A medium- and long-term runoff prediction method and system based on reservoir group operation domain

Abstract

The present invention provides a medium- and long-term runoff prediction method and system based on the dispatching domain of a reservoir group. According to the actual dispatching data of the reservoir group in the basin, the water level characteristics of different time scales and different controlling reservoir nodes are statistically analyzed to establish a reservoir dispatching domain. The flow domain affected by the dispatching of the reservoir group is analyzed based on the reservoir dispatching domain. The natural flow prediction results are superimposed on the flow domain and tendency value to achieve medium- and long-term runoff prediction under the influence of the dispatching of the reservoir group, providing important technical support for the comprehensive development and management of water resources in the basin under the influence of the dispatching of the reservoir group, the dispatching of water projects, and the prevention of water and drought disasters.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrology and water resources forecasting and dispatching, and in particular to a medium- and long-term runoff prediction method and system based on a reservoir group dispatching domain. Background Art

[0002] Medium- and long-term runoff forecasting is an important foundation for comprehensive development and management of river basin water resources, water project scheduling, and flood and drought prevention. Runoff forecasting is affected by many complex factors such as meteorological forecasting, reservoir group scheduling, and basin underlying surface characteristics. How to improve the prediction accuracy is still a technical problem that needs to be solved in the field of hydrological and water resources forecasting and scheduling technology.

[0003] At present, there are two main ways to predict medium- and long-term runoff: one is based on mathematical statistics, using climate and environmental factors and runoff to establish relevant models for prediction; the other is based on hydrological models, and the prediction results are obtained by inputting meteorological forecast products into the model, which has certain physical meanings. Both models lack objective consideration of the impact of reservoir group operation. If the predicted natural flow process can be superimposed on the reservoir group operation in the actual operation process, the water inflow process can be predicted more accurately.

[0004] The operation of a group of reservoirs is cyclical and has certain rules. By establishing a reservoir group operation domain, we can find similar operation situations on the same historical time scale, predict the impact of reservoir group operation, superimpose the operation impact on the natural water inflow forecast, and predict the actual water inflow process. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and provide a medium- and long-term runoff prediction method and system based on the reservoir group scheduling domain, analyze the feasible domain of reservoir group scheduling, and improve the rationality of medium- and long-term basin prediction.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] The present invention provides a medium- and long-term runoff prediction method based on a reservoir group dispatching domain, comprising:

[0008] S1. Collect and organize the long-sequence flood report or compilation data of the hydrological and meteorological control sections of the study basin, specifically: runoff data including the flow of hydrological stations, water level data of hydrological stations and water level stations, inflow and outflow of reservoir stations, reservoir water level data; rainfall data including the average rainfall data of each sub-area in the basin;

[0009] S2. For any reservoir station node, statistically analyze the water levels at the end of different time nodes during the dispatching period of ten days, months, seasons and years, calculate the upper and lower boundaries of the operating water levels at different time nodes during the dispatching period, and form the operating water level envelope, i.e., the reservoir dispatching domain;

[0010] S3. For any reservoir node, the average and median of the operating water level of any reservoir node are calculated to form an operating water level envelope, i.e., the reservoir dispatching tendency domain. The middle value of the tendency domain is the reservoir dispatching water level tendency value of any reservoir node.

[0011] S4. For any reservoir node and any scheduling period, the water level difference between the same water level representative values ​​at two adjacent time nodes is calculated, and the corresponding storage capacity is checked according to the water level storage capacity curve, and the storage capacity difference is calculated. Based on the storage capacity difference, the average flow intercepted or increased by any reservoir between the two adjacent time nodes is calculated, that is, the scheduling impact flow of any reservoir in the corresponding time period;

[0012] S5. Calculate the dispatching flow rate corresponding to different water level representative values, and use the maximum and minimum values ​​to determine the flow rate envelope as the flow rate domain affected by the reservoir dispatching operation; calculate the center line of the flow rate domain as the flow rate tendency value affected by the reservoir dispatching operation;

[0013] S6. Based on the natural flow of a certain section obtained from the hydrological and meteorological forecast, the flow affected by the dispatching and operation of the upstream reservoir group of a certain section is superimposed, and the predicted flow range and tendency value of a certain section are calculated to complete the medium- and long-term runoff prediction at different time scales such as ten days, months, seasons and years.

[0014] Further, in said S1,

[0015] Runoff data includes flow data from hydrological stations: ,

[0016] Water level data from hydrological stations and water level stations: ,

[0017] Inflow to the reservoir station: ,

[0018] Outbound flow: ,

[0019] Reservoir water level data: ,

[0020] The rainfall data includes the average rainfall in each sub-area of ​​the basin:

[0021] ,

[0022] If the average rainfall data of the sub-region is lacking, the weighted average rainfall data of the sub-region can be obtained by the rainfall stations in the sub-region. The equation is:

[0023] ;

[0024] in, is the number of hydrological stations; is the number of hydrological stations and water level stations; is the number of reservoirs; is the number of partitions in the basin; is the number set of rainfall stations in each sub-area of ​​the basin; For the Average rainfall in each sub-area, mm; is the corresponding weight of the rainfall station, mm; For the Partition Rainfall series for each rain gauge station, mm.

[0025] Further, the S2 is specifically:

[0026] From the reservoir water level data The final water level series set of different time nodes in the dispatching period of each reservoir at the scale of ten days, month, season and year are extracted respectively, and the equation is:

[0027] ;

[0028] ;

[0029] ;

[0030] ;

[0031] in, The water level of each reservoir at the end of ten days; The water level of each reservoir at the end of the month is aggregated; The water level of each reservoir at the end of the season is aggregated; The water level of each reservoir at the end of the year is aggregated;

[0032] Indicates Reservoir The water level sequence at the end of each decade;

[0033] Indicates Reservoir Month-end water level sequence;

[0034] Indicates Reservoir The end-of-season water level series for each season;

[0035] Indicates Reservoir The year-end water level series of each year;

[0036] , , , Respectively represent The number of ten-day, monthly, seasonal and annual scale series for each reservoir;

[0037] The equation of the reservoir operation domain is:

[0038] ;

[0039] ;

[0040] ;

[0041] ;

[0042] in, , , and They are respectively the water level regulation domains at the end of ten days, month, season and year for each reservoir within one annual cycle.

[0043] Further, the S3 is specifically:

[0044] For any reservoir station node, , , and The average and median water levels at different time points during the operation period of each reservoir at the ten-day, monthly, seasonal and annual scales are extracted to form the operation tendency domain. The equation is:

[0045] ;

[0046] ;

[0047] ;

[0048] ;

[0049] in, , , and They are the water level dispatching tendency domains of each reservoir at the end of ten days, month, season and year respectively; Indicates Reservoir The average of the water level series at the end of each decade; Indicates Reservoir The average of the month-end water level series; Indicates Reservoir The average of the end-of-season water level series for each season; Indicates Reservoir The average of the water level series at the end of each year; Indicates Reservoir The median of the water level series at the end of each decade; Indicates Reservoir The median of the water level series at the end of the month; Indicates Reservoir The median of the end-of-season water level series for each season; Indicates Reservoir The median of the year-end water level series of each year;

[0050] Find the median of the dispatch tendency domain and obtain the dispatch water level tendency value set at the end of different dispatch periods. The equation is:

[0051] ;

[0052] ;

[0053] ;

[0054] ;

[0055] in, , , and They are respectively the sets of water level dispatch tendency values ​​at the end of ten-day, monthly, seasonal and annual scales for each reservoir.

[0056] Further, the S4 is specifically:

[0057] S401, according to the water level storage capacity curve of each reservoir node, the dispatching domain, dispatching tendency domain and dispatching tendency value set of each node at different time scales are converted into storage capacity by linear interpolation;

[0058] S402. Subtract the storage of adjacent time nodes at the same reservoir, time scale and boundary to obtain the storage difference of the time scale. The storage difference is divided by the length of the time scale to convert it into the influencing flow. The equation is as follows:

[0059] ;

[0060] ;

[0061] ;

[0062] ;

[0063] ;

[0064] in, , They are the affected traffic sets calculated based on the upper and lower limits of the scheduling domain at different time scales; , They are the affected traffic sets calculated based on the upper and lower limits of the scheduling tendency domain at different time scales; It is the set of impacted flows calculated based on the dispatch tendency values ​​at different time scales; , , , Respectively represent the mth reservoir, Time scale The storage capacity, when hour, , the time scale is ten days; when hour, , the time scale is months; when hour, , the time scale is quarter; when hour, , the time scale is years.

[0065] Further, the S5 is specifically:

[0066] The dispatching flow rate corresponding to different water level representative values ​​is calculated, and the envelope of the flow rate determined by the maximum and minimum values ​​is used as the flow rate domain affected by the reservoir dispatching operation; the center line of the flow rate domain is calculated as the flow rate tendency value affected by the reservoir dispatching operation. The equation is as follows:

[0067] ;

[0068] in, , , , They respectively represent the flow tendency domain affecting different time scales, the tendency value affecting the flow, the lower limit of the flow tendency domain affecting the flow, and the upper limit of the flow tendency domain affecting the flow.

[0069] Further, the S6 is specifically:

[0070] Consider runoff data including flow data from hydrological stations , water level data from hydrological stations and water level stations , Inflow to the reservoir station , outbound traffic , rainfall data including the average rainfall in each sub-area of ​​the basin The natural flow of a certain section is obtained by using the elements of hydrological and meteorological forecast, and the impact of the dispatching and operation of the upstream reservoir group of a certain section is superimposed on the corresponding time. The forecast flow range and tendency value of a certain section are calculated to complete the medium- and long-term runoff prediction at different time scales such as ten days, months, seasons and years.

[0071] Furthermore, a medium- and long-term runoff prediction system based on a reservoir group scheduling domain comprises at least one processor and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executed by the processor, and the instructions are used to be executed by the processor to implement the medium- and long-term runoff prediction method based on the reservoir group scheduling domain. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a flowchart of a medium- and long-term runoff prediction method based on the reservoir group operation domain;

[0073] Figure 2 The topological structure diagram of the control reservoir group in the upper reaches of Hanjiang River in the embodiment;

[0074] Figure 3 This is a schematic diagram of the Shiquan Reservoir dispatching area at the mid-term scale of the embodiment;

[0075] Figure 4 Schematic diagram of the Ankang Reservoir dispatching domain in the embodiment;

[0076] Figure 5 It is a schematic diagram of the Pankou Reservoir dispatching area in the embodiment;

[0077] Figure 6 It is a schematic diagram of the Huanglongtan Reservoir dispatching area in the embodiment;

[0078] Figure 7 It is a schematic diagram of the Danjiangkou Reservoir dispatching area in the embodiment. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0080] A medium- and long-term runoff prediction method based on the reservoir group operation domain includes:

[0081] S1. Collect and organize the long-sequence flood report or compilation data of the hydrological and meteorological control sections of the study basin, specifically: runoff data including the flow of hydrological stations, water level data of hydrological stations and water level stations, inflow and outflow of reservoir stations, reservoir water level data; rainfall data including the average rainfall data of each sub-area in the basin;

[0082] S2. For any reservoir station node, statistically analyze the water levels at the end of different time nodes during the dispatching period of ten days, months, seasons and years, calculate the upper and lower boundaries of the operating water levels at different time nodes during the dispatching period, and form the operating water level envelope, i.e., the reservoir dispatching domain;

[0083] S3. For any reservoir node, the average and median of the operating water level of any reservoir node are calculated to form an operating water level envelope, i.e., the reservoir dispatching tendency domain. The middle value of the tendency domain is the reservoir dispatching water level tendency value of any reservoir node.

[0084] S4. For any reservoir node and any scheduling period, the water level difference between the same water level representative values ​​at two adjacent time nodes is calculated, and the corresponding storage capacity is checked according to the water level storage capacity curve, and the storage capacity difference is calculated. Based on the storage capacity difference, the average flow intercepted or increased by any reservoir between the two adjacent time nodes is calculated, that is, the scheduling impact flow of any reservoir in the corresponding time period;

[0085] S5. Calculate the dispatching flow rate corresponding to different water level representative values, and use the maximum and minimum values ​​to determine the flow rate envelope as the flow rate domain affected by the reservoir dispatching operation; calculate the center line of the flow rate domain as the flow rate tendency value affected by the reservoir dispatching operation;

[0086] S6. Based on the natural flow of a certain section obtained from the hydrological and meteorological forecast, the flow affected by the dispatching and operation of the upstream reservoir group of a certain section is superimposed, and the predicted flow range and tendency value of a certain section are calculated to complete the medium- and long-term runoff prediction at different time scales such as ten days, months, seasons and years.

[0087] In S1,

[0088] Runoff data includes flow data from hydrological stations: ,

[0089] Water level data from hydrological stations and water level stations: ,

[0090] Inflow to the reservoir station: ,

[0091] Outbound flow: ,

[0092] Reservoir water level data: ,

[0093] The rainfall data includes the average rainfall in each sub-area of ​​the basin:

[0094] ,

[0095] If the average rainfall data of the sub-region is lacking, the weighted average rainfall data of the sub-region can be obtained by the rainfall stations in the sub-region. The equation is:

[0096] ;

[0097] in, is the number of hydrological stations; is the number of hydrological stations and water level stations; is the number of reservoirs; is the number of partitions in the basin; is the number set of rainfall stations in each sub-area of ​​the basin; For the Average rainfall in each sub-area, mm; is the corresponding weight of the rainfall station, mm; For the Partition Rainfall series for each rain gauge station, mm.

[0098] The S2 is specifically:

[0099] From the reservoir water level data The final water level series set of different time nodes in the dispatching period of each reservoir at the scale of ten days, month, season and year are extracted respectively, and the equation is:

[0100] ;

[0101] ;

[0102] ;

[0103] ;

[0104] in, The water level of each reservoir at the end of ten days; The water level of each reservoir at the end of the month is aggregated; The water level of each reservoir at the end of the season is aggregated; The water level of each reservoir at the end of the year is aggregated;

[0105] Indicates Reservoir The water level sequence at the end of each decade; Indicates Reservoir Month-end water level sequence;

[0106] Indicates Reservoir The end-of-season water level series for each season; Indicates Reservoir The year-end water level series of each year;

[0107] , , , Respectively represent The number of ten-day, monthly, seasonal and annual scale series for each reservoir;

[0108] The equation of the reservoir operation domain is:

[0109] ;

[0110] ;

[0111] ;

[0112] ;

[0113] in, , , and They are respectively the water level regulation domains at the end of ten days, month, season and year for each reservoir within one annual cycle.

[0114] The S3 is specifically:

[0115] For any reservoir station node, , , and The average and median water levels at different time points during the operation period of each reservoir at the ten-day, monthly, seasonal and annual scales are extracted to form the operation tendency domain. The equation is:

[0116] ;

[0117] ;

[0118] ;

[0119] ;

[0120] in, , , and They are the water level dispatching tendency domains of each reservoir at the end of ten days, month, season and year respectively; Indicates Reservoir The average of the water level series at the end of each decade; Indicates Reservoir The average of the month-end water level series; Indicates Reservoir The average of the end-of-season water level series for each season; Indicates Reservoir The average of the water level series at the end of each year; Indicates Reservoir The median of the water level series at the end of each decade; Indicates Reservoir The median of the water level series at the end of the month; Indicates Reservoir The median of the end-of-season water level series for each season; Indicates Reservoir The median of the year-end water level series of each year;

[0121] Find the median of the dispatch tendency domain and obtain the dispatch water level tendency value set at the end of different dispatch periods. The equation is:

[0122] ;

[0123] ;

[0124] ;

[0125] ;

[0126] in, , , and They are respectively the sets of water level dispatch tendency values ​​at the end of ten-day, monthly, seasonal and annual scales for each reservoir.

[0127] The S4 is specifically:

[0128] S401, according to the water level storage capacity curve of each reservoir node, the dispatching domain, dispatching tendency domain and dispatching tendency value set of each node at different time scales are converted into storage capacity by linear interpolation;

[0129] S402. Subtract the storage of adjacent time nodes at the same reservoir, time scale and boundary to obtain the storage difference of the time scale. The storage difference is divided by the length of the time scale to convert it into the influencing flow. The equation is as follows:

[0130] ;

[0131] ;

[0132] ;

[0133] ;

[0134] ;

[0135] in, , They are the affected traffic sets calculated based on the upper and lower limits of the scheduling domain at different time scales; , They are the affected traffic sets calculated based on the upper and lower limits of the scheduling tendency domain at different time scales; It is the set of impacted flows calculated based on the dispatch tendency values ​​at different time scales; , , , Respectively represent the mth reservoir, Time scale The storage capacity, when hour, , the time scale is ten days; when hour, , the time scale is months; when hour, , the time scale is quarter; when hour, , the time scale is years.

[0136] The S5 is specifically:

[0137] The dispatching flow rate corresponding to different water level representative values ​​is calculated, and the envelope of the flow rate determined by the maximum and minimum values ​​is used as the flow rate domain affected by the reservoir dispatching operation; the center line of the flow rate domain is calculated as the flow rate tendency value affected by the reservoir dispatching operation. The equation is as follows:

[0138] ;

[0139] in, , , , They respectively represent the flow tendency domain affecting different time scales, the tendency value affecting the flow, the lower limit of the flow tendency domain affecting the flow, and the upper limit of the flow tendency domain affecting the flow.

[0140] S6 specifically includes: considering the runoff data including the flow data of the hydrological station , water level data from hydrological stations and water level stations , Inflow to the reservoir station , outbound traffic , rainfall data including the average rainfall in each sub-area of ​​the basin The natural flow of a certain section is obtained by using the elements of hydrological and meteorological forecast, and the impact of the dispatching and operation of the upstream reservoir group of a certain section is superimposed on the corresponding time. The forecast flow range and tendency value of a certain section are calculated to complete the medium- and long-term runoff prediction at different time scales such as ten days, months, seasons and years.

[0141] A medium- and long-term runoff prediction system based on a reservoir group scheduling domain comprises at least one processor and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executed by the processor, and the instructions are used to be executed by the processor to implement the medium- and long-term runoff prediction method based on the reservoir group scheduling domain.

[0142] Example

[0143] Taking the Danjiangkou Reservoir as an example, medium- and long-term runoff prediction calculations considering the reservoir group dispatching domain are carried out to verify the feasibility and effectiveness of the method of the present invention.

[0144] See also Figures 2 to 7 The average inflow of Danjiangkou Reservoir in late June 2024 was restored by using the method of storage variable evolution superposition. The restored natural average inflow is 439m 3 / s, and took it as the assumed forecast value. The above method was used to calculate the tendency domain and tendency value affecting the flow of Shiquan, Ankang, Pankou and Huanglongtan reservoirs respectively, see Table 1.

[0145] The actual average inflow into Danjiangkou Reservoir in late June was predicted according to the influencing flow tendency domain of the upstream controlling reservoir of Danjiangkou Reservoir in late June in Table 1, as shown in Table 2. The actual situation fell within the prediction range, and the relative error of the predicted tendency value compared with the actual situation was only 7.2%, with high prediction accuracy.

[0146] Table 1 Statistics of the influence of the upstream controlling reservoirs of Danjiangkou Reservoir on the flow trend in late June

[0147]

[0148] Table 2 Statistical table of actual inflow forecast results of Danjiangkou Reservoir in late June

[0149]

[0150] From Table 1 to Table 2, it can be seen that the method of the present invention considers the calculation idea of ​​medium- and long-term runoff prediction in the reservoir operation domain, and can quickly realize the actual water inflow prediction affected by the regulation of the upstream reservoir group. On the basis of predicting natural runoff, the prediction result is obtained by superimposing the influencing flow domain. The prediction result of late June 2024 as an example also proves the feasibility and effectiveness of this method. It can be seen that this method has a superior application effect in the medium- and long-term runoff prediction of the reservoir group operation domain.

[0151] According to the above analysis, it can be seen that the method of the present invention is highly practical and can effectively solve the problem of medium- and long-term runoff prediction under the influence of reservoir group scheduling.

[0152] In summary, the present invention has the advantages of practicality and strong operability. It can quickly realize the medium- and long-term runoff prediction in the reservoir group operation domain, obtain the medium- and long-term runoff prediction results of important sections, and provide a more scientific and efficient new method for basin hydrological forecasting and reservoir operation.

[0153] The above-mentioned embodiments only express the implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be based on the attached claims.

Claims

1. A medium- and long-term runoff prediction method based on the reservoir group dispatching domain, characterized in that: include: S1. Collect and organize the long-sequence flood report or compilation data of the hydrological and meteorological control sections of the study basin, specifically: runoff data including the flow of hydrological stations, water level data of hydrological stations and water level stations, inflow and outflow of reservoir stations, reservoir water level data; rainfall data including the average rainfall data of each sub-area in the basin; S2. For any reservoir station node, statistically analyze the water levels at the end of different time nodes during the dispatching period of ten days, months, seasons and years, calculate the upper and lower boundaries of the operating water levels at different time nodes during the dispatching period, and form the operating water level envelope, i.e., the reservoir dispatching domain; S3. For any reservoir node, the average and median of the operating water level of any reservoir node are calculated to form an operating water level envelope, i.e., the reservoir dispatching tendency domain. The middle value of the tendency domain is the reservoir dispatching water level tendency value of any reservoir node. S4. For any reservoir node and any scheduling period, the water level difference between the same water level representative values ​​at two adjacent time nodes is calculated, and the corresponding storage capacity is checked according to the water level storage capacity curve, and the storage capacity difference is calculated. Based on the storage capacity difference, the average flow intercepted or increased by any reservoir between the two adjacent time nodes is calculated, that is, the scheduling impact flow of any reservoir in the corresponding time period; S5. Calculate the dispatching flow rate corresponding to different water level representative values, and use the maximum and minimum values ​​to determine the flow rate envelope as the flow rate domain affected by the reservoir dispatching operation; calculate the center line of the flow rate domain as the flow rate tendency value affected by the reservoir dispatching operation; S6. Based on the natural flow of a certain section obtained from the hydrological and meteorological forecast, the flow affected by the dispatching and operation of the upstream reservoir group of a certain section is superimposed, and the predicted flow range and tendency value of a certain section are calculated to complete the medium- and long-term runoff prediction at different time scales such as ten days, months, seasons and years.

2. The method for medium- and long-term runoff prediction based on the reservoir group dispatching domain according to claim 1 is characterized by: In S1, Runoff data includes flow data from hydrological stations: , Water level data from hydrological stations and water level stations: , Inflow to the reservoir station: , Outbound flow: , Reservoir water level data: , The rainfall data includes the average rainfall in each sub-area of ​​the basin: 、 If the average rainfall data of the sub-region is lacking, the weighted average rainfall data of the sub-region can be obtained by the rainfall stations in the sub-region. The equation is: ； in, is the number of hydrological stations; is the number of hydrological stations and water level stations; is the number of reservoirs; is the number of partitions in the basin; is the number set of rainfall stations in each sub-area of ​​the basin; For the Average rainfall in each sub-area, mm; is the corresponding weight of the rainfall station, mm; For the Partition Rainfall series for each rain gauge station, mm.

3. The method for medium- and long-term runoff prediction based on the reservoir group dispatching domain according to claim 2 is characterized in that: The S2 is specifically: From the reservoir water level data The final water level series set of different time nodes in the dispatching period of each reservoir at the scale of ten days, month, season and year are extracted respectively, and the equation is: ； ； ； ； in, The water level of each reservoir at the end of ten days; The water level of each reservoir at the end of the month is aggregated; The water level of each reservoir at the end of the season is aggregated; The water level of each reservoir at the end of the year is aggregated; Indicates Reservoir The water level sequence at the end of each decade; Indicates Reservoir Month-end water level sequence; Indicates Reservoir The end-of-season water level series for each season; Indicates Reservoir The year-end water level series of each year; , , , Respectively represent The number of ten-day, monthly, seasonal and annual scale series for each reservoir; The equation of the reservoir operation domain is: ； ； ； ； in, , , and They are respectively the water level regulation domains at the end of ten days, month, season and year for each reservoir within one annual cycle.

4. The method for medium- and long-term runoff prediction based on the reservoir group dispatching domain according to claim 3 is characterized in that: The S3 is specifically: For any reservoir station node, , , and The average and median water levels at different time points during the operation period of each reservoir at the ten-day, monthly, seasonal and annual scales are extracted to form the operation tendency domain. The equation is: ； ； ； ； in, , , and They are the water level dispatching tendency domains of each reservoir at the end of ten days, month, season and year respectively; Indicates Reservoir The average of the water level series at the end of each decade; Indicates Reservoir The average of the month-end water level series; Indicates Reservoir The average of the end-of-season water level series for each season; Indicates Reservoir The average of the water level series at the end of each year; Indicates Reservoir The median of the water level series at the end of each decade; Indicates Reservoir The median of the water level series at the end of the month; Indicates Reservoir The median of the end-of-season water level series for each season; Indicates Reservoir The median of the year-end water level series of each year; Find the median of the dispatch tendency domain and obtain the dispatch water level tendency value set at the end of different dispatch periods. The equation is: ； ； ； ； in, , , and They are respectively the sets of water level dispatch tendency values ​​at the end of ten-day, monthly, seasonal and annual scales for each reservoir.

5. The method for medium- and long-term runoff prediction based on the reservoir group dispatching domain according to claim 4 is characterized in that: The S4 is specifically: S401, according to the water level storage capacity curve of each reservoir node, the dispatching domain, dispatching tendency domain and dispatching tendency value set of each node at different time scales are converted into storage capacity by linear interpolation; S402. Subtract the storage of adjacent time nodes at the same reservoir, time scale and boundary to obtain the storage difference of the time scale. The storage difference is divided by the length of the time scale to convert it into the influencing flow. The equation is as follows: ； ； ； ； ； in, , They are the affected traffic sets calculated based on the upper and lower limits of the scheduling domain at different time scales; , They are the affected traffic sets calculated based on the upper and lower limits of the scheduling tendency domain at different time scales; It is the set of impacted flows calculated based on the dispatch tendency values ​​at different time scales; , , , Respectively represent the mth reservoir, Time scale The storage capacity, when hour, , the time scale is ten days; when hour, , the time scale is months; when hour, , the time scale is quarter; when hour, , the time scale is years.

6. A medium- and long-term runoff prediction method based on a reservoir group dispatching domain according to claim 5, characterized in that: The S5 is specifically: The dispatching flow rate corresponding to different water level representative values ​​is calculated, and the envelope of the flow rate determined by the maximum and minimum values ​​is used as the flow rate domain affected by the reservoir dispatching operation; the center line of the flow rate domain is calculated as the flow rate tendency value affected by the reservoir dispatching operation. The equation is as follows: ； in, , , , They respectively represent the flow tendency domain affecting different time scales, the tendency value affecting the flow, the lower limit of the flow tendency domain affecting the flow, and the upper limit of the flow tendency domain affecting the flow.

7. The method for medium- and long-term runoff prediction based on the reservoir group dispatching domain according to claim 6 is characterized in that: S6 specifically includes: considering the runoff data including the flow data of the hydrological station , water level data from hydrological stations and water level stations , Inflow to the reservoir station , outbound traffic , rainfall data including the average rainfall in each sub-area of ​​the basin The natural flow of a certain section is obtained by using the elements of hydrological and meteorological forecast, and the impact of the dispatching and operation of the upstream reservoir group of a certain section is superimposed on the corresponding time. The forecast flow range and tendency value of a certain section are calculated to complete the medium- and long-term runoff prediction at different time scales such as ten days, months, seasons and years.

8. A medium- and long-term runoff prediction system based on the reservoir group dispatching domain, characterized by: It includes at least one processor and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executed by the processor, and the instructions are used to be executed by the processor to implement a medium- and long-term runoff prediction method based on a reservoir group scheduling domain as described in any one of claims 1 to 7.

Images