A Model Predictive Control-Based Flood Control Scheduling Method and System for Pump and Gate Groups

Abstract

This invention discloses a model-based predictive control (MMDC) method and system for flood control scheduling of pumping and sluice gate groups. It involves the intersection of urban water system pumping and sluice gate group scheduling technology and computational technology. Addressing the issue of inconsistent flood control scheduling standards in existing pumping and sluice gate group scheduling methods, this invention proposes a real-time scheduling strategy for pumping and sluice gate groups, considering system scheduling capacity, cost, and equipment lifespan. The basic principle of this strategy is to fully utilize SWMM software to construct a river hydraulic model, perform a flood forecasting process under designed rainfall, input the forecast results into the optimized scheduling model of the pumping and sluice gate group, and efficiently calculate the pump station opening and closing and gate adjustment methods using an artificial bee colony algorithm. It then provides scientific and reasonable suggestions and timely emergency scheduling plans to relevant departments. This invention is beneficial for maintaining stable water levels, saving operating costs of flood control and drainage pumping and sluice gate stations, reducing pumping and sluice gate station maintenance costs, and improving the risk control capabilities of urban water systems.

Description

Technical Field

[0001] This invention relates to the intersection of urban water system pumping and drainage station scheduling technology and computing technology, and in particular to a flood control scheduling strategy for pumping and drainage station groups that comprehensively considers the drainage capacity, cost and equipment operation and maintenance costs of the pumping and drainage station system. Background Technology

[0002] Urban flood control and drainage scheduling is an important part of urban flood emergency scheduling. However, there are three problems with the current scheduling of pumps and sluice gates: (1) Scheduling based on established rules makes it difficult to cope with the uncertainty of rainfall and lacks the flexibility of real-time scheduling; (2) In the study of optimized operation of pump and sluice gate groups, only the effect of drainage and flood control is usually emphasized, while indicators related to the economic and safety aspects of project operation are ignored, such as unit maintenance costs and pump start-up and shutdown times and frequency; (3) Flood control scheduling is formulated by scheduling personnel based on experience, which is highly arbitrary. There is no unified standard for real-time scheduling calculation, making it difficult to accurately control the water level of urban rivers, let alone meet specific scheduling objectives. The risk control capability is weak, and it is difficult to produce immediate effects on real-time short-term scheduling, stable equipment operation, and economic benefits.

[0003] Compared to ordinary river drainage systems, urban drainage systems are far more complex, compounded by my country's urban development philosophy. Consequently, my country places relatively little emphasis on the optimized operation and investment management of urban drainage pumping stations. Urban flooding disasters caused by torrential rains are also common.

[0004] Therefore, how to unify flood control scheduling standards based on existing pump and sluice gate group scheduling methods is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the present invention provides a method for flood control scheduling of pumping stations and sluice gates based on model predictive control. To achieve the above objectives, the present invention adopts the following technical solution:

[0006] Step 1: Construct an optimized scheduling model for the pumping and sluice gate group based on the drainage capacity, operating costs, and subsequent maintenance of the pumping and sluice gate drainage stations. The optimized scheduling model for the pumping and sluice gate group consists of scheduling objectives and constraints.

[0007] Step 2: Solve the scheduling optimization problem using the artificial bee colony algorithm, which mimics the bee colony's collection of bee resources, to obtain the combined scheme of design flood, pump station opening and closing, and gate opening.

[0008] Step 3: Set up the allocation method for pumping stations and gates, use the artificial bee colony algorithm for intelligent optimization allocation, and input the results of the pump and gate group optimization scheduling model established in Step 1 and Step 2 into the constructed urban water system river hydraulic model based on SWMM within the scheduling cycle. By configuring the opening degree of urban pumping stations and drainage stations, the flood control scheduling model is finally obtained.

[0009] Step 4: Real-time scheduling strategy for pumping and drainage stations based on artificial bee colony algorithm. By using artificial bee colony algorithm to solve the established flood control scheduling model, a real-time scheduling strategy scheme under rainfall can be obtained.

[0010] Optionally, the optimal scheduling model for pump and gate groups includes the following scheduling objectives:

[0011] (1-1) Target water level optimization:

[0012]

[0013] In the formula: ΔL1 represents the index for optimizing the target water level. For the optimal water level target within a given time period t, L t Let t be the water level during time period t, t be the scheduling time period, T be the total scheduling time period, n be the set nth level pumping station, and N be the total number of pumping stations.

[0014] (1-2) Minimal pumping costs during operation:

[0015]

[0016] In the formula: F is the pumping cost within the scheduling cycle, k is the electricity unit / kW·h, ρ is the density of water, and g is the local gravitational acceleration, taken as 9.8 m / s². 2 Q is the flow rate of the pump station during time period t, and H is the flow rate of the pump station during time period t. n,t Let ΔT be the head of the pump station during time period t, and ΔT be the length of the unit time period.

[0017] (1-3) The pump station has the fewest opening and closing times:

[0018]

[0019] In the formula: C is the sum of the number of times the pump station units start and stop during the scheduling period; L(i,j) is the number of pump station units that change in the j-th time period compared to the previous time period for the i-th pump station;

[0020] (1-4) The water level peak was lowest at the end of the scheduling period:

[0021] minΔL2=min{maxZ n,t};

[0022] In the formula: ΔL2 is the index for optimizing the peak water level at the end of the scheduling cycle of the pumping station, Z n,t Let T be the final water level of the nth pumping station at time t; T is the total scheduling period.

[0023] Optionally, the optimal scheduling model for pump and gate groups includes the following constraints:

[0024] (2-1) Water level constraint:

[0025]

[0026] This constraint is the water level constraint for the target pumping station; the water level is below [a certain value]. The target pumping station is at the dead water level, and the flood control limit water level is set at the target pumping station. As the highest water level control assessment target;

[0027] (2-2) Flow constraints:

[0028]

[0029] In the formula: These are the upper and lower limits of the flow rate of the upstream pumping station and the flow rate of the downstream pumping station for the nth pumping station, respectively.

[0030] (2-3) Pump head constraint:

[0031]

[0032] In the formula: This represents the upper and lower limits of the head of the upstream and downstream pumping stations for the nth pumping station;

[0033] (2-4) Gate opening constraint:

[0034] K i,min <K i <K i,max ;

[0035] Where: K i,min K represents the minimum gate opening. i,max This represents the maximum gate opening.

[0036] (2-5) Nonnegativity constraints:

[0037] All of the above variables are non-negative.

[0038] Optionally, the artificial bee colony algorithm in step two includes the following steps:

[0039] Step 1: Initialize the parameters of the artificial bee colony algorithm, including the maximum number of iterations (maxcycles), the threshold (limit), the new nectar source (Vi), the nectar source (Xi), the probability of selecting each nectar source (Pi), and the parameters SN and t=1.

[0040] Step 2: Generate the initial population X;

[0041] Step 3: Hire peak i to search for honey sources and generate new honey sources Vi;

[0042] Step 4: Determine if fit(Vi) is greater than fit(Xi). If it is, replace Xi with Vi; otherwise, retain the honey source Xi.

[0043] Step 5: Determine if nectar source i is less than half of SN. If so, calculate the probability Pi of each nectar source selection, observe the bees' greedy selection based on the probability, update the population, and record the optimal solution. If not, increment the number of hired bees by 1 and return to step 3 to continue the loop.

[0044] Step 6: Determine if the nectar source has reached the threshold limit. If it has, proceed to step 7. If it has not reached the threshold, generate new bees and proceed to step 7.

[0045] Step 7: Determine if the iteration count t has reached the maximum iteration count maxcycle. If it has, terminate the algorithm and output the optimal solution. If it has not reached the maximum iteration count, increment the iteration count by 1 and return to step 3 to continue the loop.

[0046] Optionally, the method for obtaining the flood control scheduling model in step three includes the following steps:

[0047] Step 1: Construct a river hydraulic model based on SWMM to provide forecasts of design flood inflow;

[0048] Step 2: Construct a prediction model based on water balance according to state variables and control variables;

[0049] Step 3: Initialize the inflow forecast results, the initial water level in front of the pumping station, and set the total time period to T, where t is time t, and also set the parameter k;

[0050] Step 4: Perform an optimization based on the pump and gate group optimization scheduling model, solve the pump and gate group optimization scheduling model to obtain k control variable sequences, but only use the control variables of the first sequence, and let t = t + 1;

[0051] Step 5: Determine if time t is less than the total time period T. If so, update the initial value and return to step 3 to continue the loop until the optimization of the entire time axis is completed. Otherwise, output the flood control scheduling model.

[0052] On the other hand, a model predictive control-based flood control scheduling system for pump and gate groups is provided, comprising the following modules:

[0053] The scheduling strategy generation module constructs an optimized scheduling model for pump and gate groups based on the drainage capacity, operating costs, and subsequent maintenance of pump and gate drainage stations.

[0054] The scheduling optimization problem-solving module uses an artificial bee colony algorithm that mimics the bee colony's collection of bee sources to solve the combined scheme of design flood, pump station opening and closing, and gate opening.

[0055] The intelligent optimization allocation module uses the artificial bee colony algorithm for intelligent optimization allocation. It inputs the results of the established pump and gate group optimization scheduling model into the constructed urban water system river hydraulic model based on SWMM within the scheduling cycle. By configuring the opening degree of urban pump and gate drainage stations, the flood control scheduling model is finally obtained.

[0056] The real-time scheduling strategy generation module uses the artificial bee colony algorithm to solve the established flood control scheduling model, which can generate a real-time scheduling strategy under rainfall.

[0057] Optionally, the scheduling strategy generation module also includes a scheduling target generation module, which is used to generate scheduling targets such as target water level optimization, minimum pumping cost during operation, minimum number of pump station start-ups and shutdowns, and minimum peak water level at the end of the scheduling period.

[0058] Optionally, the scheduling strategy generation module also includes a constraint generation module, which is used to generate water level constraints, flow constraints, pump head constraints, gate opening constraints, and non-negative constraints.

[0059] Optionally, the scheduling optimization problem-solving module also includes an artificial bee colony algorithm module, used to solve the optimal solution of the joint scheduling optimization problem of pump and gate groups.

[0060] Optionally, the intelligent optimization allocation module also includes a flood control scheduling model acquisition module, used to acquire the flood control scheduling model.

[0061] As can be seen from the above technical solutions, compared with the prior art, the present invention provides a pump and gate group flood control scheduling method and system based on model predictive control. Compared with traditional methods, by selecting optimization objectives and constraints for different scheduling needs, it is beneficial to maintain stable water levels during the scheduling cycle and reduce operating costs during scheduling operation. In addition, the present invention optimizes the start-up and closure of pump stations as an important scheduling objective, thereby reducing maintenance frequency and improving equipment lifespan. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0063] Figure 1 This is a schematic diagram of the artificial bee colony algorithm.

[0064] Figure 2 Flowchart of a real-time scheduling strategy for a certain pumping station group;

[0065] Figure 3 This is a curve showing the tidal level process.

[0066] Figure 4 This is a scheduling diagram for the upstream primary pumping station, which is designed for a 100-year flood event.

[0067] Figure 5 This is a scheduling diagram for the downstream secondary pumping station, which is designed for a 100-year flood event. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] This invention discloses a flood control scheduling method for pump and gate groups based on model predictive control, comprising the following steps:

[0070] Step 1: Construct an optimized scheduling model for the pumping and sluice gate group based on the drainage capacity, operating costs, and subsequent maintenance of the pumping and sluice gate drainage stations. The optimized scheduling model for the pumping and sluice gate group consists of scheduling objectives and constraints.

[0071] Step 2: Solve the scheduling optimization problem using the artificial bee colony algorithm, which mimics the bee colony's collection of bee resources, to obtain the combined scheme of design flood, pump station opening and closing, and gate opening.

[0072] Step 3: Set up the allocation method for pumping stations and gates, use the artificial bee colony algorithm for intelligent optimization allocation, and input the results of the pump and gate group optimization scheduling model established in Step 1 and Step 2 into the urban water system river hydraulic model built based on SWMM within the scheduling cycle. By configuring the opening degree of urban pumping stations and drainage stations, the flood control scheduling model is finally obtained.

[0073] Step 4: Real-time scheduling strategy for pump and gate groups based on artificial bee colony algorithm. Using artificial bee colony algorithm, the established flood control scheduling model is solved to obtain a real-time scheduling strategy scheme under rainfall.

[0074] In one specific embodiment, the pump and gate group optimization scheduling model includes the following scheduling objectives:

[0075] (1-1) Target water level optimization:

[0076]

[0077] In the formula: ΔL1 represents the index for optimizing the target water level. For the optimal water level target within a given time period t, L tLet t be the water level during time period t, t be the scheduling time period, T be the total scheduling time period, n be the set nth level pumping station, and N be the total number of pumping stations.

[0078] (1-2) Minimal pumping costs during operation:

[0079]

[0080] In the formula: F is the pumping cost within the scheduling cycle, k is the electricity unit / kW·h, ρ is the density of water, and g is the local gravitational acceleration, taken as 9.8 m / s². 2 Q is the flow rate of the pump station during time period t, and H is the flow rate of the pump station during time period t. n,t Let ΔT be the head of the pump station during time period t, and ΔT be the length of the unit time period.

[0081] (1-3) The pump station has the fewest opening and closing times:

[0082]

[0083] In the formula: C is the sum of the number of times the pump station units start and stop during the scheduling period; L(i,j) is the number of pump station units that change in the j-th time period compared to the previous time period for the i-th pump station;

[0084] (1-4) The water level peak was lowest at the end of the scheduling period:

[0085] minΔL2=min{maxZ n,t};

[0086] In the formula: ΔL2 is the index for optimizing the peak water level at the end of the scheduling cycle of the pumping station, Z n,t Let T be the final water level of the nth pumping station at time t; T is the total scheduling period.

[0087] In one specific embodiment, the pump and gate group optimal scheduling model includes the following constraints:

[0088] (2-1) Water level constraint:

[0089]

[0090] This constraint is the water level constraint for the target pumping station; the water level is below [a certain value]. The target pumping station is at the dead water level, and the flood control limit water level is set at the target pumping station. As the highest water level control assessment target;

[0091] (2-2) Flow constraints:

[0092]

[0093] In the formula: These are the upper and lower limits of the flow rate of the upstream pumping station and the flow rate of the downstream pumping station for the nth pumping station, respectively.

[0094] (2-3) Pump head constraint:

[0095]

[0096] In the formula: This represents the upper and lower limits of the head of the upstream and downstream pumping stations for the nth pumping station;

[0097] (2-4) Gate opening constraint:

[0098] K i,min <K i <K i,max ;

[0099] Where: K i,min K represents the minimum gate opening. i,max This represents the maximum gate opening.

[0100] (2-5) Nonnegativity constraints:

[0101] All of the above variables are non-negative.

[0102] In one specific embodiment, such as Figure 1 As shown, the artificial bee colony algorithm in step two includes the following steps:

[0103] Step 1: Initialize the parameters of the artificial bee colony algorithm, including the maximum number of iterations (maxcycles), the threshold (limit), the new nectar source (Vi), the nectar source (Xi), the probability of selecting each nectar source (Pi), and the parameters SN and t=1.

[0104] Step 2: Generate the initial population X;

[0105] Step 3: Hire peak i to search for honey sources and generate new honey sources Vi;

[0106] Step 4: Determine if fit(Vi) is greater than fit(Xi). If it is, replace Xi with Vi; otherwise, retain the honey source Xi.

[0107] Step 5: Determine if nectar source i is less than half of SN. If so, calculate the probability Pi of each nectar source selection, observe the bees' greedy selection based on the probability, update the population, and record the optimal solution. If not, increment the number of hired bees by 1 and return to step 3 to continue the loop.

[0108] Step 6: Determine if the nectar source has reached the threshold limit. If it has, proceed to step 7. If it has not reached the threshold, generate new bees and proceed to step 7.

[0109] Step 7: Determine if the iteration count t has reached the maximum iteration count maxcycle. If it has, terminate the algorithm and output the optimal solution. If it has not reached the maximum iteration count, increment the iteration count by 1 and return to step 3 to continue the loop.

[0110] In one specific embodiment, the method for obtaining the flood control scheduling model in step three includes the following steps:

[0111] Step 1: Construct a river hydrodynamic model based on SWMM to provide design flood inflow forecasts. Specifically, by using SWMM software, a river hydrodynamic model is constructed based on local topographic data, land use type, rainfall data, and river cross-section data to simulate the dynamic rainfall-runoff simulation process in the study area.

[0112] Step 2: Construct a prediction model based on water balance using state variables and control variables. The state variable is the real-time water level of the pump and gate group, and the control variables are the pump station's opening and closing status and the gate opening degree. In conjunction with actual engineering needs, during flood control scheduling in the flood season, the focus is primarily on changes in the water level itself. Corresponding engineering measures are adopted for scheduling based on the observed water level changes. Furthermore, based on the current operating water level and the inflow, outflow, and water level-reservoir capacity relationship over a future period, the model can predict the final water level change process in the study area over a future period. This allows for the establishment of a prediction model based on the principle of water balance within the optimized scheduling model of the pump and gate group.

[0113] Step 3: Initialize the inflow forecast results, the initial water level in front of the pumping station, and set the total time period to T, where t is time t, and also set the parameter k;

[0114] Step 4: Perform an optimization based on the pump and gate group optimization scheduling model, solve the pump and gate group optimization scheduling model to obtain k control variable sequences, but only use the control variables of the first sequence, and let t = t + 1;

[0115] Step 5: Determine if time t is less than the total time period T. If so, update the initial value and return to step 3 to continue the loop until the optimization of the entire time axis is completed. Otherwise, output the flood control scheduling model.

[0116] In a specific embodiment, taking a certain pump gate group as an example, its scheduling plan is formulated, and the real-time scheduling strategy flowchart is as follows: Figure 2 As shown,

[0117] The upstream pumping station has a total installed capacity of 3000kW, a net head range of 0.0–7.0m, and a design flow rate of 25m³ / h. 3 / s. Before the flood arrives, the lake water is pumped to the downstream river channel through the pumping station to free up the reservoir capacity. After the flood arrives, the pumping station is turned on to drain the water and ensure that the lake water level does not exceed the warning level.

[0118] The downstream pumping station is affected by the local tide level. When the tide level of the outer river is higher than that of the inner river, the sluice gate is closed and the drainage station is opened; if the tide level is lower, the drainage station is closed and drainage is carried out through the sluice gate. The operating flow rate of the drainage station is 120 m³ / h. 3 The water pumps are 3 units per second, with an average design flow rate of 40m³ / s. 3 / s.

[0119] A river hydrodynamic model based on SWMM is constructed to simulate the flood evolution process. The inflow is input into the joint flood control optimization scheduling model of the pump and gate group. The initial water level of the pump and gate station is set, and the joint flood control optimization scheduling model of the pump and gate group is solved within the control time domain to obtain the optimal control sequence of the initial reservoir flow. However, in the current time period t, only the instructions of the first control sequence are executed. At the same time, the final water level at time t is obtained, which is the initial water level at time t+1. Let t = t+1, and update the initial value. Repeat the above process until the optimization of the entire time axis is completed.

[0120] Rainfall settings: Based on the rainstorm intensity formula, the typical design rainfall for the above-mentioned area once every 100 years was generated using the Chicago rain pattern, with a total rainfall of 251.18 mm and an average rainfall intensity of i = 0.1744 mm / min.

[0121] Tide Level Setting: Since the opening and closing of the pumping stations and sluice gates are affected by the downstream tide level during actual operation, the tide level data selected in this paper were obtained from the tide level monitoring stations in the aforementioned areas. The tide level process curve is shown in the figure below. Figure 3 As shown.

[0122] Initial water level setting: This paper assumes that the initial water level of the first-stage pump gate is 4.15m and the initial water level of the second-stage pump gate is 4.02m.

[0123] The optimized scheduling is based on a model predictive control-based flood control scheduling method for pump station groups.

[0124] The rule is that when the upstream primary pumping station in this area reaches a river level of more than 5 meters during the flood season, the pumping station will be activated, while the secondary pumping station will maintain a level of around 4.2 meters.

[0125] This article studies the scheduling diagrams of first and second-level pumping stations for a 100-year design flood, as shown in the figure. Figure 4 , Figure 5 As shown in Tables 1 and 2, the opening and closing statistics of primary and secondary pumping stations under the joint flood control scheduling of the sluice gate group based on model predictive control are presented under the design flood of a 100-year return period. Table 3 shows the gate opening and discharge volume of the secondary pumping station under the optimized scheduling of the joint flood control of the pumping and sluice gate group based on model predictive control under the design flood of a 100-year return period.

[0126] Table 1. Start-up statistics of pumping stations and gate groups at various times based on model predictive control for optimized flood control scheduling.

[0127]

[0128] Table 2. Pump and gate group operation statistics at various times during the rule-based scheduling.

[0129]

[0130] Table 3. Downstream gate opening and discharge volume under model-predictive control-based optimized flood control scheduling of pump and gate groups under a 100-year design flood.

[0131]

[0132] As shown in the above charts, the peak flow of the upstream primary pumping station under a 100-year design flood is 147.33 m³ / h. 3 The peak flow rate of the downstream secondary pumping station is 962.08 m³ / s. 3 Both / s and s arrive in the 16th time period, coinciding with the start of high tide downstream. Figure 4 It can be seen that during the first three time periods, due to the low water inflow, it was not necessary to activate the pumps. The water level rose to 4.39m. As time progressed and rolling optimization was implemented, before the flood peak arrived at the upstream primary pumping station, the pump units at the upstream primary pumping station were activated to pre-discharge and release water, freeing up maximum reservoir capacity for the flood peak and lowering the water level at the upstream primary pumping station to 0m. When the flood peak arrived, the pumping station units were activated, ultimately controlling the highest water level at the upstream primary pumping station within the scheduling cycle to 6.15m. In contrast, under the design flood of a 100-year return period, the water level at the upstream primary pumping station had risen to 7.11m during the scheduling period through rule-based scheduling. The flood control optimization scheduling of the pumping station group based on model predictive control reduced the highest water level at the upstream primary pumping station by 13.5% compared to rule-based scheduling.

[0133] Before the flood peak arrived, the downstream secondary pumping station maintained the water level at around 4.1m by adjusting the gate opening. Due to the influence of the Minjiang River tide, only one of the drainage station gates or pumping unit could be used to keep the water level at the downstream secondary pumping station below the warning level. Figure 5 It can be seen that when the flood peak arrived at the downstream secondary pumping station, it coincided with a rise in tide level, necessitating the use of pumping station units for flood discharge. The highest water level at the downstream secondary pumping station during the scheduling cycle was 5.621m, while the peak water level at the downstream secondary pumping station under the rule-based scheduling was 5.732m. The model-based predictive control-based joint flood control optimization scheduling of the pumping station group reduced the highest water level at the downstream secondary pumping station by 1.93% compared to the rule-based scheduling.

[0134] Through observation Figure 5 It can be clearly observed that the downstream secondary pumping station with rule-based scheduling exhibits greater fluctuations in its final water level, while the downstream secondary pumping station with model-based predictive control and joint flood control optimization scheduling of the Jin'an River pumping station group shows more stable final water levels.

[0135] The statistics in Tables 1 and 2 show that the model-predictive control-based joint flood control optimization scheduling scheme involved 43 pump station starts and stops, while the rule-based scheduling scheme involved 46. The number of pump station starts and stops under the model-predictive control-based scheme was 7% lower than that under the rule-based scheme. The total pumping cost under the model-predictive control-based scheme was 23,600 yuan, while the total cost under the rule-based scheme was 24,800 yuan. The total cost under the model-predictive control-based scheme was 4.8% lower than that under the rule-based scheme.

[0136] Table 3 shows the gate opening and discharge volume of the downstream secondary pumping and sluice gates under the model-based predictive control (MMC) joint flood control optimization scheduling of the pumping and sluice gate group under a 100-year design flood. The gate opening is based on the discharge volume of the downstream secondary pumping and sluice gates optimized by the MMC joint flood control optimization scheduling of the Jin'an River. The gate opening of the downstream secondary pumping and sluice gates was repeatedly adjusted using the SWMM model and the river hydrodynamic model. When the adjusted discharge volume is consistent, the gate opening at this time is taken as the gate opening of the downstream secondary pumping and sluice gates under the MMC joint flood control optimization scheduling scheme.

[0137] The results show that the real-time dispatching process of a certain power station has a significant optimization effect, the water level process is more stable, the operating cost is significantly reduced, and most importantly, it provides scientific guidance for the real-time dispatching process of a certain power station.

[0138] In one specific embodiment, a pump and gate group flood control scheduling system based on model predictive control is disclosed, including the following modules:

[0139] The scheduling strategy generation module constructs an optimized scheduling model for pump and gate groups based on the drainage capacity, operating costs, and subsequent maintenance of pump and gate drainage stations.

[0140] The scheduling optimization problem-solving module uses an artificial bee colony algorithm that mimics the bee colony's collection of bee sources to solve the combined scheme of design flood, pump station opening and closing, and gate opening.

[0141] The intelligent optimization allocation module uses the artificial bee colony algorithm for intelligent optimization allocation. It inputs the results of the established pump and gate group optimization scheduling model into the constructed urban water system river hydraulic model based on SWMM within the scheduling cycle. By configuring the opening degree of urban pump and gate drainage stations, the flood control scheduling model is finally obtained.

[0142] The real-time scheduling strategy generation module uses the artificial bee colony algorithm to solve the established flood control scheduling model, which can generate a real-time scheduling strategy under rainfall.

[0143] Furthermore, the scheduling strategy generation module also includes a scheduling target generation module, which is used to generate scheduling targets such as optimizing the target water level, minimizing pumping costs during operation, minimizing the number of pump station start-ups and shutdowns, and minimizing the peak water level at the end of the scheduling period.

[0144] Furthermore, the scheduling strategy generation module also includes a constraint generation module, which is used to generate water level constraints, flow constraints, pump head constraints, gate opening constraints, and non-negative constraints.

[0145] Furthermore, the scheduling optimization problem-solving module also includes an artificial bee colony algorithm module, which is used to find the optimal solution to the joint scheduling optimization problem of pump and gate groups.

[0146] Furthermore, the intelligent optimization allocation module also includes a flood control scheduling model acquisition module, which is used to acquire the flood control scheduling model.

[0147] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0148] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for flood control scheduling of pumping stations and sluice gates based on model predictive control, characterized in that, include: Step 1: Construct an optimized scheduling model for the pumping and sluice gate group based on the drainage capacity, operating costs, and subsequent maintenance of the pumping and sluice gate drainage stations. The optimized scheduling model for the pumping and sluice gate group consists of scheduling objectives and constraints. Step 2: Solve the scheduling optimization problem by using the artificial bee colony algorithm, which mimics the bee colony's collection of bee sources, to obtain the combination scheme of pump station opening and closing and gate opening under the design flood. Step 3: Set up the allocation method for pumping stations and gates, use the artificial bee colony algorithm for intelligent optimization allocation, and input the results of the pump and gate group optimization scheduling model established in Step 1 and Step 2 into the urban water system river hydraulic model built based on SWMM within the scheduling cycle. By configuring the opening degree of urban pumping stations and drainage stations, the flood control scheduling model is finally obtained. Step 4: Real-time scheduling strategy for pumping and drainage stations based on artificial bee colony algorithm. By using artificial bee colony algorithm to solve the established flood control scheduling model, a real-time scheduling strategy scheme under rainfall can be obtained. The optimal scheduling model for the pump and gate group includes the following scheduling objectives: (1-1) Target water level optimization: ； In the formula: This represents the indicator for optimizing the target water level. For the set time period The optimal water level target. For time period Water level, For this scheduling period For the total scheduling period, For the set number Pumping stations and drainage stations The total number of main pumping stations and drainage stations; (1-2) Minimal pumping costs during operation: ； In the formula: The cost of pumping out water during the scheduling period. Electricity unit / kW h, Let g be the density of water, and g be the local gravitational acceleration, taken as 9.8 m / s². 2 , For pumping stations Pump flow rate during the time period For pumping stations The lift distance during the period, The duration of a unit of time; (1-3) The pump station has the fewest opening and closing times: ； In the formula: C is the sum of the number of times the pump station units start and stop during the scheduling period; For the first The first pumping station The number of pumping station units changed between the previous time period and the current time period; (1-4) The water level peak was lowest at the end of the scheduling period: ; In the formula: The index for optimizing the peak water level at the end of the scheduling cycle of the pumping station. for Time of the first Final water level at the pumping station; Total scheduling time period; The artificial bee colony algorithm described in step two includes the following steps: Step 1: Initialize the artificial bee colony algorithm parameters, including the maximum number of iterations (maxcycles) and the threshold. The parameters include: new nectar source Vi, nectar source Xi, probability Pi of selecting each nectar source, and parameters SN and t=1. Step 2: Generate the initial population X; Step 3: Hire peak i to search for honey sources and generate new honey sources Vi; Step 4: Determine if fit(Vi) is greater than fit(Xi). If it is, replace Xi with Vi; otherwise, retain the honey source Xi. Step 5: Determine if nectar source i is less than half of SN. If so, calculate the probability Pi of each nectar source selection, observe the bees' greedy selection based on the probability, update the population, and record the optimal solution. If not, increment the number of hired bees by 1 and return to step 3 to continue the loop. Step 6: Determine if the nectar source has reached the threshold limit. If it has, proceed to step 7. If it has not reached the threshold, generate new bees and proceed to step 7. Step 7: Determine if the iteration count t has reached the maximum iteration count maxcycle. If it has, terminate the algorithm and output the optimal solution. If it has not reached the maximum iteration count, increment the iteration count by 1 and return to step 3 to continue the loop.

2. The method for flood control scheduling of pump and gate groups based on model predictive control according to claim 1, characterized in that, The optimal scheduling model for the pump and gate group includes the following constraints: (2-1) Water level constraint: ； This constraint is the water level constraint for the target pumping station; the water level is below [a certain value]. The target pumping station is at the dead water level, and the flood control limit water level is set at the target pumping station. As the highest water level control assessment target; (2-2) Flow constraints: ； In the formula: The first Upper and lower limits of the flow rate of the upstream pumping station and the flow rate of the downstream pumping station for each pumping station; (2-3) Pump head constraint: ； In the formula: This represents the upper and lower limits of the head of the upstream and downstream pumping stations for the nth pumping station; (2-4) Gate opening constraint: ； In the formula: This represents the minimum gate opening. This represents the maximum gate opening. (2-5) Nonnegativity constraints: All of the above variables are non-negative.

3. The method for flood control scheduling of pump and gate groups based on model predictive control according to claim 1, characterized in that, The method for obtaining the flood control scheduling model described in step three includes the following steps: Step 1: Construct a river hydraulic model based on SWMM to provide forecasts of design flood inflow; Step 2: Construct a prediction model based on water balance according to state variables and control variables; Step 3: Initialize the inflow forecast results, the initial water level in front of the pumping station, and set the total time period to T, where t is time t, and also set the parameter k; Step 4: Perform an optimization based on the pump and gate group optimization scheduling model, solve the pump and gate group optimization scheduling model to obtain k control variable sequences, but only use the control variables of the first sequence, and let t=t+1; Step 5: Determine if time t is less than the total time period T. If so, update the initial value and return to step 3 to continue the loop until the optimization of the entire time axis is completed. Otherwise, output the flood control scheduling model.

4. A pump and gate group flood control scheduling system based on model predictive control, characterized in that, include The scheduling strategy generation module constructs an optimized scheduling model for pump and gate groups based on the drainage capacity, operating costs, and subsequent maintenance of pump and gate drainage stations. The scheduling optimization problem-solving module uses an artificial bee colony algorithm that mimics the bee colony's collection of bee sources to solve the combined scheme of pump station opening and closing and gate opening under the design flood. The intelligent optimization allocation module uses the artificial bee colony algorithm for intelligent optimization allocation. It inputs the results of the established pump and gate group optimization scheduling model into the urban water system and river hydraulic model based on SWMM within the scheduling cycle. By configuring the opening degree of urban pump and gate drainage stations, the flood control scheduling model is finally obtained. The real-time scheduling strategy generation module uses the artificial bee colony algorithm to solve the established flood control scheduling model, which can generate a real-time scheduling strategy under rainfall.

5. A pump and gate group flood control scheduling system based on model predictive control according to claim 4, characterized in that, The scheduling strategy generation module also includes a scheduling target generation module, which is used to generate scheduling targets such as target water level optimization, minimum pumping cost during operation, minimum number of pump station start-ups and shutdowns, and minimum peak water level at the end of the scheduling period. The scheduling optimization problem-solving module also includes an artificial bee colony algorithm module, which is used to solve the optimal solution of the joint scheduling optimization problem of pump and gate groups.

6. A pump and gate group flood control scheduling system based on model predictive control according to claim 4, characterized in that, The scheduling strategy generation module also includes a constraint generation module, which generates water level constraints, flow rate constraints, pump head constraints, gate opening constraints, and non-negative constraints.

7. A pump and gate group flood control scheduling system based on model predictive control according to claim 4, characterized in that, The intelligent optimization allocation module also includes a flood control scheduling model acquisition module, which is used to acquire the flood control scheduling model.

Images