Gate pump group collaborative scheduling optimization method and system based on digital twinning

Abstract

The application discloses a gate pump group collaborative scheduling optimization method and system based on digital twinning, and relates to the technical field of water conservancy engineering scheduling. The method acquires canal system topology, channel water power, control nodes, gates, pump stations, equipment available states, target water supply tasks and real-time operation data, constructs and corrects a gate pump group digital twinning model; takes gate opening, opening and closing timing, pump set operation combination and pump station target flow as scheduling variables, constructs a regulation and control response relationship, generates and simulates a candidate collaborative scheduling scheme, optimizes the scheme step by step according to target priorities, obtains a target collaborative scheduling scheme and converts the scheme into control instructions, and updates the model according to real-time operation data after execution. Thus, the collaboration, dynamic adaptability, scheme executability and water supply control accuracy of gate pump group scheduling can be improved.

Description

Technical Field

[0001] This application relates to the field of water conservancy project scheduling technology, and in particular to a method and system for optimizing the coordinated scheduling of gate and pump groups based on digital twins. Background Technology

[0002] Gates and pumping stations are crucial control facilities in water conservancy projects, crucial for water conveyance, distribution, and water level regulation. For water conservancy projects with multiple channels, control nodes, and water-using areas, it is typically necessary to coordinate the operation times and statuses of multiple gates and pumping stations based on water source conditions, canal system operation status, and water demand in each area to ensure that water is delivered to the designated areas as planned. The rationality of gate and pump group scheduling directly affects water conveyance and distribution efficiency, canal operation safety, and the water supply security level of each water-using area.

[0003] Existing gate and pumping station scheduling typically relies on historical operating experience, pre-established scheduling rules, or phased water use plans to determine gate opening and closing and pumping station operation modes. In actual operation, water inflow, regional water demand, channel water levels, flow capacity, and equipment operating status can all change over time. Furthermore, different gates, pumping stations, and channels influence each other, and changes in the operating status of local equipment can cause continuous changes in upstream and downstream water levels and flow rates. Existing scheduling methods are relatively slow to reflect real-time operating conditions, have limited coordination among multiple scheduling objects, and make it difficult to fully assess the operational effectiveness of candidate scheduling schemes before implementation. When actual operating conditions deviate from predetermined conditions, problems such as untimely adjustments to scheduling schemes, mismatch between water supply and water demand, localized water level fluctuations, or low water transmission and distribution efficiency may occur.

[0004] Therefore, in the process of scheduling gate pump groups, the large number of scheduling objects, the mutual influence of equipment operation, the dynamic changes in operating status, and the lack of coordination, adaptability, and execution accuracy of scheduling schemes have become urgent problems to be solved. Summary of the Invention

[0005] This application provides a digital twin-based collaborative scheduling optimization method and system for gate pump groups, aiming to solve the problems in the existing gate pump group scheduling process, such as the large number of scheduling objects, mutual influence of equipment operation, dynamic changes in operating status, and insufficient coordination, adaptability, and execution accuracy of scheduling schemes.

[0006] Firstly, a method for optimizing the coordinated scheduling of gate pump groups based on digital twins, the method comprising: Acquire the canal topology data, canal hydraulic parameters, gate and pump equipment availability status, target water supply tasks and real-time operation data of the water transmission and distribution system, establish the mapping relationship between physical channels, control nodes, gates and pump stations and virtual objects, and establish the hydrodynamic coupling relationship between each virtual object based on the canal topology data and the canal hydraulic parameters to obtain a digital twin model of the gate and pump group. The real-time operating data is mapped to the digital twin model of the gate pump group. Based on the deviation between the simulated state and the measured state, the hydrodynamic parameters, dynamic boundary parameters and equipment characteristic parameters of the digital twin model of the gate pump group are corrected and the model state variables are updated to obtain the corrected digital twin model of the gate pump group and the current twin operating state. Based on the target water supply task, the current twin operation status, the availability status of the gate pump equipment, and the operation constraints, scheduling variables are determined and multiple initial scheduling variable combinations are generated; based on the hydrodynamic coupling relationship, a gate pump group control response relationship is constructed, and the gate pump group control response relationship is used to filter or modify each initial scheduling variable combination to obtain a set of candidate collaborative scheduling schemes; The corrected digital twin model of the gate pump group is used to simulate the water transmission and distribution process and evaluate the schemes for each candidate coordinated scheduling scheme, and the evaluation results of the candidate schemes are obtained. From the candidate collaborative scheduling schemes that meet the operational constraints, the scheduling scheme to be optimized is determined, and the target evaluation indicators are selected in order according to the preset target priority to perform single target optimization, so as to obtain the target collaborative scheduling scheme. The target collaborative scheduling scheme is converted into control commands for gates and pumping stations, and the digital twin model of the gate-pump group and the current twin operating status are updated based on the real-time operating data after execution.

[0007] Optionally, in the above scheme, establishing the mapping relationship between physical channels, control nodes, gates, and pumping stations and their corresponding virtual objects includes: Entity object identifiers are configured for the physical channels, control nodes, gates and pumping stations respectively, and entity object association data is obtained based on the spatial location, upstream and downstream connection relationship and control relationship of each entity object; Based on the entity object association data, virtual channel objects, virtual control node objects, virtual gate objects, and virtual pump station objects corresponding to each entity object are established, and each virtual object is connected according to the upstream and downstream connection relationship and control relationship to obtain a virtual object topology network. The online, offline, or fault status of each physical gate and physical pump station is obtained. Based on the obtained status, the corresponding virtual gate object and virtual pump station object in the virtual object topology network are enabled or disabled to obtain the currently available topology network. The monitoring data fields corresponding to each physical object are associated with the corresponding virtual object, and the control command fields corresponding to each physical gate and physical pumping station are associated with the corresponding virtual gate object and virtual pumping station object to obtain the virtual-physical data mapping relationship. Based on the currently available topology network and the virtual-real data mapping relationship, the mapping relationship between physical channels, control nodes, gates and pumping stations and their corresponding virtual objects is obtained.

[0008] Optionally, in the above scheme, establishing the hydrodynamic coupling relationship between virtual objects based on the canal system topology data and the canal hydraulic parameters includes: Based on the spatial location of the control nodes and gate pump equipment in the water transmission and distribution system, each channel is divided into multiple channel calculation segments to obtain the upstream and downstream connection relationships between each channel calculation segment; Configure the channel bottom elevation, cross-sectional geometric parameters, channel length and roughness parameters for each channel calculation section to obtain the hydraulic calculation parameters corresponding to each channel calculation section; Based on the upstream and downstream connection relationship and the hydraulic calculation parameters, the water volume continuity relationship and hydrodynamic transmission relationship of each channel calculation section are established to obtain the channel hydrodynamic calculation relationship. Based on the relationship between the upstream and downstream water levels of the gate, the gate opening and the flow rate through the gate, the gate boundary calculation relationship is determined, and based on the pumping capacity of the pumping station, the pump set operation combination, the head-flow relationship and the target flow rate of the pumping station, the pumping station boundary calculation relationship is determined. The gate boundary calculation relationship and the pump station boundary calculation relationship are configured in the corresponding control nodes, and combined with the dynamic boundary conditions corresponding to upstream reservoir water release, river pump station inflow, level crossing channel inflow, branch canal water diversion outflow and discharge outflow, the hydrodynamic coupling relationship between each virtual object is obtained.

[0009] Optionally, in the above scheme, the step of correcting the hydrodynamic parameters, dynamic boundary parameters, and equipment characteristic parameters of the gate pump group digital twin model, and updating the model state variables, includes: Data on channel water level, channel flow, gate operation, and pump station operation were acquired from multiple historical scheduling periods. The data were then divided into calibration and verification datasets. The operation process of the gate pumps corresponding to the calibration dataset is input into the digital twin model of the gate pump group to obtain the historical simulated water level, historical simulated flow rate and simulated gate pump output parameters at each monitoring location; Based on the water level deviation between the historical simulated water level and the historical measured water level, the flow deviation between the historical simulated flow rate and the historical measured flow rate, and the equipment output deviation between the simulated gate pump output parameters and the measured gate pump output parameters, the channel roughness parameters, dynamic boundary parameters, gate water level-flow rate relationship parameters, and pump station head-flow rate relationship parameters are adjusted to obtain the calibrated digital twin model of the gate pump group. The verification dataset is input into the calibrated gate pump group digital twin model to obtain the model verification error. When the model verification error is greater than the preset verification error threshold, the corresponding model parameters are readjusted until the model verification error is not greater than the preset verification error threshold, and the target gate pump group digital twin model is obtained. The real-time water level, real-time flow rate, real-time gate opening, real-time pump group operation combination, and real-time pump station flow rate at the current scheduling moment are mapped to the target gate-pump group digital twin model. The corresponding model state variables are updated to obtain the corrected gate-pump group digital twin model and the current twin operation status.

[0010] Optionally, in the above scheme, generating multiple combinations of initial scheduling variables includes: The target water supply task is analyzed to obtain the target water supply volume and target water supply sequence of each scheduling unit within the target scheduling period, and the target water supply process corresponding to each scheduling unit is determined based on the target water supply volume and the target water supply sequence. Based on the target water supply process, the canal topology between each scheduling unit, and the water loss parameters, the target water diversion process of the canal head during the target scheduling period is calculated in reverse. Based on the current twin operation status and the availability status of the gate pump equipment, the available water conveyance paths between the canal head and each dispatching unit are determined, and based on the target water intake process and the target water supply process, the target water conveyance process corresponding to each available water conveyance path is determined; Based on the target water conveyance process, the opening value range and opening and closing sequence range of each gate, as well as the available pump group operation combination and target flow range of each pumping station, are determined to obtain the feasible domain of scheduling variables. Based on the gate's maximum flow capacity, the gate's opening rate of change, the pump station's maximum water lifting capacity, the pump set's start-stop interval, the channel's flow capacity, and the control node's allowable water level range, the feasible domain of the scheduling variables is constrained, and multiple initial scheduling variable combinations are generated within the constrained feasible domain of the scheduling variables.

[0011] Optionally, in the above scheme, the construction of the gate pump group control response relationship and the screening or modification of each initial scheduling variable combination include: Preset continuous variable increments are applied to the gate opening and the target flow of the pumping station, and the gate opening and closing sequence and pump group operation combination are switched to adjacent feasible discrete states to obtain multiple scheduling variable test states. Using the corrected digital twin model of the gate pump group, the changes in water level, flow rate, and operating parameters of the associated gate pump equipment relative to the test state of each scheduling variable relative to the corresponding initial state are calculated to obtain the scheduling variable response data. Based on the correspondence between the changes in each scheduling variable and the changes in the water level, flow rate, and operating parameters of the associated gate pump equipment at the control node, a gate pump group control response matrix is ​​constructed to obtain the gate pump group control response relationship. By inputting the combination of initial scheduling variables into the gate pump group control response relationship, the control node water level, control node flow and gate pump equipment operating parameters corresponding to each combination of initial scheduling variables are predicted to obtain the initial state prediction results. Based on the initial state prediction results, combinations of initial scheduling variables that exceed the allowable water level range of the control node, the channel flow capacity, or the operating range of the gate pump equipment are filtered out. When the distance between the initial state prediction result and any constraint boundary is less than the preset safety margin, the corresponding initial scheduling variable combination is corrected according to the gate pump group control response matrix to obtain the candidate collaborative scheduling scheme set.

[0012] Optionally, in the above scheme, obtaining the evaluation results of the candidate scheme includes: According to the preset simulation time step, the gate opening degree, gate opening and closing status, pump group operation combination and pump station target flow corresponding to each candidate collaborative scheduling scheme are sequentially loaded into the corrected gate-pump group digital twin model to obtain the water level and flow calculation results and equipment operation parameter calculation results corresponding to each simulation time. Based on the hydrodynamic coupling relationship between each virtual object, the calculation results of water level and flow rate and the calculation results of equipment operating parameters are transmitted in time sequence to obtain the evolution process of water level and flow rate of control nodes, the water supply completion process of scheduling unit and the operation process of gate pump equipment corresponding to each candidate collaborative scheduling scheme. Based on the relationship between the cumulative simulated water supply and the target water supply of each scheduling unit, the water supply task completion rate of each candidate coordinated scheduling scheme is determined, and the water use efficiency of each scheduling unit is determined based on the water supply time, water wastage, cumulative simulated water supply and target water supply of each scheduling unit. The water transmission and distribution efficiency of each candidate coordinated dispatch scheme is determined based on the water intake at the head of the canal, the effective water intake of the dispatching unit, and the water discharge. The operating cost of each pumping station is determined based on its flow rate, head, operating time, and number of pump start-ups and shutdowns. The operating cost of each gate is determined based on the number of gate actions and the amount of gate opening adjustment. The operating costs of the pumping stations and the operating costs of the gates are then correlated to obtain the operating cost of the gate pump equipment. Based on the completion rate of the water supply task, the water use efficiency of each scheduling unit, the water transmission and distribution efficiency, the operating cost of gate pump equipment, and the water level satisfaction of control nodes, the evaluation results of the candidate schemes corresponding to each candidate collaborative scheduling scheme are formed.

[0013] In the above scheme, optionally, the target scheduling period is determined according to the target water supply task, the canal topology relationship between each scheduling unit is determined according to the canal topology data, and adjacent scheduling units that share upstream channels or have water volume competition relationships are identified according to the canal topology relationship and the water supply completion process of the scheduling unit, so as to obtain multiple local collaborative scheduling unit groups. For each local coordinated scheduling unit group, the local water supply effectiveness of each scheduling unit is determined based on the water supply time, water abandonment, target water supply, simulated water supply, and the water level and flow status of the corresponding channel. Using the gate flow rate and pumping flow rate corresponding to each scheduling unit as local scheduling strategies, the local water supply utility as the strategy benefit, and the minimum and maximum allowable water levels of the corresponding channels as local scheduling constraints, a local water volume game matrix corresponding to each local collaborative scheduling unit group is constructed. According to the time sequence within the target scheduling period, the local water quantity game matrix is ​​evolved and solved to obtain the local equilibrium water quantity allocation results corresponding to each local collaborative scheduling unit group. Based on the local equilibrium water allocation results, the gate flow rate, gate opening and closing sequence, pump station water lifting flow rate, and pump group operation combination in the corresponding candidate coordinated scheduling scheme are adjusted to obtain the candidate coordinated scheduling scheme after local coordination.

[0014] Optionally, in the above scheme, the step of selecting target evaluation indicators sequentially according to preset target priority for single-target optimization, and updating the results based on real-time operating data after the control command is executed, includes: Based on the importance of the target water supply task, determine the priority order of the target tasks corresponding to the water supply task completion rate, the water level satisfaction of control nodes, water transmission and distribution efficiency, the operating cost of gate pump equipment, and water supply time. Select the target evaluation index corresponding to the current target priority, and use the allowable change range of each target evaluation index with higher than the current target priority as a constraint to adjust the scheduling variables associated with the current target evaluation index to obtain the current level scheduling scheme; The current level scheduling scheme is input into the corrected gate pump group digital twin model to simulate the water transmission and distribution process, and the corresponding scheduling variables are adjusted according to the simulation results until the current target evaluation index meets the corresponding preset termination condition to obtain the current level optimization result. The current level optimization result is used as the initial scheduling scheme for the next target priority. The single-target optimization of each target evaluation index is completed in sequence, and the current level optimization result corresponding to the last target priority is determined as the target collaborative scheduling scheme. After the target collaborative scheduling scheme is executed, the measured water level, measured flow rate, actual water supply and actual operating status of the gate pump equipment are obtained, and they are compared with the simulation results corresponding to the digital twin model of the gate pump group to obtain the scheduling execution deviation. Based on the scheduling execution deviation, the model parameters of the gate pump group digital twin model, the current twin operating state, and the initial range of scheduling variables corresponding to the next scheduling cycle are updated to obtain the initial conditions of the model for the next scheduling cycle.

[0015] Secondly, a gate pump group collaborative scheduling optimization system based on digital twins, the system comprising: The digital twin model construction module is used to acquire the canal topology data, channel hydraulic parameters, gate and pump equipment availability status, target water supply tasks and real-time operation data of the water transmission and distribution system, establish the mapping relationship between physical channels, control nodes, gates and pump stations and virtual objects, and establish the hydrodynamic coupling relationship between each virtual object based on the canal topology data and the channel hydraulic parameters to obtain the gate and pump group digital twin model. The twin state correction module is used to map the real-time operating data to the gate pump group digital twin model. Based on the deviation between the simulated state and the measured state, it corrects the hydrodynamic parameters, dynamic boundary parameters and equipment characteristic parameters of the gate pump group digital twin model and updates the model state variables to obtain the corrected gate pump group digital twin model and the current twin operating state. The candidate scheme generation module is used to determine scheduling variables and generate multiple initial scheduling variable combinations based on the target water supply task, the current twin operation status, the availability status of the gate pump equipment, and operation constraints; construct the gate pump group control response relationship based on the hydrodynamic coupling relationship, and use the gate pump group control response relationship to filter or modify each initial scheduling variable combination to obtain a set of candidate collaborative scheduling schemes; The simulation evaluation module is used to simulate the water transmission and distribution process and evaluate the schemes of each candidate coordinated scheduling scheme using the corrected digital twin model of the gate pump group, and obtain the evaluation results of the candidate schemes. The collaborative scheduling optimization module is used to determine the scheduling scheme to be optimized from the candidate collaborative scheduling schemes that meet the operational constraints, and select target evaluation indicators in order according to the preset target priority to perform single-target optimization and obtain the target collaborative scheduling scheme. The control feedback module is used to convert the target coordinated scheduling scheme into control commands for gates and pumping stations, and update the digital twin model of the gate-pump group and the current twin operating status based on the real-time operating data after execution.

[0016] Compared with the prior art, this application has at least the following beneficial effects: This application, based on further analysis and research of existing technical problems, recognizes the shortcomings of existing gate-pump group scheduling processes, including a large number of scheduling objects, mutual influence of equipment operation, dynamic changes in operating status, and insufficient coordination, adaptability, and execution accuracy of scheduling schemes. By acquiring canal topology data, channel hydraulic parameters, control node parameters, gate parameters, pump station parameters, gate-pump equipment availability status, target water supply tasks, and real-time operating data of the water transmission and distribution system, this application establishes mapping relationships and hydrodynamic coupling relationships between physical channels, control nodes, gates, and pump stations and their corresponding virtual objects. This allows the dispersed gate-pump equipment and their upstream and downstream hydraulic influences to be uniformly represented in a digital twin model of the gate-pump group. By mapping real-time operating data to the digital twin model of the gate-pump group and correcting model parameters and state variables based on the deviation between simulated and measured states, subsequent scheduling calculations can adapt to the dynamic changes in the operating boundaries and equipment performance of the water transmission and distribution system. By using gate opening degree, gate opening and closing sequence, pump group operation combination, and pump station target flow as collaborative scheduling variables, a gate-pump group control response mechanism is constructed. By screening or modifying the initial combination of scheduling variables, scheduling combinations that may cause abnormalities in water level, flow rate, or equipment operation can be eliminated or corrected in advance. Furthermore, a digital twin model is used to simulate the water supply and distribution process of candidate coordinated scheduling schemes. Constraint judgments and evaluations are made based on the evolution of water level and flow rate at control nodes, the water supply completion status of scheduling units, and the operation status of gate and pump equipment. Then, step-by-step optimization is implemented according to target priorities, ensuring that the final target coordinated scheduling scheme simultaneously considers the target water supply task, channel operation safety, water supply and distribution efficiency, and the operating costs of gate and pump equipment. Finally, the target coordinated scheduling scheme is converted into gate and pump station control commands, and the digital twin model and its operational status are updated based on real-time operational data after execution. This forms a closed loop of scheduling pre-simulation, optimization decision-making, control execution, and feedback correction, improving the coordination, dynamic adaptability, executability, and accuracy of gate and pump group scheduling. This addresses the problems of a large number of scheduling objects, mutual influence of equipment operation, dynamic changes in operational status, and insufficient coordination, adaptability, and execution accuracy of scheduling schemes. Attached Figure Description

[0017] Figure 1This is a flowchart illustrating a digital twin-based gate pump group collaborative scheduling optimization method provided in one embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] In one embodiment, such as Figure 1 As shown, a digital twin-based collaborative scheduling optimization method for gate and pump group is provided. It can be executed by an irrigation district scheduling platform, model calculation server, edge control device or cloud-edge collaborative computing platform. It is used to jointly optimize the operation of multiple gates and multiple pump stations according to the target water supply task and the real-time operating status of the water transmission and distribution system.

[0020] A water transmission and distribution system may include reservoirs, headworks, main canals, branch canals, tributary canals, control gates, diversion gates, spillway gates, culverts, riverside pumping stations, pumping stations, level crossings, dispatching units, and water level and flow monitoring equipment. A dispatching unit may be a spatial area where water supply is controlled by one or more channel control nodes.

[0021] Step S201: Obtain the canal topology data, canal hydraulic parameters, control node parameters, gate parameters, pump station parameters, gate and pump equipment availability status, target water supply task, and real-time operation data of the water supply and distribution system. Establish the mapping relationship between physical channels, control nodes, gates, and pump stations and their corresponding virtual objects. Based on the canal topology data and canal hydraulic parameters, establish the hydrodynamic coupling relationship between each virtual object to obtain a digital twin model of the gate and pump group.

[0022] Canal system topology data can include the spatial location of canals and hydraulic structures, the connections between canals, the connections between canals and control nodes, the upstream and downstream relationships of control nodes, the controlled objects of gates and pumping stations, the distribution of dispatching units, and the water supply routes. Canal system topology data can be determined based on irrigation district digital elevation data, river and canal network layout maps, gate and pump equipment layout maps, engineering design data, and field survey data.

[0023] Channel hydraulic parameters may include channel bottom elevation, channel length, cross-sectional shape, bottom width, side slope coefficient, cross-sectional area of ​​flow passage, free water surface width, hydraulic radius, longitudinal slope, and Manning roughness. Control node parameters may include control node location, node type, normal water level, minimum allowable water level, maximum allowable water level, and design flow capacity.

[0024] Gate parameters may include gate type, gate width, gate orifice height, upper and lower limits of gate opening, design flow rate, water level-flow rate curve, maximum opening rate of change, and allowable operating frequency. Pump station parameters may include pump station design water lifting capacity, number of pump sets, rated flow rate of a single pump, head-flow rate relationship, efficiency curve, rated power, minimum operating time of pump sets, minimum downtime of pump sets, and maximum number of start-stop cycles.

[0025] The availability status of gate and pumping equipment indicates whether each gate and pumping station can currently participate in scheduling, and can include online, offline, fault, and maintenance status. Gate and pumping equipment in offline, fault, or maintenance status will not participate in the generation of candidate schemes for the current scheduling cycle.

[0026] The target water supply task can be derived from water supply and demand forecasting models, water resource allocation models, or manual scheduling plans. The target water supply task includes the target water supply volume and target water supply sequence for each scheduling unit within the target scheduling period. For example, the target water supply task may include the total water demand of the i-th scheduling unit in the m-th ten-day period and the target water supply allocation process for each day and hour within that ten-day period.

[0027] Real-time operational data can include reservoir discharge flow, cross-river flow, dynamic boundary outflow, channel real-time water level, channel real-time flow, gate real-time opening degree, gate opening and closing status, pump set real-time operation combination, pump station real-time flow, pump station head, and pump station power.

[0028] The digital twin model of the gate and pump station group includes a virtual object topology network, a hydrodynamic calculation model, an equipment characteristic model, a status database, and a virtual-physical data interface. Monitoring data of the physical objects updates the status of the corresponding virtual objects through the virtual-physical data interface, and the target operating parameters output by the virtual objects can be converted into control commands for the physical gates and pump stations through the control interface.

[0029] Hydrodynamic coupling is used to describe the water level and flow transmission relationships between upstream channels, downstream channels, control nodes, gates, and pumping stations.

[0030] Step S202: Map the real-time operating data to the digital twin model of the gate pump group to obtain the simulated hydraulic state and simulated equipment state. Correct the model parameters and update the model state variables according to the deviation between the simulated state and the measured state to obtain the corrected digital twin model of the gate pump group and the current twin operating state.

[0031] Simulated hydraulic conditions can include simulated water levels at each control node, simulated flow rates at each channel calculation section, and simulated water intake at each dispatching unit. Simulated equipment conditions can include simulated gate opening, simulated gate flow rate, simulated pump unit operation combinations, simulated pump station flow rate, simulated head, and simulated power.

[0032] For deviations in channel water level or flow rate, the channel roughness, dynamic boundary flow rate, initial water level, or initial flow rate can be adjusted; for deviations in gate output, the gate flow coefficient or water level-flow rate relationship parameter can be adjusted; for deviations in pump station output, the pump station head-flow rate relationship parameter or pump group efficiency parameter can be adjusted.

[0033] The current twin operating status is used as the initial state for the simulation of candidate scheduling schemes, including the current water level of each control node, the flow rate of each channel, the opening degree of each gate, the operating status of each pump group, and the flow rate of each pumping station.

[0034] Step S203: Based on the target water supply task, the current twin operation status, the availability status of the gate pump equipment, and the equipment operation constraints, determine the scheduling variables and generate multiple combinations of initial scheduling variables.

[0035] The scheduling variables include gate opening degree, gate opening and closing sequence, pump unit operation combination, and pump station target flow rate. Gate opening degree and pump station target flow rate can be continuous scheduling variables, while gate opening and closing sequence and pump unit operation combination can be discrete scheduling variables.

[0036] Based on the target water supply task, the target water supply process for each dispatching unit is determined. Then, considering the water loss in the canal and the location of each dispatching unit, the target water intake process at the head of the canal is determined in reverse. Based on the target water intake process and the target water supply process, the target flow rate in each water supply path is determined, thereby forming the range of values ​​for the dispatching variables of each gate and pumping station.

[0037] Step S204: Construct the gate pump group control response relationship based on the hydrodynamic coupling relationship, and use the gate pump group control response relationship to screen or modify the initial scheduling variable combination to obtain a set of candidate collaborative scheduling schemes.

[0038] The gate-pump group control response relationship is used to characterize the impact of changes in scheduling variables on the water level, flow rate, and operating parameters of associated gate-pump equipment at the control nodes. The control response matrix of the gate-pump group can be formed by separately changing each scheduling variable and then calling the digital twin model of the gate-pump group to calculate the state changes.

[0039] The gate pump group control response matrix can be used to quickly predict whether the initial combination of scheduling variables may lead to water level exceeding limits, channel overflow, or equipment operating parameter exceeding limits. Combinations that clearly do not meet the constraints can be directly eliminated; combinations close to the constraint boundaries can be directionally corrected based on the response matrix.

[0040] Step S205: Input each candidate collaborative scheduling scheme into the corrected digital twin model of the gate pump group to simulate the water transmission and distribution process, and obtain the evolution process of water level and flow at the control node, the water supply completion process of the scheduling unit, and the operation process of the gate pump equipment.

[0041] The time step for the water conveyance and distribution process simulation can be 1 minute, and the spatial calculation resolution can be 10 meters. For the ten-day hourly or daily hourly scheduling scheme, the simulation process loads the gate opening degree, gate opening and closing status, pump group operation combination and pump station target flow at each time step, and calculates the dynamic propagation of water level and flow in the canal system.

[0042] The water supply completion process of the dispatching unit includes the instantaneous water supply flow rate, cumulative simulated water supply volume, and the percentage of target water supply completion for each dispatching unit. The operation process of the gate pump equipment includes the gate operation process, the flow rate through the gate process, the pump set start-up and shutdown process, the pump station flow rate process, the head process, and the power process.

[0043] Step S206: Perform constraint judgment and evaluation on the candidate collaborative scheduling schemes to obtain the evaluation results of the candidate schemes.

[0044] The constraints include target water supply task constraints, control node water level constraints, channel flow capacity constraints, and gate pump equipment operation constraints. Target water supply task constraints are used to limit insufficient water supply and oversupply; control node water level constraints are used to ensure that the channel water level is between the minimum and maximum allowable water levels; channel flow capacity constraints are used to ensure that the channel flow does not exceed the design flow; and gate pump equipment operation constraints are used to ensure that the opening degree, number of start-stop cycles, pump station flow rate, and head are within the allowable range.

[0045] The evaluation indicators for candidate schemes include the completion rate of water supply tasks, water use efficiency of dispatching units, water transmission and distribution efficiency, operating costs of gate pump equipment, and water level satisfaction at control nodes. Evaluation indicators can be assigned different weights based on dispatching and management objectives, or they can be initially assigned the same weights.

[0046] Step S207: Determine the scheduling scheme to be optimized from the candidate cooperative scheduling schemes that meet the constraints, and perform single-objective optimization in sequence according to the preset target priority to obtain the target cooperative scheduling scheme.

[0047] The water supply task completion rate, control node water level satisfaction, water transmission and distribution efficiency, gate pump equipment operating costs, and water supply time can be prioritized according to their importance. When optimizing the current level of objectives, the high-priority objectives that have already been optimized are restricted to within the allowable range of variation to avoid subsequent objective optimizations significantly degrading the performance of high-priority objectives.

[0048] Step S208: Convert the target collaborative scheduling scheme into control commands, and update the gate pump group digital twin model and the current twin operation status based on the real-time operation data after the control commands are executed.

[0049] Gate control commands may include equipment identification, target opening degree, start adjustment time, opening change rate, and duration of maintenance; pump station control commands may include pump station identification, target pump group operation combination, start and stop time, target flow rate, and target head.

[0050] After the control command is executed, the actual water level, actual flow rate, actual water supply, and actual operating status of the gate pump equipment are obtained and compared with the simulation results of the target coordinated scheduling scheme. Based on the comparison results, the model parameters, state variables, and the initial range of scheduling variables for the next scheduling cycle are updated to achieve rolling optimization.

[0051] Through the above implementation methods, channels, control nodes, gates, and pumping stations are established as digital twin objects that can be dynamically updated according to the actual operating status. Before the control command is issued, the coordinated scheduling scheme of the gate and pump group is subjected to hydrodynamic pre-simulation, constraint judgment, and hierarchical optimization, so that the gate regulation and pumping station operation can be coordinated and determined according to the changes in water level and flow rate of upstream and downstream. At the same time, the model state is continuously updated through the control execution results, thereby improving the coordination of gate and pump group scheduling, the adaptability of operating status, the feasibility of the scheme, and the accuracy of water supply task completion.

[0052] In one implementation, entity object identifiers are configured for physical channels, control nodes, gates, and pumping stations. The entity object identifier may consist of an object type code, a channel code, a spatial region code, and an equipment serial number, used to uniquely identify the corresponding entity object.

[0053] Based on the spatial location, upstream and downstream connections, and control relationships of each entity object, entity object association data is obtained. Entity object association data may include upstream object identifiers, downstream object identifiers, controlling object identifiers, controlled object identifiers, and adjacent object identifiers.

[0054] Create virtual objects corresponding to each physical object. For a long physical channel, it can be split into multiple virtual channel objects according to the spatial calculation resolution; for multiple small buildings with the same function, they can be merged into a single virtual control node object according to the model generalization requirements.

[0055] Virtual objects are connected according to the upstream and downstream connections and control relationships of physical objects to form a virtual object topology network. The virtual object topology network can be stored using a directed graph structure, where virtual channel objects, virtual control node objects, virtual gate objects, and virtual pump station objects are graph nodes, and the water flow direction and control relationship are directed edges.

[0056] Obtain the equipment status of physical gates and pumping stations. For online equipment, enable the corresponding virtual object and its control variables; for offline, faulty, or under maintenance equipment, disable the control function of the corresponding virtual object, and set the object as a closed boundary, fixed flow boundary, or unavailable boundary according to the actual project status, thereby obtaining the currently available topology network.

[0057] Associate the monitoring data fields of physical objects with the corresponding virtual objects. For example, associate the water level gauge field with the water level status variable of the virtual control node object, associate the flow meter field with the flow status variable of the virtual channel object or virtual gate object, associate the gate opening field with the virtual gate object, and associate the pump station flow, head, and power fields with the virtual pump station object.

[0058] Associate the control command fields of physical gates and physical pumping stations with the corresponding virtual gate objects and virtual pumping station objects. The control command fields may include target opening degree, start / stop command, target flow rate, pump set start / stop command, and pump set combination code.

[0059] The virtual-to-real data mapping relationship can also record data units, data precision, data update time, communication address, and data quality identifiers. When physical devices or monitoring points are added, deleted, or their locations change, the virtual object topology network can be adjusted by updating the associated data of the physical objects and the virtual-to-real data mapping relationship, without needing to rebuild the entire digital twin model.

[0060] Through the above implementation methods, different types of entity objects in the water transmission and distribution system can be uniformly mapped into virtual objects with topological relationships, status fields, and control fields. The available topology can be dynamically adjusted according to the online, offline, fault, or maintenance status of gate pump equipment, so that the digital twin model can adapt to changes in equipment status and the addition or deletion of engineering objects, thereby improving the model maintenance efficiency and the accuracy of scheduling object identification.

[0061] In one implementation, each channel is divided into multiple channel calculation segments based on the spatial location of control nodes and gate pump equipment within the water distribution system. The division locations of the channel calculation segments may include channel cross-sectional change locations, gate locations, pump station locations, water distribution locations, water collection locations, and water level and flow monitoring locations.

[0062] Configure hydraulic calculation parameters for each channel calculation section. For trapezoidal channels, cross-sectional geometric parameters may include channel bottom width and side slope coefficient; for rectangular channels, cross-sectional geometric parameters may include channel width; for complex cross-section channels, the parameters of each cross-section component can be recorded in separate sections.

[0063] Based on the upstream and downstream connections and hydraulic calculation parameters, the continuity of water volume and the hydrodynamic transmission relationship of the channel calculation section are established. The hydrodynamic control equations can be discretized and solved using the finite difference method, finite volume method, or method of characteristics. Adjacent channel calculation sections are connected by water level continuity conditions and flow balance conditions at common nodes.

[0064] Dynamic boundary conditions can include upstream reservoir discharge, inflow from riverside pumping stations, inflow from level crossings, outflow from tributary channels, and discharge from spillways. When discharge culverts, tributary channel intake gates, or spillway gates are closed, their outflow is set to zero; when open, the flow rate can be determined by real-time monitoring of the corresponding flowmeter or by calculation based on the gate boundary relationship. Inflow from level crossings can be determined by using the output of a flood control model or by real-time monitoring of the flow rate.

[0065] Model inputs can be organized according to a ten-day, hourly structure. For example, the target total water supply of the i-th scheduling unit in the m-th ten-day period, the runoff of the j-th level crossing channel at time h in the m-th ten-day period, the outflow of the k-th dynamic boundary at time h in the m-th ten-day period, and the water release of the upstream reservoir at time h in the m-th ten-day period can all be used as time series inputs for hydrodynamic simulation.

[0066] Through the above implementation method, the non-steady flow dynamic relationship of the channel calculation section is uniformly coupled with the gate boundary, pump station boundary and external dynamic boundary, which can simulate the propagation process of gate opening and pump station flow changes in the upstream and downstream channels, and provide an executable hydrodynamic model basis for calculating the water level, flow and equipment operation results of the gate and pump group coordinated scheduling scheme.

[0067] In one implementation, channel water level data, channel flow data, gate operation data, and pump station operation data from multiple historical scheduling periods are acquired and divided into calibration datasets and verification datasets.

[0068] The calibration dataset can include historical data under different inflow conditions, different water supply loads, and different gate pump operating modes. The validation dataset should be independent of the calibration dataset in time to verify whether the calibrated model is applicable to data that were not involved in parameter adjustments.

[0069] For channel sections with the same channel conditions, at least one roughness calibration point can be set. The roughness calibration point can be set at a location close to the flow meter and where there is no obvious water distribution or collection nearby, so as to reduce the interference of local water volume changes on the roughness calibration results.

[0070] Input the gate opening process, pump group operation combination, pump station flow process, upstream water inflow process and dynamic boundary process in the calibration dataset into the digital twin model of the gate and pump group, and calculate the historical simulated water level, historical simulated flow and simulated gate and pump output parameters at each monitoring location.

[0071] The simulated gate pump output parameters can include simulated gate flow rate, simulated pump station flow rate, simulated head, and simulated power. The measured gate pump output parameters are the actual output parameters obtained from flow meters, pressure sensors, and power metering equipment within the corresponding historical time period.

[0072] Model error can be evaluated using root mean square error, mean absolute error, relative error, or Nash efficiency coefficient.

[0073] Based on the water level deviation, flow rate deviation, and equipment output deviation, the channel roughness parameter, dynamic boundary parameter, gate flow coefficient, and pump station head-flow relationship parameter are adjusted to obtain the calibrated digital twin model of the gate-pump group.

[0074] The calibration of dynamic boundary parameters can be performed on typical boundary nodes with water level and flow monitoring conditions, and the generalized results of the dynamic boundary can be compared with the measured water level and flow processes. For a specific channel, roughness parameters can be calibrated separately under different operating water levels or different flow ranges.

[0075] The validation dataset is input into the calibrated model to obtain the model validation error. When any validation error exceeds the preset validation error threshold, the corresponding error-sensitive parameter is determined and the parameter is readjusted until the validation error meets the requirements, thus obtaining the digital twin model of the target gate pump group.

[0076] At the start of the current scheduling cycle, real-time water level, real-time flow rate, real-time gate opening, real-time pump unit operation combination, and real-time pump station flow rate are mapped to the target gate-pump group digital twin model. For state variables with measured data, the measured data is used for updating; for state variables lacking measured data, the current twin operating state can be estimated based on the states of adjacent nodes and hydrodynamic calculation results.

[0077] Through the above implementation methods, the channel, dynamic boundary, and equipment characteristic parameters are calibrated using historical measured water levels, flow rates, and gate pump output parameters, and verified using independent data. Then, the model state is updated using current real-time data. This can reduce simulation deviations caused by model generalization, changes in equipment performance, and changes in operating boundaries, and improve the accuracy of the digital twin model in representing the current state of the water transmission and distribution system.

[0078] In one implementation, the target water supply task is analyzed to obtain the target water supply volume and target water supply sequence for each scheduling unit within the target scheduling period. Based on the target water supply volume and target water supply sequence for each scheduling unit, the target water supply process corresponding to each scheduling unit is determined. The target water supply process can be a ten-day hourly water supply process or a daily hourly water supply process.

[0079] When the target water supply task only gives the total water supply for ten days, the total water supply for ten days can be decomposed into daily water supply based on crop irrigation period, historical water supply pattern, canal operating period and water supply priority of each dispatching unit, and then the daily water supply can be decomposed into hourly target water supply flow.

[0080] By combining the canal topology and water loss parameters between the scheduling units, the target water supply process of each scheduling unit is accumulated from downstream to upstream to obtain the target water diversion process at the head of the canal.

[0081] Based on the current twin operation status and the availability of gate pump equipment, search for available water conveyance paths between the headworks and each dispatching unit. Paths containing unavailable equipment are eliminated; for dispatching units with multiple available paths, path priority can be determined based on path length, water conveyance loss, flow capacity, and equipment operating costs.

[0082] Based on the target water intake process at the head of the canal and the target water supply process of the dispatching unit, the target flow is allocated to the corresponding available water conveyance paths to obtain the target water conveyance process.

[0083] The opening range and opening / closing sequence range of each gate are determined based on the target water conveyance process. The target flow rate through the gate can be converted into the initial opening range using the gate water level-flow relationship, and the gate opening time can be determined based on the water conveyance delay.

[0084] The available pump unit operating combinations and target flow ranges for each pumping station are determined based on the target water conveyance process. The available flow ranges for single-pump operation, multi-pump parallel operation, and different pump unit combinations can be compared sequentially to select the pump unit operating combination that can cover the target flow.

[0085] The feasible domain of scheduling variables is also limited by the maximum flow capacity of the gate, the rate of change of the gate opening, the maximum water lifting capacity of the pumping station, the start-stop interval of the pumping unit, the flow capacity of the channel, and the allowable water level range of the control node.

[0086] Within the feasible region of the constrained scheduling variables, multiple initial scheduling variable combinations can be generated using methods such as rule combination, random sampling, Latin hypercube sampling, or perturbation based on historical scheduling schemes. Initial scheduling variable combinations can include regular scheduling combinations and emergency scheduling combinations.

[0087] Conventional dispatching combinations can follow dispatching rules such as maintaining constant water levels in front of key control gates, even distribution, prioritizing water supply to severely water-scarce areas, and utilizing water from priority areas. Emergency dispatching combinations can adjust the operation of gates and pumps in response to heavy rain, equipment failure, or sudden water supply demands. For example, under conditions of extremely heavy rain, some key control gates can be closed, and flood discharge can be carried out through diversion gates and spillway gates.

[0088] Through the above implementation method, the headwater water intake process is determined in reverse according to the target water supply process of the scheduling unit. Then, the feasible domain of scheduling variables is formed by combining the equipment availability status, water conveyance path and engineering operation constraints. This can avoid generating initial schemes that obviously do not have water supply capacity or violate equipment constraints, and provide an initial scheduling variable combination that covers a reasonable operating range for the rapid screening of candidate schemes.

[0089] In one implementation, preset continuous variable increments are applied to continuous scheduling variables such as gate opening and target flow rate of the pumping station. For example, the gate opening can be increased or decreased by preset opening values, and the target flow rate of the pumping station can be increased or decreased by preset flow rate values.

[0090] For discrete scheduling variables such as gate opening and closing timing and pump group operation combinations, they are switched to adjacent feasible discrete states. For example, the gate opening time can be advanced or delayed by one time step, and the pump group operation combination can be switched from a single pump state to an adjacent dual pump state.

[0091] Each time, only one or a group of associated scheduling variables are changed, while keeping other scheduling variables unchanged. The changes in water level, flow rate, and equipment operating parameters of the control nodes are calculated using the corrected digital twin model of the gate pump group.

[0092] Changes in equipment operating parameters can include changes in gate flow, pump station flow, pump station head, and pump station power. The correspondence between changes in scheduling variables and changes in response status forms the scheduling variable response data.

[0093] For discrete scheduling variables, the response differences between different discrete states can be recorded separately, and these response differences can be configured in a discrete response table. The control response relationship of the gate pump group can include a continuous response matrix and a discrete response table.

[0094] By combining the initial scheduling variables and inputting them into the gate pump group control response relationship, the water level, flow rate, and operating parameters of the corresponding control node and gate pump equipment can be quickly predicted to obtain the initial state prediction results.

[0095] Initial scheduling variable combinations that are predicted to exceed the allowable range for water level, exceed the channel's flow capacity, or exceed the operating range for equipment are excluded.

[0096] The preset safety margin can be a preset proportion of the allowable range width, or it can be set according to water level measurement error, model prediction error, and operational risk level. When the distance between the predicted state and any constraint boundary is less than the preset safety margin, the response matrix is ​​used to determine the adjustment direction of the scheduling variables that can move the predicted state away from the constraint boundary, and the initial combination of scheduling variables is corrected.

[0097] The corrected combination is then input into the gate pump group control response relationship for prediction until it meets the safety margin requirements or reaches the preset number of corrections. The combinations that pass the screening and correction are determined as the candidate coordinated scheduling scheme set.

[0098] Through the above implementation methods, gate pump control response relationships are established for continuous and discrete scheduling variables, respectively. By using these response relationships, over-limit schemes and critical schemes can be quickly screened out and corrected before the complete hydrodynamic simulation. This can reduce the hydrodynamic simulation resources occupied by obviously infeasible schemes, improve the efficiency of candidate scheme generation, and increase the operational safety margin of candidate schemes.

[0099] In one implementation, according to a preset simulation time step, the gate opening degree, gate opening and closing status, pump group operation combination and pump station target flow in the candidate collaborative scheduling scheme are sequentially loaded into the corrected gate-pump group digital twin model.

[0100] The preset simulation time step can be 1 minute, or it can be set to a time step less than or greater than 1 minute depending on the channel length, flow velocity, and numerical stability requirements. The water level and flow rate calculation results at each simulation moment are used as the initial state for the next simulation moment.

[0101] Based on the hydrodynamic coupling relationship, time-series calculations were performed to obtain the evolution process of water level and flow at the control node, the water supply completion process of the dispatching unit, and the operation process of the gate pump equipment.

[0102] The cumulative simulated water supply of each scheduling unit can be determined by time integration of the simulated flow rate entering that scheduling unit.

[0103] The water efficiency of the dispatching unit can be determined by the degree of water supply completion, the water supply time, and the amount of water wasted. The shorter the water supply time, the higher the degree of water supply completion, and the less water wasted, the higher the corresponding water efficiency.

[0104] The operating cost of a gate can be determined based on the number of gate operations and the amount of opening adjustment. The operating cost of a gate pump system can be the sum of the pump station operating cost and the gate operating cost, or it can be configured with corresponding weights according to equipment maintenance requirements.

[0105] The control node water level can be determined based on whether the water level is between the minimum and maximum allowable water levels at each simulation time, or based on the cumulative degree of deviation of the water level from the target water level.

[0106] The evaluation results of candidate schemes can include evaluation index values, constraint satisfaction indicators, and scheme ranking information. The evaluation index system can also include cumulative water allocation completion status, water use efficiency of the dispatching unit, water level compliance status, and comparison of simulated and measured values ​​after scheme implementation. The initial weight of each evaluation index can be set to 25%, or it can be adjusted according to the importance of the dispatching task.

[0107] Through the above implementation method, the dynamic water transmission and distribution process of each candidate coordinated scheduling scheme is fully simulated, and the scheme is evaluated from aspects such as water supply completion, water use efficiency, water transmission and distribution efficiency, equipment operation cost and water level safety. This avoids judging the merits of the schemes based solely on static flow allocation, and allows the evaluation results of the candidate schemes to simultaneously reflect the water supply effect, channel operation status and gate pump equipment operation burden.

[0108] In one implementation, a target scheduling period is determined based on the target water supply task, and the canal topology relationship between each scheduling unit is determined based on canal topology data; based on the canal topology relationship between each scheduling unit and the water supply completion process of the scheduling unit, adjacent scheduling units that share upstream channels or have water volume competition relationships are identified.

[0109] Water competition can include two scheduling units sharing the same upstream water distribution node, sharing the same pumping capacity, one scheduling unit increasing its flow rate reducing the flow rate available to the other scheduling unit, or the water supply periods of the two scheduling units overlapping.

[0110] Scheduling units that compete for water resources are grouped into local coordinated scheduling unit groups. A scheduling unit may belong to one or more local coordinated scheduling unit groups.

[0111] The local water supply effectiveness is determined based on the water supply time, water wastage, target water supply, simulated water supply, and channel water level and flow status of each dispatching unit.

[0112] Using the gate flow rate and pumping station flow rate as local scheduling strategies, the corresponding local water supply utility is calculated based on different scheduling unit strategy combinations, resulting in a local water quantity game matrix.

[0113] The local water quantity game matrix is ​​solved through evolution according to the time sequence within the target scheduling period. For the current time step, a strategy combination in which no scheduling unit can improve its own utility by changing its own strategy can be selected as the local equilibrium strategy for the current time step.

[0114] The local equilibrium water allocation result at the current time step is used as the initial condition for the next time step, thereby obtaining the local equilibrium water allocation result for the entire target scheduling period along the time dimension.

[0115] Based on the results of local equilibrium water allocation, the gate flow rate, gate opening and closing sequence, pump station water lifting flow rate, and pump group operation combination in the candidate collaborative scheduling scheme are adjusted, and the digital twin model is re-called for hydrodynamic verification.

[0116] When there are no pumping stations participating in water volume regulation in a specific local coordinated scheduling unit group, only the corresponding gate parameters are adjusted; when a pumping station participates in water supply from an adjacent scheduling unit, the pumping flow rate of the pumping station and the pump group operation combination are used as local scheduling strategies to solve together.

[0117] Through the above implementation method, local utility is established for adjacent scheduling units with water competition relationship based on water supply time, water supply completion degree and water abandonment, and the local equilibrium water allocation result is solved in time sequence under water level constraint. This can reduce water competition, water abandonment and water supply time conflicts between adjacent scheduling units and improve the coordination of local water allocation.

[0118] In one implementation, the priority order of multiple target evaluation indicators is determined according to the importance of the target water supply task.

[0119] In conventional water supply scenarios, the priority can be set from highest to lowest as follows: water supply task completion rate, control node water level satisfaction, water transmission and distribution efficiency, gate pump equipment operating cost, and water supply time. In emergency water supply scenarios, priority can be given to ensuring the completion rate of water supply tasks for key dispatch units; in flood control and drainage scenarios, priority can be given to ensuring the water level of control nodes and the safety of channel flow.

[0120] The scheme with the better comprehensive evaluation is selected from the candidate cooperative scheduling schemes that meet the constraints as the initial scheduling scheme for the first objective level.

[0121] Select the target evaluation index corresponding to the current target priority, and adjust the scheduling variables associated with the target evaluation index. For example, when optimizing the completion of water supply tasks, the opening degree of the diversion gate and the corresponding pump station flow can be adjusted; when optimizing the water level satisfaction, the opening degree of the control gate and the upstream pump station flow can be adjusted; when optimizing operating costs, the pump group operation combination and peak and valley period operation arrangement can be adjusted.

[0122] The evaluation metrics for high-priority objectives that have been optimized are restricted to a range of permissible variation. This range can be set based on the optimized values ​​of the high-priority objectives. For example, if the high-priority objective is a maximization metric, its subsequent optimization results must not be lower than a preset percentage of the optimized value; if the high-priority objective is a minimization metric, its subsequent optimization results must not be higher than the sum of the optimized value and the preset tolerance.

[0123] After each adjustment of the scheduling variables, the current level scheduling scheme is input into the corrected digital twin model of the gate pump group to simulate the water transmission and distribution process, and the corresponding evaluation indicators are recalculated.

[0124] Preset termination conditions may include the current target evaluation index reaching the preset target value, the improvement margin between two adjacent iterations being less than the preset improvement threshold, reaching the maximum number of iterations, or not obtaining a better solution within a consecutive preset number of iterations.

[0125] After the current target level is optimized, the optimization result of the current level is used as the initial scheduling scheme for the next target level. After the optimization of the last target level is completed, the optimization result corresponding to the last target level is determined as the target collaborative scheduling scheme. The optimization results of each intermediate level can be saved as alternative schemes for scheduling personnel to discuss and select.

[0126] After the target collaborative scheduling scheme is executed, the measured water level, measured flow rate, actual water supply and actual operating status of gate pump equipment are obtained, and the results are aligned with the simulation results according to the same equipment identifier and timestamp to obtain the scheduling execution deviation.

[0127] Dispatch execution deviations can include water level execution deviations, flow rate execution deviations, water supply execution deviations, gate opening execution deviations, pump station flow rate execution deviations, and pump group status execution deviations.

[0128] The digital twin model of the gate and pump group is updated based on the scheduling execution deviation. For persistent water level and flow deviations, the channel roughness or dynamic boundary parameters can be adjusted; for gate flow deviations, the gate flow coefficient can be adjusted; and for pump station flow or power deviations, the pump station characteristic curve parameters can be adjusted.

[0129] The initial range of scheduling variables for the next scheduling cycle is updated based on the scheduling execution deviation. For example, when the actual flow rate of a pump group is lower than the simulated flow rate multiple times, the upper limit of the target flow rate corresponding to that pump group can be reduced; when the actual action time of a gate is significantly longer than the set time, the opening and closing time of the next cycle can be advanced.

[0130] The updated model parameters, the current twin running status, and the initial range of scheduling variables form the initial conditions for the next scheduling cycle, thus enabling the rolling compilation of scheduling schemes.

[0131] Through the above implementation method, multiple scheduling objectives are optimized step by step according to their priority, and the optimization results of high-priority objectives are maintained by utilizing the allowable range of change. This avoids the situation where important objectives are offset by secondary objectives due to simple weighting of multiple objectives. At the same time, the model parameters and the range of variables for the next cycle are corrected according to the scheduling execution results, so that the subsequent scheduling scheme can adapt to the changes in the actual operating capacity of the channel and gate pump equipment.

[0132] In one implementation, the gate pump group collaborative scheduling optimization system based on digital twins includes a digital twin model construction module, a twin state correction module, a candidate scheme generation module, a simulation evaluation module, a collaborative scheduling optimization module, and a control feedback module.

[0133] The digital twin model building module is used to acquire canal topology data, canal hydraulic parameters, control node parameters, gate parameters, pump station parameters, equipment availability status, target water supply tasks, and real-time operation data, and to establish a mapping relationship between physical objects and virtual objects.

[0134] The digital twin model building module can include a topology management unit, a virtual-physical mapping unit, a hydrodynamic calculation unit, a gate boundary unit, and a pump station boundary unit. The topology management unit supports the addition, deletion, activation, and deactivation of virtual objects to adapt to changes in the status of engineering objects and equipment.

[0135] The twin-state calibration module is used to calibrate historical data, verify independent data, and update the status in real time. The twin-state calibration module can save model parameters corresponding to different channel sections, different water level ranges, and different equipment operating combinations.

[0136] The candidate scheme generation module is used to determine the target water supply process, reverse the target water diversion process at the head of the canal, search for available water conveyance paths, establish the feasible domain of scheduling variables, generate the initial combination of scheduling variables, and construct the control response relationship of the gate pump group.

[0137] The simulation evaluation module is used to simulate the water transmission and distribution process at a preset simulation time step, and outputs the hourly flow rate of the diversion gate, the hourly flow rate of the control gate, the daily water supply of the dispatching unit, the cumulative completion status of water distribution, the water level of the channel, and the comparison results between the simulated and measured values ​​of water supply.

[0138] The simulation evaluation module can include constraint judgment units, water supply task evaluation units, water use efficiency evaluation units, water level evaluation units, water transmission and distribution efficiency evaluation units, and equipment operation cost evaluation units. Evaluation results can be output through reports, time series curves, canal system thematic maps, and early warning information.

[0139] The collaborative scheduling optimization module is used to implement water quantity coordination and multi-objective hierarchical cyclic optimization of local scheduling units. The collaborative scheduling optimization module can generate regular scheduling schemes, emergency scheduling schemes, ten-day hourly scheduling schemes, and daily hourly scheduling schemes, and output the opening and closing times and target flow rates of each key scheduling node.

[0140] The control feedback module is used to convert the target coordinated scheduling scheme into gate and pump station control commands, which are issued through the gate control system and pump station control system, and to receive water level, flow rate and equipment status feedback after the control is executed.

[0141] The system may also include a model database and standardized interfaces. The model database is used to store canal system topology, canal and structure parameters, historical scheduling data, real-time monitoring data, candidate scheduling schemes, model parameters, evaluation results, and control execution results.

[0142] The standardized interface can connect with water supply and demand forecasting models, water resource allocation models, channel flood control models, and real-time monitoring systems. The water supply and demand forecasting models and water resource allocation models provide target water supply data for the scheduling unit; the channel flood control model provides data on inflow into level crossings and flood control boundaries; and the real-time monitoring system provides water level, flow rate, and equipment operation data.

[0143] The model input can be organized in array or JSON format, including the total water demand of the scheduling unit for ten days, the hourly inflow of the intersecting river channel for ten days, the hourly outflow of the dynamic boundary for ten days, and the hourly discharge of the upstream reservoir for ten days. The model can return a success flag when it runs successfully and a corresponding error code when it fails.

[0144] The system's hydrodynamic simulation has a time resolution of 1 minute and a spatial resolution of 10 meters. The system can be deployed on the Zhongke Fangde operating system or other server operating systems, and supports deployment at both the higher-level water administration unit and the irrigation district management unit. The two locations can synchronize basic model data, scheduling plans, and operational results.

[0145] The system can also be configured with a presentation layer to display the canal topology, gate and pump equipment status, canal water level and flow rate processes, water allocation completion status of dispatching units, evaluation results of candidate schemes, and target collaborative dispatching schemes. Dispatchers can view recommended and alternative schemes through the presentation layer and conduct manual consultations for confirmation when necessary.

[0146] Through the above implementation methods, each module can be executed collaboratively according to a unified canal system topology, virtual-real mapping relationship and model state. It forms a complete closed loop from target water supply task input, digital twin state update, candidate scheme generation and pre-drilling, to collaborative optimization, instruction issuance and execution feedback. It can also interact with prediction, configuration, flood control and monitoring systems through standardized interfaces, thereby improving the model consistency, operation automation and rolling scheduling capability of the gate pump group scheduling system.

[0147] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A gate pump group collaborative scheduling optimization method based on digital twins, characterized in that, The method includes: Acquire the canal topology data, canal hydraulic parameters, gate and pump equipment availability status, target water supply tasks and real-time operation data of the water transmission and distribution system, establish the mapping relationship between physical channels, control nodes, gates and pump stations and virtual objects, and establish the hydrodynamic coupling relationship between each virtual object based on the canal topology data and the canal hydraulic parameters to obtain a digital twin model of the gate and pump group. The real-time operating data is mapped to the digital twin model of the gate pump group. Based on the deviation between the simulated state and the measured state, the hydrodynamic parameters, dynamic boundary parameters and equipment characteristic parameters of the digital twin model of the gate pump group are corrected and the model state variables are updated to obtain the corrected digital twin model of the gate pump group and the current twin operating state. Based on the target water supply task, the current twin operation status, the availability status of the gate pump equipment, and the operation constraints, scheduling variables are determined and multiple initial scheduling variable combinations are generated; based on the hydrodynamic coupling relationship, a gate pump group control response relationship is constructed, and the gate pump group control response relationship is used to filter or modify each initial scheduling variable combination to obtain a set of candidate collaborative scheduling schemes; The corrected digital twin model of the gate pump group is used to simulate the water transmission and distribution process and evaluate the schemes for each candidate coordinated scheduling scheme, and the evaluation results of the candidate schemes are obtained. From the candidate collaborative scheduling schemes that meet the operational constraints, the scheduling scheme to be optimized is determined, and the target evaluation indicators are selected in order according to the preset target priority to perform single target optimization, so as to obtain the target collaborative scheduling scheme. The target collaborative scheduling scheme is converted into control commands for gates and pumping stations, and the digital twin model of the gate-pump group and the current twin operating status are updated based on the real-time operating data after execution.

2. The gate pump group collaborative scheduling optimization method based on digital twin as described in claim 1, characterized in that, The establishment of the mapping relationship between physical channels, control nodes, gates and pumping stations and corresponding virtual objects includes: Entity object identifiers are configured for the physical channels, control nodes, gates and pumping stations respectively, and entity object association data is obtained based on the spatial location, upstream and downstream connection relationship and control relationship of each entity object; Based on the entity object association data, virtual channel objects, virtual control node objects, virtual gate objects, and virtual pump station objects corresponding to each entity object are established, and each virtual object is connected according to the upstream and downstream connection relationship and control relationship to obtain a virtual object topology network. The online, offline, or fault status of each physical gate and physical pump station is obtained. Based on the obtained status, the corresponding virtual gate object and virtual pump station object in the virtual object topology network are enabled or disabled to obtain the currently available topology network. The monitoring data fields corresponding to each physical object are associated with the corresponding virtual object, and the control command fields corresponding to each physical gate and physical pumping station are associated with the corresponding virtual gate object and virtual pumping station object to obtain the virtual-physical data mapping relationship. Based on the currently available topology network and the virtual-real data mapping relationship, the mapping relationship between physical channels, control nodes, gates and pumping stations and their corresponding virtual objects is obtained.

3. The gate pump group collaborative scheduling optimization method based on digital twin as described in claim 1, characterized in that, The step of establishing the hydrodynamic coupling relationship between virtual objects based on the canal system topology data and the canal hydraulic parameters includes: Based on the spatial location of the control nodes and gate pumps in the water transmission and distribution system, each channel is divided into multiple channel calculation segments to obtain the upstream and downstream connection relationships between each channel calculation segment; Configure the channel bottom elevation, cross-sectional geometric parameters, channel length and roughness parameters for each channel calculation section to obtain the hydraulic calculation parameters corresponding to each channel calculation section; Based on the upstream and downstream connection relationship and the hydraulic calculation parameters, the water volume continuity relationship and hydrodynamic transmission relationship of each channel calculation section are established to obtain the channel hydrodynamic calculation relationship. Based on the relationship between the upstream and downstream water levels of the gate, the gate opening and the flow rate through the gate, the gate boundary calculation relationship is determined, and based on the pumping capacity of the pumping station, the pump set operation combination, the head-flow relationship and the target flow rate of the pumping station, the pumping station boundary calculation relationship is determined. The gate boundary calculation relationship and the pump station boundary calculation relationship are configured in the corresponding control nodes, and combined with the dynamic boundary conditions corresponding to upstream reservoir water release, river pump station inflow, level crossing channel inflow, branch canal water diversion outflow and discharge outflow, the hydrodynamic coupling relationship between each virtual object is obtained.

4. The gate pump group collaborative scheduling optimization method based on digital twin as described in claim 3, characterized in that, The process of correcting the hydrodynamic parameters, dynamic boundary parameters, and equipment characteristic parameters of the digital twin model of the gate pump group, and updating the model state variables, includes: Data on channel water level, channel flow, gate operation, and pump station operation were acquired from multiple historical scheduling periods. The data were then divided into calibration and verification datasets. The operation process of the gate pumps corresponding to the calibration dataset is input into the digital twin model of the gate pump group to obtain the historical simulated water level, historical simulated flow rate and simulated gate pump output parameters at each monitoring location; Based on the water level deviation between the historical simulated water level and the historical measured water level, the flow deviation between the historical simulated flow rate and the historical measured flow rate, and the equipment output deviation between the simulated gate pump output parameters and the measured gate pump output parameters, the channel roughness parameters, dynamic boundary parameters, gate water level-flow rate relationship parameters, and pump station head-flow rate relationship parameters are adjusted to obtain the calibrated digital twin model of the gate pump group. The verification dataset is input into the calibrated gate pump group digital twin model to obtain the model verification error. When the model verification error is greater than the preset verification error threshold, the corresponding model parameters are readjusted until the model verification error is not greater than the preset verification error threshold, and the target gate pump group digital twin model is obtained. The real-time water level, real-time flow rate, real-time gate opening, real-time pump group operation combination, and real-time pump station flow rate at the current scheduling moment are mapped to the target gate-pump group digital twin model. The corresponding model state variables are updated to obtain the corrected gate-pump group digital twin model and the current twin operation status.

5. The gate pump group collaborative scheduling optimization method based on digital twin as described in claim 1, characterized in that, The generation of multiple initial scheduling variable combinations includes: The target water supply task is analyzed to obtain the target water supply volume and target water supply sequence of each scheduling unit within the target scheduling period, and the target water supply process corresponding to each scheduling unit is determined based on the target water supply volume and the target water supply sequence. Based on the target water supply process, the canal topology between each scheduling unit, and the water loss parameters, the target water diversion process of the canal head during the target scheduling period is calculated in reverse. Based on the current twin operation status and the availability status of the gate pump equipment, the available water conveyance paths between the canal head and each dispatching unit are determined, and based on the target water intake process and the target water supply process, the target water conveyance process corresponding to each available water conveyance path is determined; Based on the target water conveyance process, the opening value range and opening and closing sequence range of each gate, as well as the available pump group operation combination and target flow range of each pumping station, are determined to obtain the feasible domain of scheduling variables. Based on the gate's maximum flow capacity, the gate's opening rate of change, the pump station's maximum water lifting capacity, the pump set's start-stop interval, the channel's flow capacity, and the control node's allowable water level range, the feasible domain of the scheduling variables is constrained, and multiple initial scheduling variable combinations are generated within the constrained feasible domain of the scheduling variables.

6. The gate pump group collaborative scheduling optimization method based on digital twin as described in claim 5, characterized in that, The construction of the gate pump group control response relationship and the selection or modification of each initial scheduling variable combination include: Preset continuous variable increments are applied to the gate opening and the target flow of the pumping station, and the gate opening and closing sequence and pump group operation combination are switched to adjacent feasible discrete states to obtain multiple scheduling variable test states. Using the corrected digital twin model of the gate pump group, the changes in water level, flow rate, and operating parameters of the associated gate pump equipment relative to the test state of each scheduling variable relative to the corresponding initial state are calculated to obtain the scheduling variable response data. Based on the correspondence between the changes in each scheduling variable and the changes in the water level, flow rate, and operating parameters of the associated gate pump equipment at the control node, a gate pump group control response matrix is ​​constructed to obtain the gate pump group control response relationship. By inputting the combination of initial scheduling variables into the gate pump group control response relationship, the control node water level, control node flow and gate pump equipment operating parameters corresponding to each combination of initial scheduling variables are predicted to obtain the initial state prediction results. Based on the initial state prediction results, combinations of initial scheduling variables that exceed the allowable water level range of the control node, the channel flow capacity, or the operating range of the gate pump equipment are filtered out. When the distance between the initial state prediction result and any constraint boundary is less than the preset safety margin, the corresponding initial scheduling variable combination is corrected according to the gate pump group control response matrix to obtain the candidate collaborative scheduling scheme set.

7. The gate pump group collaborative scheduling optimization method based on digital twin as described in claim 1, characterized in that, The evaluation results of the candidate solutions obtained include: According to the preset simulation time step, the gate opening degree, gate opening and closing status, pump group operation combination and pump station target flow corresponding to each candidate collaborative scheduling scheme are sequentially loaded into the corrected gate-pump group digital twin model to obtain the water level and flow calculation results and equipment operation parameter calculation results corresponding to each simulation time. Based on the hydrodynamic coupling relationship between each virtual object, the calculation results of water level and flow rate and the calculation results of equipment operating parameters are transmitted in time sequence to obtain the evolution process of water level and flow rate of control nodes, the water supply completion process of scheduling unit and the operation process of gate pump equipment corresponding to each candidate collaborative scheduling scheme; Based on the relationship between the cumulative simulated water supply and the target water supply of each scheduling unit, the water supply task completion rate of each candidate coordinated scheduling scheme is determined, and the water use efficiency of each scheduling unit is determined based on the water supply time, water wastage, cumulative simulated water supply and target water supply of each scheduling unit. The water transmission and distribution efficiency of each candidate coordinated dispatch scheme is determined based on the water intake at the head of the canal, the effective water intake of the dispatching unit, and the water discharge. The operating cost of each pumping station is determined based on its flow rate, head, operating time, and number of pump start-ups and shutdowns. The operating cost of each gate is determined based on the number of gate actions and the amount of gate opening adjustment. The operating costs of the pumping stations and the operating costs of the gates are then correlated to obtain the operating cost of the gate pump equipment. Based on the completion rate of the water supply task, the water use efficiency of each scheduling unit, the water transmission and distribution efficiency, the operating cost of gate pump equipment, and the water level satisfaction of control nodes, the evaluation results of the candidate schemes corresponding to each candidate collaborative scheduling scheme are formed.

8. The gate pump group collaborative scheduling optimization method based on digital twin according to claim 7, characterized in that, The optimization of the candidate cooperative scheduling scheme also includes: The target scheduling period is determined based on the target water supply task. The canal topology relationship between each scheduling unit is determined based on the canal topology data. Based on the canal topology relationship and the water supply completion process of the scheduling unit, adjacent scheduling units that share upstream channels or have water volume competition relationships are identified, resulting in multiple local collaborative scheduling unit groups. For each local coordinated scheduling unit group, the local water supply effectiveness of each scheduling unit is determined based on the water supply time, water abandonment, target water supply, simulated water supply, and the water level and flow status of the corresponding channel. Using the gate flow rate and pumping flow rate corresponding to each scheduling unit as local scheduling strategies, the local water supply utility as the strategy benefit, and the minimum and maximum allowable water levels of the corresponding channels as local scheduling constraints, a local water volume game matrix corresponding to each local collaborative scheduling unit group is constructed. According to the time sequence within the target scheduling period, the local water quantity game matrix is ​​evolved and solved to obtain the local equilibrium water quantity allocation results corresponding to each local collaborative scheduling unit group; Based on the local equilibrium water allocation results, the gate flow rate, gate opening and closing sequence, pump station water lifting flow rate, and pump group operation combination in the corresponding candidate coordinated scheduling scheme are adjusted to obtain the candidate coordinated scheduling scheme after local coordination.

9. The gate pump group collaborative scheduling optimization method based on digital twin according to claim 1, characterized in that, The step of selecting target evaluation indicators sequentially according to preset target priority for single-target optimization, and updating the results based on real-time operating data after the control command is executed, includes: Based on the importance of the target water supply task, determine the priority order of the target tasks corresponding to the water supply task completion rate, the water level satisfaction of control nodes, water transmission and distribution efficiency, the operating cost of gate pump equipment, and water supply time. Select the target evaluation index corresponding to the current target priority, and use the allowable change range of each target evaluation index with higher than the current target priority as a constraint to adjust the scheduling variables associated with the current target evaluation index to obtain the current level scheduling scheme; The current level scheduling scheme is input into the corrected gate pump group digital twin model to simulate the water transmission and distribution process, and the corresponding scheduling variables are adjusted according to the simulation results until the current target evaluation index meets the corresponding preset termination condition to obtain the current level optimization result. The current level optimization result is used as the initial scheduling scheme for the next target priority. The single-target optimization of each target evaluation index is completed in sequence, and the current level optimization result corresponding to the last target priority is determined as the target collaborative scheduling scheme. After the target collaborative scheduling scheme is executed, the measured water level, measured flow rate, actual water supply and actual operating status of the gate pump equipment are obtained, and they are compared with the simulation results corresponding to the digital twin model of the gate pump group to obtain the scheduling execution deviation. Based on the scheduling execution deviation, the model parameters of the gate pump group digital twin model, the current twin operating state, and the initial range of scheduling variables corresponding to the next scheduling cycle are updated to obtain the initial conditions of the model for the next scheduling cycle.

10. A gate pump group collaborative scheduling optimization system based on digital twin, characterized in that, The system includes: The digital twin model construction module is used to acquire the canal topology data, channel hydraulic parameters, gate and pump equipment availability status, target water supply tasks and real-time operation data of the water transmission and distribution system, establish the mapping relationship between physical channels, control nodes, gates and pump stations and virtual objects, and establish the hydrodynamic coupling relationship between each virtual object based on the canal topology data and the channel hydraulic parameters to obtain the gate and pump group digital twin model. The twin state correction module is used to map the real-time operating data to the gate pump group digital twin model. Based on the deviation between the simulated state and the measured state, it corrects the hydrodynamic parameters, dynamic boundary parameters and equipment characteristic parameters of the gate pump group digital twin model and updates the model state variables to obtain the corrected gate pump group digital twin model and the current twin operating state. The candidate scheme generation module is used to determine scheduling variables and generate multiple initial scheduling variable combinations based on the target water supply task, the current twin operation status, the availability status of the gate pump equipment, and operation constraints; construct the gate pump group control response relationship based on the hydrodynamic coupling relationship, and use the gate pump group control response relationship to filter or modify each initial scheduling variable combination to obtain a set of candidate collaborative scheduling schemes; The simulation evaluation module is used to simulate the water transmission and distribution process and evaluate the schemes of each candidate coordinated scheduling scheme using the corrected digital twin model of the gate pump group, and obtain the evaluation results of the candidate schemes. The collaborative scheduling optimization module is used to determine the scheduling scheme to be optimized from the candidate collaborative scheduling schemes that meet the operational constraints, and select target evaluation indicators in order according to the preset target priority to perform single-target optimization and obtain the target collaborative scheduling scheme. The control feedback module is used to convert the target coordinated scheduling scheme into control commands for gates and pumping stations, and update the digital twin model of the gate-pump group and the current twin operating status based on the real-time operating data after execution.

Images