Multi-objective optimal scheduling method for multi-channel water supply and distribution system

Abstract

The application discloses a multi-target optimization scheduling method for a multi-channel water supply and distribution system, and proposes a two-layer coupling framework of "monitoring and prediction driven object screening and merging mechanism + water supply order optimization + double-target gate scheduling": the outer layer obtains future water consumption based on real-time monitoring daily water intake and historical law prediction, calculates "tension degree" and "merging index", determines a single water supply object set and optimizes the water supply order; the inner layer calls an existing water supply module to simulate water head propagation under the candidate order, solves the discharge flow sequence of each channel gate by using a DP-POA or equivalent algorithm, and then reversely calculates the gate opening area and opening and closing time. The merging water supply mechanism and the double targets (minimum total water consumption and minimum total time) are unified into the same optimization framework, so that the repeated water lifting and the fixed loss of single water supply are reduced. Meanwhile, the water supply order of single water supply is optimized by the model, the urgency of water users and the structural water lifting cost of the system are considered, and the total water intake and the completion time are reduced.

Description

Technical Field

[0001] This invention belongs to the field of water resource optimization and water conservancy project scheduling technology, specifically relating to a multi-objective optimization scheduling method for a multi-channel water transmission and distribution system. Background Technology

[0002] Long-distance open water conveyance channels are often divided into several series of channels by multiple gates, distributing water to multiple users along the route. Since water supply must meet certain head and level conditions, supplying water to downstream users often requires raising the water levels of several upstream channels first, and the tailrace must be consumed after supply to reduce losses and risks. Existing technologies are mostly based on experience or single-objective optimization, making it difficult to simultaneously balance water consumption and supply efficiency (time). Furthermore, water supply is typically only initiated when water shortages occur, leading to an excessive number of water supply cycles per year, frequent repeated water lifting, and accumulated fixed losses from each water supply. While real-time monitoring capabilities have improved, allowing for the acquisition of daily water intake, reservoir capacity, and emergency water levels for each user, a universal framework for system optimization that integrates "real-time monitoring + historical pattern prediction + threshold-based combined water supply" is lacking. Summary of the Invention

[0003] To address the aforementioned problems, embodiments of this invention propose a multi-objective optimization scheduling method for multi-channel water transmission and distribution systems. A two-layer coupled framework is proposed: a monitoring and prediction-driven object selection and merging mechanism + water supply sequence optimization + dual-objective gate scheduling. The outer layer predicts future water consumption based on real-time monitoring of daily water intake and historical patterns, calculates the "stress level θ" and "merging index η," and determines the set of objects for a single water supply. It also optimizes the water supply sequence; the inner layer calls the existing water conveyance module to simulate water head propagation under the candidate sequence, and uses DP-POA or equivalent algorithm to solve the gate discharge flow sequence of each channel section, and then reverses the gate opening area and opening and closing time.

[0004] The multi-objective optimization scheduling method for multi-segment water conveyance and distribution systems of the present invention is applicable to water conveyance trunk canal systems consisting of n series-connected canal segments, wherein each canal segment i corresponds to at least one water user. The method includes the following steps:

[0005] S1. Obtain system topology and parameters: Obtain the series relationship of n canal segments, the control gate information of each canal segment, and the water level-storage function F of each canal segment. i (Z i ), and the water demand information of each water user j and the corresponding canal segment i(j);

[0006] S2. Real-time monitoring and water consumption forecasting: Real-time monitoring of each water user's daily water intake / consumption (d) j (d) and combined with historical patterns, the water consumption within the future prediction window (L) is predicted to obtain the predicted water consumption. ;

[0007] S3. Calculate the degree of water shortage;

[0008] S4. Calculate the combined water supply indicators;

[0009] S5. Determine the set and order of objects for a single water supply;

[0010] S6. Construct a multi-objective optimization model;

[0011] S7. Call the water conveyance module and solve the problem;

[0012] S8. Multiple rounds of optimization throughout the year, repeating S2 to S7 within the annual planning cycle, and merging water supply index thresholds η. merg The triggered water supply consolidation mechanism will include water users who are not short of water but are close to the threshold into the current water supply cycle, in order to reduce the number of water supply cycles within the year.

[0013] The predicted water consumption It is the daily water intake monitoring sequence d j (d) The water consumption within the future prediction window is predicted by combining historical pattern models, which include one or a combination of seasonal segmented mean, similar day matching, moving average, or time series models.

[0014] The water scarcity level is calculated based on the current reservoir capacity, critical reservoir capacity, and water consumption during the propagation time of the water user's water storage unit. The formula for calculating the water scarcity level is:

[0015]

[0016] In the formula, θ j To determine the degree of water scarcity, C j (T) represents the water consumption of the water storage unit to which user j belongs during the propagation time T, resW j Based on the current reservoir capacity of the water storage unit to which water user j belongs, resW emg, j Based on the emergency storage capacity of the water storage unit to which water user j belongs, τ j For the equivalent propagation time, t0 is the scheduling starting point.

[0017] The combined water supply index is calculated based on the water consumption within the predicted window, and the calculation formula is as follows:

[0018]

[0019] In the formula, η j In order to merge water supply quotas, To predict water consumption within the window, Let L be the daily water consumption of the j-th water user, predicted by combining real-time monitoring of daily water intake with historical patterns. Let L be the prediction window. jThis is the prediction window for the j-th water user.

[0020] The set of objects requiring single water supply includes the set of mandatory water supply and the set of combined water supply, as shown in the formula:

[0021]

[0022] In the formula, Let j be the set of objects supplied with water in a single instance. urg For the necessary water supply collection, j merg To merge the water supply sets, θ tres η is the threshold for water scarcity. merg To merge water supply indicator thresholds.

[0023] The multi-objective optimization model uses the source water diversion flow, the discharge flow of each canal section, the water intake flow of each water user, and the water supply order variables as decision variables. Under the conditions of water balance, water level-storage function, storage before supply, tailwater consumption, gate overflow constraint, and water head propagation constraint, a multi-objective optimization model is established with the objectives of minimizing the total water supply and water consumption and minimizing the total water supply time.

[0024] The multi-objective optimization model includes the optimal candidate water supply order. and optimal gate water level trajectory ,in The calculation formula is:

[0025]

[0026] The calculation formula is:

[0027]

[0028] In the formula, π represents the water supply sequence, Z represents the gate trajectory, and F... W It is the water consumption during the water supply process, F T It is the water supply time, λ1 is F W The weights, λ2 is F T The weight.

[0029] The F W The calculation formula is:

[0030]

[0031] F T The calculation formula is:

[0032]

[0033] In the formula, Q Y(t) represents the water diversion flow at the source gate or source boundary during time period t, Δt represents the time step, and β i The penalty coefficient is between 0 and 1. Let represent the water storage capacity of the i-th canal segment at time t. This represents the minimum allowable water storage capacity for the i-th canal segment. The penalty for tailwater / remaining water storage is calculated using the following formula: , This represents the total time required to complete one water transmission and distribution process, determined by the maximum water transmission time for all water users in this round.

[0034] The formula for calculating the water diversion flow rate of the source gate or source boundary in time period t is as follows:

[0035]

[0036] In the formula, q i (t) represents the water intake flow rate of user j in time period t (the flow rate supplied to the water storage unit / tributary), L i (t) represents the flow loss of channel segment i in time period t: leakage, evaporation, water wastage, management losses, etc. (calculated according to the empirical coefficients or models available in the project), △t represents the time step, and t+△t represents the next time period. The last term on the right side of the equation represents: the "net water storage change rate" of channel segment i (converted to flow rate).

[0037] The aforementioned The calculation formula is:

[0038]

[0039] In the formula, The formula for calculating the water delivery time for the j-th water user is as follows:

[0040]

[0041] In the formula, Indicates the propagation time of the canal section. This represents the time it takes for the k-th canal segment to reach the water supply threshold level, expressed as a function of canal segment water level – storage capacity (S). i =F i (H i The inflow rate of the canal section is calculated by back-calculation. This indicates the time it takes for the user to fill the water tank themselves.

[0042] The beneficial effects of this invention are: (1) Optimizing the order of single water supply. Based on real-time monitoring and water use prediction data of each water user, the urgency of water use is proposed, and the water supply order of each water user in each canal section is initially arranged; a combined water supply threshold is proposed, and water users that reach the threshold are merged into the current round, thus optimizing the order of single water supply; (2) Reducing the number of water supply rounds within the year. Through the optimization model, the fixed losses of repeated water filling and single water supply in the canal section are reduced, and the multi-objective optimization of "minimum total water consumption" and "minimum total water supply time" is achieved, thereby reducing the number of water supply rounds within the year, and an executable gate scheduling command is output. (3) Unifying the combined water supply mechanism and the dual objectives (minimum total water consumption and minimum total time) into the same optimization framework, reducing repeated water lifting and fixed losses of single water supply. At the same time, the water supply order of single water supply is optimized by the model, taking into account the urgency of water users and the structural water lifting cost of the system, thus achieving a lower total water diversion volume and a shorter completion time. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the open channel water diversion system of the present invention.

[0044] Figure 2 This is a schematic diagram illustrating the relationship between canal water storage and reservoir filling in this invention.

[0045] Figure 3 This is a flowchart of the optimization calculation process for the long-distance water conveyance channel water distribution process of the present invention. Detailed Implementation

[0046] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0047] 1. For example Figures 1-3 As shown, the multi-objective optimization scheduling method for multi-segment water conveyance and distribution systems of the present invention is applicable to water conveyance trunk canal systems consisting of n series-connected canal segments, wherein each canal segment i corresponds to at least one water user. The method includes the following steps:

[0048] S1. Obtain system topology and parameters: Obtain the series relationship of n canal segments, the control gate information of each canal segment, and the water level-storage function F of each canal segment. i (Z i ), and the water demand information of each water user j and the corresponding canal segment i(j).

[0049] S2. Real-time monitoring and water consumption forecasting: Real-time monitoring of each water user's daily water intake / consumption (d) j (d) and combined with historical patterns, the water consumption within the future prediction window (L) is predicted to obtain the predicted water consumption. .

[0050] Historical pattern models include one or a combination of seasonal segmented mean, similar day matching, moving average, or time series models.

[0051] Predicted water consumption C during propagation time j Based on equivalent propagation time τ j calculate, or .

[0052] S3. Calculate the water shortage level.

[0053] The degree of water scarcity is calculated based on the current reservoir capacity, critical reservoir capacity, and water consumption during the propagation time of the water user's water storage unit. The formula for calculating the degree of water scarcity is:

[0054] (1)

[0055] In the formula, θ j To determine the degree of water scarcity, C j (T) represents the water consumption of the water storage unit to which user j belongs during the propagation time T, resW j Based on the current reservoir capacity of the water storage unit to which water user j belongs, resW emg, j Based on the emergency storage capacity of the water storage unit to which water user j belongs, τ j For the equivalent propagation time, t0 is the scheduling starting point.

[0056] S4. Calculate the combined water supply indicators.

[0057] The combined water supply index is calculated based on the water consumption within the forecast window, and the calculation formula is as follows:

[0058] (2)

[0059] In the formula, η j In order to merge water supply quotas, To predict water consumption within the window, Let L be the daily water consumption of the j-th water user, predicted by combining real-time monitoring of daily water intake with historical patterns. Let L be the prediction window. j This is the prediction window for the j-th water user.

[0060] S5. Determine the set and order of objects for a single water supply.

[0061] According to θ j With water stress threshold θ tres Determine the necessary water supply collection According to η j Combined with the water supply index threshold η merg Determine the merged water supply group ; Form a set of objects for a single water supply The system generates an initial water supply sequence based on the urgency of the situation.

[0062] S6. Construct a multi-objective optimization model.

[0063] The multi-objective optimization model is based on the source water diversion flow rate Q. Y (t), discharge flow rate q in each channel section i (t), water intake flow rate q of each water user j Using (t) and water supply order variables as decision variables, a multi-objective optimization model is established with the objectives of minimizing total water consumption and minimizing total water supply time, under the conditions of water balance, water level-storage function, storage before supply, tailwater consumption, gate overflow constraint, and water head propagation constraint.

[0064] The constraint of "storage before supply" is: when water is supplied to water user j at any time, the water level of the channel segment i(j) to which the water user belongs and all upstream channel segments k ≤ i(j) shall not be lower than the water supply level threshold H. req (k, j).

[0065] The tailwater consumption constraint is: after the last water user in channel segment i completes the water supply in this round, the water storage S in channel segment i will be... i Reduced to the lower operating limit S i, min Nearby, or S i With S i, min The deviation is used as a penalty term in the objective function.

[0066] The multi-objective optimization outputs a Pareto optimal solution set, and a compromise solution is selected according to management preferences. The multi-objective algorithm includes NSGA-II, NSGA-III, or a multi-objective mixed integer programming solver.

[0067] S7. Call the water delivery module and solve the problem.

[0068] In the optimization iteration, the existing water conveyance hydraulic module is called to simulate the water head propagation and update the water level of the channel section. The optimal water supply sequence, source water diversion process, and opening area and opening and closing time of each control gate are obtained for a single water supply.

[0069] S8. Multiple rounds of optimization throughout the year, repeating S2 to S7 within the annual planning cycle, and through η merg The triggered water supply consolidation mechanism will include water users who are not short of water but are close to the threshold into the current water supply cycle, in order to reduce the number of water supply cycles within the year.

[0070] 2. Candidate generation and evaluation selection of outer layer water supply order π (enumeration / search)

[0071] To achieve combined optimization of the water supply sequence π and the gate water level trajectory Z, this invention employs a two-layer solution strategy: "outer layer enumeration / search of π + inner layer DP-POA solution of Z". The outer layer is responsible for generating the candidate water supply sequence set Π and selecting the optimal candidate water supply sequence. The inner layer solves for the optimal gate water level trajectory under a given π condition. And calculate the objective function value accordingly.

[0072] (1) Determining the set of water supply targets: based on the threshold of water shortage level. With merging threshold Determine the set of water supply recipients for this round. .

[0073] The set of objects requiring a single water supply includes the set of objects requiring water supply and the set of objects requiring combined water supply, as shown in the formula:

[0074] (3)

[0075] In the formula, Let j be the set of objects supplied with water in a single instance. urg For the necessary water supply collection, j merg To merge the water supply sets, θ tres η is the threshold for water scarcity. merg To merge water supply indicator thresholds.

[0076] (2) Generation of candidate order set Π: Select one or a combination of the following methods based on the system size:

[0077] Small-scale full enumeration: when | When the value is relatively small (e.g., ≤7~8), it can be used for Perform a full permutation enumeration, where Π contains all permutations. All sequences;

[0078] Large-scale heuristic search: when | When | is large, Π is generated from the neighborhood of the initial order π0 (by swapping, inserting, inversion, etc.), or a finite candidate set is generated by using genetic algorithms / simulated annealing / tacit search, etc.

[0079] Group pruning: First, prune according to the canal section number i(j) or the propagation time τ. j Group the water users, first enumerate the "group order", then sort them by θ within each group. j Sorting or small-scale perturbations can significantly reduce the number of candidates and decrease repeated water lifting.

[0080] The initial sequence π0 can be determined according to the level of tension θ. j Generate in descending order and use the merging index η j and propagation time τ jAs a criterion .

[0081] (3) Inner evaluation of candidate π: For any candidate π∈Π, generate a staged water supply switch y according to π. j (t) and water intake process q j (t), and call the inner DP-POA to solve. Subsequently, the objective function value (F) under π is obtained through the water conveyance module Φ (the existing hydrodynamic model). W , F T ).

[0082] Only when y j Only when (t)=1 is it strictly required that the upstream water level reach the water supply threshold.

[0083] in The calculation formula is:

[0084] (4)

[0085] In the formula, F W It is the water consumption during the water supply process, F T It refers to the time it takes to supply water.

[0086] (4) Outer layer selection rules:

[0087] Output the Pareto non-dominated order set for the candidate set Π, and select the final order based on management preferences (water conservation priority or efficiency priority). .

[0088] The calculation formula is:

[0089] (5)

[0090] In the formula, π represents the water supply sequence, Z represents the gate trajectory, and F... W It is the water consumption during the water supply process, F T It is the water supply time, λ1 is F W The weights, λ2 is F T The weight.

[0091] By coupling the outer enumeration / search with the inner DP-POA solution, the combined optimization of water supply sequence π and gate trajectory Z can be achieved within the calculable scale of the project, and the number of rounds can be reduced in the annual rolling decision-making.

[0092] 3. DP-POA Optimization Process (Single Water Supply Optimization)

[0093] This application employs a coupled dynamic programming (DP) and stepwise optimization (POA) approach to solve the Z component of u (u=(π,Z), i.e., outer layer optimization / enumeration of π; inner layer DP-POA to calculate Z), in order to reduce the computational complexity of multi-gate joint control. The coupling process between DP-POA and the water conveyance module Φ is as follows:

[0094] Step 1, water level discretization. For each channel segment i (at the downstream control gate), in K water points were obtained by internal discretization. , where H i,min H represents the lower limit of the permissible water level before the impact in section i. i,max Let represent the upper limit of the pre-sluice water level allowed for channel segment i, and k be the number of discrete points of the downstream gate corresponding to channel segment i. For the optimization round of POA, Indicates the first The k discrete points of the i-th gate during round-robin optimization.

[0095] Step 2, DP single-gate optimization. This is a phased state; the next phase water level will be... For phased decision-making; given the inflow and water intake conditions, the feasible q can be derived from equation (6). i (t), if the flow / water level / gate passage constraints are not met, it is judged as infeasible.

[0096] (6)

[0097] In the formula, Z i (t) represents the water level in front of the gate of channel segment i during time period t. Indicates the first The k discrete points of the i-th gate during round-robin optimization. q i (t) represents the water intake flow rate of user j in time period t, L i Q(t) represents the flow loss of channel segment i in time period t. i,in (t) represents the inflow entering channel segment i (upstream discharge and lateral inflow, etc.), Δt represents the time step, t+Δt represents the next time period, and S i (t) represents the water storage volume of the i-th canal segment at time t.

[0098] Step 3, recursion and iteration. The target is scalarized. Transform into stage variable g i,t A dynamic programming recursion is established to backtrack and obtain the initial feasible trajectory for a single gate. .

[0099] (7)

[0100] In the formula, Let represent the minimum residual cost from time period t to the endpoint. express The cost within a calculation step (including the cost of water supply and the cost of time) is calculated using the following equation:

[0101] (8)

[0102] In the formula, This represents the state Z during time interval t in a DP search. i (t) to the next time t+Δt state Z i (t+Δt), the immediate cost / stage cost generated, S i (t) represents the water storage capacity of the i-th canal segment at time t, and λ1 is the water storage capacity of F. w The weights, λ2 is F T The weight.

[0103] Indicates this time period Water diversion volume within, β represents the penalty value for high water level in section i of the canal during this time period. i This is the penalty coefficient, ranging from 0 to 1.

[0104] Step 4, POA multi-gate coordination. (The rest of the text appears to be incomplete and requires further context.) } represents the initial trajectory. The sequence of gates (channel segments) i=1…n is repeated: The trajectories of all gates except the i-th gate are fixed, and steps 2-3 are repeated to update Z. i A single full gate update is recorded as one iteration.

[0105] Step 5, Φ Coupling Update. After each update of any gate trajectory, the water conveyance module Φ is called to update Q. i,in (t), H i (t) and the water head propagation process, and by F i (H i Update S i (t), ensuring that DP-POA always evaluates the target under consistent hydraulic conditions.

[0106] Q i,in (t) represents the inflow rate at the i-th upstream boundary (i.e., the outflow rate from the i-th gate); H i (t) represents the water depth in front of the i-th gate, F i (H i S represents the "water level ~ water storage" relationship function for the i-th gate; i (t) represents the water storage capacity of canal section i.

[0107] Step 6, Convergence and Output. When the improvement ΔJ of the objective after two iterations is less than the threshold ε or the iteration limit Rmax is reached, output... And obtained from equation (6) Then, the gate opening area / opening degree and opening and closing time are deduced from the gate flow rate formula.

[0108] 4. Multi-objective optimization model

[0109] (1) Objective function

[0110] (9)

[0111] (10)

[0112] In the formula, Q Y (t) represents the water diversion flow rate at the source gate or source boundary during time period t, in m³ / s, where Δt represents the time step, and β i The penalty coefficient is between 0 and 1. This represents the water storage capacity of the i-th canal segment at time t, in ten thousand m³. This represents the minimum allowable water storage capacity for the i-th canal segment. The penalty for tailwater / remaining water storage is calculated using the following formula: , The total time taken to complete one water transmission and distribution process is expressed in hours (h), and is determined by the maximum water transmission time taken by all water users in this round.

[0113] The water diversion flow rate Q at the source gate or source boundary during time period t Y The formula for calculating (t) is:

[0114] (11)

[0115] In the formula, q j (t) represents the water intake flow rate of user j within the canal section during time period t, L i (t) represents the loss, such as leakage or evaporation, △t represents the time step, and t+△t represents the next time period.

[0116] The The calculation formula is:

[0117] (12)

[0118] In the formula, The water delivery time (h) for the j-th water user is represented by the following formula:

[0119] (13)

[0120] In the formula, The propagation time in a channel segment can be calculated using one-dimensional hydrodynamic equations or the simple Chezy formula for open channels. This represents the time it takes for the k-th canal segment to reach the water supply threshold level, expressed as a function of canal segment water level – storage capacity (S). i =F i (H i The inflow rate of the canal section is calculated by back-calculation (where H represents the water depth). This indicates the time it takes for a water user (water storage unit) to "fully fill / meet its needs".

[0121] (2) Decision variables

[0122] The decision variable is the discrete point of water level at each control gate in each channel section at each time period, i.e., the trajectory of the discrete point of water level. Other process quantities (including source water diversion flow rate Q) Y (t), discharge flow rate q from each gate i (t), water intake flow rate q of each water user j (t), water supply switches, and their sequence relationships, etc., are all determined by Through DP-POA, water delivery module Φ, and the "water level ~ storage capacity" function S i (t) =F i (H i This was derived.

[0123] (3) Recurrence equation

[0124] (14)

[0125] in, It can be derived from the formula for the flow rate through the gate:

[0126] (15)

[0127] In the formula, Let represent the water storage capacity of the i-th canal section at time t+1, in ten thousand m³. It can be obtained from the "water storage capacity ~ water level" relationship curve of the i-th canal section, i.e., S i (t) = F i (Z i Q i,in (t) represents the inflow (upstream discharge and lateral inflow, etc.) entering channel segment i. i,out (t) represents the discharge flow rate of channel segment i (controlled by the gate). q j (t) represents the water intake flow rate of water user j within the canal section. i (t) represents loss (leakage, evaporation, etc.). These represent the comprehensive gate coefficient, unit width, and gate opening of the i-th gate, respectively. , These represent the water level before and after the i-th gate, respectively.

[0128] (4) Constraint equations

[0129] 1) Upper and lower limits of water level in the canal section:

[0130] (16)

[0131] In the formula, H i,min H represents the lower limit of the permissible water level before the impact in section i. i,max This indicates the permitted operating water level for channel segment i before impact.

[0132] 2) Water level constraints in the canal section:

[0133] (17)

[0134] In the formula, Let k be the water depth in section k of the canal. The threshold water level that canal section k must reach when supplying water to j, where M is a constant and takes a value greater than or equal to... and The difference, This is a water supply switch, used during the water supply period. j =1, otherwise y j =0, , The highest and lowest water depths of the canal sections are specified.

[0135] 3) Cumulative water supply constraints:

[0136] (18)

[0137] In the formula, This represents the cumulative water supply demand for user j in this round, in tens of thousands of m³. This represents the maximum water intake capacity (flow limit) for user j.

[0138] 4) Throughflow constraint:

[0139] (19)

[0140] In the formula, For the flow rate through the gate, This is the minimum allowable flow rate through the gate. This represents the maximum allowable flow rate through the gate. The result of the flow rate through the gate is shown in equation (13).

[0141] 5) Source water diversion flow constraints:

[0142] The water flow rate at the source must be less than the available water volume.

[0143] 6) Flow continuity constraint:

[0144]

[0145] In the formula, It can be a piecewise function, with the main constraint being the sum of the filling flow of the first channel segment and the filling flow of the corresponding reservoir within the channel segment; the sum of the filling flow of the second channel segment and the filling flow of the corresponding reservoir within the channel segment; and so on up to the sum of the filling flow of the nth channel segment and the filling flow of the corresponding reservoir within the channel segment.

[0146] 7) Nonnegativity constraint:

[0147] All variables are non-negative.

[0148] (5) Multiple rounds of reduction mechanism within the year

[0149] Monitoring and forecasts will be updated on a rolling basis throughout the year's planning period, and d will be updated accordingly. j (d) resW j And predict the future window L If there exists j satisfying θ j ≥ θ tres If a cycle is triggered, a single water supply is initiated; simultaneously, if there exists j satisfying η j ≥ η merg Even if there is no current water shortage, the water supply will be included in the current water supply cycle set to reduce repeated water lifting and fixed losses in future separate water supply. Through the "trigger-merge-tailwater consumption" closed loop, the number of water supply cycles per year can be reduced while ensuring safe water supply.

[0150] Although the above embodiments have been shown and described, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Any changes, modifications, substitutions and variations made to the above embodiments by those skilled in the art are within the protection scope of the present invention.

Claims

1. A multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system, characterized in that, Applicable to a water conveyance trunk canal system consisting of n series-connected canal segments, wherein each canal segment i corresponds to at least one water user, the method includes the following steps: S1. Obtain system topology and parameters: Obtain the series relationship of n canal segments, the control gate information of each canal segment, and the water level-storage function F of each canal segment. i (Z i ), and the water demand information of each water user j and the corresponding canal segment i(j); S2. Real-time monitoring and water consumption forecasting: Real-time monitoring of each water user's daily water intake / consumption (d) j (d) and combined with historical patterns, the water consumption within the future prediction window (L) is predicted to obtain the predicted water consumption. ; S3. Calculate the degree of water shortage; S4. Calculate the combined water supply indicators; S5. Determine the set and order of objects for a single water supply; S6. Construct a multi-objective optimization model; S7. Call the water conveyance module and solve; S8. Multiple rounds of optimization throughout the year, repeating S2 to S7 within the annual planning cycle, and merging water supply index thresholds η. merg The triggered water supply consolidation mechanism will include water users who are not short of water but are close to the threshold into the current water supply cycle, in order to reduce the number of water supply cycles within the year.

2. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 1, characterized in that, The predicted water consumption It is the daily water intake monitoring sequence d j (d) The water consumption within the future prediction window is predicted by combining historical pattern models, which include one or a combination of seasonal segmented mean, similar day matching, moving average, or time series models.

3. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 1, characterized in that, The water scarcity level is calculated based on the current storage capacity, emergency storage capacity, and water consumption during the propagation time of the water user's water storage unit. The formula for calculating the water scarcity level is as follows: In the formula, θ j To determine the degree of water scarcity, C j (T) represents the water consumption of the water storage unit to which user j belongs during the propagation time T, resW j Based on the current reservoir capacity of the water storage unit to which water user j belongs, resW emg, j Based on the emergency storage capacity of the water storage unit to which water user j belongs, τ j For the equivalent propagation time, t0 is the scheduling starting point.

4. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 3, characterized in that, The combined water supply index is calculated based on the water consumption within the predicted window, and the calculation formula is as follows: In the formula, η j In order to merge water supply quotas, To predict water consumption within the window, Let L be the daily water consumption of the j-th water user, predicted by combining real-time monitoring of daily water intake with historical patterns. Let L be the prediction window. j This is the prediction window for the j-th water user.

5. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 4, characterized in that, The set of objects requiring single water supply includes the set of mandatory water supply and the set of combined water supply, as shown in the formula: In the formula, Let j be the set of objects supplied with water in a single instance. urg For the necessary water supply collection, j merg To merge the water supply sets, θ tres η is the threshold for water scarcity. merg To merge water supply indicator thresholds.

6. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 1, characterized in that, The multi-objective optimization model uses the source water diversion flow, the discharge flow of each canal section, the water intake flow of each water user, and the water supply order variables as decision variables. Under the conditions of water balance, water level-storage function, storage before supply, tailwater consumption, gate overflow constraint, and water head propagation constraint, a multi-objective optimization model is established with the objectives of minimizing the total water supply and water consumption and minimizing the total water supply time.

7. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 1, characterized in that, The multi-objective optimization model includes the optimal candidate water supply order. and optimal gate water level trajectory ,in The calculation formula is: The calculation formula is: In the formula, π represents the water supply sequence, Z represents the gate trajectory, and F... W It is the water consumption during the water supply process, F T It is the water supply time, λ1 is F W The weights, λ2 is F T The weight.

8. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 7, characterized in that, The F W The calculation formula is: F T The calculation formula is: In the formula, Q Y (t) represents the water diversion flow at the source gate or source boundary during time period t, Δt represents the time step, and β i The penalty coefficient is... Let represent the water storage capacity of the i-th canal segment at time t. This represents the minimum allowable water storage capacity for the i-th canal segment. The penalty for tailwater / remaining water storage is calculated using the following formula: , This represents the total time required to complete one water transmission and distribution process, determined by the maximum water transmission time for all water users in this round.

9. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 8, characterized in that, The formula for calculating the water diversion flow rate of the source gate or source boundary in time period t is as follows: In the formula, q i (t) represents the water intake flow rate of user j in time period t, L i (t) represents the flow loss of channel segment i in time period t, △t represents the time step, and t+△t represents the next time period.

10. The multi-objective optimization scheduling method for a multi-channel water conveyance and distribution system according to claim 8, characterized in that, The aforementioned The calculation formula is: In the formula, The formula for calculating the water delivery time for the j-th water user is as follows: In the formula, Indicates the propagation time of the canal section. This indicates the time it takes for the k-th canal segment to reach the water supply threshold level. This indicates the time it takes for the user to fill the water tank themselves.

Images