them as the basic input data for the water resource optimization module. Multiple inflow and upstream power generation scenarios are then set up to analyze the impact of inflow randomness and management policies on optimal water resource allocation, thus obtaining multiple sets of results for decision-makers to choose from; specific output results are as follows... Figure 2 As shown.

[0063] Step B: With the goal of maximizing the efficiency of unit water resource utilization, and taking into account constraints such as reservoir water balance, irrigation and power generation water demand, and salinity leaching, a water resource optimization model is constructed. The specific implementation steps are as follows:

[0064] Step B1: Considering the uncertainties in optimal water resource allocation, construct the model objective function. Simultaneously, due to the discrepancy between the expected water allocation target and the water demand, economic penalties are necessary to address water shortages in the allocation system—an economic penalty will occur when the expected water allocation target cannot be met. Therefore, the model objective function is constructed as follows:

[0065] Max f = (a) / (b)

[0066]

[0067]

[0068] In the formula: i represents the specific crop type (i = 1, 2, 3, 4, 5); j represents the specific region (j = 1, 2, 3); t represents the specific period (t = 1, 2, 3, 4, 5); h represents different water inflow levels (h = 1, 2, 3, 4, 5); f is the marginal benefit (in US dollars); NB ij Net benefit of crop i in region j; TA ij Irrigation targets for crop i in region j; TEg t The target for hydropower generation during period t; NP t The net benefit of hydropower generation during period t; Eg t p represents the reservoir's power generation during period t; th C represents the probability of water level h occurring in period t; ij Economic penalties for failing to meet irrigation targets for crop i in region j; CP t The economic penalty when the hydropower generation target for period t cannot be met; AC ijth Planting area for each crop; SW ijth Irrigation water allocation for various crops; GW th This refers to the water allocation for hydropower generation.

[0069] In step B1 of this embodiment, the objective function includes (a) the system benefits generated by hydropower generation and agricultural irrigation activities, including expected benefits and economic penalties (economic losses) when the expected objectives cannot be met; and (b) the total water consumption.

[0070] Step B2: Considering natural resource reserves, pollutant control, and soil salinity leaching, set constraints such as water resource availability, water quality protection, salinity leaching, and food security:

[0071] Water availability constraints:

[0072]

[0073] In the formula, LW ijth WR is used to determine the amount of water to be used for leaching each crop based on its salinity. th Let h be the reservoir discharge at time t under the scenario of inflow.

[0074] Water quality requirements and constraints:

[0075]

[0076] In the formula, MF ij Fertilizer application rate for crop i in region j during period t; PP k The content of pollutant K in fertilizer; TP represents the emission rate of pollutant k during period t. kt Let k be the total emissions of pollutant during period t.

[0077] Salinity leaching constraint;

[0078]

[0079] In the formula, LF ijt The salinity leaching fraction of crop i in region j during period t.

[0080] Food security constraints:

[0081]

[0082]

[0083] In the formula, TAC j This represents the maximum permissible arable land area in region j. The maximum area ratio of crop i in region j; The minimum area ratio of crop i in region j

[0084] Step C: The model is optimized using a coupled fixed-combination stochastic programming and fractional programming method. Based on the optimization results, the water allocation for each sector is determined, and the hydropower generation and crop planting patterns are arranged according to the water shortage under different water inflow and power generation levels to achieve the maximum unit water allocation benefit.

[0085] Step C is further divided into:

[0086] Step C1: Based on the basic water resources situation within the basin and the actual water demand of users, a fixed-combination stochastic fractional programming method is developed to optimize the water allocation for each user. Compared with traditional methods, this algorithm can simultaneously handle the uncertainties represented by probability functions in the model and the contradictory relationship between the two objectives.

[0087] Step C2: Select 5 inflow levels (h=1,2,3,4,5) and 5 hydropower generation levels (α=1,0,75,0.6,0.45,0.3), and use LINGO programming software to calculate sub-models under different level combinations to obtain reservoir water resource optimization allocation schemes, thereby alleviating the water use contradiction between upstream hydropower generation and downstream agricultural irrigation.

[0088] The results show that hydropower generation can generate higher economic benefits, therefore, water resources should be prioritized for allocation to hydropower plants. When the inflow level decreases, agricultural irrigation water allocation decreases first, while ensuring food security (a 30.4% reduction in agricultural irrigation water allocation occurs under the scenario of low inflow versus high inflow). By adjusting different upstream power generation levels, a water allocation scheme that maximizes the benefit per unit of water allocation can be found, thereby balancing upstream hydropower water use and downstream agricultural irrigation water use. Scenario analysis results indicate that under extreme water shortage conditions, to ensure food demand, the upstream hydropower generation level should be controlled at α = 0.45. Figure 2 This figure shows the water resource allocation for each user under certain scenarios in the example. As can be seen from the figure, Dashaguz receives the most water, while Khorezm receives the least. For example, under the high inflow level and α = 0.45 scenario, the water allocation for Dashaguz and Khorezm is 63.58 × 10⁻⁶. 8 m 3 and 38.60×10 8 m 3 The results of crop cultivation also vary significantly with changes in water levels: cotton is the main crop when water levels are high, while grapes are the main crop when water levels are low. Figure 3 The figure compares the FSFP of this invention with the traditional fixed-combination stochastic programming FSP, where a is the overall curve, b is a magnified local curve, and c is a magnified local curve. The figure shows that the total water resource allocation calculated by the method of this invention is lower than that of the traditional method, especially for irrigation water allocation. For example, under the scenario of high inflow level and α = 0.45, the irrigation water allocation for FSFP and FSP is 15.54 × 10⁻⁶, respectively. 9 m 3 and 16.64×10 9 m 3 Overall, the FSFP model consumes less water and has higher marginal benefits (12.51 × 10⁻⁶ times higher than the traditional method). -3 US$ / m 3 This indicates that the FSFP model is more adaptable to changes in inflow and upstream hydropower levels. The water-saving potential of FSFP not only brings significant system benefits in the current period but also alleviates water shortages in the next period through reservoir regulation, helping managers to better utilize limited water resources.

[0089] This invention combines fixed-combination stochastic programming with fractional programming to generate a fixed-combination stochastic fractional programming method. This method is then applied to a reservoir water resource optimization allocation system, which can effectively handle: (1) the random uncertainty of the available water resources represented by a probability function; (2) the contradictory relationship between two different objectives; and (3) the conflict between different upstream power production water allocation objectives and downstream agricultural irrigation, thereby providing a more reliable decision-making solution to alleviate the contradiction between upstream and downstream water use.

Claims

1. A method for optimal allocation of reservoir water resources based on multi-stage and multi-objective approaches, characterized in that, This method is a water resource optimization allocation method based on fixed-combination stochastic fractional programming, and includes the following steps: Step A: Predict the available water resources and water demand within the watershed, and use these as input data for the water resources optimization allocation module; Step B: With the goal of maximizing the efficiency of unit water resource utilization, and with the constraints of reservoir water balance, irrigation, power generation water demand and salinity leaching, construct a water resource optimization model; Step C: Optimize the model using coupled fixed-combination stochastic programming and fractional programming; Based on the optimization results, the water allocation for each department is determined, and based on the water shortage under different water inflow and power generation levels, hydropower generation and crop planting patterns are arranged to achieve the maximum water allocation benefit per unit. In step B, the objective function Maxf for maximizing marginal benefits in the constructed reservoir water resource optimization allocation model includes (a) the system benefits generated by hydropower and agricultural irrigation activities, including expected benefits and economic penalties (economic losses) when the expected goals cannot be met; and (b) total water consumption, specifically represented as follows: ； in, ， ， In the formula: i represents the specific crop type (i = 1,2,3,4,5); j represents the specific region (j = 1,2,3); t represents the specific period (t = 1,2,3,4,5); h represents different water inflow levels (h = 1,2,3,4,5); f is the marginal benefit (USD); NB ij Net benefit of crop i in region j; TA ij Irrigation targets for crop i in region j; TEg t The target for hydropower generation during period t; NP t The net benefit of hydropower generation during period t; Eg t p represents the reservoir's power generation during period t; th C represents the probability of water flow level h occurring in period t; ij Economic penalties for failing to meet irrigation targets for crop i in region j; CP t The economic penalty when the hydropower generation target for period t cannot be met; AC ijth Planting area for each crop; SW ijth Irrigation water allocation for each crop; GW th Water allocation for hydropower generation; m jt Channel delivery efficiency in region j during period t; The constraints are B01~B09 B01) Water balance constraint ； In the formula, S th S(t) represents the reservoir's water storage capacity at the level of the inflow at the end of period t; t-1),h q represents the reservoir's water storage capacity at the initial inflow level h during period t; th WR represents the inflow volume of the reservoir at the inflow level of time t (h). th E represents the reservoir discharge at the inflow level of time t (h). th The reservoir evapotranspiration rate under the inflow level of h during period t; B02) Water resource availability constraints ； In the formula, LW ijth Calculate the amount of water to be used for leaching each crop based on its salinity. B03) Constraints on available irrigable water resources ； In the formula, WPC ijt Let θ be the water requirement per unit area of ​​crop i in region j at period t; jt For the field irrigation efficiency in region j during period t; B04) Soil salinity leaching constraints ； In the formula, LF ijt Salinity leaching fraction of crop i in region j during period t B05) Water resource constraints for power generation ； In the formula, GW min,t The minimum water demand for hydropower generation during period t; GW max,t The water demand corresponding to the installed capacity of the hydropower station in period t; B06) Power Supply Constraints ； In the formula, PED ijt The electricity required per unit of water transport for crop i in region j during period t; EP jh for Maximum irrigation power distribution in region j during period t; B07) Farmland Area Constraints ； B08) Food Security Constraints ； ； In the formula, TAC j This represents the maximum permissible arable land area in region j. The maximum area ratio of crop i in region j; Let i be the minimum area ratio of crop i in region j; B09) Agricultural Permitted Non-point Source Pollution Constraints ； In the formula, MF ij Fertilizer application rate for crop i in region j during period t; PP k The content of pollutant K in fertilizer; φ ikt TP represents the emission rate of pollutant k during period t. kt Let k be the total emissions of pollutant during period t; Step C includes: Step C1: Based on the basic water resources situation in the basin and the actual water demand of users, a fixed-combination stochastic fractional programming method is developed to optimize the water allocation for each user. Compared with traditional methods, this method can simultaneously handle the uncertainty represented by probability functions in the model and the contradictory relationship between the two objectives. Step C2: Select 5 inflow levels (h = 1, 2, 3, 4, 5) and 5 hydropower generation levels (α = 1, 0, 75, 0.6, 0.45, 0.3), and use Lingo programming software to calculate sub-models under different level combinations to obtain reservoir water resource optimization allocation schemes, thereby alleviating the water use contradiction between upstream hydropower generation and downstream agricultural irrigation.

2. The reservoir water resource optimization allocation method based on multi-stage and multi-objective as described in claim 1, characterized in that, Step A includes: Step A1: Based on the economic development of the study area and the impact of human activities, predict the water resource demand of each user in the watershed, and then set the upper and lower limits of the water demand of each user. Step A2: Based on the natural geographical conditions within the basin, predict the available water resources within the basin, and use the obtained available water resources as the input boundary of the optimization model. Step A3: Use the obtained water resource demand and available water resources of each user as the basic input data for the water resource optimization and allocation module.

Images