A method for determining multi-dimensional drought resistance early warning parameters of a cascade reservoir group

Abstract

The present application belongs to the technical field of hydraulic engineering, and discloses a method for determining multi-dimensional drought resistance early warning parameters of cascade reservoir groups, which comprises the following steps: constructing multiple drought scenarios according to the basins to which the cascade reservoir groups belong, and constructing a conventional scheduling model and a drought resistance scheduling model based on the cascade reservoir groups; inputting each drought scenario into the drought resistance scheduling model, and determining the limited supply early warning water level of each reservoir under each drought scenario according to the solution of the drought resistance scheduling model; inputting each drought scenario into the conventional scheduling model, and determining the drought resistance early warning period and the limited water supply coefficient of each reservoir under each drought scenario according to the scheduling comparison results of the cascade reservoir groups under the two scheduling models. The present application determines the limited supply early warning water level of the cascade reservoir groups in the basin, determines the drought resistance early warning period and the limited water supply coefficient, constructs a drought early warning system, and predicts the drought trend in advance, so as to implement artificial drought and limited water supply measures, effectively store the water quantity of the reservoirs, and reduce the impact of drought on the balance between water supply and demand.

Description

TECHNICAL FIELD

[0001] The application discloses a method for determining multi-dimensional drought resistance early warning parameters of a cascade reservoir group, and belongs to the technical field of water conservancy projects. BACKGROUND

[0002] Drought disaster is one of the most common and most serious natural disasters in the world, and its severity cannot be ignored. It not only destroys the natural environment and affects agricultural production, but also poses a major threat to the sustainable development of society and economy. According to statistics, the economic loss caused by drought each year is as high as 6-8 billion US dollars, far more than other meteorological disasters.

[0003] Reservoirs, as key engineering measures for basin drought resistance and disaster mitigation, play an irreplaceable role in relieving drought and ensuring water supply through their storage capacity. With the help of the storage function of reservoirs, the natural runoff process of the main stream and tributaries of the basin can be scientifically adjusted to match the production, life and ecological water demand process of different regions, thereby effectively alleviating the loss caused by drought.

[0004] However, there are obvious deficiencies in the construction of current reservoir drought resistance early warning systems. Especially in the joint drought resistance of cascade reservoir groups, there is a lack of systematic methods from the overall perspective of the basin and comprehensive consideration of water supply and demand on both sides. The existing early warning system mainly focuses on the monitoring and prediction of a single reservoir or a local area, ignoring the synergistic effect between multiple reservoirs in the basin and the hydrological connection between upstream and downstream, resulting in low efficiency of water resource allocation and difficulty in achieving optimal allocation and efficient use.

[0005] In addition, the traditional early warning mode also cannot fully reflect the complex hydrological process in the basin and the mutual influence between the reservoirs, making the early warning information lack accuracy and timeliness. SUMMARY

[0006] The purpose of the present application is to provide a method for determining multi-dimensional drought resistance early warning parameters of a cascade reservoir group, which aims to solve the limitations in current technology. Traditional early warning modes mainly focus on the monitoring and prediction of a single reservoir or a local area, often ignoring the synergistic effect between multiple reservoirs in the entire basin and the close hydrological connection between upstream and downstream, thereby making it difficult to achieve optimal allocation and efficient use of water resources. In order to overcome these technical difficulties, the present application proposes a method for determining multi-dimensional drought resistance early warning parameters of a cascade reservoir group, the specific scheme being as follows:

[0007] A method for determining multi-dimensional drought resistance early warning parameters of a cascade reservoir group, comprising:

[0008] Step 1, constructing multiple drought scenarios according to the basin to which the cascade reservoir group belongs, and a conventional scheduling model and a drought resistance scheduling model based on the cascade reservoir group;

[0009] Step 2, input each drought scenario into the drought-resistant scheduling model, and determine the limited supply warning water level of each reservoir under each drought scenario according to the solving result of the drought-resistant scheduling model;

[0010] Step 3, input each drought scenario into the conventional scheduling model, and determine the drought-resistant warning period and the limited water supply coefficient of each reservoir under each drought scenario according to the scheduling comparison result of the cascade reservoir group under the two scheduling models.

[0011] Preferably, a plurality of drought scenarios are constructed, specifically including:

[0012] A plurality of drought scenarios are constructed according to the hydrological data of the basin to which the cascade reservoir group belongs and the preset drought tolerance corresponding to different scenarios in the basin.

[0013] Preferably, a plurality of drought scenarios are constructed according to the hydrological data of the basin to which the cascade reservoir group belongs and the preset drought tolerance corresponding to different scenarios in the basin, specifically including:

[0014] A plurality of typical years corresponding to different drought levels are selected according to the historical hydrological data of the basin to which the cascade reservoir group belongs.

[0015] A plurality of drought scenarios are constructed according to the hydrological process of the plurality of typical years and the preset drought tolerance corresponding to different scenarios in the basin.

[0016] Preferably, a plurality of typical years corresponding to different drought levels are selected according to the historical hydrological data, specifically including:

[0017] A plurality of typical years corresponding to different drought levels are selected according to the difference between the water demand process corresponding to different scenarios and the inflow of the cascade reservoir group.

[0018] Preferably, the optimization target of the drought-resistant scheduling model is:

[0019] The sum of the anomaly values of the water supply satisfaction degree of different scenarios in the basin is minimum.

[0020] Preferably, the constraint condition of the drought-resistant scheduling model is:

[0021] Reservoir water level constraint, reservoir group water balance constraint, reservoir outflow constraint, warning water shortage rate threshold constraint and non-negative constraint.

[0022] Preferably, the warning water shortage rate threshold is determined according to the warning level under the drought level corresponding to the drought scenario.

[0023] Preferably, the step 3 specifically includes:

[0024] Each drought scenario is input into the conventional scheduling model, and the scheduling comparison result of the cascade reservoir group under the conventional scheduling model and the drought-resistant scheduling model under each drought scenario is determined.

[0025] determining the drought resistance warning period and the limited water supply coefficient of each reservoir under each drought scenario according to the scheduling comparison result.

[0026] Preferably, the drought resistance warning period of each reservoir under each drought scenario is determined according to the scheduling comparison result, and specifically includes:

[0027] In the scheduling comparison result, the period in which the outflow of each reservoir in the drought resistance scheduling model is less than the corresponding period of the conventional scheduling model is recorded as the drought resistance warning period of the reservoir under the corresponding drought scenario.

[0028] Preferably, the limited water supply coefficient of each reservoir under each drought scenario is determined according to the scheduling comparison result, and specifically includes:

[0029] In the drought resistance warning period, the proportion of the simulated water supply of each reservoir to each water-using unit in the drought resistance scheduling model under the corresponding drought scenario is recorded as the limited water supply coefficient of the reservoir under the corresponding drought scenario.

[0030] Beneficial effects: The multi-dimensional warning parameter determination method of the present application establishes a drought resistance scheduling model of a cascade reservoir group in a river basin to obtain an optimized reservoir scheduling mode, guides the reservoir to release stored water resources flexibly and efficiently, and converts the original possible “concentrated destruction” into “wide and shallow destruction”, thereby reducing the impact of drought on the balance between water supply and demand, protecting the stable operation of regional social and economic activities and the sustainability of the ecological environment. By predicting the drought trend and implementing artificial drought and limited water supply measures, the water quantity in the reservoir can be effectively stored. The present application uses the limited supply warning water level, the drought resistance warning period and the limited water supply coefficient of the reservoir as the parameters of the multi-dimensional warning mechanism. In the drought resistance warning mode, the decision maker can make early warning and take effective measures through the determination method of the warning parameters to reduce the loss caused by drought, so that the overall water supply situation of the water demand department is better. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 is the water demand process in a light drought and a medium drought typical year in the embodiment of the present application, wherein (a) is the water demand process in a light drought typical year, and (b) is the water demand process in a medium drought typical year;

[0032] Figure 2 is the water demand process in a heavy drought and a severe drought typical year in the embodiment of the present application, wherein (c) is the water demand process in a heavy drought typical year, and (d) is the water demand process in a severe drought typical year;

[0033] Figure 3 is the limited supply warning water level of Longyangxia Reservoir in the embodiment of the present application;

[0034] Figure 4 is the limited supply warning water level of Liujiaxia Reservoir in the embodiment of the present application;

[0035] Figure 5 is the limited supply early warning water level of Xiaolangdi Reservoir in the embodiment of the present application;

[0036] Figure 6 is the scheduling process of Longyangxia Reservoir under two models in the embodiment of the present application;

[0037] Figure 7 is the scheduling process of Liujiaxia Reservoir under two models in the embodiment of the present application;

[0038] Figure 8 is the scheduling process of Xiaolangdi Reservoir under two models in the embodiment of the present application;

[0039] Figure 9 is the limited supply water coefficient corresponding to different drought scenarios in the embodiment of the present application;

[0040] Figure 10 is a flowchart of a method for determining multi-dimensional early warning parameters of joint drought resistance of a cascade reservoir group according to the present application. DETAILED DESCRIPTION

[0041] In order to make the objectives, technical solutions and advantages of the present application clearer and more comprehensible, the present application will be further described in detail below with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the protection scope of the present application.

[0042] The present application proposes a method for determining multi-dimensional early warning parameters of joint drought resistance of a cascade reservoir group as shown in Figure 10 , which aims to improve the drought resistance capacity in the basin through scientific and reasonable prediction and scheduling. Specifically, the present embodiment will take the Yellow River Basin as an example to describe the specific implementation of the present application:

[0043] Step 1, constructing multiple drought scenarios according to the hydrological data of the basin to which the cascade reservoir group belongs, and constructing a conventional scheduling model and a drought resistance scheduling model of the cascade reservoir group;

[0044] Further, multiple drought scenarios are constructed, specifically including: constructing multiple drought scenarios according to the hydrological data of the basin to which the cascade reservoir group belongs and the preset drought tolerance corresponding to different scenarios in the basin. The preset drought tolerance corresponding to different scenarios represents the preset drought tolerance of each water user in the basin.

[0045] Further, multiple drought scenarios are constructed according to the hydrological data of the basin to which the cascade reservoir group belongs and the preset drought tolerance of each water user in the basin, specifically including: obtaining historical hydrological data of the basin to which the cascade reservoir group belongs, selecting multiple typical years corresponding to different drought levels according to the historical hydrological data; constructing multiple drought scenarios according to the hydrological processes of the multiple typical years and the preset drought tolerance of each water user in the basin.

[0046] In this embodiment, the data includes consulting the hydrological yearbook and other data of the Yellow River Basin, and collecting long-term water inflow and water demand information of the main and tributary rivers in the basin. Specifically, this includes the Longyangxia Reservoir, Liujiaxia Reservoir, and Xiaolangdi Reservoir, as well as the Tao River, Huangshui River, Datong River, Fen River, Wei River, Qin River, Yiluo River, and Dawen River connected to the Yellow River mainstream. Taking the hydrological year as the statistical period, the water shortage rate of each hydrological year is calculated based on the water demand of water users. The calculation formula of the water shortage rate is as follows:

[0047]

[0048] Where:

[0049] for Water shortage rate in the hydrological year;

[0050] and They are Water demand and water inflow in the hydrological year.

[0051] Furthermore, the drought level of a typical year is determined according to the water shortage rate of the typical year.

[0052] After obtaining the water shortage rate for each hydrological year, the drought level for each hydrological year is determined according to the drought level classification standard shown in Table 1. Finally, multiple typical years of different drought levels are selected based on the water demand process corresponding to different scenarios and the water inflow difference of the cascade reservoir group. That is, based on the principle of the most unfavorable water demand process of the water user (that is, the water demand process of each water user is the most different from the water inflow of the cascade reservoir group), a typical year is selected from each warning level under each drought level. In this embodiment, the water inflow and water demand process of the typical years of different drought levels are as follows: Figure 1 and Figure 2 shown.

[0053] Table 1 Drought levels at different water shortage rates

[0054]

[0055] Subsequently, considering the water demand characteristics and ability to withstand water shortages of water-using units in the Yellow River Basin, the drought tolerance of each water-using unit was preset, and different drought scenarios were constructed based on the drought tolerance of each water-using unit and various typical year combinations.

[0056] Constructing a conventional scheduling model and a drought-resistant scheduling model, specifically including: collecting data on the cascade reservoir groups in the Yellow River Basin, as well as the needs of water users in different time periods, and constructing a conventional scheduling model and a drought-resistant scheduling model.

[0057] The construction of the conventional scheduling model is as follows:

[0058] Considering the water inflow of the main and tributary rivers in the Yellow River Basin, the water demand of each reservoir is calculated using the following formula:

[0059]

[0060] Where:

[0061] is the simulation period;

[0062] for Reservoir in Water storage capacity during the period;

[0063] for The amount of water inflow from the main stream during the period;

[0064] for The tributaries within the reservoir's controlled catchment area are The amount of water inflow during the period.

[0065] A conventional scheduling model is established based on standardized operating rules and various constraints. The conventional scheduling model is used to simulate the operation of cascade reservoirs. The conventional scheduling model calculates the relationship between the water storage and water demand of each reservoir in the Yellow River Basin during the simulation period in each drought scenario; it simulates the operation process of each reservoir in the Yellow River Basin. If the reservoir water storage is greater than the water demand, water will be supplied on demand; if the reservoir water storage is less than the water demand, the reservoir will use all the water above the dead water level for water supply; if the reservoir water storage is greater than the reservoir's maximum water storage capacity, the reservoir will increase the discharge to keep the water storage at the maximum level.

[0066] The constraints of the conventional scheduling model include:

[0067] (1) Water balance constraint, as shown in the following formula:

[0068]

[0069] Where:

[0070] It is The year The storage capacity of the reservoir at the end of the time period;

[0071] It is years The inflow to the reservoir during the period;

[0072] It is Year Downflow within a period of time.

[0073] (2) Reservoir water level constraint, as shown in the following formula:

[0074]

[0075] Where:

[0076] It is the dead water level of the reservoir;

[0077] It is The year The water level of the reservoir at the end of the period;

[0078] In the non-flood season, it indicates the normal water storage level of the reservoir; in the flood season, it indicates the flood limit water level.

[0079] (3) Downflow flow constraint, as shown in the following formula:

[0080]

[0081] Where:

[0082] yes Minimum downstream flow requirement within the time period;

[0083] It is years The outflow from the reservoir at the end of the period;

[0084] yes The maximum downstream flow demand during the period.

[0085] (4) Non-negative constraints: all variables are non-negative.

[0086] The construction of drought resistance scheduling model specifically includes:

[0087] Taking the water level and supply limit warning water level of each reservoir in the cascade reservoir group in the Yellow River basin as decision variables, and minimizing the sum of the water supply satisfaction deviation values ​​of each water user in the Yellow River basin (minimizing intra-year fluctuation) as the optimization objective, a drought-resistant scheduling model is constructed based on the optimization objectives and constraints.

[0088] The optimization objective of the drought resistance scheduling model is:

[0089]

[0090] Where:

[0091] The goal for model optimization;

[0092] For the Period Water supply satisfaction in each region;

[0093] for Average water supply satisfaction level in the region;

[0094] The average value of the annual water supply satisfaction deviation value;

[0095] is the total number of simulation periods;

[0096] is the total number of regions.

[0097] The constraints of the drought-resistant scheduling model include reservoir water level constraints, reservoir group water balance constraints, reservoir discharge flow constraints, warning water shortage rate threshold constraints, and non-negative constraints, as follows:

[0098] (1) Water balance constraint, as shown in the following formula:

[0099]

[0100] Where:

[0101] It is The year The storage capacity of the reservoir at the end of the time period;

[0102] It is years The inflow to the reservoir during the time period;

[0103] It is Year Downflow within a period of time.

[0104] (2) Reservoir water level constraint, as shown in the following formula:

[0105]

[0106] Where:

[0107] It is the dead water level of the reservoir;

[0108] It is The year The water level of the reservoir at the end of the period;

[0109] In the non-flood season, it indicates the normal water storage level of the reservoir; in the flood season, it indicates the flood limit water level.

[0110] (3) the outflow constraint, as shown in the following formula:

[0111]

[0112] In the formula:

[0113] is the minimum outflow requirement in the period;

[0114] is the outflow of the reservoir at the end of the period in the year;

[0115] is the maximum outflow requirement in the period. (4) the early warning water shortage rate threshold constraint, as shown in the following formula:

[0116]

[0117] In the formula:

[0118]

[0119] is the water shortage rate of the basin under the early warning level of the i-th drought level; is the water shortage rate of the basin under the early warning level of the i-th drought level.

[0120] is the water shortage rate of the basin under the early warning level of the i-th drought level. (5) the non-negative constraint, all variables are non-negative.

[0121] The above early warning water shortage rate threshold is determined according to the early warning level under the drought level corresponding to the drought scenario, and the early warning level specifically includes I, II, III, and IV, and the early warning water shortage rate threshold represents the threshold corresponding to the simulation and scheduling of the drought-resistant scheduling model under each early warning level.

[0122] Step 2, input each drought scenario into the drought-resistant scheduling model, and determine the limited supply early warning water level of each reservoir under the drought scenario according to the solving result of the drought-resistant scheduling model;

[0123] Specifically, a certain drought scenario is input into the conventional scheduling model to simulate the scheduling process of the cascade reservoir group in the Yellow River Basin, and the drought scenario is input into the drought-resistant scheduling model to simulate the drought-resistant scheduling process of the cascade reservoir group in the Yellow River Basin.

[0124]

[0125] ​​​​​​Using the Gurobi solver package in Python, the conventional and drought-resistant operation models were compiled and solved, simulating the conventional and drought-resistant operation processes of the cascade reservoirs for the input drought scenario. The reservoir water levels obtained during the drought-resistant operation process are recorded as the supply restriction warning water levels corresponding to the warning level for the drought severity of the scenario. The supply restriction warning water levels serve as indicative indicators for issuing drought warning signals in the basin. Water restriction is used to provide early warnings during the initial reservoir operation phase, allowing for artificial water restrictions to increase water supply in the later stages of the drought. Higher drought severity requires greater advance water storage. Therefore, the supply restriction warning water levels should vary with the severity of the drought. Furthermore, for the same drought severity, the supply restriction warning water levels should also differ for different drought scenarios. This means that different supply restriction warning water levels correspond to different warning levels and therefore require different drought mitigation measures.

[0126] In this embodiment, the water levels corresponding to the different warning levels of Longyangxia, Liujiaxia and Xiaolangdi reservoirs in the Yellow River Basin under different drought levels are as follows: Figures 3-5 shown.

[0127] Step 3: Input each drought scenario into the conventional scheduling model, and determine the drought warning period and restricted water supply coefficient of each reservoir under each drought scenario based on the scheduling comparison results of the cascade reservoir group under the two scheduling models.

[0128] Furthermore, the step 3 specifically includes:

[0129] Each drought scenario is input into the conventional scheduling model to determine a comparison of the scheduling results of the cascade reservoir group under the conventional scheduling model and the drought relief scheduling model for each drought scenario. Based on these scheduling comparison results, the drought relief warning period and restricted water supply coefficient for each reservoir in each drought scenario are determined. Specifically, the period in the scheduling comparison results where the outflow from each reservoir in the drought relief scheduling model is less than the corresponding period in the conventional scheduling model is recorded as the drought relief warning period for that reservoir in that drought scenario. Furthermore, during the drought relief warning period, the proportion of simulated water supply from each reservoir to each user in the drought relief scheduling model for that drought scenario is recorded as the restricted water supply coefficient for that reservoir in that drought scenario. The drought relief warning period provides flexibility to coordinate water needs for drought relief in the later stages of the drought. A longer warning period allows for more drought relief water to be reserved for the later stages of the drought, making it more likely that wide-shallow damage will occur throughout the scheduling period. A shorter warning period can exacerbate irreversible drought losses in the early stages of the drought if drought event forecasts are significantly inaccurate. The water supply restriction coefficient is a drought mitigation measure corresponding to the water supply restriction level. By artificially restricting water supply during the warning period, the amount of water available later in the season is increased. Different water supply restriction coefficients correspond to different warning water levels.

[0130] The height of the limited supply early warning water level affects the drought resistance early warning period and the limited supply proportion in the early warning period. The longer the early warning period is, the smaller the limited supply coefficient is; the shorter the early warning period is, the larger the limited supply coefficient is.

[0131] Specifically, the scheduling process and the drought resistance scheduling process of the same drought scenario are compared, and the regular scheduling process and the drought resistance scheduling process of the reservoir in the two scheduling models are compared. If the discharge flow of the reservoir in the drought resistance scheduling process is less than that in the regular scheduling process, the period is the drought resistance early warning period. Specifically, in the embodiment, the regular scheduling process and the drought resistance scheduling process of the Longyangxia Reservoir, the Liujiaxia Reservoir and the Xiaolangdi Reservoir under the same drought scenario are plotted in the same graph, and the plotted graph of each reservoir under different drought grades is as shown in Figures 6-8 .

[0132] The simulation water supply rate of different water units in the drought resistance early warning period is recorded as the limited supply coefficient. Specifically, in the embodiment, the limited supply coefficients of the basins under different drought grades are as shown in Figure 9 . The limited supply coefficients of the tributaries of the Yellow River Basin in the embodiment are as shown in Table 2.

[0133] Table 2: Limited supply coefficients of tributaries

[0134]

[0135] The application determines the drought resistance early warning period and the limited supply coefficient by calculating the limited supply early warning water level of the cascade reservoir group in the basin, constructs a drought early warning system, predicts the drought trend in advance, implements artificial drought and limited water supply measures, and effectively stores the water quantity of the reservoir. When drought occurs, the reservoir under the guidance of the model can release the stored water resources more flexibly and efficiently, and convert the original possible "concentrated destruction" into "wide and shallow destruction", thereby reducing the impact of drought on water supply and demand balance, protecting the stable operation of regional social and economic activities and the sustainability of the ecological environment. Reservoir decision makers can take effective measures to reduce drought losses by early warning, so that the overall water supply of the water demand department is better.

[0136] The above is only a few embodiments of the application, and does not limit the application in any form. Although the application is disclosed as above with preferred embodiments, it is not intended to limit the application. Any skilled person in the art can make some changes or modifications to the disclosed technical content without departing from the scope of the technical solution of the application, which are equivalent to equivalent embodiments, and belong to the scope of the technical solution.

Claims

1. A method for determining multi-dimensional early warning parameters for joint drought relief in a cascade reservoir group, characterized in that: include: Step 1: Construct multiple drought scenarios and conventional and drought-resistant operation models based on the cascade reservoir groups according to the river basins to which the cascade reservoir groups belong; The conventional scheduling model is used to simulate the operation process of the cascade reservoir group according to the supply and demand relationship between the water storage capacity and water demand of each reservoir during the simulation period in each drought scenario; The drought-resistant scheduling model is used to simulate the operation process of the cascade reservoir group with the optimization goal of minimizing the sum of the water supply satisfaction deviation values ​​of each water user in the basin; Step 2: Input each drought scenario into the drought relief scheduling model, and determine the supply limit warning water level of each reservoir under each drought scenario according to the solution of the drought relief scheduling model; Step 3: Input each drought scenario into the conventional scheduling model, and determine the drought warning period and water supply restriction coefficient of each reservoir under each drought scenario based on the scheduling comparison results of the cascade reservoir group under the two scheduling models; Specifically, the period in which the outflow of each reservoir in the drought-resistant scheduling model is less than the corresponding period in the conventional scheduling model in the scheduling comparison result is recorded as the drought-resistant warning period of the reservoir in the corresponding drought scenario; During the drought warning period, the proportion of simulated water supply from each reservoir to each water user under the corresponding drought scenario in the drought scheduling model is recorded as the restricted water supply coefficient of the reservoir under the corresponding drought scenario.

2. The method for determining multi-dimensional early warning parameters for joint drought relief of cascade reservoirs according to claim 1, characterized in that: Build multiple drought scenarios, including: Multiple drought scenarios are constructed based on the hydrological data of the basin to which the cascade reservoir group belongs and the preset drought tolerance corresponding to different scenarios in the basin.

3. The method for determining multi-dimensional early warning parameters for joint drought relief of cascade reservoirs according to claim 2, characterized in that: Multiple drought scenarios are constructed based on the hydrological data of the basin to which the cascade reservoirs belong and the preset drought tolerance corresponding to different scenarios in the basin, including: Selecting multiple typical years corresponding to different drought levels based on historical hydrological data of the basin to which the cascade reservoir group belongs; A plurality of drought scenarios are constructed according to the hydrological processes of the plurality of typical years and preset drought tolerances corresponding to different scenarios in the basin.

4. The method for determining multi-dimensional early warning parameters for joint drought relief of cascade reservoirs according to claim 3 is characterized in that: Based on the historical hydrological data, several typical years corresponding to different drought levels are selected, including: A plurality of typical years of different drought levels are selected according to the water demand processes corresponding to different scenarios and the difference in water inflow of the cascade reservoir group.

5. The method for determining multi-dimensional early warning parameters for joint drought relief of cascade reservoirs according to claim 1, characterized in that: The constraints of the drought resistance scheduling model are: Reservoir water level constraints, reservoir group water balance constraints, reservoir discharge flow constraints, warning water shortage rate threshold constraints and non-negative constraints.

6. The method for determining multi-dimensional early warning parameters for joint drought relief of cascade reservoirs according to claim 5, characterized in that: The warning water shortage rate threshold is determined according to the warning level under the drought level corresponding to the drought scenario.

Images