Sub-basin-based distributed flood forecasting and dispatching model construction method and system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XI'AN UNIVERSITY OF ARCHITECTURE AND TECHNOLOGY
- Filing Date
- 2026-05-26
- Publication Date
- 2026-08-07
AI Technical Summary
[0004]本发明的目的在于克服现有水库群泄流削峰调度方法未能充分结合山洪由产流、汇流至河道传播的动力演进过程，导致各水库泄流时序和泄流量难以与洪峰传播及叠加规律精准匹配的问题，提供基于子流域的分布式洪水预报调度模型构建方法及系统
本发明通过获取山洪动力演进结果，结合各水库的空间位置、入库支流范围、出库汇入河道及其与指定控制断面之间的水力传导关系，确定各水库泄流水量到达指定控制断面的传播时长，能够使水库泄流调度不再仅依据单座水库自身库水位或入库流量进行，而是与山洪在流域内的传播过程相匹配，从而提高泄流调度的针对性和准确性。本发明通过将各水库候选泄流时段与传播时长进行匹配，计算指定控制断面的预测叠加流量，并以预测叠加流量不超过安全流量、不同水库泄流水量在指定控制断面错峰叠加为约束生成协同泄流方案，可以避免多个水库泄流水量与天然洪峰或其他水库泄流过程在同一时段集中叠加，降低指定控制断面的洪峰流量，提高流域整体削峰效果。本发明通过将协同泄流方案发送至泄流闸门执行端，并在泄流过程中采集实际出库流量、闸门运行状态和指定控制断面的实测流量进行动态调整，能够及时修正预测误差和执行偏差，提高水库群协同调度的实时性、可靠性和防洪安全性。
Smart Images
Figure CN122528748A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of water conservancy engineering technology, specifically relating to a method and system for constructing a distributed flood forecasting and scheduling model based on sub-basins. Background Technology
[0002] Flash floods are typically triggered by short-duration, heavy rainfall in hilly areas, characterized by short formation times, rapid peak rise, and strong localized destructiveness. For small watersheds in hilly areas with multiple reservoirs, the reservoir system is not only a crucial engineering facility for intercepting floods and reducing peak flows, but also a significant regulatory factor influencing the downstream flood superposition process. Existing flash flood control technologies typically utilize hydrological models, hydrodynamic models, or empirical statistical methods to simulate the watershed's rainfall-runoff process, and formulate reservoir scheduling plans based on the simulated inflow flood process or changes in flow at control sections. Regarding reservoir scheduling, existing methods often determine the discharge flow based on the reservoir's water level, inflow, outflow, and established scheduling rules for a single reservoir. In multi-reservoir scenarios, some methods allocate discharge tasks among multiple reservoirs through a unified scheduling platform or optimization algorithms.
[0003] However, the spatial distribution of rainfall, topographic slope, tributary confluence sequence, and hydraulic distance between reservoirs and downstream control sections within hilly watersheds all affect flood arrival time and peak overlap patterns. Existing reservoir group coordinated scheduling methods often only use flash flood simulation results as scheduling input, failing to fully consider the dynamic evolution process of flash floods from generation and confluence to river channel propagation. They also insufficiently consider the arrival time, superposition relationship, and peak-shaving effect of different reservoir discharge volumes at downstream control sections, easily leading to the problem of multiple reservoir discharge processes converging at the same control section, forming a new flood peak. Therefore, there is an urgent need for a reservoir group coordinated discharge peak-shaving scheduling method that can combine the dynamic evolution process of flash floods, coordinate and determine the discharge time period, discharge volume, and discharge sequence of each reservoir, and dynamically adjust based on real-time feedback. Summary of the Invention
[0004] The purpose of this invention is to overcome the problem that existing reservoir group discharge peak shaving scheduling methods fail to fully incorporate the dynamic evolution process of mountain torrents from generation and confluence to river propagation, resulting in the difficulty in accurately matching the discharge timing and discharge volume of each reservoir with the propagation and superposition patterns of flood peaks. This invention provides a method and system for constructing a distributed flood forecasting and scheduling model based on sub-basins.
[0005] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for constructing a distributed flood forecasting and scheduling model based on sub-basins, comprising the following steps: The study aims to obtain the dynamic evolution results of flash floods in small watersheds in hilly areas, and the basic scheduling information of each reservoir participating in the discharge and peak shaving operation. The dynamic evolution results of flash floods include the inflow flood process of each reservoir, the flow process of river sections within the watershed, and the predicted flood process of designated control sections. The basic scheduling information includes the spatial location of each reservoir, the range of inflow tributaries, the outflow confluence rivers, scheduling constraint parameters, and the hydraulic transmission relationship between each reservoir and the designated control sections. Based on the hydraulic conduction relationship between each reservoir and the designated control section, determine the propagation time of the water discharge from each reservoir to the designated control section; Based on the inflow flood process, scheduling constraints, propagation time, and safe flow at designated control sections of each reservoir, the candidate discharge time and candidate discharge volume for each reservoir are determined. Match the candidate discharge periods of each reservoir with the corresponding propagation time to determine the arrival time of the discharge volume of each reservoir at the designated control section; Calculate the predicted superimposed flow rate at the designated control section based on the arrival time of the discharge from each reservoir at the designated control section; With the constraint that the predicted superimposed flow does not exceed the safe flow of the designated control section and that the discharge volumes of different reservoirs are staggered and superimposed at the designated control section, a coordinated discharge scheme for the reservoir group is generated. The coordinated discharge scheme for the reservoir group includes the discharge period, discharge volume allocation value and discharge sequence of each reservoir. The coordinated discharge plan for the reservoir group is sent to the discharge gate execution terminals of each reservoir, so that each reservoir can discharge according to the corresponding discharge period, discharge volume allocation value and discharge sequence. During the discharge process, the actual outflow of each reservoir, the gate operation status, and the measured flow at the designated control section are obtained. Based on the actual outflow, gate operation status, and measured flow at the designated control section, the coordinated discharge scheme of the reservoir group is dynamically adjusted.
[0006] A further improvement of this invention is that the results of flash flood dynamic evolution are output using a flash flood dynamic evolution model. The flash flood dynamic evolution model is used to simulate small watersheds in hilly areas and outputs the outlet flow of each sub-watershed, the cross-sectional flow of the river, the time series data of flood evolution, and the inflow flood process of each reservoir.
[0007] A further improvement of this invention lies in the following method for determining the propagation time of the discharge volume from each reservoir to the designated control section based on the hydraulic conduction relationship between each reservoir and the designated control section: Based on the rivers into which the water flows from each reservoir, determine the location where the water discharged from each reservoir enters the main river or tributary river. Based on the hydraulic distance from each reservoir to the designated control section and the flood evolution time, the propagation time of the water discharge from each reservoir to the designated control section is determined.
[0008] A further improvement of this invention lies in the following method for determining the candidate discharge period and candidate discharge volume for each reservoir based on the inflow flood process, scheduling constraint parameters, propagation time, and safe flow at the designated control section: Based on the inflow flood process of each reservoir, determine the required discharge flow of each reservoir at different times; Based on the scheduling constraint parameters of each reservoir, determine the allowable discharge range of each reservoir at different times; Based on the predicted flood process and safe flow at the designated control section, determine the allowable discharge margin of the designated control section at different time periods; Based on the propagation time of each reservoir, the allowable discharge margin of the designated control section is back-matched to the candidate discharge time period of each reservoir to obtain the candidate discharge time period and candidate discharge volume of each reservoir.
[0009] A further improvement of this invention lies in the following method for calculating the predicted superimposed flow rate at a designated control section based on the arrival time of the discharge from each reservoir at that section: Obtain the predicted flood discharge at a specified control section during the target time period; Determine the discharge volume of each reservoir reaching the designated control section during the target time period; The predicted flood flow is superimposed with the discharge from each reservoir reaching the designated control section to obtain the predicted superimposed flow for the target period.
[0010] A further improvement of this invention lies in the following method for generating a coordinated discharge scheme for a group of reservoirs, constrained by the prediction that the superimposed flow does not exceed the safe flow at a designated control section and the staggered superposition of the discharge volumes of different reservoirs at the designated control section: Within the same target time period, the sum of the discharge from each reservoir reaching the designated control section and the predicted flood flow at the designated control section shall not exceed the safe flow of the designated control section. The discharge capacity of any reservoir is less than or equal to the maximum outflow capacity of that reservoir; The discharge capacity of any reservoir is greater than or equal to the minimum outflow capacity of the reservoir under discharge conditions; The variation in outflow from any reservoir during adjacent scheduling periods shall be less than or equal to the maximum allowable variation in outflow from that reservoir per unit time period. The water level of any reservoir shall not exceed the highest water level of that reservoir.
[0011] A further improvement of the present invention is that the reservoirs participating in the discharge and peak shaving scheduling include the main scheduling reservoir and the auxiliary scheduling reservoir; The main reservoir adjusts its discharge flow based on the predicted flood process and safe flow at the designated control section, and coordinates the discharge timing of the auxiliary reservoirs. The auxiliary reservoirs coordinate with the main reservoirs to carry out discharge and peak shaving operations based on the range of their inflow tributaries, the rivers into which they flow, and the propagation time to the designated control section.
[0012] A further improvement of the present invention is that, after the coordinated discharge scheme of the reservoir group is sent to the discharge gate execution terminal of each reservoir, the discharge gate execution terminal of each reservoir determines the gate opening degree according to the discharge flow allocation value of the corresponding reservoir and the current reservoir water level, and controls the discharge gate to operate according to the gate opening degree. The discharge gate actuators of each reservoir provide feedback on the gate's operating status and actual outflow at a preset frequency.
[0013] A further improvement of this invention lies in the following specific method for dynamically adjusting the coordinated discharge scheme of the reservoir group based on the actual outflow, gate operating status, and measured flow at the designated control section: When the deviation between the actual outflow of any reservoir and the corresponding discharge allocation value exceeds the first preset threshold, the discharge allocation value and discharge sequence of that reservoir and other reservoirs that have a hydraulic transmission relationship with that reservoir are recalculated. When the deviation between the measured flow and the predicted superimposed flow at the designated control section exceeds the second preset threshold, the current flash flood dynamic evolution results, reservoir operation data and measured flow at the designated control section are reacquired, and the coordinated discharge scheme of the reservoir group is regenerated. The regenerated coordinated discharge scheme for the reservoir group is sent to the discharge gate execution terminals of each reservoir.
[0014] Secondly, this invention provides a system for constructing a distributed flood forecasting and scheduling model based on sub-basins, comprising: The data acquisition module is used to acquire the dynamic evolution results of flash floods in small watersheds in hilly areas, and to acquire the basic scheduling information of each reservoir participating in the discharge and peak shaving scheduling. Among them, the dynamic evolution results of flash floods include the inflow flood process of each reservoir, the flow process of the river section within the watershed, and the predicted flood process of the designated control section. The basic scheduling information includes the spatial location of each reservoir, the range of inflow tributaries, the outflow confluence river, the scheduling constraint parameters, and the hydraulic transmission relationship between each reservoir and the designated control section. The propagation time acquisition module is used to determine the propagation time of the water discharge from each reservoir to the designated control section based on the hydraulic conduction relationship between each reservoir and the designated control section. The discharge acquisition module is used to determine the candidate discharge time period and candidate discharge volume for each reservoir based on the inflow flood process, scheduling constraint parameters, propagation time and safe flow of the designated control section. The arrival time acquisition module is used to match the candidate discharge time periods of each reservoir with the corresponding propagation time to determine the arrival time of the discharge volume of each reservoir at the specified control section. The superimposed flow acquisition module is used to calculate the predicted superimposed flow at a specified control section based on the arrival time of the discharge from each reservoir at the specified control section. The discharge scheme acquisition module is used to generate a coordinated discharge scheme for a group of reservoirs, with the constraint that the predicted superimposed flow does not exceed the safe flow of the specified control section and the discharge volume of different reservoirs is staggered and superimposed at the specified control section. The coordinated discharge scheme for a group of reservoirs includes the discharge period, discharge volume allocation value and discharge sequence of each reservoir. The discharge scheme execution module is used to send the coordinated discharge scheme of the reservoir group to the discharge gate execution terminal of each reservoir, so that each reservoir can perform discharge according to the corresponding discharge period, discharge volume allocation value and discharge sequence. The dynamic adjustment module is used to acquire the actual outflow of each reservoir, the gate operating status, and the measured flow at the designated control section during the discharge process. Based on the actual outflow, gate operating status, and measured flow at the designated control section, the module dynamically adjusts the coordinated discharge scheme of the reservoir group.
[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention, by acquiring the dynamic evolution results of flash floods and combining them with the spatial location of each reservoir, the range of inflow tributaries, the outflow channels, and the hydraulic transmission relationship between them and designated control sections, determines the propagation time of the discharged water from each reservoir to the designated control section. This allows reservoir discharge scheduling to no longer rely solely on the water level or inflow of a single reservoir, but rather to match it with the propagation process of flash floods within the watershed, thereby improving the targeting and accuracy of discharge scheduling. This invention matches the candidate discharge periods of each reservoir with the propagation time, calculates the predicted superimposed flow at the designated control section, and generates a coordinated discharge scheme with constraints such as the predicted superimposed flow not exceeding the safe flow and the staggered superposition of discharge volumes from different reservoirs at the designated control section. This avoids the concentrated superposition of discharge volumes from multiple reservoirs with natural flood peaks or other reservoir discharge processes during the same period, reducing the peak flow at the designated control section and improving the overall peak reduction effect of the watershed. This invention sends the coordinated discharge plan to the discharge gate execution end and dynamically adjusts it by collecting the actual outflow, gate operation status and measured flow at the designated control section during the discharge process. This can promptly correct prediction errors and execution deviations, and improve the real-time performance, reliability and flood control safety of reservoir group coordinated scheduling. Attached Figure Description
[0016] Figure 1 This is a flowchart of the present invention; Figure 2 This is a system diagram of the present invention. Detailed Implementation
[0017] To further understand the content of this invention, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are merely illustrative and not limiting of the invention.
[0018] Example 1: See Figure 1 This embodiment uses a small watershed in a hilly area as the application scenario. This watershed is mainly characterized by hilly terrain with significant topographic relief. Rainfall is concentrated and often consists of short-duration, heavy rainfall, making it prone to flash floods. Several reservoirs of varying sizes have been built within this watershed, forming a reservoir group covering the entire area. Hydraulic transmission paths are formed between the reservoirs via main channels and tributaries. Water flowing out of the reservoirs propagates downstream through corresponding tributaries or main channels. Designated control sections are located downstream of the watershed. These control sections are used to monitor flood flow at the watershed outlet or in key protected areas and are the core control points for coordinated flood discharge and peak shaving operations of the reservoir group.
[0019] Before the flood season arrives, the mountain torrent dynamic evolution model and reservoir group collaborative scheduling framework are debugged, and the data acquisition unit, scheduling decision unit and execution unit are checked to ensure that they are operating normally. The initial reservoir water level of each reservoir is adjusted to near the normal storage level. At the same time, the operation status of each reservoir's spillway gate, gate actuator, communication link and flow monitoring equipment is checked to ensure that the spillway gate can complete the opening and closing actions according to the scheduling instructions.
[0020] In this embodiment, the dynamic evolution results of flash floods in a small watershed in a hilly area are obtained. These results include the inflow flood processes of each reservoir, the cross-sectional flow processes of rivers within the watershed, and the predicted flood processes at designated control sections. Specifically, the inflow flood processes of each reservoir reflect the inflow situation at different times; the cross-sectional flow processes of rivers reflect the flood propagation within tributaries and main channels; and the predicted flood processes at designated control sections reflect the flood change trends at key downstream control locations during future scheduling periods.
[0021] Obtain basic scheduling information for each reservoir participating in the discharge and peak-shaving operation. This information includes the spatial location of each reservoir, the range of inflow tributaries, the outflow confluence with rivers, scheduling constraint parameters, and the hydraulic transmission relationship between each reservoir and the designated control section. The spatial location of each reservoir determines its relative distribution within the watershed; the range of inflow tributaries determines the incoming water area controlled by the reservoir; the outflow confluence with rivers determines the location where the discharged water enters the river system; scheduling constraint parameters define the safety boundaries of the reservoir discharge process; and the hydraulic transmission relationship represents the propagation path and time of the outflow from the reservoir to the designated control section.
[0022] The reservoir group collaborative scheduling framework comprises a data acquisition unit, a scheduling decision-making unit, and an execution unit. The data acquisition unit is deployed at each reservoir, watershed hydrological station, rainfall station, evaporation monitoring station, and river control section to collect real-time data such as reservoir water level, inflow, outflow, reservoir storage, rainfall, evaporation, soil moisture content, real-time river flow, and river water level. The collected data, after filtering, noise reduction, missing value completion, and format standardization, is transmitted to the scheduling decision-making unit and the flash flood dynamic evolution model.
[0023] The scheduling decision-making unit is deployed at the basin scheduling center and includes a constraint module, an optimization calculation module, and a data storage module. The constraint module stores the minimum and maximum reservoir water levels, normal storage levels, dead storage levels, maximum outflow rates, minimum outflow rates, and the maximum allowable outflow rate variation within a unit time period for each reservoir. It also stores the safe flow rate, warning flow rate, and guaranteed water level for designated control sections. The optimization calculation module uses the results of flash flood dynamics evolution and combines them with reservoir scheduling constraint parameters and designated control section control parameters to calculate coordinated discharge schemes for the reservoir group. The data storage module stores collected real-time data, model parameters, historical scheduling processes, and calculated discharge schemes, and supports real-time querying and historical tracing.
[0024] After obtaining the above information, the propagation time of the water discharge from each reservoir to the designated control section is determined based on the hydraulic conduction relationship between each reservoir and the designated control section. The propagation time can be determined based on the hydraulic distance from each reservoir to the designated control section, the flood evolution time, and historical flood propagation data.
[0025] Based on the inflow flood process, scheduling constraints, propagation time, and safe flow at designated control sections for each reservoir, candidate discharge periods and candidate discharge volumes are determined. Specifically, if a reservoir experiences a large inflow and discharge demand within a given scheduling period, the feasible discharge volume range for that reservoir during that period is determined based on its minimum reservoir level, maximum reservoir level, normal storage level, dead storage level, maximum outflow volume, minimum outflow volume, and the allowable maximum outflow volume variation within a unit time period. Simultaneously, based on the predicted flood process and safe flow at the designated control sections, it is determined whether the discharge from the reservoir during that period will cause the flow at the designated control sections to exceed limits in subsequent periods.
[0026] After determining the candidate discharge periods and candidate discharge volumes, the candidate discharge periods for each reservoir are matched with their corresponding propagation durations to determine the arrival time of the discharge volume at the designated control section. This allows us to determine in which subsequent period a discharge volume from a reservoir will affect the flow rate at the designated control section after the reservoir discharges at a certain time.
[0027] Based on the arrival time of the discharge from each reservoir at the designated control section, the predicted superimposed discharge at the designated control section is calculated. The predicted superimposed discharge includes the predicted flood discharge at the designated control section and the discharge contribution formed after the discharge from each reservoir reaches the designated control section.
[0028] When generating a coordinated discharge scheme for a group of reservoirs, the constraints are set as follows: the predicted superimposed flow does not exceed the safe flow at the designated control section, and the discharge volumes from different reservoirs are staggered at the designated control section. This determines the discharge period, discharge volume allocation, and discharge sequence for each reservoir. This method ensures that the discharge volumes from different upstream reservoirs do not overlap at the designated control section during the same time period, thereby preventing the formation of new flood peaks.
[0029] After generating the coordinated discharge plan for the reservoir group, the plan is sent to the discharge gate execution terminals of each reservoir. Execution units are deployed at the control terminals of each reservoir's discharge gates, and each discharge gate is equipped with a corresponding execution mechanism. The execution units receive the discharge plan instructions from the scheduling decision unit, control the gate opening, and ensure that each reservoir discharges according to the corresponding discharge period, discharge volume allocation, and discharge sequence. Simultaneously, they collect real-time data on gate operating status and actual outflow, and feed this data back to the scheduling decision unit.
[0030] During the discharge process, the actual outflow from each reservoir, the gate operating status, and the measured flow at the designated control section are continuously acquired. When there is a deviation between the actual outflow, gate operating status, or measured flow at the designated control section and the original reservoir group coordinated discharge plan, the reservoir group coordinated discharge plan is dynamically adjusted, the discharge period, discharge flow allocation value, and discharge sequence of each reservoir are re-determined, and the adjusted plan is sent to the discharge gate execution terminal of each reservoir.
[0031] This embodiment enables the reservoir group's discharge process to match the propagation pattern of flash floods within the watershed, allowing the discharge volumes of different reservoirs to arrive at designated control sections at staggered times, reducing the peak flow at the control sections, and improving the safety and real-time adaptability of coordinated peak shaving scheduling of the reservoir group.
[0032] Example 2: In this embodiment, the flash flood dynamic evolution results are output by the flash flood dynamic evolution model. The flash flood dynamic evolution model uses a small watershed in a hilly area as the simulation scope to simulate the process of flood propagation from slope runoff, tributary confluence, to the main river channel under short-duration heavy rainfall conditions.
[0033] The input data for the flash flood dynamics evolution model includes watershed underlying surface parameters and basic meteorological and hydrological parameters. Watershed underlying surface parameters are extracted using geographic information processing (GIS) techniques and include parameters such as watershed topographic slope, soil type, vegetation cover, river network density, river length, river gradient, and reservoir catchment area. Basic meteorological and hydrological parameters include data on historical rainfall, evaporation, runoff depth, initial soil moisture content, initial reservoir water level, and initial river flow. For areas with sparse monitoring stations, spatial distribution data on rainfall, vegetation cover, land use, and surface moisture status can be supplemented by combining satellite remote sensing data.
[0034] The flash flood dynamic evolution model comprises a series of modules: evapotranspiration calculation module, water storage saturation runoff calculation module, water source distribution calculation module, and runoff confluence calculation module. Parameters are exchanged between these modules via a data interface. The evapotranspiration calculation module calculates the evapotranspiration of water surface, vegetation, and soil within the watershed; the water storage saturation runoff calculation module calculates the runoff after the soil moisture content reaches saturation; the water source distribution calculation module divides the runoff into surface runoff, interflow, and groundwater runoff; and the runoff confluence calculation module integrates the runoff from each sub-watershed into the main channel and simulates flood evolution within the channel.
[0035] The runoff calculation module employs a lag algorithm for sub-basin runoff calculation and a segmented Muskingen method for channel runoff calculation. Sub-basins can be divided based on the distribution of reservoir catchment areas, hydrological station control areas, and river control sections within the basin. Each sub-basin corresponds to a unique runoff outlet, which establishes a runoff connection with the main channel or adjacent sub-basins.
[0036] After the flash flood dynamics evolution model is run, it outputs the outlet flow of each sub-basin, the cross-sectional flow of the river, the flood evolution time series data, and the inflow flood process of each reservoir. The flood evolution time series data includes the propagation time of the flood between tributaries, the main channel, and designated control sections, the time of peak flood occurrence, and the flow variation process. The inflow flood process of each reservoir can be further included, including the peak inflow flood value, the time of peak value occurrence, and the flood duration.
[0037] Before applying the model, its parameters can be calibrated and validated using historical flash flood events over many years. During calibration, historical rainfall, evaporation, initial reservoir water levels, and historical river discharges are input into the model. The deviations between the model's output control section discharge process and the measured discharge process are compared, and runoff generation, confluence, and river propagation parameters are adjusted. During validation, historical flood events not involved in calibration are used to test the model's output results, ensuring the model can better reflect the flood dynamics evolution of small watersheds in hilly areas.
[0038] In actual scheduling, the scheduling decision-making unit uses the data output by the flash flood dynamic evolution model, rather than primarily aiming to rebuild the model. The scheduling decision-making unit mainly uses the inflow flood process of each reservoir, the flow process of each river section, and the predicted flood process of designated control sections as the calculation basis for the reservoir group's discharge peak shaving scheduling.
[0039] In this embodiment, the results of flash flood dynamic evolution are no longer used only as general flood forecast results, but are further used as input basis for the coordinated discharge and peak shaving scheduling of reservoir groups, so that the scheduling scheme can fit the flash flood propagation process.
[0040] Example 3: In this embodiment, the location where the discharge water from each reservoir enters the main channel or tributary channel is first determined based on the outflow channels of each reservoir. For reservoirs located upstream of tributaries, their discharge water first enters the corresponding tributary channel, then flows into the main channel, and finally reaches the designated control section. For reservoirs located near the main channel, their discharge water can directly enter the main channel and propagate along the main channel to the designated control section.
[0041] After determining the location where the discharged water enters the river channel, the propagation time of the discharged water from each reservoir to the designated control section is determined based on the hydraulic distance from each reservoir to the designated control section, the flood evolution time, and historical flood propagation data. The propagation time reflects the time required for the discharged water to travel from the reservoir outlet through the corresponding tributary or main channel to the designated control section.
[0042] In another implementation, the propagation duration can also be determined based on the flood evolution time series data in the flash flood dynamics evolution results. For example, if the flash flood dynamics evolution model has output the time difference between the arrival of the flood peak at the confluence of a certain tributary at a designated control section, then this time difference can be used as the propagation duration of the corresponding reservoir discharge.
[0043] After determining the propagation duration, for any reservoir, after determining its initial discharge period, the arrival time of the discharged water volume at the designated control section is determined based on the corresponding propagation duration. In this way, a correspondence can be established between the discharge operation at the reservoir end and the flow response at the control section end, enabling the dispatching decision unit to determine in which future period a discharge from a reservoir will affect the flow at the designated control section.
[0044] Example 4: In this embodiment, the process of inflow flooding into each reservoir is first used to determine whether each reservoir has a discharge requirement at different times. For example, for a certain reservoir, if its inflow rate continues to increase, the current reservoir water level is close to the maximum reservoir water level, or the reservoir water level is expected to continue to rise in subsequent periods, it is determined that the reservoir has a discharge requirement.
[0045] Subsequently, based on the scheduling constraint parameters of each reservoir, the permissible discharge range for each reservoir at different time periods is determined. These scheduling constraint parameters include the minimum reservoir water level, maximum reservoir water level, normal storage level, dead storage level, maximum outflow, minimum outflow, and the maximum permissible variation in outflow within a unit of time period. The scheduling decision unit uses these parameters to determine whether a reservoir is permitted to discharge water during a given time period, the permissible discharge range, and whether corresponding changes in outflow are permissible within adjacent scheduling periods.
[0046] Then, based on the predicted flood process and safe flow at the designated control section, the range of discharge impact that the designated control section can withstand at different time periods is determined. If the predicted flood flow at the designated control section is close to the safe flow at a certain time period, the discharge from each reservoir should be restricted from reaching the designated control section during that time period; if the designated control section still has a large flood control margin at a certain time period, then a portion of the reservoir discharge can be scheduled to reach the designated control section during that time period.
[0047] After determining the carrying capacity of the designated control section at different time periods, the safety constraints of the designated control section are back-matched to the candidate discharge periods of each reservoir based on the propagation time of the discharge from each reservoir to the designated control section. This yields the candidate discharge periods and candidate discharge volumes for each reservoir. Only if the discharge from a reservoir during a candidate period does not cause the flow rate at the designated control section to exceed the safe flow rate, are the candidate discharge periods and candidate discharge volumes retained.
[0048] This embodiment does not simply release water immediately based on the reservoir's own inflow situation, but instead reverses the flood control safety requirements of the designated control section for future periods to the release time of each reservoir, thereby reducing the risk of the release process overlapping with the downstream flood peak process.
[0049] Example 5: In this embodiment, when calculating the predicted superimposed flow at a specified control section, the predicted flood flow at the specified control section during the target time period is first obtained. This predicted flood flow can be obtained from the predicted flood process at the specified control section in the flash flood dynamic evolution results, and is used to represent the flood flow at the specified control section without considering the impact of additional discharges from each reservoir under the current scheduling scheme.
[0050] Then, the discharge volume of each reservoir reaching the designated control section during the target time period is determined. For any reservoir, after it discharges during a candidate discharge period, the target time period for its discharge volume to reach the designated control section can be determined based on the corresponding propagation time of that reservoir. The scheduling decision unit then identifies the discharge volumes of multiple reservoirs that may reach the designated control section within the same target time period.
[0051] After determining the discharge volume of each reservoir reaching the designated control section, the lag, attenuation, and diffusion effects during river propagation are comprehensively considered to obtain the flow contribution of each reservoir discharge to the designated control section. Subsequently, the predicted flood flow at the designated control section is superimposed with the discharge contribution of each reservoir to obtain the predicted superimposed flow at the designated control section.
[0052] If the predicted superimposed flow exceeds the safe flow of the designated control section, the scheduling decision-making unit will adjust the discharge period or discharge volume of at least one reservoir, so that some of the reservoir's discharge volume avoids reaching the designated control section during the target period. If the predicted superimposed flow does not exceed the safe flow of the designated control section, the corresponding discharge period and discharge volume can be considered as part of a feasible solution.
[0053] This embodiment allows for the prediction of the cumulative effect of water discharge from each reservoir at the downstream control section before reservoir discharge, providing a basis for generating staggered discharge schemes.
[0054] Example 6: In this embodiment, when generating a coordinated discharge scheme for a group of reservoirs, the safe flow rate at a designated control section is used as the core constraint, requiring that the discharge volume of each reservoir reaching the designated control section, when combined with the predicted flood flow rate at the designated control section, does not exceed the safe flow rate.
[0055] At the same time, constraints are imposed on the scheduling safety of each reservoir. The discharge of any reservoir shall not exceed its maximum outflow or be lower than its minimum outflow under discharge conditions; the change in outflow of any reservoir within adjacent scheduling periods shall not exceed the maximum allowable change in outflow within a unit time period; and the water level of any reservoir shall not exceed its highest water level.
[0056] The aforementioned scheduling constraint parameters can be uniformly stored and retrieved by the constraint condition module in the scheduling decision unit. Constraint parameters can be set separately for reservoirs of different sizes. For example, the main scheduling reservoir can have a larger allowable discharge range and stronger regulation capacity; the constraint parameters for auxiliary scheduling reservoirs and small reservoirs are set separately according to their reservoir capacity, discharge capacity, and flood control tasks. The standard parameters for basin flood control include the safe flow, warning flow, and guaranteed water level at designated control sections.
[0057] When a candidate discharge scheme simultaneously meets the above constraints, it is further determined whether the discharge volumes of different reservoirs form a concentrated superposition at the designated control section. If the discharge volumes of multiple reservoirs will arrive at the designated control section in the same target time period, and the predicted superposition flow is close to or exceeds the safe flow, the discharge time period of at least one of the reservoirs will be adjusted so that its discharge volume arrives at the designated control section earlier or later.
[0058] This embodiment can simultaneously address the flood control safety of the control section, the operational safety of the reservoir itself, and the demand for staggered peak discharge from multiple reservoirs.
[0059] Example 7: In this embodiment, the reservoirs participating in the discharge and peak shaving scheduling include main scheduling reservoirs and auxiliary scheduling reservoirs. The auxiliary scheduling reservoirs may further include medium-sized reservoirs and small reservoirs. The main scheduling reservoir can be a reservoir with a large storage capacity, strong regulation capacity, or significant impact on the designated control section. The auxiliary scheduling reservoirs can be reservoirs located in the upstream, midstream, or local control areas of tributaries.
[0060] In one implementation, the main reservoir serves as a first-level dispatching node, medium-sized reservoirs as second-level dispatching nodes, and small reservoirs as third-level dispatching nodes. The first-level dispatching nodes determine the overall discharge strategy for the basin based on the predicted flood process and safe flow at designated control sections; the second-level dispatching nodes execute zoned discharge and peak shaving based on the inflow flood process and propagation duration of their respective tributaries and their own dispatching constraints; and the third-level dispatching nodes execute auxiliary peak shaving or discharge restriction based on the inflow from local tributaries and instructions from higher-level dispatching authorities.
[0061] The main reservoir adjusts its discharge flow based on the predicted flood process and safe flow at the designated control section, and coordinates the discharge sequence of the auxiliary reservoirs. Specifically, when the predicted flood process at the designated control section indicates that the downstream flood peak is about to form, the main reservoir first determines whether it needs to pre-discharge or reduce its discharge; at the same time, it determines the discharge sequence of each auxiliary reservoir based on the propagation time from each auxiliary reservoir to the designated control section.
[0062] Auxiliary reservoirs coordinate with the main reservoir to implement discharge and peak-shaving operations based on the range of their inflow tributaries, the channels into which they flow out, and the propagation time to the designated control section. For example, auxiliary reservoirs located on tributaries near the control section have a shorter propagation time to the designated control section, and should avoid large-volume discharges before and after the peak flood period at the control section. Auxiliary reservoirs located on tributaries far from the control section have a longer propagation time to the designated control section, and can discharge water in advance based on reverse matching results to avoid the period of discharge from the main reservoir.
[0063] The main dispatching reservoir, auxiliary dispatching reservoirs, and small reservoirs establish information exchange processes and command transmission paths through communication links. The dispatching decision-making unit sends discharge plans to the main dispatching reservoir and each auxiliary dispatching reservoir based on the dynamic evolution of flash floods. The main dispatching reservoir provides feedback on its own reservoir water level, outflow, and gate status, while the auxiliary dispatching reservoirs and small reservoirs provide their respective real-time operational data. Based on the feedback results, the dispatching decision-making unit determines whether there is a risk of overlapping discharge volumes among the reservoirs at each level and makes unified adjustments to the discharge sequence.
[0064] This embodiment enables the main reservoir, auxiliary reservoirs, and small reservoirs to form a hierarchical collaborative relationship, avoiding downstream flow overlap and conflict caused by independent discharge from each reservoir.
[0065] Example 8: In this embodiment, the coordinated discharge scheme for the reservoir group includes the discharge period, discharge volume allocation, and discharge sequence for each reservoir. After the scheduling decision unit sends the coordinated discharge scheme to the discharge gate execution terminals of each reservoir, the discharge gate execution terminals of each reservoir determine the gate opening degree based on the discharge volume allocation value of the corresponding reservoir and the current reservoir water level.
[0066] Specifically, for any reservoir, the spillway gate actuator receives the discharge allocation value sent by the scheduling decision unit and reads the current reservoir water level. Based on the pre-stored correspondence between gate opening and discharge capacity, the spillway gate actuator determines the gate opening corresponding to the discharge allocation value, and then controls the spillway gate to operate at that gate opening.
[0067] During operation, the discharge gate actuators of each reservoir provide feedback on the gate's operating status and actual outflow at a preset frequency. The gate's operating status includes the gate opening degree, gate opening and closing status, whether the gate is properly executed, and gate fault information. The actual outflow can be obtained through flow sensors, the correlation between water level and flow rate, or the correlation between gate opening degree and flow rate.
[0068] The data acquisition unit uses water level sensors, flow sensors, rainfall sensors, and evaporation sensors to collect data at fixed frequencies. Water level sensors collect real-time reservoir water levels and river control section water levels; flow sensors collect inflow, outflow, and river section flow; rainfall sensors collect basin rainfall; and evaporation sensors collect basin evaporation. After preprocessing, the collected data is transmitted to the scheduling and decision-making unit, providing a real-time data foundation for the execution and adjustment of the spillway plan.
[0069] The data storage module of the scheduling decision-making unit records each scheduling process. The records include the scheduling start time, initial reservoir water levels, inflow flood process, discharge plan, gate operation status, actual outflow, measured flow at designated control sections, and subsequent adjustment instructions. This data is used for scheduling process tracing, model parameter correction, and as a reference for scheduling similar flood events in the future.
[0070] This embodiment can transform the coordinated discharge scheme of reservoir groups into specific gate execution actions, and ensure the feasibility of the scheduling scheme through real-time feedback.
[0071] Example 9: In this embodiment, during the discharge process, the actual outflow of each reservoir, the gate operation status, and the measured flow at the designated control section are continuously acquired, and the above real-time data are compared with the original reservoir group coordinated discharge scheme.
[0072] When the deviation between the actual outflow from any reservoir and the corresponding discharge allocation value exceeds a first preset threshold, the discharge allocation values and discharge sequence of that reservoir and other reservoirs with hydraulic transmission relationships with it are recalculated. At this time, the dispatching decision unit can revise the subsequent discharge plan by combining the current reservoir water level, gate operating status, actual outflow, and discharge arrangements of its associated reservoirs.
[0073] When the deviation between the measured flow rate and the predicted superimposed flow rate at the designated control section exceeds the second preset threshold, the current moment's flash flood dynamic evolution results, reservoir operation data, and measured flow rate at the designated control section are reacquired, and a new coordinated discharge plan for the reservoir group is generated. At this time, the dispatching decision unit uses the current moment as the new dispatching starting point to reassess the timing of the discharge volume from each reservoir reaching the designated control section within the future dispatching period and the possible superimposed effects.
[0074] When regenerating the coordinated discharge plan for the reservoir group, the scheduling decision unit redetermines the discharge period, discharge volume allocation value and discharge sequence of each reservoir according to the principles that the predicted superimposed flow does not exceed the safe flow of the designated control section, the staggered superposition of discharge volumes from different reservoirs, and each reservoir meets its own scheduling constraints.
[0075] After the regenerated coordinated discharge plan for the reservoir group is sent to the discharge gate execution terminals of each reservoir, each reservoir adjusts the gate opening according to the new discharge plan and continues to carry out discharge peak shaving scheduling.
[0076] When the dynamic evolution results of flash floods indicate that the flood runoff process within the basin tends to stabilize, the measured flow at the designated control section is lower than the safe flow, the inflow into each reservoir continues to decrease, and the water level in each reservoir remains stable below the normal storage level, the dispatch decision unit generates a dispatch termination command. The actuators of each reservoir's spillway gate then control the spillway gates to close or restore them to normal operation based on the dispatch termination command.
[0077] After the scheduling is completed, the data storage module saves all data from the entire scheduling process, including collected hydrological and meteorological data, reservoir operation data, flash flood dynamics evolution results, discharge plans, adjustment plans, gate execution status, measured flow changes at designated control sections, and the final peak shaving effect. The saved data can be used for subsequent scheduling reviews, model parameter correction, scheduling rule optimization, and historical case retrieval.
[0078] This embodiment enables the real-time correction of the reservoir group coordinated discharge scheme when the actual flood process, gate operation status, or reservoir outflow is inconsistent with the original prediction results, and completes the scheduling closed loop after the flood process ends, thereby improving the reliability, safety, and traceability of the scheduling process.
[0079] Example 10: See Figure 2 A system for constructing a distributed flood forecasting and scheduling model based on sub-basins includes: The data acquisition module is used to acquire the dynamic evolution results of flash floods in small watersheds in hilly areas and to acquire the basic scheduling information of each reservoir participating in the discharge and peak shaving scheduling. The dynamic evolution results of flash floods include the inflow flood process of each reservoir, the flow process of the river section in the watershed, and the predicted flood process of the designated control section. The basic scheduling information includes the spatial location of each reservoir, the range of inflow tributaries, the outflow confluence river, the scheduling constraint parameters, and the hydraulic transmission relationship between each reservoir and the designated control section.
[0080] The propagation time acquisition module is used to determine the propagation time of the water discharge from each reservoir to the designated control section based on the hydraulic conduction relationship between each reservoir and the designated control section.
[0081] The discharge acquisition module is used to determine the candidate discharge time period and candidate discharge volume for each reservoir based on the inflow flood process, scheduling constraint parameters, propagation time, and safe flow at the designated control section.
[0082] The arrival time acquisition module is used to match the candidate discharge time periods of each reservoir with the corresponding propagation duration to determine the arrival time of the discharge volume of each reservoir at the specified control section.
[0083] The superimposed flow acquisition module is used to calculate the predicted superimposed flow at a specified control section based on the arrival time of the discharge from each reservoir at the specified control section.
[0084] The discharge scheme acquisition module is used to generate a coordinated discharge scheme for a group of reservoirs, with the constraint that the predicted superimposed flow does not exceed the safe flow of the specified control section and that the discharge volumes of different reservoirs are staggered and superimposed at the specified control section. The coordinated discharge scheme for a group of reservoirs includes the discharge period, discharge volume allocation value and discharge sequence of each reservoir.
[0085] The discharge scheme execution module is used to send the coordinated discharge scheme of the reservoir group to the discharge gate execution terminal of each reservoir, so that each reservoir can perform discharge according to the corresponding discharge period, discharge volume allocation value and discharge sequence.
[0086] The dynamic adjustment module is used to acquire the actual outflow of each reservoir, the gate operating status, and the measured flow at the designated control section during the discharge process. Based on the actual outflow, gate operating status, and measured flow at the designated control section, the module dynamically adjusts the coordinated discharge scheme of the reservoir group.
[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for constructing a distributed flood forecasting and scheduling model based on sub-basins, characterized in that, Includes the following steps: The study aims to obtain the dynamic evolution results of flash floods in small watersheds in hilly areas, and the basic scheduling information of each reservoir participating in the discharge and peak shaving operation. The dynamic evolution results of flash floods include the inflow flood process of each reservoir, the flow process of river sections within the watershed, and the predicted flood process of designated control sections. The basic scheduling information includes the spatial location of each reservoir, the range of inflow tributaries, the outflow confluence rivers, scheduling constraint parameters, and the hydraulic transmission relationship between each reservoir and the designated control sections. Based on the hydraulic conduction relationship between each reservoir and the designated control section, determine the propagation time of the water discharge from each reservoir to the designated control section; Based on the inflow flood process, scheduling constraints, propagation time, and safe flow at designated control sections of each reservoir, the candidate discharge time and candidate discharge volume for each reservoir are determined. Match the candidate discharge periods of each reservoir with the corresponding propagation time to determine the arrival time of the discharge volume of each reservoir at the designated control section; Calculate the predicted superimposed flow rate at the designated control section based on the arrival time of the discharge from each reservoir at the designated control section; With the constraint that the predicted superimposed flow does not exceed the safe flow of the designated control section and that the discharge volumes of different reservoirs are staggered and superimposed at the designated control section, a coordinated discharge scheme for the reservoir group is generated. The coordinated discharge scheme for the reservoir group includes the discharge period, discharge volume allocation value and discharge sequence of each reservoir. The coordinated discharge plan for the reservoir group is sent to the discharge gate execution terminals of each reservoir, so that each reservoir can discharge according to the corresponding discharge period, discharge volume allocation value and discharge sequence. During the discharge process, the actual outflow of each reservoir, the gate operation status, and the measured flow at the designated control section are obtained. Based on the actual outflow, gate operation status, and measured flow at the designated control section, the coordinated discharge scheme of the reservoir group is dynamically adjusted.
2. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, The results of flash flood dynamic evolution are output using the flash flood dynamic evolution model, which is used to simulate small watersheds in hilly areas and outputs the outlet flow, cross-sectional flow, flood evolution time series data, and inflow flood process of each reservoir within the watershed.
3. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, Based on the hydraulic conduction relationship between each reservoir and the designated control section, the specific method for determining the propagation time of the discharge from each reservoir to the designated control section is as follows: Based on the rivers into which the water flows from each reservoir, determine the location where the water discharged from each reservoir enters the main river or tributary river. Based on the hydraulic distance from each reservoir to the designated control section and the flood evolution time, the propagation time of the water discharge from each reservoir to the designated control section is determined.
4. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, Based on the inflow flood process, scheduling constraints, propagation time, and safe flow at designated control sections of each reservoir, the specific methods for determining the candidate discharge periods and candidate discharge volumes for each reservoir are as follows: Based on the inflow flood process of each reservoir, determine the required discharge flow of each reservoir at different times; Based on the scheduling constraint parameters of each reservoir, determine the allowable discharge range of each reservoir at different times; Based on the predicted flood process and safe flow at the designated control section, determine the allowable discharge margin of the designated control section at different time periods; Based on the propagation time of each reservoir, the allowable discharge margin of the designated control section is back-matched to the candidate discharge time period of each reservoir to obtain the candidate discharge time period and candidate discharge volume of each reservoir.
5. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, The specific method for calculating the predicted superimposed flow at a designated control section based on the arrival time of the discharge from each reservoir at the designated control section is as follows: Obtain the predicted flood discharge at a specified control section during the target time period; Determine the discharge volume of each reservoir reaching the designated control section during the target time period; The predicted flood flow is superimposed with the discharge from each reservoir reaching the designated control section to obtain the predicted superimposed flow for the target period.
6. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, The specific method for generating a coordinated discharge scheme for a group of reservoirs, constrained by the premise that the predicted superimposed flow does not exceed the safe flow at the designated control section and that the discharge volumes of different reservoirs are staggered and superimposed at the designated control section, is as follows: Within the same target time period, the sum of the discharge from each reservoir reaching the designated control section and the predicted flood flow at the designated control section shall not exceed the safe flow of the designated control section. The discharge capacity of any reservoir is less than or equal to the maximum outflow capacity of that reservoir; The discharge capacity of any reservoir is greater than or equal to the minimum outflow capacity of the reservoir under discharge conditions; The variation in outflow from any reservoir during adjacent scheduling periods shall be less than or equal to the maximum allowable variation in outflow from that reservoir per unit time period. The water level of any reservoir shall not exceed the highest water level of that reservoir.
7. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, The reservoirs participating in the discharge and peak shaving operation include the main dispatch reservoir and the auxiliary dispatch reservoir; The main reservoir adjusts its discharge flow based on the predicted flood process and safe flow at the designated control section, and coordinates the discharge timing of the auxiliary reservoirs. The auxiliary reservoirs coordinate with the main reservoirs to carry out discharge and peak shaving operations based on the range of their inflow tributaries, the rivers into which they flow, and the propagation time to the designated control section.
8. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, After the coordinated discharge plan of the reservoir group is sent to the discharge gate execution terminal of each reservoir, the discharge gate execution terminal of each reservoir determines the gate opening degree according to the discharge flow allocation value of the corresponding reservoir and the current reservoir water level, and controls the discharge gate to operate according to the gate opening degree; The discharge gate actuators of each reservoir provide feedback on the gate's operating status and actual outflow at a preset frequency.
9. The method for constructing a distributed flood forecasting and scheduling model based on sub-basins according to claim 1, characterized in that, The specific method for dynamically adjusting the coordinated discharge scheme of the reservoir group based on the actual outflow, gate operating status, and measured flow at the designated control section is as follows: When the deviation between the actual outflow of any reservoir and the corresponding discharge allocation value exceeds the first preset threshold, the discharge allocation value and discharge sequence of that reservoir and other reservoirs that have a hydraulic transmission relationship with that reservoir are recalculated. When the deviation between the measured flow and the predicted superimposed flow at the designated control section exceeds the second preset threshold, the current flash flood dynamic evolution results, reservoir operation data and measured flow at the designated control section are reacquired, and the coordinated discharge scheme of the reservoir group is regenerated. The regenerated coordinated discharge scheme for the reservoir group is sent to the discharge gate execution terminals of each reservoir.
10. A system for constructing a distributed flood forecasting and scheduling model based on sub-basins, characterized in that, include: The data acquisition module is used to acquire the dynamic evolution results of flash floods in small watersheds in hilly areas, and to acquire the basic scheduling information of each reservoir participating in the discharge and peak shaving scheduling. Among them, the dynamic evolution results of flash floods include the inflow flood process of each reservoir, the flow process of the river section within the watershed, and the predicted flood process of the designated control section. The basic scheduling information includes the spatial location of each reservoir, the range of inflow tributaries, the outflow confluence river, the scheduling constraint parameters, and the hydraulic transmission relationship between each reservoir and the designated control section. The propagation time acquisition module is used to determine the propagation time of the water discharge from each reservoir to the designated control section based on the hydraulic conduction relationship between each reservoir and the designated control section. The discharge acquisition module is used to determine the candidate discharge time period and candidate discharge volume for each reservoir based on the inflow flood process, scheduling constraint parameters, propagation time and safe flow of the designated control section. The arrival time acquisition module is used to match the candidate discharge time periods of each reservoir with the corresponding propagation time to determine the arrival time of the discharge volume of each reservoir at the specified control section. The superimposed flow acquisition module is used to calculate the predicted superimposed flow at a specified control section based on the arrival time of the discharge from each reservoir at the specified control section. The discharge scheme acquisition module is used to generate a coordinated discharge scheme for a group of reservoirs, with the constraint that the predicted superimposed flow does not exceed the safe flow of the specified control section and the discharge volume of different reservoirs is staggered and superimposed at the specified control section. The coordinated discharge scheme for a group of reservoirs includes the discharge period, discharge volume allocation value and discharge sequence of each reservoir. The discharge scheme execution module is used to send the coordinated discharge scheme of the reservoir group to the discharge gate execution terminal of each reservoir, so that each reservoir can perform discharge according to the corresponding discharge period, discharge volume allocation value and discharge sequence. The dynamic adjustment module is used to acquire the actual outflow of each reservoir, the gate operating status, and the measured flow at the designated control section during the discharge process. Based on the actual outflow, gate operating status, and measured flow at the designated control section, the module dynamically adjusts the coordinated discharge scheme of the reservoir group.