Long-distance cascade gate pump water transfer project layered optimization scheduling method

Abstract

The application discloses a long-distance cascade gate pump water transfer project layered optimization scheduling method, according to the large system decomposition coordination principle, the large-scale cascade gate pump group is spatially layered, the global optimization of the cascade pump station and the local optimization of the partition series gate station are carried out, and the layered optimization scheduling model is constructed according to the time and space scales of each level, the water supply safety of the whole line and the local is ensured, the water supply demand response efficiency is improved, the difficulty of optimization calculation and the requirement for computing power are reduced, the scheduling working condition requirements such as the water filling stage, the water source adjustment, the water distribution change, the water stop stage and emergency response in the actual operation of the long-distance cascade gate pump water transfer project are met, the optimal scheduling scheme of the whole line is generated, and the rolling scheduling decision support according to the real-time monitoring data is realized. The application has the advantages of general structure, strong universality, easy customization, effectively solves the shortage that the traditional scheduling method cannot quickly respond to various water supply working condition requirements.

Description

Technical Field

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

[0002] Long-distance water transfer projects, as a core means of alleviating the imbalance in water resource distribution, are generally characterized by spanning multiple river basins and being large-scale. They require the deployment of numerous control structures along the route, such as gates, pumping stations, and water distribution points, forming a complex cascade gate and pumping system. To achieve efficient water transfer and distribution, current commonly used scheduling methods are mainly divided into three categories: step-by-step water level control, segmented and zoned scheduling, and global optimization scheduling. Each of these methods has its own technical adaptability and significant limitations.

[0003] The step-by-step water level control method is a basic scheduling mode in engineering operations. Its core advantage lies in its simple and intuitive operation process, which can be implemented based on traditional hydraulic experience. Furthermore, it takes the stability of the water level at a single node as the control objective, resulting in strong system operation safety and high controllability. However, this method has significant efficiency shortcomings: control needs to be carried out step by step from the water source or the end distribution point along the process. The lag effect of water flow transmission and parameter adjustment is obvious, which means that it takes a long time to reach a stable operating condition across the entire line, making it difficult to quickly respond to dynamic water supply demands.

[0004] Segmented and zoned scheduling methods often divide control units based on administrative divisions, with each unit having an independent management and control system. This results in a clear and well-defined scheduling management structure, facilitating the implementation of responsibilities and refined control. However, this zoning model has high coordination costs: due to the coupling of hydraulic conditions in different sections, differences in water demand and operating conditions can easily lead to conflicts, requiring frequent coordination meetings between sections. This not only increases management complexity but may also cause a decrease in system operating efficiency due to coordination delays.

[0005] The global optimization scheduling method, with efficient water conveyance and precise water supply as its core objectives, integrates data from the entire system for comprehensive planning. It can maximize the balance between water conveyance efficiency, water supply accuracy, and operating costs, making it an ideal scheduling direction for modern water transfer projects. However, its technical threshold is extremely high: it requires stringent accuracy in the hydraulic simulation of the entire system, necessitates powerful computing support for multi-objective optimization, and sets standards for the control execution accuracy of the scheduling scheme that far exceed those of traditional methods. Due to technical limitations and cost constraints, it is difficult to widely implement in various projects. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a hierarchical optimization scheduling method for long-distance cascade gate pump water diversion projects. This method spatially stratifies a large-scale cascade gate pump group and constructs a hierarchical optimization scheduling model for each level's spatiotemporal scale, generating the optimal scheduling scheme for the entire line. This method effectively ensures the safety of water supply for the entire line and local areas, improves the efficiency of water supply demand response, and reduces the difficulty of optimization calculations and the computational power requirements.

[0007] This invention is implemented as follows:

[0008] A hierarchical optimization scheduling method for long-distance cascade pumping station water diversion projects includes three steps: setting multi-objective optimization parameters for cascade pumping stations, setting multi-objective optimization parameters for zoned series pumping stations, and hierarchical optimization scheduling calculation. The steps are as follows:

[0009] Step 1, set the multi-objective optimization parameters for the cascade pumping station:

[0010] The water volume of the cascade canal section is selected as the state variable, and the average daily flow of each cascade pumping station is selected as the decision variable to optimize the setting of various scheduling parameters.

[0011] Step 11, Determine the optimization objective:

[0012] To meet water supply demand, the highest water supply guarantee rate is represented by the lowest water distribution point; to ensure safe operation, the minimum cumulative fluctuation of the water level in the forebay of each pumping station is represented; and to enable rapid switching of operating conditions, the minimum cumulative change in the overall storage capacity is represented.

[0013] Step 12, Select constraints:

[0014] The constraints include the maximum flow capacity of the project, water level constraints at key sections, and water level fluctuation constraints.

[0015] Step 13, Water Quantity Dispatch Simulation:

[0016] By constructing a steady flow hydraulic model for a cascade canal section, the hydraulic element curves corresponding to different working conditions of each cascade canal section are calculated, simplifying the simulation of the hydraulic characteristics in the water transfer response of the cascade pumping station canal section.

[0017] Step 2, set the multi-objective optimization parameters for the zoned series gate station:

[0018] Step 21, Determine the optimization objective:

[0019] Rapid water diversion response is characterized by the minimum cumulative delay response time for all water diversion needs at all diversion points; ensuring safe operation is characterized by the minimum cumulative fluctuation of water levels in front of each sluice gate; overall rapid stabilization is characterized by the shortest cumulative stabilization time at each sluice gate.

[0020] Step 22, Select constraints:

[0021] The constraints include prohibitions on water level and flow rate inversion, maximum water level constraints, and water level fluctuation constraints.

[0022] Step 23, Hydraulic transmission simulation:

[0023] In each cascade canal section, several canal pools are further divided by multiple gate stations, and the boundary conditions are determined by the first and last two pump stations. In the real-time control scenario with hourly step, an integral time delay model of the series canal pools in each partition is constructed. The connection between the canal pools is achieved by using the gate submerged outflow formula, thus realizing rapid hydraulic simulation of the series canal pools.

[0024] Step 3, hierarchical optimization of scheduling computation:

[0025] Real-time data and future scheduling plans of long-distance cascade gate pump water diversion projects are obtained as inputs for hierarchical optimization scheduling calculations. The real-time data includes the water level before and after each control project and the flow rate through each pump station. The future scheduling plans include the water source diversion plan represented by the daily average flow sequence and the water supply plan of each branch point. A stepwise optimization algorithm is used to solve the calculations and realize multi-objective short-term optimization scheduling of the series channel sections.

[0026] Furthermore, in step 11, the rapid operating condition switching is characterized by minimizing the cumulative change in overall storage capacity, and the calculation formula is as follows:

[0027]

[0028] In the formula: W change,i,t Let m be the change in storage capacity of the i-th canal segment during time period t. 3 Minimal cumulative change in overall storage capacity enables rapid switching of overall operating conditions and ensures a smooth switching process.

[0029] Furthermore, in step 12, the calculation formula for the key section water level constraint is as follows:

[0030]

[0031] In the formula: Z sec,t Let be the water level at a certain cross-section at time t, in meters (m). This is the highest allowable water level at this cross-section, which is generally the design water level, in meters.

[0032] Furthermore, in step 12, the water level fluctuation constraint is calculated using the following formula:

[0033]

[0034] In the formula: Let m be the maximum allowable rise in water level during period t. Let m be the maximum allowable drop in water level during time period t.

[0035] Furthermore, in step 13, the hydraulic steady flow model is constructed using a water balance model that considers the hydraulic response characteristics of the cascade canal sections as a simulation model, achieving the purpose of rapid simulation instead of hydrodynamic simulation. The state vector is composed of the upstream flow and storage capacity of each canal section. The calculation formulas for the state parameters of the cascade canal section, with system water balance as the core, are as follows:

[0036]

[0037] In the formula: W t Let m be the canal storage capacity at time t. 3 ; The initial storage capacity of the canal section is m. 3 ; The average inflow rate of the canal section during the time period, including the average flow rate of the upstream pumping station. and interval inflow, m 3 / s; The average outflow of the canal section during the time period, including the average flow of the downstream pumping station. Average water distribution flow of the canal section m 3 / s, f1(S) represents the downstream water level of the canal section during time period t, i.e., the water level in front of the downstream pumping station, in meters. t The curve representing the flow rate, water level, and storage capacity of this canal section is used in this formula to find the steady-state water level downstream of the canal section. f2(S) represents the upstream water level of the canal section during time period t, i.e., the water level after the inlet station of the canal section, in meters. t ) represents the flow-water level-head difference relationship curve for this channel section, which is used in this formula to find the head difference between the upstream and downstream water levels of the channel section.

[0038] Furthermore, in step 13, the hydraulic element corresponding curves are the 'upstream flow-downstream water level-storage curve' and the 'upstream flow-downstream water level-head difference curve'.

[0039] Furthermore, in step 23, the integral time delay model uses the following formula to describe the relationship between the control point water level deviation and the changes in the inlet and outlet flow rates of the channel pool:

[0040]

[0041] In the formula: e is the deviation of the water level at the downstream water level control point of the canal from the target water level, in meters (m); f is time, in seconds (s); q in Let m be the change in the inlet flow rate of the canal / pool relative to the initial state. 3 / s;q out Let m be the change in the outlet flow rate of the canal / pool relative to the initial state. 3 / s;q dLet m be the change in water distribution in the canal / pool relative to the initial state. 3 / s;A d The area of ​​the backwater zone is in meters. 2 ;τ d Let A be the lag time of the canal pool, s, and parameter A be the lag time. d and τ d It can be obtained through formula calculation, simulation result relationship fitting, or measured data relationship fitting.

[0042] Furthermore, in step 23, the floodgate outflow formula is as follows:

[0043]

[0044] In the formula: Q is the flow rate through the gate, m 3 / s; M is the comprehensive flow coefficient under submerged orifice flow conditions; e g B represents the gate opening, in meters (m). g ΔZ is the gate width, in meters; I Let m be the water level difference before and after the i-th sluice gate.

[0045] Furthermore, in step 23, the hydraulic state parameters of the series-connected channel pool are calculated using the following formula:

[0046]

[0047] In the formula: Let t be the downstream water level of the gate station at time t, which is the upstream water level of the next canal pool, in meters.

[0048] Furthermore, in step 3, the stepwise optimization algorithm steps are as follows:

[0049] First layer: Global optimization of cascade pumping stations:

[0050] Step S1, Obtain basic data for the cascade pumping station:

[0051] Acquire real-time data (water level before and after each control project, and flow rate through each pumping station) and future scheduling plans (water source diversion plan and water supply plan for each branch point, expressed as daily average flow rate sequence) for long-distance cascade gate pumping water diversion projects.

[0052] Step S2, generate the initial scheduling scheme:

[0053] Based on the input real-time data and scheduling plan, a cascade pump station scheduling scheme is randomly generated;

[0054] Step S3, Water Quantity Dispatch Simulation:

[0055] Substituting the initial scheduling scheme into the water allocation simulation, we obtain the corresponding multi-objective function values;

[0056] Step S4, iteratively optimize the algorithm:

[0057] Within the set number of iterations, the stepwise optimization algorithm optimizes the original scheme and the corresponding objective function value within the constraints, generates a new scheduling scheme, and repeats the water scheduling simulation to obtain a new objective function value.

[0058] Step S5, output the globally optimal solution:

[0059] After iteration, the optimal cascade pump station scheduling scheme is generated and output, which is used as the scheduling objective and boundary condition for the second layer of optimization.

[0060] The second layer involves local optimization of zoned, interconnected gate stations:

[0061] Step S7, Input partition data and boundary conditions:

[0062] Obtain the real-time data and scheduling plan of the partition, and input the optimal cascade pump station scheduling scheme output by the first layer (as boundary conditions);

[0063] Step S8: Generate the initial scheme for the partitioned series gate station scheduling:

[0064] Based on the real-time data and scheduling plan input by the partition, a random partition series gate station scheduling scheme is generated;

[0065] Step S9, Hydraulic transmission simulation:

[0066] Substituting the initial scheme for the scheduling of the zoned series gate stations into the hydraulic transmission simulation, the corresponding objective function value is obtained;

[0067] Step S10, iteratively optimize the algorithm:

[0068] Within the set number of iterations, the stepwise optimization algorithm optimizes the original scheme based on the corresponding objective function value, generates a new scheduling scheme, and repeats the hydraulic transmission simulation to obtain the new objective function value.

[0069] Step S11, output the optimal solution for the zoned series gate station:

[0070] After the iteration is completed, the optimal partitioned series gate station scheduling scheme is generated and output; the above steps are executed in parallel for each partition in the project, and finally the optimal scheduling scheme of the whole project is formed in both "global + local".

[0071] The beneficial effects of this invention are as follows: This invention provides a hierarchical optimization scheduling method for long-distance cascade gate pump water transfer projects. Based on the principle of large-scale system decomposition and coordination, it spatially hierarchically divides a large-scale cascade gate pump group into layers and constructs a hierarchical optimization scheduling model for each layer's spatiotemporal scale. This effectively ensures the safety of water supply along the entire line and in specific areas, improves the efficiency of water supply demand response, and reduces the difficulty of optimization calculations and the computational power requirements. It can meet the needs of various scheduling conditions in the actual operation of long-distance cascade gate pump water transfer projects, such as the water filling stage, water source adjustment, water distribution changes, water outage stage, and emergency response, generating the optimal scheduling scheme for the entire line. It can also realize rolling scheduling decision support based on real-time monitoring data. This invention has a universal structure, strong versatility, and is easy to customize; it effectively solves the shortcomings of traditional scheduling methods that cannot quickly respond to various water supply conditions.

[0072] The present invention will be explained in detail below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0073] Figure 1 This is the overall flowchart of the hierarchical optimization scheduling method of the present invention;

[0074] Figure 2 This is a hierarchical overview diagram of the hierarchical optimization scheduling method of the present invention;

[0075] Figure 3 This is a graph showing the upstream flow rate-downstream water level-storage capacity curve and the upstream flow rate-downstream water level-head difference curve of the canal section of this invention;

[0076] Figure 4 This is a schematic diagram of the partitioned series channel-pool integral time delay model of the hierarchical optimization scheduling method of the present invention;

[0077] Figure 5 This is a flowchart of the hierarchical optimization scheduling calculation in the hierarchical optimization scheduling method of the present invention. Detailed Implementation

[0078] Example 1;

[0079] This embodiment presents a hierarchical optimization scheduling method for long-distance cascade gate pump water diversion projects, such as... Figure 1-2 As shown, the scheme includes three steps: setting multi-objective optimization parameters for cascade pumping stations, setting multi-objective optimization parameters for zoned series gate stations, and hierarchical optimization scheduling calculation. The steps are as follows:

[0080] Step 1, Set the multi-objective optimization parameters for the cascade pumping station:

[0081] The first layer involves global optimization of the cascade pumping stations. Water volume in the cascade canal sections is selected as the state variable, and the average daily flow rate of each pumping station is used as the decision variable. An optimization algorithm is employed for calculation. The specific optimization scheduling parameter settings include three steps: optimization objective, constraints, and water volume scheduling simulation method.

[0082] (1-1) Optimization objective:

[0083] Optimization objective 1: To meet water supply demand, characterized by the highest water supply guarantee rate at the lowest water distribution point.

[0084] The goal is to maximize the water supply security level of the "most unfavorable" water distribution point, rather than maximizing the average guarantee rate of all water distribution points. It reflects the principle of backup guarantee in water transfer projects, that is, in the event of insufficient total water resources or disruption to the water transfer process, priority is given to avoiding extreme situations such as supply interruptions or severe water shortages at individual water distribution points, thus achieving a balance between fairness and stability in water supply. The calculation formula is as follows:

[0085]

[0086] In the formula: This represents the minimum water supply guarantee rate for the branch outlets. This value is found by traversing all branch outlets and is used as the target for dynamic optimization. This ensures that even when the overall water supply is insufficient, some branch outlets will not experience extreme water shortages.

[0087] Optimization objective 2, ensuring safe operation, is characterized by minimizing the cumulative fluctuation of the water level in the forebay of each pumping station:

[0088] Pumping stations are the core of water diversion projects, and their operational status directly affects project safety. The objective is to control the fluctuation range of the water level in the pumping station's forebay to avoid equipment damage caused by frequent switching of operating conditions. This is because the forebay water level is a key parameter affecting the pumping station's water intake conditions; sudden rises and falls in water level can lead to problems such as pump cavitation and vibration, and even safety accidents such as unit shutdown and equipment failure. The calculation formula is as follows:

[0089]

[0090] In the formula: The maximum fluctuation in the forebay water level of the pumping station is expressed in meters (m). This value is obtained by iterating through the water levels of all pumping station forebays. Minimizing this value ensures smoother switching of pumping station operating conditions and guarantees safe operation of the project.

[0091] Optimization objective 3: Rapid switching between operating conditions, characterized by minimizing the cumulative change in overall storage capacity.

[0092] Long-distance water transfer projects often require switching from one water transfer condition (e.g., low-flow water transfer) to another (e.g., high-flow water transfer) based on changes in water source inflow and adjustments to water demand along the route (e.g., agricultural irrigation season, peak urban water demand). The core objective is to achieve a rapid and smooth transition between operating conditions by controlling the overall changes in channel storage, avoiding problems such as disordered water levels and excessive flow fluctuations throughout the entire line during the switching process. The calculation formula is as follows:

[0093]

[0094] In the formula: W change,i,t Let m be the change in storage capacity of the i-th canal segment during time period t. 3 Minimizing the cumulative change in overall storage capacity enables rapid switching of overall operating conditions and ensures that the switching process is as smooth as possible.

[0095] (1-2) Constraints:

[0096] In the optimized scheduling of long-distance water transfer projects, constraints are rigid boundaries that ensure the feasibility, safety, and compliance of the scheduling scheme. The achievement of any optimization goal must be based on meeting these constraints. The three major constraints—maximum flow capacity, water level at key sections, and water level fluctuation—define the operational boundaries from three dimensions: the upper limit of water conveyance capacity, the safety of key nodes, and the protection of the engineering structure.

[0097] Constraint 1, Maximum current carrying capacity of the project:

[0098] This constraint refers to the maximum flow rate that various water conveyance structures (such as canals, control gates, and pumping stations) in a water transfer project can pass under the premise of safe operation. It is an inherent property determined by the structural dimensions, design standards, and hydraulic characteristics of the structures, and is a rigid boundary that cannot be breached in the flow allocation scheme. The calculation formula is as follows:

[0099]

[0100] In the formula: Q en,t Let m be the flow rate of a certain project at time t. 3 / s; The maximum current carrying capacity of this project is m. 3 / s.

[0101] Constraint 2, Water level constraint at key cross-sections:

[0102] This constraint means that at key monitoring sections of a water diversion project (such as the forebay of a pumping station, before and after the gate, etc.), the water level must be controlled below the upper limit (maximum allowable water level). The water level at these sections is directly related to the stable operation of core equipment and the safety of project operation, and is a key indicator that needs to be monitored in real time during the scheduling process. The calculation formula is as follows:

[0103]

[0104] In the formula: Z sec,t Let be the water level at a certain cross-section at time t, in meters (m). This is the highest allowable water level at this cross-section, which is generally the design water level, in meters.

[0105] Constraint 3, Water level fluctuation constraint:

[0106] This constraint is a dynamic water level control requirement specifically designed for the unique characteristics of open channel water conveyance. It stipulates that the water level fluctuation (the magnitude of increase or decrease) at the channel cross-section must not exceed a certain limit within a unit of time. Its core objective is to protect the channel lining structure. During open channel operation, the linings on both sides are simultaneously subjected to inward groundwater pressure and outward water pressure. This constraint aims to prevent lining damage due to sudden rises or falls in water level, thereby extending the project's service life. The formula for calculating the water level fluctuation constraint is as follows:

[0107]

[0108] In the formula: Let m be the maximum allowable rise in water level during period t. Let m be the maximum allowable drop in water level during time period t.

[0109] (1-3) Water allocation simulation:

[0110] In the optimized scheduling of long-distance cascade gate pump water transfer projects, water volume scheduling simulation serves as a bridge connecting the scheduling scheme with the actual response of the project. It simulates the water transport process of the project through mathematical models, accurately predicts core parameters such as water level, flow rate, and storage capacity under different scheduling commands, and provides data support for solving the optimization objectives and verifying the constraints.

[0111] By constructing a steady-flow hydraulic model for a cascade canal section, a simplified steady-flow model using the one-dimensional Saint-Venant equations can be adopted. The calculated hydraulic element curves for each cascade canal section under various upstream flow and downstream water level operating conditions are: the 'upstream flow-downstream water level-storage curve' and the 'upstream flow-downstream water level-head difference curve', as shown below. Figure 3 As shown, the hydraulic characteristics in the water transfer response of a cascade pumping station section are simplified.

[0112] Simplifying the model avoids the computational complexity of overly detailed hydraulic simulations, while allowing for a certain range of simulation deviations increases the feasibility of the optimization algorithm converging. Subsequent second-layer scheduling can prevent excessively large or unreasonable scheduling deviations.

[0113] The steady flow model of hydraulics in cascade canal sections is a fundamental tool for simulation. It characterizes the hydraulic transmission law of cascade canal sections (including control projects such as pumping stations and gates) under the assumption of relatively stable water flow.

[0114] A water balance model considering hydraulic response characteristics is constructed as the simulation model. While ensuring accuracy, a simplified water conservation logic that accounts for the hydraulic response characteristics of cascade canal sections is used to replace the complex hydrodynamic equations, thus achieving rapid simulation of the hydrodynamics of cascade canal sections. The state vector is composed of the upstream flow and storage capacity of each canal section. The state parameters of the cascade canal section, with system water balance as the core, are calculated as follows:

[0115]

[0116] In the formula: W t Let m be the canal storage capacity at time t. 3 ; The initial storage capacity of the canal section is m. 3 ; The average inflow rate of the canal section during the time period, including the average flow rate of the upstream pumping station. and interval inflow, m 3 / s; The average outflow of the canal section during the time period, including the average flow of the downstream pumping station. Average water distribution flow of the canal section m 3 / s. f1(S) represents the downstream water level of the canal section during time period t, i.e., the water level in front of the downstream pumping station, in meters. t The curve representing the flow rate, water level, and storage capacity of this canal section is used in this formula to find the steady-state water level downstream of the canal section. f2(S) represents the upstream water level of the canal section during time period t, i.e., the water level after the inlet station of the canal section, in meters. t ) represents the flow-water level-head difference relationship curve for this channel section, which is used in this formula to find the head difference between the upstream and downstream water levels of the channel section.

[0117] Step 2, set the multi-objective optimization parameters for the zoned series gate station:

[0118] In long-distance water transfer projects, the series of canals and pools formed by the intervals between tandem gates are the core units of water conveyance (such as the channel section between two adjacent control gates). These canals and pools do not operate independently, but are formed by strong hydraulic connections through the regulation of the gates and stations. Changes in the water level and flow rate of one canal or pool will be rapidly transmitted to the upstream and downstream canals and pools through the water flow. Especially during transitions in operating conditions (such as switching from low flow rate to high flow rate or adjusting the water supply at the branch outlet), this mutual influence will be amplified and may cause hydraulic fluctuations throughout the entire line.

[0119] The second layer involves local optimization of the series-connected gate stations, leveraging the strong hydraulic connections between the series-connected channels and pools to understand their mutual influence and transmission relationships during operational transitions. Therefore, the optimization scheduling steps in this layer are similar to those in the previous layer, with the main differences being: 1) a series-connected channel-pool integral time-delay model that reliably simulates the actual hydraulic process under unsteady conditions; and 2) the decision variable is changed from the average flow rate over the step-time period to the water level at the end of the step-time period, i.e., the hourly water level at each gate.

[0120] The integral time delay model is a simulation tool for strong hydraulic connections in series canals and pools. It reliably captures the delay and cumulative effects of water flow transmission under unsteady conditions, overcoming the limitation of steady-state models that cannot reflect dynamic transition processes. This model can realistically replicate the complete chain of upstream disturbance → midstream transmission → downstream response during the transition of operating conditions, providing simulation data that is closer to engineering reality for scheduling decisions.

[0121] Decision variables are parameters that can be actively controlled in a scheduling scheme. The difference between the two directly determines the accuracy and dynamic adaptability of the scheduling. For example, when adjusting the water supply at the branch outlet, if the average flow rate is used as the decision variable, there may be fluctuations in the water level, such as a sudden rise and fall within a period of time. However, if the "water level at the end of the period" is used as the target, the water level can be ensured to approach the target value steadily through real-time fine-tuning of the flow rate, thus avoiding the transmission and amplification of hydraulic disturbances between the series channels and pools.

[0122] (2-1) Optimization objective:

[0123] Optimization objective 1: Rapid water distribution response, characterized by minimizing the cumulative delay response time of all water distribution point demands.

[0124] This objective focuses on the time difference between the demand at the water distribution point and the actual water supply. It uses the minimum cumulative delayed response time of all water distribution points as a quantitative indicator, enabling the water conveyance system to quickly respond to changes in water demand along the route. This reduces the lag problem of demand changing but supply not being adjusted. It ensures that the series-connected gates can quickly respond to the overall switching of operating conditions at both upstream and downstream pumping stations, while utilizing the water storage capacity of the canal and reservoir and the hydraulic transmission effect to achieve rapid water distribution response. The calculation formula is as follows:

[0125]

[0126] In the formula: Q outlet,i,t Let m be the water distribution flow rate at time t for the i-th water distribution point. 3 / s;Q demand,i For the water distribution demand of the i-th water distribution point, m 3 / s. The above formula minimizes the cumulative delay response time of the water distribution demand of all water outlets by accumulating the difference between the water distribution flow and the water distribution demand at each time period.

[0127] Optimization objective 2, ensuring safe operation, is characterized by minimizing the cumulative fluctuation of water level in front of each sluice gate:

[0128] The goal is to minimize the cumulative fluctuation of the water level in front of each gate station as a quantitative indicator. By controlling the water level fluctuations at key nodes, damage to gate station equipment and destruction of canal and pool structures can be avoided.

[0129] Optimization objective 3: Overall rapid and stable operation, characterized by the shortest cumulative stabilization time for each gate station.

[0130] This objective focuses on the final result, using the shortest cumulative stabilization time of each gate station as a quantitative indicator, so that the entire series gate station-channel pool system can quickly transition and remain stable, reducing the duration of hydraulic fluctuations.

[0131]

[0132] In the formula: Q gate,i,t Let m be the flow rate at time t for the i-th gate. 3 / s;Q gate,i,final Let m be the flow rate at which the i-th gate finally reaches a steady state. 3 / s. The goal is to minimize the difference between the cumulative flow rate of all gate stations at each time period and the final steady-state flow rate, representing the shortest cumulative stabilization time for each gate station. This goal enables rapid switching of overall operating conditions while ensuring the smoothest possible switching process.

[0133] (2-2) Constraints:

[0134] Constraint 1: Water level and flow rate inversion is not allowed.

[0135] Unlike cascade pumping stations, this constraint is a hydraulic constraint on the basis of the series gate and canal pool system. It means that during the entire operation of the system (especially during the transition of operating conditions), the water level of the upstream canal pool must be higher than that of the downstream canal pool, and the flow rate of the upstream gate station must not be less than that of the downstream gate station. In other words, it is strictly forbidden for the downstream water level and flow rate to be higher than that of the upstream.

[0136]

[0137] Constraint 2, Highest water level constraint:

[0138] This constraint means that the water level in front of all gate stations in the system (i.e., the water level at the end of the upstream channel pool) must not exceed the set maximum allowable water level. This water level is usually based on the design water level, and a certain safety margin is reserved for some dangerous sections.

[0139] Constraint 3: Water level fluctuation constraint:

[0140] The variation range of water level before and after each sluice gate within a unit time period shall not exceed the water level variation constraint (set limit), and the hourly water level variation constraint is generally 15cm.

[0141] (2-3) Hydraulic transmission simulation:

[0142] In the real-time control of a series gate-sluice gate-canal-pool system, hydraulic transmission simulation is a technology for understanding the dynamic transmission law of water flow between the gate-sluice gate and the canal-pool, and for supporting scheduling decisions.

[0143] Within each cascade canal section, multiple gate stations further divide the area into several canal pools, while the daily average flow rates of the first and last pumping stations determine the upstream and downstream boundary conditions. In a real-time control scenario with hourly step sizes for the series gate stations (i.e., scheduling decisions need to be updated hourly to adapt to dynamic changes in operating conditions), an integral time-delay model of the series canal pools for each section is constructed, such as... Figure 4 As shown in the figure, this model can characterize the cumulative change of water storage in the channel pool with the inflow and outflow through the integral effect, and restore the time delay of water flow transmission between the channel pool through the time delay effect, thus capturing unsteady hydraulic characteristics. The outflow of the gate station is calculated by the flood discharge formula of the gate, and the connection between the channel pool is realized to achieve rapid hydraulic simulation of the series channel pool.

[0144] The integral time-delay model describes the relationship between the water level deviation at the control point and the changes in the inlet and outlet flow rates of the canal / pond. The calculation formula is as follows:

[0145]

[0146] In the formula: e is the deviation of the water level at the downstream water level control point of the canal from the target water level, in meters (m); t is time, in seconds (s); q in Let m be the change in the inlet flow rate of the canal / pool relative to the initial state. 3 / s;q out Let m be the change in the outlet flow rate of the canal / pool relative to the initial state. 3 / s;q d Let m be the change in water distribution in the canal / pool relative to the initial state. 3 / s;A d The area of ​​the backwater zone is in meters. 2 ;τ d Let A be the lag time of the canal pool, in seconds. d and τ d It can be obtained through formula calculation, fitting simulation results, or fitting measured data. The calculation formula is as follows:

[0147]

[0148] In the formula: n is the Manning roughness coefficient; L is the channel length, m; R is the hydraulic radius, m; S is the riverbed slope, m; K is the ratio of the change in downstream water depth over time.

[0149] This formula requires calculation and calibration of the canals and pools between each gate and pump project, and the engineering conditions of the water conveyance canals and pools of the same water diversion project are not significantly different.

[0150] The formula for calculating the submerged outflow of a flat slab gate is as follows:

[0151]

[0152] In the formula: Q is the flow rate through the gate, m 3 / s; M is the comprehensive flow coefficient under submerged orifice flow conditions; e g B represents the gate opening, in meters (m). g ΔZ is the gate width, in meters; I Let m be the water level difference before and after the i-th sluice gate.

[0153] When selecting the control step size for series gate stations, if the step size is too small, the next calculation will be performed before the previous gate action has been hydraulically transmitted. At this time, the system cannot accurately perceive the actual effect of the previous control, which will cause the entire series gate station system to collapse because it cannot accurately perceive the hydraulic control process.

[0154] If the step size is too large, the actions of several upstream gates will have a cumulative effect in the hydraulic transmission, causing the entire system to become unstable and collapse. Ideally, the number of control operations at each gate in a series-connected gate system should be as small as possible; therefore, the recommended gate control step size is within [τ]. d ,2τ d Choose from within ], where τ d The hydraulic transmission characteristic time is the average time required for the control effect of the upstream gate to be transmitted to the adjacent downstream gate station (determined by factors such as the length of the channel and the water flow velocity).

[0155] In addition, the selection of the gate control step size also needs to consider two practical factors: first, the gate action time (the time it takes for the gate to adjust to the target opening from receiving the command; the step size must be greater than this time to ensure that the operation is completed); second, the actual control requirements (such as when the water demand at the water distribution point is frequent, a smaller step size can be selected to improve the response speed, while a larger step size can be selected when the operating conditions are stable to reduce the operation).

[0156] If the gate control step size is appropriately selected, the calculation formulas for the hydraulic state parameters of the series channel pool based on the integral time delay model are as follows, using equations (11) and (12):

[0157]

[0158] In the formula: Let t be the downstream water level of the gate station at time t, which is the upstream water level of the next canal pool, in meters.

[0159] Step 3, hierarchical optimization of scheduling computation:

[0160] Real-time data and future scheduling plans for long-distance cascade gate pumping projects are acquired as inputs for hierarchical optimization scheduling calculations. The real-time data includes the real-time water levels before and after each control project, and the flow rate through each pumping station. The future scheduling plans include water source diversion plans expressed as daily average flow sequences and water supply plans for each branch point.

[0161] A stepwise optimization algorithm is employed to solve the problem and achieve multi-objective short-term optimal scheduling of cascaded canal sections. This algorithm possesses good global optimization capabilities, relatively low requirements on problem constraints and dimensionality, and wide applicability. The specific calculation steps are as follows: Figure 5 As shown.

[0162] First, based on the input real-time data and scheduling plan, a cascade pumping station scheduling scheme is randomly generated within constraints. Through water volume scheduling simulation, the corresponding multi-objective function values ​​are obtained. Within a set number of iterations, a stepwise optimization algorithm continuously optimizes the scheme based on the original scheme and the corresponding objective function values, resulting in a new scheduling scheme within the constraints. Finally, the optimal cascade pumping station scheduling scheme, i.e., the daily average flow sequence of each cascade pumping station, is generated. This scheme is then used as the scheduling objective and boundary conditions for the optimized scheduling calculation of the series-connected gate pumping stations in each zone.

[0163] Subsequently, a randomized scheduling scheme for the cascade pumping stations in each zone is generated based on the real-time data and scheduling plan input by the zone. The objective function value is obtained through hydraulic transmission simulation. Within a set number of iterations, a stepwise optimization algorithm continuously optimizes the scheme based on the original scheme and the corresponding objective function value, resulting in a new scheduling scheme. Finally, the optimal scheduling scheme for the cascade pumping stations in each zone is generated, representing the hourly upstream water level sequence for each cascade pumping station. Each zone in this layer is calculated in parallel with clear boundaries, representing the scheme results for the first layer (pumping station flow rate, water level). The needs of each zone are comprehensively planned within the first-layer cascade pumping station scheduling scheme, effectively avoiding computational conflicts. First layer: Global optimization of cascade pumping stations:

[0164] Step S1, Obtain basic data for the cascade pumping station:

[0165] Acquire real-time data (water level before and after each control project, and flow rate through each pumping station) and future scheduling plans (water source diversion plan and water supply plan for each branch point, expressed as daily average flow rate sequence) for long-distance cascade gate pumping water diversion projects.

[0166] Step S2, generate the initial scheduling scheme:

[0167] Based on the input real-time data and scheduling plan, a cascade pump station scheduling scheme is randomly generated within the constraints.

[0168] Step S3, Water Quantity Dispatch Simulation:

[0169] Substituting the initial scheduling scheme into the water volume scheduling simulation yields the corresponding multi-objective function values.

[0170] Step S4, iteratively optimize the algorithm:

[0171] Within the set number of iterations (500-2000), the stepwise optimization algorithm will perform optimization calculations according to the water supply plans of each water distribution point and in accordance with the priority of multiple objectives. Ultimately, it will generate a better scheduling scheme within the constraints, and repeatedly perform water scheduling simulations to continuously optimize the objective function value. The specific optimization logic and parameter settings are as follows:

[0172] Prioritize achieving the optimal goal 1: meeting water supply demand as the core primary goal. Through algorithmic iteration, first find the optimal solution for this goal, that is, ensure that the water supply guarantee rate of each branch water outlet reaches 100%, which is the optimal state and the basis for subsequent optimization.

[0173] Optimize subsequent objectives within the optimal range of objective 1: Within the constraint of achieving a 100% water supply guarantee rate at the water distribution point, sequentially optimize subsequent objectives:

[0174] Objective 2: Ensure safe operation. The optimization direction is to minimize the cumulative fluctuation of the water level in the forebay of each pumping station. Under the premise of meeting the water supply demand, find the combination of scheduling parameters that minimizes the fluctuation of the water level.

[0175] Within the constraints of objectives 1 and 2, optimize objective 3 for rapid switching between operating conditions: with the goal of minimizing the cumulative change in overall storage capacity, further optimize the scheduling scheme within the range of simultaneously satisfying the constraints of objectives 1 and 2, thereby improving the efficiency of switching between operating conditions.

[0176] Step S5, output the globally optimal solution:

[0177] After the iteration is completed, the optimal cascade pump station scheduling scheme is generated and output, which is used as the scheduling objective and boundary condition for the second-level optimization. This condition is a rigid constraint.

[0178] The second layer involves local optimization of zoned, interconnected gate stations:

[0179] Step S6, Input partition data and boundary conditions:

[0180] Obtain the real-time data and scheduling plan of the partition, and input the optimal cascade pump station scheduling scheme output by the first layer (as boundary conditions).

[0181] Step S7: Generate the initial scheme for the partitioned series gate station scheduling:

[0182] Based on the real-time data and scheduling plan input by the partition, a partition-connected gate station scheduling scheme is randomly generated within the constraints.

[0183] Step S8, Hydraulic transmission simulation:

[0184] Substituting the initial scheme for the scheduling of the sectional series gate stations into the hydraulic transmission simulation, the corresponding objective function value is obtained.

[0185] Step S9, iteratively optimize the algorithm:

[0186] Within the set number of iterations, the stepwise optimization algorithm will be based on the original scheduling scheme and the corresponding initial objective function value, and will carry out iterative optimization in the order of multi-objective priority: first focus on the optimization of the core objective, and then promote the optimization of subsequent objectives within the constraints of the preceding objectives. After each iteration generates a new scheduling scheme, hydraulic transmission simulation is repeated to obtain the objective function value corresponding to the new scheme, and the iteration continues until the optimization requirements are met.

[0187] The number of iterations in this layer is set according to the actual engineering conditions (interval channel length, engineering design flow rate, etc.). While meeting the requirements, the number of iterations should be set as low as possible to ensure calculation efficiency. For example, in the Jiaodong section of the East Line (interval between adjacent projects is 6-10 kilometers, and the design flow rate is 25-32 cubic meters per second), the number of iterations is set to 500.

[0188] The specific optimization logic and target setting are as follows: Prioritize achieving target 1 optimally, with rapid water distribution response as the primary optimization objective. The optimal standard is 100% achievement of the rapid response requirement, specifically characterized by minimizing the cumulative delay response time of all water distribution outlets. The algorithm will iteratively search for optimization until this objective reaches its optimal state; this serves as a prerequisite for subsequent optimizations.

[0189] Optimize Objective 2 within the optimal range of Objective 1: Within the constraint of minimizing the cumulative delayed response time of Objective 1, and already achieving the optimal range, further optimize Objective 2 to ensure safe operation. This objective has no fixed optimal value; it needs to be optimized within a reasonable range. Specifically, the optimization direction is to minimize the cumulative fluctuation of the water level in the forebay of each sluice gate, ensuring the safe operation of the sluice gates during scheduling.

[0190] Optimize objective 3 within the constraints of objectives 1 and 2: Under the premise of simultaneously satisfying the optimality of objective 1 and the minimum reasonable range of cumulative water level fluctuation in objective 2, conduct optimization of objective 3 for overall rapid and stable operation. This objective also needs to be optimized within the specified range, specifically characterized by minimizing the cumulative stabilization time of each gate station, further improving the overall stability and response efficiency of the dispatching system.

[0191] Step S10: Output the optimal solution for the zoned series gate station:

[0192] After iteration, the optimal partitioned series gate station scheduling scheme is generated and output. The above steps are executed in parallel for each partition in the project, and finally the optimal scheduling scheme of the whole project is formed, namely the daily average flow sequence of each cascade pump station and the hourly gate front water level sequence of each series gate station.

[0193] 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 long-distance stepped gate pump water transfer project stratified optimization scheduling method, characterized in that, The method comprises three steps of setting multi-objective optimization parameters of cascade pumping stations, setting multi-objective optimization parameters of zoned cascade gate stations and hierarchical optimization scheduling calculation, and the steps are as follows. Step 1, setting multi-objective optimization parameters of cascade pumping stations: Selecting water quantity of cascade channel sections as state variables and daily average flow of each cascade pumping station as decision variables, the parameters of optimization scheduling are set; Step 11, determining optimization objectives: Meeting water supply demand, with the highest water supply guarantee rate of the lowest water distribution port as the representation; ensuring safe operation, with the minimum cumulative variation of water level in front of each pumping station as the representation; and switching fast working conditions, with the minimum cumulative change of overall storage as the representation; Step 12, selecting constraint conditions: The constraint conditions include maximum flow capacity of the project, key section water level constraint and water level variation constraint; Step 13, water quantity scheduling simulation: Through construction of a cascade channel section hydraulics constant flow model, the corresponding curves of hydraulic elements of each cascade channel section under different working conditions are calculated, so as to simplify the hydraulic characteristics in simulation of cascade pumping station channel section water quantity transmission response; Step 2, setting multi-objective optimization parameters of zoned cascade gate stations: Step 21, determining optimization objectives: Fast water distribution response, with the cumulative minimum delay response time of water distribution demand of all water distribution ports as the representation; ensuring safe operation, with the minimum cumulative variation of water level in front of each gate station as the representation; and overall fast and stable, with the minimum cumulative stable time of each gate station as the representation; Step 22, selecting constraint conditions: The constraint conditions include water level and flow hysteresis, maximum water level constraint and water level variation constraint; Step 23, hydraulic transmission simulation: In each cascade channel section, a plurality of channel pools are further divided by a plurality of gate stations, and the first and last pumping stations determine boundary conditions; in a real-time control scene with an hour step, an integral time delay model of each zoned cascade channel pool is constructed, the channel pools are connected through a gate submerged outflow flow formula, and hydraulic fast simulation of the zoned cascade channel pools is realized; Step 3, hierarchical optimization scheduling calculation: Real-time data and future scheduling plan of the long-distance cascade gate and pump water transfer project are obtained as inputs of the hierarchical optimization scheduling calculation, wherein the real-time data include water level in front of each control project, water level behind each control project, flow through each pumping station, the future scheduling plan includes water source diversion plan represented by daily average flow sequence and water supply plan of each water distribution port; A step-by-step optimization algorithm is used for solving and calculation, so as to realize multi-objective short-term optimization scheduling of the zoned cascade channel section. In step 11, the switching fast working conditions, with the minimum cumulative change of overall storage as the representation, has the following calculation formula:

2. The long-distance cascade gate pump water transfer project layered optimization scheduling method according to claim 1, characterized in that, In step 12, the key section water level constraint has the following calculation formula: In the formula: is the storage volume change of the ith channel section at the t period, unit m³, and the overall storage volume cumulative change is minimized to enable rapid switching of the overall working condition and ensure that the switching process is as smooth as possible.

3. The long-distance cascade gate pump water transfer project layered optimization scheduling method according to claim 1, characterized in that, In step 12, the water level variation constraint has the following calculation formula: In the formula: is the water level of a certain section at time t, in meters; is the highest allowable water level of the section, which is the design water level, in meters.

4. The long-distance cascade gate pump water transfer project stratified optimization scheduling method according to claim 1, characterized in that, In step 13, the corresponding curves of hydraulic elements are 'upstream flow-downstream water level-storage curve' and 'upstream flow-downstream water level-head difference curve'. In the formula: is the maximum allowed rise of water level in the period t, unit m; is the maximum allowed fall of water level in the period t, unit m.

5. The long-distance cascade gate pump water transfer project stratified optimization scheduling method according to claim 1, characterized in that, In step 13, the hydraulic constant flow model is constructed by considering the water balance model of the hydraulic response characteristics of the cascade channel section as a simulation model to achieve the purpose of replacing the water dynamic rapid simulation, and the upstream flow and storage state vector of each channel section are used as the state vector The cascade channel section state parameters with the system water balance as the core are calculated by the following formula: wherein: is the storage of the canal section at time t, in m³; is the storage of the canal section at initial time, in m³; is the average inflow of the canal section during the time interval, including the average flow of the upstream pumping station and the inflow of the section, in m³ / s; is the average outflow of the canal section during the time interval, including the average flow of the downstream pumping station and the average diversion flow of the canal section , in m³ / s; is the downstream water level of the canal section at time t, i.e. the water level before the downstream pumping station, in m; is the flow-water level-storage relationship curve of the canal section, used in the present formula to find the downstream steady-state water level of the canal section; is the upstream water level of the canal section at time t, i.e. the water level after the inlet works of the canal section, in m; is the flow-water level-head difference relationship curve of the canal section, used in the present formula to find the head difference between the upstream and downstream water levels of the canal section.

6. The long-distance cascade gate pump water transfer project stratified optimization scheduling method according to claim 1, characterized in that, In step 23, the integral time delay model describes the relationship between water level deviation of a control point and change amount of inlet and outlet flow of a channel pool, and has the following calculation formula:

7. The long-distance cascade gate pump water transfer project stratified optimization scheduling method according to claim 1, characterized in that, In step 23, the gate submerged outflow flow formula is: In the formula: is the deviation of the water level of the control point downstream of the canal from the target water level, with units of m; is time, with units of s; is the change in the inflow of the canal from the initial state, with units of m 3 / s; is the change in the outflow of the canal from the initial state, with units of m 3 / s; is the change in the water distribution of the canal from the initial state, with units of m 3 / s; is the area of the backwater zone, with units of m 2 ; is the lag time of the canal, with units of s, and parameters and may be obtained through formula calculation, simulation result relationship fitting, or measured data relationship fitting.

8. The long-distance cascade gate pump water transfer project stratified optimization scheduling method according to claim 1, wherein, In step 23, the hydraulic state parameters of the zoned cascade channel pool have the following calculation formula: wherein: Q is the discharge, in m 3 / s; Ci is the comprehensive discharge coefficient for the submerged orifice flow condition; H is the gate opening, in m; B is the gate width, in m; hi is the head difference before and after the i-th gate, in m.

9. The long-distance cascade gate pump water transfer project stratified optimization scheduling method according to claim 1, characterized in that, In step 3, the step-by-step optimization algorithm has the following steps: In the formula: is the water level behind the gate at the downstream gate station of the canal pool at time t, i.e. the water level at the upstream of the next canal pool, in meters.

10. The long-distance cascade gate pump water transfer project stratified optimization scheduling method according to claim 1, wherein, First layer, global optimization of cascade pumping stations: ​ Step S1, obtain the basic data of the cascade pumping station: Obtain the real-time data of the long-distance cascade gate-pump water transfer project, including the water level before and after the station, the passing flow of each pumping station, and the future scheduling plan including the water source diversion plan represented by the daily average flow sequence and the water supply plan of each water outlet; Step S2, generate the initial scheduling scheme: According to the input real-time data and scheduling plan, randomly generate the cascade pumping station scheduling scheme; Step S3, water quantity scheduling simulation: Substitute the initial scheduling scheme into the water quantity scheduling simulation to obtain the corresponding multi-objective function value; Step S4, step-by-step optimization algorithm iteration: Within the set number of iterations, the step-by-step optimization algorithm optimizes the original scheme and the corresponding objective function value within the constraint condition to generate a new scheduling scheme, and repeats the water quantity scheduling simulation to obtain a new objective function value; Step S5, output the global optimal scheme: After iteration, generate and output the optimal cascade pumping station scheduling scheme, which is used as the scheduling target and boundary condition of the second layer optimization; Second layer, local optimization of zonal cascade gate station: Step S7, input the zonal data and boundary conditions: Obtain the real-time data and scheduling plan of the zone, and input the optimal cascade pumping station scheduling scheme output by the first layer as the boundary condition; Step S8, generate the initial scheduling scheme of the zonal cascade gate station: According to the input real-time data and scheduling plan of the zone, randomly generate the zonal cascade gate station scheduling scheme; Step S9, hydraulic transmission simulation: Substitute the initial scheduling scheme of the zonal cascade gate station into the hydraulic transmission simulation to obtain the corresponding objective function value; Step S10, step-by-step optimization algorithm iteration: Within the set number of iterations, the step-by-step optimization algorithm optimizes the original scheme and the corresponding objective function value to generate a new scheduling scheme, and repeats the hydraulic transmission simulation to obtain a new objective function value; Step S11, output the optimal scheme of the zonal cascade gate station: After iteration, generate and output the optimal zonal cascade gate station scheduling scheme; perform the above steps in parallel for each zone in the project to finally form the optimal scheduling scheme of the whole project "global + local".

Images