Nested feedback regulation and control method for long-distance cascade gate pump water transfer project

Abstract

The invention provides a nested feedback regulation and control method for a long-distance cascade gate pump water transfer project, which adopts regulation and control logic of cascade pump station daily-level overall planning and check gate group hour-level cooperation, and accurately matches pump station water delivery and gate group water distribution demands based on whole-line daily water demand prediction and hour-level real-time monitoring data; a dynamic response mechanism based on channel pool storage capacity control deviation is established, execution deviation is judged in real time, feedback calculation is carried out at the corresponding regulation and control scale, a whole-line optimal control strategy is generated in a rolling mode, whole-line and local water supply safety is guaranteed, the water supply demand response efficiency is improved, accurate execution of a whole-line water quantity scheduling scheme is guaranteed, and the water supply efficiency is improved. The distribution error and the local deviation are greatly reduced; the defect that a traditional regulation and control method cannot quickly respond to local control deviation is effectively overcome; the monitoring equipment and operation and maintenance cost is reduced, and the engineering cost performance is improved; and the adaptive capacity is enhanced, emergencies are dynamically handled, and hydraulic risks are prejudged and avoided to prolong the service life of facilities.

Description

Technical Field

[0001] This invention relates to the field of water conservancy and water diversion engineering technology, and in particular to a nested feedback control method for long-distance cascade gate pump water diversion projects. Background Technology

[0002] Long-distance water transfer projects play a significant role in alleviating regional water shortages. Inter-basin water transfer projects are generally characterized by long distances, large flow rates, and low elevation differences, often requiring 1-2 weeks to transport water from the source area to the receiving area. The actual operation of such large-scale projects relies on the coordinated scheduling and control of numerous control structures along the route, such as gates and pumping stations. To achieve precise operation control across the entire line, the currently commonly used control methods are mainly divided into three categories: flow feedback control, canal and pool water volume feedforward control, and water level feedback control.

[0003] Existing control technologies typically employ three methods to achieve engineering operation control: flow feedback control, which best aligns with scheduling logic, operates based on water demand and water loss coefficients along the route, offering high controllability and operational efficiency; canal and reservoir water feedforward control fully leverages the canal's storage capacity, maximizing water storage and regulation within the canal's safe operating range to achieve rapid and stable switching between operating conditions; and water level feedback control, the simplest and most economical method, utilizes easily monitored real-time water levels for control, with a simple and safe judgment process. All three methods require the coordinated operation of control works such as gates and pumping stations along the route to achieve overall line scheduling.

[0004] Each of the three existing control technologies has significant limitations: Flow feedback control requires high-frequency feedback judgment and high-precision monitoring due to rapid and large fluctuations in flow, necessitating the deployment of numerous flow monitoring devices, demanding extremely high accuracy and reliability from the equipment, and incurring enormous construction and maintenance costs; Canal and pool water feedforward control requires a high level of capability in formulating water allocation plans for the entire line, and canal and pool water volume is difficult to monitor in real time, requiring high-precision canal hydraulic simulation and continuous maintenance of simulation parameters; Water level feedback control is affected by factors such as hydraulic propagation time delay and local loss changes, with many influencing factors on water level changes and high difficulty in target calculation, requiring a scientific and reliable control feedback mechanism. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a nested feedback control method for long-distance cascade gate pump water diversion projects. Based on the overall water volume scheduling scheme, the large-scale gate pump project group is decomposed into spatiotemporal components, establishing multi-scale nested feedback control models at the real-time level for single stations, the hourly level for multiple stations, and the daily level for the entire line. This method generates the optimal control method for each station along the entire line in real time. It can effectively ensure the accurate execution of the overall water volume scheduling scheme and also cope with local execution deviations. Through real-time-hourly-daily scale hierarchical and feedback closed-loop, it makes up for the shortcomings of single control technologies and significantly improves the precise water supply operation capability of the project.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is as follows:

[0007] A nested feedback control method for long-distance cascade gate pump water diversion projects includes the following steps:

[0008] Step 1, Real-time control triggering of cascade gate pumps:

[0009] The control target and initial state are obtained through manual active control or timed rolling triggering.

[0010] Step 2, generate the optimal control strategy across the entire line:

[0011] Based on the overall water volume scheduling plan and real-time operation monitoring data, the system takes the daily scale control strategy of the cascade pumping station to coordinate the total amount and the hourly scale control strategy of the control gate group to make precise allocation. By setting the benchmark with daily scale control and adjusting the deviation with hourly scale control, and linking the deviation of canal and pool storage, the system achieves the balance of water supply and demand and safe and efficient operation of the entire line.

[0012] Step 3, Real-time dynamic response to control deviation:

[0013] A predictive control model for a single gate station is constructed. The gate station control strategy issues control commands to each gate station, driving the real-time control model of each gate station to perform parallel calculations. The real-time control opening degree of the gate station is determined. A dynamic response mechanism is constructed for each step of the calculation at multiple scales, including single-station real-time level, multi-station hourly level, and full-line daily level.

[0014] Furthermore, in step 1, the control targets are formulated based on the water use plan of the water receiving area, the water diversion plan of the water source area, and the engineering operation capacity, including the water source diversion target, the water distribution and supply target, and the water exchange of each section / key section.

[0015] Furthermore, in step 1, the initial state refers to the real-time flow, water level, and opening data of all control works (including pumping stations and control gates) along the entire water diversion project.

[0016] Furthermore, in step 2, the daily-scale control strategy for the cascade pumping stations controls the daily water throughput based on the control flow rate of each cascade pumping station. It considers the economic operation and changes in operating conditions of the pumping station units, adjusting operation in different time periods. The calculation formula for the control flow rate of each cascade pumping station is as follows:

[0017]

[0018] In the formula: Let be the planned water flow of the i-th cascade pumping station on day d, in m³ / s; this value is obtained by the cumulative flow of the downstream water exchange section of the cascade pumping station or the water supply of each branch in the water dispatching scheme. Let be the average flow rate of the i-th cascade pumping station on day d. .

[0019] Furthermore, in step 2, the hourly-scale control strategy of the control gate group is to construct a hydraulic steady flow model of the series canal pool, calculate the constructed flow-water level-storage curve and flow-water level-head difference curve, simulate the hydraulic characteristics of the water transfer response of the series canal pool, and achieve the goals of controlling the water level of the canal section to be stable, the flow rate to be adapted to the downstream hourly water demand, and the response to be timely.

[0020] Furthermore, for long-distance water diversion projects with long channels and varying bottom slopes, calculating the flow-water-storage curve and the flow-water-water-head difference curve requires calculating the flow rate and water surface line. The formula for calculating the flow rate is as follows:

[0021]

[0022] In the formula: Q is the flow rate, A represents the water flow area. n is the Manning roughness coefficient; R is the hydraulic radius, m; R = A / P, where P is the wetted perimeter, m; The slope of the canal bottom;

[0023] The formula for calculating the water level line is as follows:

[0024]

[0025] Where: Z is the bottom elevation of the cross-section plus water depth, m; α is the kinetic energy correction coefficient (usually taken as 1.0~1.1); For head loss along the route, (Average friction slope); This is a localized head loss; The friction slope is calculated using Manning's formula: .

[0026] Furthermore, in step 3, the predictive control model of the single gate station takes the future regulation strategies of the upstream and downstream stations of the control station as the prediction boundary, takes the scheduling target of the station as the regulation target, and considers the hydraulic connection between adjacent regulation structures, and incorporates future scheduling information and water level constraints.

[0027] Furthermore, the control objectives are to minimize water level deviation and the number of gate actions. A one-dimensional hydrodynamic unsteady flow model is used as the simulation tool, and a multi-objective optimization algorithm is employed to generate a real-time control strategy. The calculation formula is as follows:

[0028]

[0029] In the formula: The objective function is... for Weighting of water level control deviation at any given time; for Constantly control the weight of actions; For prediction in the time domain; For control time domain; control time domain is generally less than or equal to prediction time domain; For the model based on Predicting the future The deviation between the water level in front of the sluice gate and the target water level at any given time; For the model in The future determined by time The deviation between the water level in front of the gate and the target water level at any given time is the target value. When it is required to keep the water level at the target water level and the target water level does not change over time, it is always zero. In order to be in The future determined by time The magnitude of flow adjustment at the upstream gate station at any given time.

[0030] Furthermore, in step 3, the control command issued by the gate control strategy to each gate drives the parallel calculation of the real-time control model of each gate to determine the real-time control opening degree of the gate. The calculation logic of the real-time control model of the j-th gate at time t is summarized as follows:

[0031]

[0032] In the formula: the first two terms in parentheses are the control targets input into the gate station scheduling scheme; the third to fifth terms are the boundary conditions input into the gate station scheduling scheme, representing the inflow rate of the upstream canal pool, the outflow rate of the downstream canal pool, and the downstream water level of the downstream canal pool, respectively; The first three values ​​represent the water distribution flow rates at each branch point within the two canals before and after this sluice gate, derived from the water allocation plan. The last three values ​​represent the initial conditions of the model from real-time monitoring, representing the initial flow rate of this sluice gate, the initial upstream water level of this sluice gate, and the initial opening of this sluice gate, respectively. These are the conditions obtained through the real-time control model of this sluice gate. Calculate and obtain the control opening degree .

[0033] Furthermore, in step 3, the dynamic response mechanism is constructed using the deviation between the actual storage change and the target storage change as heuristic information for the dynamic response stage. The calculation formula is as follows:

[0034]

[0035] In the formula: The time step of the scheduling scheme is d at the cascade pump station scheduling level and h at the series gate station control level; the integral of the former term represents the actual execution, and is discretized according to the model step of this level. In order to be in Future plans formulated based on water allocation schemes The range of adjustment of the canal's storage capacity at any given moment;

[0036] Set allowable deviation limits The storage capacity deviation is related to the allowable variation in downstream water level when the canal / pond operates under this flow condition. Based on the hydraulic characteristic curve of this canal / pond, the calculation formula is as follows:

[0037]

[0038] In the formula: The maximum allowable water level fluctuation in this canal / pond during time period k is expressed in meters. Deviation at the step size of the upper-level scheduling scheme If the deviation is within the allowable limit, the storage deviation information will be fed back to the scheduling target of the next scheduling step. By utilizing past deviation information, it automatically adjusts operational decisions for future periods through adaptive feedback responses.

[0039] The beneficial effects of this invention are:

[0040] 1. Significantly improve the precision of water supply operation: Adopting a control logic of daily overall planning of cascade pumping stations and hourly coordination of control gate groups, based on daily water demand forecasts and hourly real-time monitoring data for the entire line, the system accurately matches the water delivery demand of pumping stations with the water distribution demand of the control gate groups; establishes a dynamic response mechanism based on the control deviation of canal and pool storage, judges the execution deviation in real time, and calculates the feedback at the corresponding control scale, continuously generating the optimal control strategy for the entire line, ensuring the water supply security of the entire line and local areas, improving the water supply demand response efficiency, ensuring the accurate execution of the water volume scheduling plan for the entire line, and significantly reducing allocation errors and local deviations; effectively solves the shortcomings of traditional control methods in that they cannot quickly respond to local control deviations, and overcomes the problems of high-frequency feedback of flow deviations, difficulty in calculating water level changes, and insufficient monitoring of canal and pool status.

[0041] 2. Cost Reduction and Efficiency Enhancement: Relying on daily-hourly hierarchical data verification, there is no need to deploy high-frequency real-time monitoring equipment, reducing equipment procurement, installation, and maintenance costs, and resolving the high-cost problem of traditional flow feedback control. Automated closed-loop control reduces the frequency of manual intervention. The combination of stable daily operation of cascade pumping stations and refined adjustment of control gate groups reduces equipment start-up and shutdown losses and energy waste, significantly improving the cost-effectiveness of the project.

[0042] 3. Enhanced Adaptability: The model dynamically adapts to unexpected situations such as fluctuations in water flow and equipment failures, enabling rapid response through daily fine-tuning and hourly emergency linkage mechanisms to prevent problems from escalating. It quantifies risks such as hydraulic time lag, providing daily prediction and hourly avoidance to reduce blind operations and extend facility lifespan.

[0043] 4. Improved Engineering Safety and Management: A closed-loop management system was established, encompassing daily forecasting, hourly control, and next-day review, ensuring process traceability and optimization. Daily stable operation of the pumping station and hourly control of the sluice gates provide dual safety guarantees, driving the transformation of the project from extensive operation and maintenance to layered and precise management.

[0044] The present invention will be explained in detail below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the nested feedback control method for long-distance cascade gate pump water diversion projects in Embodiment 1 of the present invention;

[0046] Figure 2 This is a diagram showing the flow rate-water level-storage capacity curve and the flow rate-water level-head difference curve of the canal pool in Embodiment 1 of the present invention;

[0047] Figure 3 This is a schematic diagram of the real-time control model algorithm for a single gate station in Embodiment 1 of the present invention;

[0048] Figure 4 This is a layout diagram of a long-distance water conveyance project in a certain water diversion section of Embodiment 2 of the present invention;

[0049] Figure 5 This is a graph showing the flow rate-water level-storage curve and the relationship between flow rate-water level-head difference for the channel pool in section 3-3 to 3-4 of the present invention in Embodiment 2.

[0050] Figure 6 This is a graph showing the flow rate-water level-storage curve and the relationship between flow rate-water level-head difference for the channel pool in section 3-4 to 3-5 of the present invention in Embodiment 2.

[0051] Figure 7 This is a graph showing the flow rate-water level-storage curve and the relationship between flow rate-water level-head difference for the channel pool in section 3-5 to 3-6 of the present invention in Embodiment 2.

[0052] Figure 8 This is a graph showing the flow rate-water level-storage curve and the relationship between flow rate-water level-head difference for the channel pool in section 3-6 to 3-7 of the present invention in Embodiment 2.

[0053] Figure 9 This is a diagram of the hourly-scale control strategy for each control gate group in Pump-3 to Pump-4 segments in Embodiment 2 of the present invention;

[0054] Figure 10 This is a schematic diagram of the nested feedback control results of the cascade gate pump project in Embodiment 2 of the present invention. Detailed Implementation

[0055] Example 1:

[0056] This embodiment presents a nested feedback control method for long-distance cascade gate pump water diversion projects, such as... Figure 1 As shown, it includes the following steps:

[0057] Step 1, Real-time control triggering of cascade gate pumps:

[0058] The control targets and initial states are obtained through active manual triggering or timed rolling triggering. The real-time operation control of long-distance water transfer projects requires taking the overall water volume scheduling plan as the core control target and the real-time operation monitoring data of the project as the initial state to carry out overall control, so as to ensure the safe, efficient and stable operation of the water transfer project.

[0059] (1-1) Obtaining the control target:

[0060] The overall water allocation plan needs to be formulated based on the water use plan of the water receiving area, the water diversion plan of the water source area, and the actual operational capacity of the project, such as the water conveyance capacity of the cascade pumping stations and the water distribution capacity of the control gate group.

[0061] Target setting sequence: Before the start of the scheduling period, the water allocation plan for the entire scheduling period (including the total water diversion quota of the water source and the total water supply quota of the water receiving area) needs to be determined; during the scheduling period, the water allocation plan can be flexibly adjusted according to the actual situation.

[0062] Objectives include: ① Water source diversion objectives; ② Water distribution and supply objectives; ③ Water exchange between different sections / key sections, providing direct control basis for the formulation of daily operation plans for cascade pumping stations and the calculation of hourly opening of control gate groups.

[0063] (1-2) Obtain the initial state:

[0064] The system acquires real-time flow, water level, and opening data of all control works (including pumping stations and control gates) along the entire water diversion project, which serve as the initial conditions for real-time control calculations and support the formulation of the optimal control strategy for the entire project.

[0065] Step 2, generate the optimal control strategy across the entire line:

[0066] Based on the overall water allocation plan and real-time operation monitoring data, the optimal control strategy for the entire line is generated. The core strategy is to coordinate the total amount of water supply and demand through the daily-scale control strategy of the cascade pumping stations and the precise allocation through the hourly-scale control strategy of the control gate group. Through the closed-loop logic of setting the benchmark with the daily plan, adjusting the deviation at the hourly level, and linking the deviation of the canal and pool storage, the water supply and demand of the entire line is balanced and the operation is safe and efficient.

[0067] (2-1) Daily-scale control strategy for cascade pumping stations:

[0068] Daily scale control targets for cascade pumping stations:

[0069] ① Total Balance: The deviation between the total water delivery volume of the cascade pumping stations on the day and the planned water flow demand on a daily scale is ≤±10%;

[0070] ② Optimal energy consumption: The energy consumption of the pump station group is optimized while meeting the water supply demand;

[0071] ③ Equipment safety: The number of start-ups and shutdowns of a single pump station shall be ≤ 2 times per day to avoid high-frequency impact losses.

[0072] The core of daily-scale regulation is to ensure that the daily water flow meets the standard. The overall target can be determined by using the daily average regulation flow as a benchmark, and the accumulation of time-period deviations can be avoided.

[0073] Each station in the cascade pumping station is a multi-unit water lifting station. Multi-unit water lifting stations typically operate at several discrete flow rates. By combining the economic operating characteristics of the units, peak and valley load differences, and changes in operating conditions, and while ensuring that the total daily water flow meets the standards, the flow targets for each time period are broken down based on the daily average controlled flow rate. The decomposed targets must be consistent with the discrete values ​​that the units can operate at. Real-time feedback of water flow deviations clarifies the timing of flow adjustments, avoiding excessive cumulative deviations over time periods. This ensures stable water delivery throughout the entire line and achieves efficient operation of the cascade pumping stations.

[0074] The average daily controlled flow rate of each cascade pumping station is calculated based on the daily flow rate, using the following formula:

[0075] (1)

[0076] In the formula: Let be the planned water flow of the i-th cascade pumping station on day d, in m³ / s; this value is obtained by the cumulative flow of the downstream water exchange section of the cascade pumping station or the water supply of each branch in the water dispatching scheme. Let be the average flow rate of the i-th cascade pumping station on day d, in m³ / s.

[0077] (2-2) Hourly-scale control strategy for control gate groups:

[0078] Control gate group hourly scale target:

[0079] ① Water level stability: The water level fluctuation range of each control gate section is ≤0.2m (canal and pool water level);

[0080] ② Flow rate adaptation: The deviation between the flow rate through the control gate and the hourly demand of downstream water users is ≤ ±10%;

[0081] ③ Timely response: In response to hydraulic fluctuations, the gate opening adjustment response time is ≤5 minutes to ensure control accuracy;

[0082] In long-distance water transfer projects, channels are typically divided into multiple pools (each pool located between two control structures). For each pool, under steady flow conditions, the flow rate remains constant along the flow path, and the water depth and velocity are determined by Manning's formula, calculated as follows:

[0083] (2)

[0084] In the formula: Q is the flow rate, m³ / s; A is the cross-sectional area, m². 2 ; n is the Manning roughness coefficient; R is the hydraulic radius, m; R=A / P, where P is the wetted perimeter, m; S0 is the channel bottom slope.

[0085] For long-distance water transfer projects with long channels and varying bottom slopes, it is necessary to calculate the water surface line. In steady flow, the water surface line is governed by the energy equation, taking head loss into account. The energy equation calculation formulas for upstream section 1 and downstream section 2 of the channel are as follows:

[0086] (3)

[0087] Where: Z is the bottom elevation of the cross-section plus water depth, m; α is the kinetic energy correction coefficient (usually taken as 1.0~1.1); For head loss along the route, (Average friction slope); This is a localized head loss; The friction slope is calculated using Manning's formula: .

[0088] A steady-flow hydraulic model of a series-connected canal and pool was constructed, and the hydraulic element curves corresponding to different operating conditions of each series-connected canal and pool were calculated: the flow-water-storage curve and the flow-water-water-head difference curve, as shown below. Figure 2 As shown, the hydraulic characteristics in the simplified simulation of the water transfer response of a series channel pool are illustrated.

[0089] The state vector is composed of the upstream flow and storage capacity of each canal section. The calculation formula for the hourly-scale control strategy of the control gate group, with system water balance as the core, is as follows:

[0090] (4)

[0091] In the formula: The channel storage capacity at time t is m³; Let t be the engineering control water level, such as The water level at time h represents the control level of the j-th control gate, m; To store the initial volume of the channel, m³; The average inflow rate of the channel during the time period, including the average inflow rate of upstream projects. Inflow within the interval, m³ / s; The average outflow from the channel during the time period, including the average flow from downstream projects. Average water distribution flow in channels , m³ / s; Let t be the downstream water level of the channel, in meters. The flow-water-storage relationship curve for this channel is used in this formula to find the steady-state water level downstream of the channel. Let t be the upstream water level of the channel, i.e., the water level after the channel inlet engineering station, in meters; This is the flow-water-head difference curve for this canal section, which is used in this formula to find the head difference between the upstream and downstream water levels of the canal section.

[0092] Step 3, Real-time dynamic response to control deviation:

[0093] Once the water level and flow scheduling process for each sluice gate is determined, real-time control will be implemented. In this step, the real-time control opening degree of each sluice gate needs to be determined, and then dynamic response control based on deviations will be implemented according to the measured data, as follows:

[0094] (3-1) Real-time dynamic response adjustment strategy:

[0095] ① Define the quantification deviation threshold:

[0096] Daily storage changes in cascade canal ponds The storage range limit corresponding to the allowable daily water level fluctuation is set according to the actual situation of the project. ;

[0097] Hourly storage changes of cascade canal pools The storage range limit corresponding to the allowable hourly water level fluctuation is set according to the actual situation of the project. ;

[0098] ② Hourly-level control execution rules:

[0099] Monitor hourly changes in the hourly storage capacity of the cascade canal pools (t mod d=0). If in Within the specified range, the hourly-scale control strategy of the control gate group is executed normally;

[0100] like Exceeding The hourly standard control strategy for the control gate group is suspended, the daily standard control strategy is triggered, and the hourly standard control strategy for the control gate group is regenerated based on the daily standard control strategy of the cascade pumping station.

[0101] ③ Daily-scale control implementation rules:

[0102] After the daily regulation ends (t mod d=0), calculate the daily storage change of the cascade canal pools. If in Within the scope, the original water allocation plan for the entire line (water source diversion, pump station water conveyance, and gate group water distribution) will be maintained for the following day.

[0103] like Exceeding The daily water supply scheduling plan for the entire line was recalculated: the available water supply and the demand of the water receiving area were recalculated, the daily water delivery volume of the pumping station and the hourly flow distribution benchmark were adjusted, and the water distribution target of the control gate group was updated simultaneously to solve the problem of total imbalance from the source.

[0104] ④ Priority rules for special scenarios:

[0105] If the daily and hourly storage changes of the cascade canal pools both exceed the design range, a daily-scale control strategy will be implemented, and an hourly-scale control strategy for the control gate group will be generated based on the daily-scale control strategy of the cascade pumping stations.

[0106] (3-2) Control strategy execution

[0107] The prediction boundary is set at the future control strategies of upstream and downstream stations of the control station, and the control target is set at the station's own control objective. Considering the hydraulic connection between adjacent control structures, future control information and water level constraints are substituted into the prediction model to construct a predictive control model for a single gate station.

[0108] A one-dimensional hydrodynamic unsteady flow model was used as the simulation tool, and a multi-objective optimization algorithm was employed to generate a real-time control strategy. The control objectives were to minimize water level deviation and the number of gate actions.

[0109] (5)

[0110] In the formula: The objective function is... for Weighting of water level control deviation at any given time; for Constantly control the weight of actions; For prediction in the time domain; For control time domain; control time domain is generally less than or equal to prediction time domain; For the model based on Predicting the future The deviation between the water level in front of the sluice gate and the target water level at any given time; For the model in The future determined by time The deviation between the water level in front of the gate and the target water level at any given time is the target value. When it is required to keep the water level at the target water level and the target water level does not change over time, it is always zero. In order to be in The future determined by time The magnitude of flow adjustment at the upstream gate station at any given time.

[0111] The optimal control action can be obtained by solving the problem. Only the first step of the control action The adjustment amount of the gate opening is obtained by back-calculation using equation (5). For example... Figure 3 As shown.

[0112] (3-3) Execution Deviation Feedback

[0113] The gate control strategy issues control commands to each gate, driving the parallel calculation of the real-time control models of each gate to determine the real-time control opening degree of the gate. The calculation logic of the real-time control model of the j-th gate at time t can be summarized as follows:

[0114] (6)

[0115] In the formula: the first two terms are the control targets input to the gate station scheduling scheme; the third to fifth terms are the boundary conditions input to the gate station scheduling scheme, representing the inflow rate of the upstream canal pool, the outflow rate of the downstream canal pool, and the downstream water level of the downstream canal pool, respectively. The first three values ​​represent the water distribution flow rates at each branch point within the two canals before and after this sluice gate, derived from the water allocation plan. The last three values ​​represent the initial conditions of the model from real-time monitoring, representing the initial flow rate of this sluice gate, the initial upstream water level of this sluice gate, and the initial opening of this sluice gate, respectively. These are the conditions obtained through the real-time control model of this sluice gate. Calculate and obtain the control opening degree .

[0116] (3-4) Dynamic response correction

[0117] In actual operation, deviations from the scheduling target may occur due to factors such as engineering conditions and control errors. Therefore, a dynamic response mechanism needs to be constructed at each calculation step across multiple scales, including single-station real-time, multi-station hourly, and full-line daily levels. To this end, the state variable, storage capacity, is used as the core control variable, and the deviation between the actual storage capacity change and the target storage capacity change serves as the heuristic information (direction and basis signal for adjustment decisions) for the dynamic response stage. The calculation formula is as follows:

[0118] (7)

[0119] In the formula: The time step of the scheduling scheme is d at the cascade pump station scheduling level and h at the series gate station control level; the integral of the former term represents the actual execution, and is discretized according to the model step of this level. In order to be in Future plans formulated based on water allocation schemes The range of adjustment of the canal's storage capacity at any given moment.

[0120] Set allowable deviation limits This is related to the storage capacity deviation corresponding to the allowable variation in downstream water level when the canal / reservoir is operating under this flow condition. The calculation is based on the hydraulic characteristic curve of this canal / reservoir as follows:

[0121] (8)

[0122] In the formula: The maximum allowable water level fluctuation in this canal / pond during time period k is expressed in meters. Deviation at the step size of the upper-level scheduling scheme If the deviation is within the allowable limit, the storage deviation information will be fed back to the scheduling target of the next scheduling step. The aim is to utilize past deviations to automatically adjust operational decisions for future periods through adaptive feedback responses.

[0123] This invention achieves precise water supply and response efficiency through a hierarchical control logic that combines daily coordination of cascade pumping stations with hourly collaboration of control gate groups. This is achieved through a dynamic response mechanism and hierarchical data verification, resulting in accurate supply and demand matching, reduced allocation deviations, and rapid response to local control deviations. It also reduces monitoring equipment and maintenance costs, lowers energy consumption and equipment wear, and improves the cost-effectiveness of the project. Furthermore, it enhances adaptability, dynamically handles emergencies, anticipates and mitigates hydraulic risks, and extends facility lifespan. Finally, it improves project safety and management, establishes a closed-loop management system, and promotes the transformation of the project towards hierarchical and precise governance.

[0124] Example 2:

[0125] This embodiment uses a section of the South-to-North Water Diversion Project as an application scenario to verify its regulation effect under conditions of sudden changes in flow at important water diversion points, such as... Figure 4 As shown, this section has 5 pumping stations, 30 sluice gates, and 14 water diversion gates, with a total length of 197.42 kilometers.

[0126] Step 1, Real-time control triggering of cascade gate pumps:

[0127] When the water use plan needs to be adjusted, the flow rate at the target water distribution point 3-3 should be adjusted to 7.5 m³ / s at 8:00 AM on day T. The water allocation scheme should be adjusted in response to the control method described in Example 1 above to ensure the safe, efficient and stable operation of the water transfer project.

[0128] (1-1) Obtaining the control target:

[0129] Obtain the adjusted water allocation plan for the entire line, including the daily water source diversion plan, the water supply plan for each branch point, and the water exchange at each pumping station section.

[0130] (1-2) Obtain the initial state:

[0131] Obtain real-time flow, water level, and opening data of all pumping stations and control gates along the entire project line.

[0132] Step 2, generate the optimal control strategy across the entire line:

[0133] Based on the overall water allocation plan and real-time operation monitoring data, the optimal control strategy for the entire line is generated, including the daily-scale control strategy for cascade pumping stations and the hourly-scale control strategy for the control gate group.

[0134] (2-1) Daily-scale control strategy for cascade pumping stations:

[0135] The controlled flow rates for each cascade pumping station are determined according to the calculation method provided in Example 1. The daily average flow rates of pumping stations Pump-1 through Pump-3 remain unchanged. Due to the flow rate change at water distribution point 3-3, the daily average flow rate of pumping station Pump-4 is reduced by 3.4 on the first day. The average flow rate decreased by 4.2 on the second day. Pump-5's average flow rate decreased by 7.5 on the second day. .

[0136] (2-2) Hourly-scale control strategy for control gate groups:

[0137] By constructing a steady-flow hydraulic model for a series of canals and pools, the hydraulic element curves corresponding to different operating conditions of each series canal and pool were calculated: the flow-level-storage curve and the flow-level-head difference curve. The hydraulic characteristic curves of the four series canals and pools located downstream of the flow-changing branch point 3-3 in the third canal section are shown below. Figures 5-8 As shown in the figure, the flow-water-storage curve clearly defines the safe flow range corresponding to the target water level of each cascade canal and pool, providing a direct basis for flow adaptation; the flow-water-head difference curve provides the maximum allowable head difference under the target water level, constraining the upper limit of flow.

[0138] State vectors are constructed based on upstream flow and storage in each canal section. With system water balance as the core logic, and combined with the hydraulic characteristics of the series-connected canals and pools, the correlation between flow, water level, storage, and head difference is analyzed. Following the method in Example 1, hourly-scale control strategies for the control gate groups are generated. The hourly-scale control strategies for each control gate group in the Pump-3-Pump-4 section, where key water distribution changes occur, are as follows: Figure 9 As shown: Gate 3-1, located after Pump-3 outlet, is not regulated. Gate 3-2, upstream of the diversion point 3-3 where the flow rate changes, is regulated synchronously with the diversion point at 8:00 on day T. The opening degree of each downstream gate station is reduced in turn, and then fine-tuned according to the hydraulic transmission to control the water level change within a safe range.

[0139] Step 3, Real-time dynamic response control deviation

[0140] Once the water level and flow scheduling process for each sluice gate is determined, real-time control is implemented. Using the method of this invention for real-time nested feedback regulation, the water level changes of the five pumping stations over 48 hours from 00:00 on day T, and the water level changes of each control gate in the Pump-3-Pump-4 section over 40 hours from 00:00 on day T, are as follows: Figure 10 As shown:

[0141] In this embodiment, under the control of the method of the present invention, the water level deviations of the cascade pumping stations and the series gate stations can be effectively controlled within ±0.1m. This indicates that in actual operation, the deviation nested feedback control mechanism constructed by the method of the present invention to address scheduling and control deviations at various scales can meet the requirements of short adjustment time and few control operations in the control of large-scale cascade pumping station groups.

[0142] Finally, it should be noted that the above is 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention (such as the application of various formulas, the order of steps, etc.) without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A nested feedback control method for long-distance cascade gate pump water diversion projects, characterized in that, Includes the following steps: Step 1, Real-time control triggering of cascade gate pumps: The control target and initial state are obtained through manual active control or timed rolling triggering. Step 2, generate the optimal control strategy across the entire line: Based on the overall water volume scheduling plan and real-time operation monitoring data, the system takes the daily scale control strategy of cascade pumping stations to coordinate the total amount and the hourly scale control strategy of control gate groups to accurately allocate water volume as the core. By setting the benchmark through daily scale control and adjusting the deviation through hourly scale control, and by linking the deviation of canal and pool storage, the system achieves the balance of water supply and demand and safe and efficient operation of the entire line. Step 3, Real-time dynamic response to control deviation: A predictive control model for a single gate station is constructed. The gate station control strategy issues control commands to each gate station, driving the real-time control model of each gate station to perform parallel calculations. The real-time control opening degree of the gate station is determined. A dynamic response mechanism is constructed for each step of the calculation at multiple scales, including single-station real-time level, multi-station hourly level, and full-line daily level.

2. The nested feedback control method for long-distance cascade gate pump water diversion projects according to claim 1, characterized in that, In step 1, the control targets are formulated based on the water use plan of the water receiving area and the water diversion plan of the water source area, combined with the engineering operation capacity, including the water source diversion target, the water distribution and supply target, and the water exchange of each section / key section.

3. The nested feedback control method for long-distance cascade gate pump water diversion projects according to claim 1, characterized in that, In step 1, the initial state refers to the real-time flow, water level, and opening data of all control works (including pumping stations and control gates) along the entire water diversion project.

4. The nested feedback control method for long-distance cascade gate pump water diversion projects according to claim 1, characterized in that, In step 2, the daily-scale control strategy for the cascade pumping stations controls the daily water throughput based on the control flow rate of each cascade pumping station. It considers the economic operation and changes in operating conditions of the pumping station units, adjusting operation in different time periods. The calculation formula for the average daily control flow rate of each cascade pumping station is as follows: In the formula, The planned water throughput of the i-th cascade pumping station on day d is... This value is obtained by accumulating the excess flow rate at the downstream water exchange section of the cascade pumping station or the water supply volume at each branch point in the water allocation scheme. Let be the average flow rate of the i-th cascade pumping station on day d. .

5. The nested feedback control method for long-distance cascade gate pump water diversion projects according to claim 1, characterized in that, In step 2, the hourly-scale control strategy of the control gate group is to construct a hydraulic steady flow model of the series canal pool, calculate the flow-water level-storage curve and the flow-water level-head difference curve, simulate the hydraulic characteristics of the water transfer response of the series canal pool, and achieve the goals of controlling the water level of the canal section to be stable, the flow rate to be adapted to the downstream hourly water demand, and the response to be timely.

6. The nested feedback control method for long-distance cascade gate pump water diversion projects according to claim 5, characterized in that, For long-distance water transfer projects with long channels and varying bottom slopes, calculating the flow-water-storage curve and the flow-water-water-head difference curve requires calculating the flow rate and water surface line. The formula for calculating the flow rate is as follows: In the formula, Q represents the flow rate. ; A represents the water flow area. n is the Manning roughness coefficient; R is the hydraulic radius, m; R = A / P, where P is the wetted perimeter, m; The slope of the canal bottom; The formula for calculating the water level line is as follows: ； In the formula, Z is the bottom elevation of the cross-section plus water depth, in meters; α is the kinetic energy correction coefficient (usually taken as 1.0~1.1). For head loss along the route, (Average friction slope); This is a localized head loss; The friction slope is calculated using Manning's formula: .

7. The nested feedback control method for long-distance cascade gate pump water diversion projects according to claim 1, characterized in that, In step 3, the predictive control model of the single gate station takes the future regulation strategies of the upstream and downstream stations of the control station as the prediction boundary, takes the scheduling target of the station as the regulation target, and considers the hydraulic connection between adjacent regulation structures, and incorporates future scheduling information and water level constraints.

8. The nested feedback control method for long-distance cascade gate pump water diversion projects according to claim 7, characterized in that, The control objectives are to minimize water level deviation and the number of gate actions. A one-dimensional hydrodynamic unsteady flow model is used as the simulation tool, and a multi-objective optimization algorithm is employed to generate a real-time control strategy. The calculation formula is as follows: ； In the formula, The objective function is... for Weighting of water level control deviation at any given time; for Constantly control the weight of actions; For prediction in the time domain; For control time domain; control time domain is generally less than or equal to prediction time domain; For the model based on Predicting the future The deviation between the water level in front of the sluice gate and the target water level at any given time; For the model in The future determined by time The deviation between the water level in front of the gate and the target water level at any given time is the target value. When it is required to keep the water level at the target water level and the target water level does not change over time, it is always zero. In order to be in The future determined by time The magnitude of flow adjustment at the upstream gate station at any given time.

9. The nested feedback control method for long-distance cascade gate pump water diversion projects as described in claim 1, characterized in that, In step 3, the gate control strategy issues control commands to each gate, driving the real-time control models of each gate to perform parallel calculations and determine the real-time control opening degree of the gate. The calculation logic of the real-time control model of the j-th gate at time t is summarized as follows: In the formula, the first two terms in parentheses are the control targets input to the gate station scheduling scheme; the third to fifth terms are the boundary conditions input to the gate station scheduling scheme, representing the inflow of the upstream canal pool, the outflow of the downstream canal pool, and the downstream water level of the downstream canal pool, respectively. The first three values ​​represent the water distribution flow rates at each branch point within the two canals before and after this sluice gate, derived from the water allocation plan; the last three values ​​represent the initial conditions of the model from real-time monitoring, namely the initial flow rate of this sluice gate, the initial water level upstream of the sluice gate, and the initial opening of the sluice gate, respectively, after being processed by the real-time control model of this sluice gate. Calculate and obtain the control opening degree .

10. The nested feedback control method for long-distance cascade gate pump water diversion projects as described in claim 1, characterized in that, In step 3, the dynamic response mechanism is constructed using the deviation between the actual storage change and the target storage change as heuristic information for the dynamic response stage. The calculation formula is as follows: In the formula, The time step of the scheduling scheme is d at the cascade pump station scheduling level and h at the series gate station control level; the integral of the former term represents the actual execution, and is discretized according to the model step of this level. In order to be in The future is determined based on the water allocation plan. The range of adjustment of the canal's storage capacity at any given moment; Set allowable deviation limits The storage capacity deviation is related to the allowable variation in downstream water level when the canal / pond operates under this flow condition. Based on the hydraulic characteristic curve of this canal / pond, the calculation formula is as follows: In the formula, The maximum allowable water level fluctuation during time period k in this canal pool is expressed in meters. like Deviation at the step size of the upper-level scheduling scheme This will trigger the upper layer to redefine the scheduling plan; If the deviation is within the allowable limit, the storage deviation information will be fed back to the scheduling target of the next scheduling step, i.e. ; By utilizing past deviation information, operational decisions for future periods are automatically adjusted through adaptive feedback responses.

Images