Simulation method and system for water resource management in irrigation area based on digital twinning technology

Abstract

The present application relates to the technical field of digital twin of irrigation district water conservancy, in particular to an irrigation district water resource management simulation method and system based on digital twin technology, comprising: establishing a digital mirror model containing pipe network topology, equipment parameters and spatial information, synchronously feeding back current and voltage of water pumps of integrated equipment in the irrigation district, opening degree of valves, pressure difference between inlet and outlet of filters, injection flow of fertilizer machines, and soil humidity and meteorological parameters distributed in the irrigation district; injecting data into the model to simulate dynamic hydraulic response of the water network, forming a global simulation state of pressure field, flow distribution and water quality concentration field; comparing operation efficiency with benchmark target to identify deviation, carrying out multiple rounds of virtual regulation and control simulation in the model and screening the optimal strategy, and finally executing closed-loop regulation and control on site equipment. The present application realizes real-time mapping of physical entities and digital models, improves the integrity of global simulation of the water network, optimizes regulation and control actions through virtual rehearsal, and ensures stable and efficient operation of the irrigation district water network.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology for irrigation district water conservancy, and in particular to a simulation method and system for irrigation district water resource management based on digital twin technology. Background Technology

[0002] Traditional irrigation district water resource management relies primarily on decentralized monitoring, using independent sensors to collect data on pipeline flow, pressure, soil moisture, and meteorological conditions. Pumps, valves, fertilizer applicators, and filters operate under local, independent control. Integrated skid-mounted systems only achieve partial hardware integration, failing to synchronize operational data across the entire area. Some systems use simplified pipeline models to display parameters, presenting only single-point monitoring values ​​and failing to construct a digital mirror model that fully matches the physical entity.

[0003] The sensed data are isolated and cannot reflect the overall hydraulic relationship of the pipeline network. They cannot simulate the dynamic changes of pressure field, flow distribution, and water quality concentration field. There is a temporal and spatial disconnect between equipment operating parameters and model simulation. Control relies on manual experience or single-parameter threshold feedback, directly adjusting the equipment on site without virtual pre-simulation or optimization. The control accuracy is insufficient, and it is prone to over-adjustment or lag. The water network operation efficiency deviates continuously from the preset target.

[0004] Unable to rely on multi-source data synchronization from skid-mounted equipment to drive full-domain dynamic hydraulic simulation, unable to conduct multi-strategy virtual calculations in the digital space, unable to form a closed-loop control from virtual simulation optimization to physical equipment execution, the pipeline network operation status is difficult to stably match the preset performance target. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a simulation method and system for irrigation district water resource management based on digital twin technology.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a simulation method for irrigation district water resource management based on digital twin technology, comprising: Establish a digital mirror model of the physical entity of the irrigation area, the digital mirror model including the pipeline network topology, equipment attribute parameters and spatial distribution information; Data is synchronously and in real time sensed from the integrated skid-mounted equipment at the irrigation area site. The real-time sensed data includes the current and voltage of the water pump, the valve opening feedback, the inlet and outlet pressure difference of the filter, the injection flow of the fertilizer applicator, and the soil moisture and meteorological parameters distributed in the irrigation area. The real-time sensing data is injected into the digital mirror model, which drives the digital mirror model to simulate the dynamic hydraulic response of the irrigation area water network under the current equipment state, generating a full system simulation state including pressure field, flow distribution, and water quality concentration field. Based on the simulation state of the entire system, the actual operating efficiency index of the current irrigation district water network is calculated and compared with the preset operating efficiency benchmark target to identify the deviation. Based on the identified deviations, multiple rounds of virtual control simulation with preset strategies are performed in the digital mirror model, and the optimal set of control strategies that meet the operational performance benchmark target is selected from the results of the multiple rounds of virtual control simulation. The optimal control strategy set is distributed to the integrated skid-mounted equipment at the irrigation area site for execution, completing one closed-loop management cycle.

[0007] As a further aspect of the present invention, the real-time sensing of data from the integrated skid-mounted equipment at the irrigation area includes: By using data acquisition interfaces deployed on water pump motor drivers, electric regulating valve controllers, differential pressure sensor transmitters, fertilizer metering pump controllers, soil moisture sensor nodes, and weather stations, raw equipment signals are acquired through a combination of periodic polling and event triggering. The original device signal is subjected to a signal quality assessment, which includes checking for data jumps, out-of-bounds values, and communication timeouts. The original equipment signals that have passed the quality assessment are standardized and converted, converting current and voltage into pump operating power and efficiency, valve opening feedback into equivalent flow area, filter inlet and outlet pressure difference into clogging coefficient, fertilizer injector flow rate into instantaneous fertilizer concentration, and soil moisture and meteorological parameters into regional evapotranspiration water demand. The standardized data is aligned and packaged according to preset spatiotemporal labels to form a real-time sensing data package with timestamps and geographic coordinates.

[0008] As a further aspect of the present invention, the driving digital mirror model simulates the dynamic hydraulic response of the irrigation district water network under the current equipment state, generating a full-system simulation state including pressure field, flow distribution, and water quality concentration field, including: The operating power and efficiency of the water pump, the equivalent flow area of ​​the valve, and the clogging coefficient of the filter in the real-time sensing data packet are used as boundary conditions and internal parameters to update the attribute parameters of the corresponding devices in the digital mirror model. Based on the updated digital mirror model, the physical control equations describing water flow, mass conservation and solute transport in the irrigation network are solved, and the coupled simulation calculations of transient hydraulics and water quality are carried out. In the simulation calculation process, the regional evapotranspiration water demand in the real-time sensing data packet is taken as a dynamic field water consumption term, and the instantaneous value of fertilizer solution concentration is taken as a solute source term of water source, and substituted into the physical control equation. After completing the transient simulation calculation for a preset duration, the pressure, flow, and water concentration values ​​of all nodes and pipe segments in the digital mirror model are extracted to form a pressure field distribution map, flow distribution map, and water concentration field distribution map covering the entire irrigation area, which together constitute the simulation state of the entire system.

[0009] As a further aspect of the present invention, based on the simulation state of the entire system, the actual operational efficiency index of the current irrigation district water network is calculated and compared with the preset operational efficiency benchmark target to identify deviations, including: From the pressure field distribution map of the entire system simulation state, the pressure values ​​of key water-using nodes are extracted, and the water supply pressure guarantee rate and pressure uniformity index are calculated as sub-indicators of water supply efficiency. From the flow distribution map of the entire system simulation state, the flow value of each field inlet is extracted and compared with the crop water requirement of the corresponding field. The irrigation water satisfaction and distribution uniformity index are calculated as the irrigation efficiency sub-indicators. From the water quality concentration field distribution map of the entire system simulation state, extract the fertilizer concentration value at the inlet of each field, compare it with the preset fertilizer formula concentration, and calculate the fertilizer uniformity and accuracy index as a sub-index of fertilizer efficiency. The water supply efficiency sub-indicators, irrigation efficiency sub-indicators, and fertilization efficiency sub-indicators are integrated according to preset weights to calculate the comprehensive operational efficiency index value of the current irrigation district water network. The comprehensive operational efficiency index value is compared with the preset operational efficiency benchmark target value, the difference is calculated, and the difference is used as the deviation.

[0010] As a further aspect of the present invention, based on the identified deviation, multiple rounds of virtual control simulation with preset strategies are performed in the digital mirror model. From the results of these multiple rounds of virtual control simulation, an optimal set of control strategies that satisfies the operational performance benchmark is selected, including: Analyzing the composition of the aforementioned deviations, it was determined that the main factors leading to the failure of the comprehensive operational efficiency index to meet the standards were insufficient water supply efficiency, insufficient irrigation efficiency, or insufficient fertilization efficiency. Based on the main factors, one or more candidate control strategies are selected from the preset strategy knowledge base. The candidate control strategies include adjusting the pump speed combination, adjusting the valve opening combination, starting the filter backwashing program, modifying the fertilizer applicator injection ratio, and switching the field irrigation group. In the digital mirror model, each group of candidate control strategies is loaded sequentially, and the state changes of the integrated skid-mounted device after executing the candidate control strategies are simulated. Based on the equipment state after the simulation execution, the digital mirror model is re-driven to perform coupled simulation calculations of transient hydraulics and water quality, to obtain the new full-system simulation state corresponding to each group of candidate control strategies, and to calculate its new comprehensive operating efficiency index value. The new comprehensive operational efficiency index values ​​corresponding to all candidate control strategies are compared with the operational efficiency benchmark target. Strategies that enable the new comprehensive operational efficiency index values ​​to reach or exceed the benchmark target are selected, and these strategies are combined to form the optimal control strategy set.

[0011] As a further aspect of the present invention, the step of selecting one or more sets of candidate control strategies from a preset strategy knowledge base includes: The strategy knowledge base is constructed from historical successful control cases, domain expert rules, and equipment operation constraints. When the main factor is insufficient water supply efficiency, the selected candidate control strategies focus on increasing the pressure of the pipeline network, including activating standby pumps, increasing the frequency of operating pumps, and closing downstream non-critical branch valves. When the main factor is insufficient irrigation efficiency, the selected candidate control strategies focus on optimizing flow distribution, including extending the rotation irrigation time in high water demand areas, adjusting the opening sequence of the rotation irrigation group, and increasing the opening degree of the branch pipe valves supplying water to water-scarce areas. When the main factor is insufficient fertilization efficiency, the selected candidate control strategies focus on stabilizing fertilizer solution concentration, including calibrating the closed-loop control parameters of the fertilizer applicator's injection volume, switching to the backup fertilization channel, and adding a pressure stabilizing device to the upstream pipeline of the fertilizer applicator.

[0012] As a further aspect of the present invention, the optimal control strategy set is distributed to the integrated skid-mounted equipment at the irrigation area for execution, including: Each strategy instruction in the optimal control strategy set is translated into a standardized control command that can be executed by specific equipment. The standardized control command includes the target frequency and start / stop of the water pump, the target opening degree of the valve, the backwash trigger signal of the filter, the target injection ratio and start / stop of the fertilizer applicator, and the group switching sequence of the field solenoid valve. Through a secure industrial communication protocol, the standardized control commands are sequentially sent to the pump control unit, valve control unit, filter control unit, fertilizer control unit, and field valve controller at the irrigation area according to a preset execution sequence. After the command is issued, the real-time sensing data returned from the irrigation area is continuously monitored to verify whether the standardized control command is executed correctly and whether the equipment status meets the strategy expectations. The verified actual execution results are compared with the virtual control simulation prediction results in the digital mirror model to generate a strategy execution consistency report.

[0013] As a further aspect of the present invention, the method further includes correcting the digital mirror model based on the policy execution consistency report, including: When the strategy execution consistency report shows that the actual pressure, flow rate or concentration changes differ from the simulation prediction values ​​in a continuous manner and exceed the allowable range, it is determined that the digital mirror model has model errors. Extract the real-time sensing data during the period in which the difference occurs, and the internal simulation calculation parameters of the digital mirror model during the corresponding period; The source of the model error was analyzed to determine whether it stemmed from inaccurate pipeline resistance parameters, drift of pump characteristic curves, deviation of valve flow characteristics, or inaccurate soil infiltration parameters. Based on the error analysis results, the corresponding equipment attribute parameters, pipeline topology parameters, or field water consumption model parameters in the digital mirror model are reverse-calibrated and updated to make the model output closer to the physical entity response.

[0014] As a further aspect of the present invention, reverse calibration and updating are performed on the corresponding equipment attribute parameters, pipeline topology parameters, or field water consumption model parameters in the digital mirror model, including: If the error originates from the pipeline resistance parameters, the difference between the actual monitored pipeline pressure drop and the simulated pressure drop is used to invert and update the friction coefficient and local resistance coefficient of the corresponding pipe segment in the digital mirror model using an iterative optimization algorithm. If the error originates from the drift of the pump characteristic curve, the actual operating point of the pump is fitted with the sample curve to generate the current actual pump head-flow characteristic curve, and the pump performance curve data in the digital mirror model is updated accordingly. If the error originates from valve characteristic deviation, then the flow coefficient curve of the valve in the digital mirror model is corrected based on the actual flow rate and simulated flow rate at different valve openings. If the error originates from the field water consumption model, the crop coefficient or soil hydrodynamic parameters of the corresponding field in the digital mirror model are adjusted by using the measured changes in soil moisture and the soil moisture consumption calculated by the model.

[0015] As a further aspect of the present invention, the present invention also includes an irrigation district water resources management simulation system based on digital twin technology. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the irrigation district water resources management simulation method based on digital twin technology as described above.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: The integrated skid-mounted equipment can simultaneously collect pump current and voltage, valve opening feedback, filter inlet and outlet pressure difference, fertilizer applicator injection flow, irrigation area soil moisture, and meteorological parameters. Multiple types of sensor data are transmitted synchronously in real time and fully integrated into a digital mirror model containing pipeline topology, equipment attribute parameters, and spatial distribution information. Driven by multi-source real-time data, the digital mirror model dynamically simulates the hydraulic response process of the irrigation network under corresponding equipment operating conditions, generating a coupled simulation state of the global pressure field, flow distribution, and water concentration field. This fully reflects the continuous changes in hydraulic transmission, flow distribution, and water and fertilizer concentration diffusion within the water network, eliminating data fragmentation caused by decentralized monitoring, achieving real-time correspondence between the physical irrigation area and the digital model, accurately restoring the overall operational characteristics of the water network, and improving the comprehensiveness and realism of the water network's status representation.

[0017] The operational efficiency indicators of the irrigation district's water network are obtained through full-system simulation. These indicators are then compared with preset operational efficiency benchmarks to accurately pinpoint operational deviations. Multiple independent virtual simulations are conducted within a digital mirror model for various preset control schemes. The simulation results are used to quantitatively evaluate the control effects of different schemes, selecting the optimal set of control strategies that matches the operational efficiency benchmarks. This optimal set of control strategies is then distributed to integrated skid-mounted equipment, driving field equipment to perform corresponding adjustments. This avoids the randomness and lag of direct field control, reduces irrational fluctuations in water network pressure, flow, and concentration, minimizes ineffective losses during equipment adjustment, and constructs a coherent closed loop between virtual simulation optimization and physical equipment execution, ensuring that the irrigation district's water network operation continuously approaches the preset efficiency targets. Attached Figure Description

[0018] Figure 1 This is a flowchart of the irrigation district water resource management simulation method based on digital twin technology described in this invention; Figure 2 A flowchart for synchronizing real-time sensing data; Figure 3 A flowchart for calculating operational performance indicators and identifying deviations. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0021] See Figure 1 This invention provides a simulation method for irrigation district water resource management based on digital twin technology, the specific implementation of which is as follows: A digital mirror model of the irrigation district's physical entities is established, comprehensively representing the topological relationships of the irrigation network, the attribute parameters of equipment such as pumps and valves, and their spatial distribution information. Real-time sensing data is synchronously acquired from the integrated skid-mounted equipment in the irrigation district, covering pump current and voltage, valve opening feedback, filter inlet and outlet pressure difference, fertilizer applicator injection flow rate, and distributed soil moisture and meteorological parameters. This real-time sensing data is injected into the digital mirror model, driving the model to simulate the dynamic hydraulic response of the irrigation network under the current equipment state, generating a full-system simulation state including pressure field, flow distribution, and water quality concentration field. Based on the full-system simulation state, the actual operating efficiency indicators of the irrigation network are calculated, and deviations are identified by comparing them with preset operating efficiency benchmarks. Multiple rounds of virtual control simulations with preset strategies are executed in the digital mirror model to select the optimal set of control strategies that meet the benchmark targets. Finally, this set is distributed to the integrated skid-mounted equipment for closed-loop management.

[0022] In one embodiment of the present invention, see [reference] Figure 2The system acquires raw equipment signals through hardware data acquisition interfaces deployed on pump motor drivers, electric regulating valve controllers, differential pressure sensor transmitters, fertilizer metering pump controllers, soil moisture sensor nodes, and weather stations, using a periodic polling and event-triggered mechanism. Signal quality assessment is performed on the raw equipment signals to detect abnormal data jumps, out-of-range values, and communication timeouts. Signals passing the quality assessment enter a standardized conversion process: current and voltage are converted into pump operating power and efficiency; valve opening feedback is mapped to equivalent flow area; filter inlet and outlet pressure difference is used to derive a clogging coefficient; fertilizer injector flow rate is converted into instantaneous fertilizer concentration; and soil moisture and meteorological parameters are converted into regional evapotranspiration water demand. Standardized data is aligned and packaged according to preset spatiotemporal labels to generate real-time sensing data packets with timestamps and geographic coordinates. The pump operating power and efficiency, valve equivalent flow area, and filter clogging coefficient from these data packets are used as boundary conditions to update the corresponding equipment attribute parameters in the digital mirror model. Based on the updated model, the physical control equations for water flow, mass conservation, and solute transport in the irrigation network are solved, and transient hydraulic and water quality coupled simulation calculations are performed. During the simulation, regional evapotranspiration water demand is used as a dynamic field water consumption term, and instantaneous fertilizer solution concentration is used as a solute source term and substituted into the control equations. After completing the transient calculation for a set duration, the pressure, flow rate, and concentration values ​​of all nodes and pipe sections in the model are extracted to generate pressure field distribution maps, flow rate distribution maps, and water quality concentration field distribution maps covering the entire irrigation area, which together constitute the simulation state of the entire system.

[0023] In practical implementation, this embodiment relies on a series of data acquisition interfaces deployed on-site in the irrigation area. A Modbus RTU communication module is installed on the water pump motor driver to read three-phase current and voltage. A Profibus-DP interface is provided on the electric regulating valve controller to obtain the valve opening percentage. A 4-20mA analog input channel is configured on the differential pressure sensor transmitter to collect the pressure difference between the filter inlet and outlet. A pulse counting function is integrated on the fertilizer metering pump controller to count the injection flow rate. The soil moisture sensor node uses the LoRa wireless transmission protocol to report volumetric water content. Temperature, humidity, and wind speed data are transmitted via an Ethernet interface at the weather station. These interfaces poll every 10 seconds and trigger immediate reporting when a sudden change in value exceeds a threshold, thereby acquiring raw equipment signals. In practice, the signal quality assessment module checks the acquired raw equipment signals. For example, a phase imbalance of more than 15% in the three-phase current is marked as a data jump, a valve opening feedback value greater than 100% is judged as a value exceeding the limit, and a sensor node without response for three consecutive sampling cycles is recorded as a communication timeout. The raw equipment signals that pass the quality assessment enter the standardization conversion process, including the three-phase current of the water pump. , , With line voltage , , Through formula Converted to operating power:

[0024] in: Represents the average line voltage. Represents the average phase current. The power factor is used; the 0-100% linear mapping of valve opening feedback is used as the equivalent flow area; the filter inlet and outlet pressure difference is divided by the initial clean pressure difference to obtain the clogging coefficient; the fertilizer injector flow rate is divided by the total flow rate of the main pipeline to calculate the instantaneous fertilizer concentration; soil moisture and meteorological parameters are converted into regional evapotranspiration water demand using the Penman-Monteith model. In specific implementation, the standardized and converted data are assigned a unified timestamp and WGS-84 geographic coordinates. For example, at a certain moment, the pump operating power is 55kW, efficiency is 0.88, the equivalent flow area of ​​the DN200 valve is 0.0314㎡, the filter clogging coefficient is 0.35, the instantaneous fertilizer concentration is 1.2g / L, and the regional evapotranspiration water demand is 3.5mm / d. These data are aligned according to the pipeline node ID and field number and packaged into a structured real-time sensing data package.

[0025] In some embodiments, the pump operating power and efficiency, valve equivalent flow area, and filter clogging coefficient from the real-time sensing data packet are used as boundary conditions and internal parameters, and directly written into the attribute fields of the corresponding equipment in the digital mirror model. For example, the rated power of the pump is updated to the current actual operating power, the opening-area relationship in the valve characteristic table is adjusted to the current equivalent flow area, and the filter resistance parameter is corrected by associating it with the clogging coefficient. In a specific implementation, the updated digital mirror model calls a transient hydraulic solver to solve for the pressure at the network nodes and the flow rate in the pipe section based on the principles of mass conservation and momentum conservation. At the same time, it couples the solute transport equation to calculate the propagation of fertilizer concentration. During the simulation calculation, the regional evapotranspiration water demand is extracted from the real-time sensing data packet as a dynamic field water consumption term and applied to the terminal field node. The instantaneous value of fertilizer concentration is extracted as a solute source term for the water source and injected into the starting end of the main pipeline. In the specific implementation, the simulation duration was set to 24 hours, with a time step of 1 minute. After the solution was completed, the pressure, flow, and water concentration values ​​of all 256 nodes and 312 pipe segments were extracted from the model to generate pressure field distribution maps, flow distribution maps, and water concentration field distribution maps covering the entire irrigation area. The pressure field distribution map shows that the pressure at the beginning of the main pipe is 0.45 MPa and the pressure at the end is 0.28 MPa. The flow distribution map shows that the flow rate distributed in each branch pipe is between 15-25 m³ / h. 3 The water quality concentration field distribution map shows that the inlet concentration of the field plot fluctuates between 1.05 and 1.32 g / L, which together constitute the simulation state of the entire system.

[0026] In one embodiment of the present invention, see [reference] Figure 3 The pressure values ​​of key water-using nodes are extracted from the pressure field distribution map of the entire system simulation, and the water supply pressure guarantee rate and pressure uniformity index are calculated as sub-indicators of water supply efficiency. The inlet flow rate values ​​of each field are extracted from the flow distribution map and compared with the corresponding crop water requirements to calculate the irrigation water quantity satisfaction and distribution uniformity index, as sub-indicators of irrigation efficiency. The inlet fertilizer concentration values ​​of each field are extracted from the water quality concentration field distribution map and compared with the pre-formulated fertilizer formula concentration to calculate the fertilizer uniformity and accuracy index, as sub-indicators of fertilizer efficiency. These three types of sub-indicators are integrated according to preset weights to obtain the current comprehensive operational efficiency index value of the irrigation district's water network. This index value is compared with the preset operational efficiency benchmark target value, and the difference is calculated as the identified deviation.

[0027] In practical implementation, pressure values ​​of key water-using nodes are extracted from the pressure field distribution map of the entire system simulation. For example, 12 farmland inlets at the end of the irrigation district are selected as key nodes, and the pressure sequence of each node is recorded during the simulation period. Based on these pressure data, the percentage of time when the node pressure is not lower than the design minimum pressure of 0.25 MPa is calculated, and the water supply pressure guarantee rate is 92%. At the same time, the ratio of the standard deviation to the mean of the pressure of all key nodes is calculated, and the pressure uniformity index is 0.18. These two values ​​are used as sub-indicators of water supply efficiency. In practical implementation, the flow rate values ​​of each field inlet are extracted from the flow distribution map of the entire system simulation. For example, the irrigation district is divided into 36 fields, and the average flow rate of each field inlet during the simulation period is 18 cubic meters per hour. The average flow rate of each field is compared with the corresponding crop water requirement of 20 cubic meters per hour, and the percentage of fields whose flow rate meets the crop water requirement is calculated, resulting in an irrigation water satisfaction rate of 83%. At the same time, the coefficient of variation of the inlet flow rate of all fields is calculated, and the distribution uniformity index is 0.22. These two values ​​are used as sub-indicators of irrigation efficiency. In practical implementation, the fertilizer concentration values ​​at the inlet of each field are extracted from the water quality concentration field distribution map of the whole system simulation. For example, if the preset fertilizer formula concentration is 1.0 g / L, the simulation results show that the inlet concentration of each field fluctuates between 0.95 and 1.08 g / L. The average absolute error between the inlet concentration of each field and the preset concentration is calculated to obtain the fertilizer accuracy index of 0.03 g / L. At the same time, the standard deviation of the inlet concentration of all fields is calculated to obtain the fertilizer uniformity index of 0.06 g / L. These two values ​​are used as sub-indicators of fertilizer efficiency.

[0028] In some embodiments, the water supply efficiency sub-indicators, irrigation efficiency sub-indicators, and fertilization efficiency sub-indicators are integrated according to preset weights, wherein the weight of water supply efficiency is set to 0.4, the weight of irrigation efficiency is set to 0.4, and the weight of fertilization efficiency is set to 0.2; the comprehensive operational efficiency index value of the current irrigation district water network is calculated using a linear weighted formula.

[0029] in: Represents the comprehensive operational efficiency index value. , , The weights corresponding to the effectiveness of water supply, irrigation, and fertilization are respectively. , , These are the normalized sub-indicators for water supply efficiency, irrigation efficiency, and fertilization efficiency. Normalization converts each sub-indicator into a relative value between 0 and 1. For example, a water supply pressure guarantee rate of 92% is normalized to 0.92, pressure uniformity of 0.18 is normalized to 0.82, irrigation water quantity satisfaction of 83% is normalized to 0.83, distribution uniformity of 0.22 is normalized to 0.78, fertilization accuracy of 0.03 is normalized to 0.97, and fertilization uniformity of 0.06 is normalized to 0.94. The weighted sum yields a comprehensive operational efficiency index of 0.85. In practice, this comprehensive operational efficiency index of 0.85 is compared with the preset operational efficiency benchmark target value of 0.90, and the difference is calculated to be -0.05. This difference is used as the identified deviation. It is understandable that the selection of key water-using nodes can be adjusted according to the actual layout of the irrigation area. For example, nodes connecting the main pipe and branch pipe can be included in the set of key nodes, or the zoning valves of large sprinkler irrigation areas can be used as key nodes, as long as they can reflect the pressure distribution characteristics of the pipeline network. It is also understandable that the normalization method can adopt linear scaling or piecewise function mapping. For example, when mapping the pressure uniformity index value to the 0-1 interval, the optimal uniformity of 0.1 can be set to correspond to a normalized value of 1.0, the worst uniformity of 0.3 can correspond to a normalized value of 0, and the intermediate values ​​can be linearly interpolated, as long as the various sub-indices are kept at the same order of magnitude for weighted integration.

[0030] In one embodiment of the present invention, the components of the deviation are analyzed to determine that the root cause of the failure to meet the comprehensive operational efficiency index belongs to one of the following categories: insufficient water supply efficiency, insufficient irrigation efficiency, or insufficient fertilization efficiency. Based on the primary cause of the deviation, corresponding candidate control strategies are retrieved from a preset strategy knowledge base, which is constructed from historical successful control cases, domain expert rules, and equipment operation constraints. If the primary cause is insufficient water supply efficiency, the candidate strategies focus on increasing pipeline pressure, including activating standby pumps, increasing the frequency of operating pumps, and closing downstream non-critical branch valves. If the primary cause is insufficient irrigation efficiency, the candidate strategies focus on optimizing flow distribution, including extending the rotation irrigation time in high water demand areas, adjusting the rotation irrigation group's opening sequence, and increasing the opening degree of branch valves in water-scarce areas. If the primary cause is insufficient fertilization efficiency, the candidate strategies focus on stabilizing fertilizer concentration, including calibrating the closed-loop control parameters of the fertilizer applicator's injection volume, switching to a standby fertilization channel, and adding a pressure stabilizing device upstream of the fertilizer applicator. Each group of candidate control strategies is sequentially loaded into a digital mirror model to simulate the state changes of the integrated skid-mounted equipment after execution. Based on the simulated equipment state, the model is re-driven to perform transient hydraulic and water quality coupled simulations, obtaining the new full-system simulation state corresponding to each strategy and calculating its new comprehensive operational efficiency index value. All new index values ​​are compared with the operational efficiency benchmark target, and strategies that enable the index values ​​to reach or exceed the benchmark target are selected and combined to form the optimal control strategy set.

[0031] In the specific implementation, the analysis module identified a deviation of -0.05 between the comprehensive operational efficiency index value of 0.85 and the benchmark target of 0.90. Further analysis of the deviation revealed that the irrigation water satisfaction rate was only 83%, significantly lowering the overall index. Therefore, the main factor was determined to be insufficient irrigation efficiency. Based on this primary factor, three sets of candidate control strategies were retrieved from the preset strategy knowledge base. This knowledge base stores historical irrigation optimization cases, rotational irrigation rules recommended by agronomic experts, and operational limits for water pumps and valves: The first set of candidate control strategies extended the rotational irrigation time in high-water-demand areas, increasing the rotational irrigation time in the corn-growing area from 4 hours to 5 hours; the second set adjusted the opening sequence of the rotational irrigation groups, prioritizing the opening of the western rotational irrigation groups located further from the water source; the third set increased the opening of the branch pipe valves supplying water to water-scarce areas, increasing the opening of the branch pipe valves supplying the southeast hilly area from 60% to 75%.

[0032] In practical implementation, the digital mirror model sequentially loads three sets of candidate control strategies to simulate changes in equipment status after execution: When loading the first set of strategies, the model updates the opening sequence of solenoid valves in the corn planting area while keeping the rotation irrigation plan for other areas unchanged; when loading the second set of strategies, the model adjusts the starting sequence of the rotation irrigation groups, advancing the opening of the solenoid valve group in the western area to the first round; when loading the third set of strategies, the model sets the opening degree of the branch pipe valve in the southeast hill area to 75%, maintaining the status of the main pipe valve. In practical implementation, based on the simulated equipment status, the digital mirror model is re-driven to perform transient hydraulic and water quality coupled simulation. The simulation duration for each set of strategies is 24 hours, and new comprehensive operational efficiency index values ​​are calculated, as shown in Table 1. Table 1: Comparison of Simulation Results of Candidate Regulation Strategies

[0033] In practical implementation, all new comprehensive operational efficiency index values ​​are compared with the operational efficiency benchmark target of 0.90. A new index value of 0.91 is selected to achieve the benchmark target for increasing the opening of branch pipe valves in water-scarce areas, and this value constitutes a separate set of optimal control strategies. In some embodiments, if deviation analysis determines that the main factor is insufficient water supply efficiency, such as a water supply pressure guarantee rate of only 82%, candidate strategies such as activating standby pumps and increasing the operating pump frequency are selected from the strategy knowledge base. If the main factor is insufficient fertilization efficiency, such as a fertilization uniformity index below 0.80, candidate strategies for calibrating the closed-loop control parameters of the fertilizer applicator are selected. In practical implementation, the domain expert rules in the strategy knowledge base define that when water supply efficiency is insufficient, adjusting the pump frequency is prioritized over closing valves. Equipment operation constraints limit the frequency of a single pump to no more than 10% of its rated value to prevent equipment overload. It is understood that the number of candidate control strategies can be adjusted according to the size of the strategy knowledge base. For example, more than five sets of candidate strategies can be set for complex irrigation areas, or only two sets of core strategies can be retained for small irrigation areas, as long as the control directions corresponding to the main influencing factors are covered. It is understandable that the criteria for forming the optimal set of control strategies can be configured. For example, it may be required that the new comprehensive operational efficiency index value exceeds the benchmark target by at least 0.02, or the strategy with the smallest adjustment range may be selected first, in order to adapt to different management preferences.

[0034] In one embodiment of the present invention, each strategy instruction in the optimal control strategy set is translated into a device-level standardized control command, including the target frequency and start / stop command for the water pump, the target opening degree command for the valve, the backwashing trigger signal for the filter, the target injection ratio and start / stop command for the fertilizer applicator, and the switching sequence of the field solenoid valve group. Through a secure industrial communication protocol, the standardized control commands are sequentially sent to the pump control unit, valve control unit, filter control unit, fertilizer control unit, and field valve controller at the irrigation site according to a preset execution sequence. After the commands are sent, the real-time sensing data returned from the site is continuously monitored to verify whether the standardized control commands are correctly executed and whether the equipment status meets the strategy expectations. The verified actual execution results are compared with the virtual control simulation prediction results in the digital mirror model to generate a strategy execution consistency report.

[0035] In specific implementation, this embodiment receives a single strategy command from the optimal control strategy set: "Increase the opening of the Southeast Gangdi branch pipe valve to 75%". The command translation module parses it into a standardized control command at the equipment level, specifying the target equipment as the Southeast Gangdi branch pipe electric regulating valve (equipment code VLV-SE-203), the command content as setting the opening value to 75%, and the effective time as the irrigation start time of the next day at 08:00. In specific implementation, the structure of the standardized control command follows the OPCUA information model, containing four types of fields: equipment identifier, parameter type, set value, and timestamp. For example, the equipment identifier is "ns=3;s=VLV_SE_203_Position", the parameter type is "Double", the set value is 75.0, and the timestamp is "2025-03-15T08:00:00Z". In practical implementation, standardized control commands are sent to the irrigation area's field equipment according to a preset execution sequence via the MQTToverTLS industrial communication protocol based on certificate authentication: a command is sent to the edge gateway at 07:59:50, and after verifying the device's permissions and parameter range, the gateway forwards it to the valve control unit of the Southeast Gangdi branch pipe precisely at 08:00:00. After the command is issued, the monitoring system continuously collects real-time sensing data returned from the field to verify the execution results: the valve opening sensor feedback value reached 74.8% at 08:01:30, with an absolute error of 0.2% from the target value of 75%, within the allowable tolerance range of ±1%; the downstream pressure sensor reading of the branch pipe increased from 0.26 MPa to 0.31 MPa, and the flow meter showed an increase in flow rate from 18 cubic meters per hour to 23 cubic meters per hour, consistent with the hydraulic response law after the valve opening increases, confirming that the standardized control command was correctly executed and the equipment status met the strategy expectations.

[0036] In practical implementation, the verified actual execution results are compared with the virtual control simulation prediction results in the digital mirror model to generate a strategy execution consistency report: the consistency between the actual valve opening of 74.8% and the simulation prediction of 75% is 99.73%; the relative deviation between the actual branch pipe pressure of 0.31MPa and the simulation prediction of 0.305MPa is 1.64%; and the relative deviation between the actual flow rate of 23 cubic meters per hour and the simulation prediction of 23.5 cubic meters per hour is 2.13%. The consistency quantification uses the following formula:

[0037] in: This represents the percentage of similarity for the k-th type of parameter. The number of sampling points. This represents the actual monitored value at the i-th sampling point. These are the simulated predicted values ​​for the corresponding sampling points. This refers to the range of this type of parameter (e.g., valve opening range of 0-100%); the calculation results are summarized in Table 2: Table 2: Strategy Implementation Conformity Report Data Table

[0038] In some embodiments, the standardized control command issuance protocol can be adapted to the industrial standards supported by the field equipment. For example, the Modbus TCP protocol can be used to send register write instructions to a traditional PLC controller, or the PROFINET protocol can be used to send analog setting signals to a smart valve positioner, as long as the secure and reliable transmission of equipment control commands is achieved. In specific implementations, the execution timing control can be configured to immediate execution or scheduled execution mode. For example, for emergency pressure regulation needs, an immediate execution command can be issued directly, while for periodic irrigation plans, a scheduled command can be issued 24 hours in advance to adapt to the needs of different regulation scenarios.

[0039] It is understandable that the calculation cycle for strategy execution consistency can be adjusted according to the speed of the control response. For example, the consistency calculation for valve opening adjustment uses a 1-minute high-frequency sampling, while the consistency calculation for fertilizer concentration changes uses a 5-minute low-frequency sampling, to match the dynamic characteristics of different processes. It is also understandable that the consistency report can be presented in a format that can be expanded into visual charts or structured log files. For example, it can generate time-series comparison curves to show the actual and simulated pressure change trends, or export JSON format reports for use by upper-level systems to meet different analytical needs.

[0040] In one embodiment of the present invention, when the strategy execution consistency report shows that the actual pressure, flow rate, or concentration changes differ from the simulation prediction values ​​that continuously exceed the allowable range, it is determined that the digital mirror model has model errors. Real-time sensing data and internal simulation calculation parameters of the corresponding time period are extracted during the period of difference. The source of the error is analyzed to determine if it falls into a specific category, such as inaccurate pipeline resistance parameters, pump characteristic curve drift, valve flow characteristic deviation, or inaccurate soil infiltration parameters. If the error originates from pipeline resistance parameters, the difference between the actual monitored pipeline pressure drop and the simulated pressure drop is used to inversely update the friction coefficient and local resistance coefficient of the corresponding pipe section in the model using an iterative optimization algorithm. If the error originates from pump characteristic curve drift, the actual operating point of the pump is fitted with the sample curve to generate the current actual head-flow characteristic curve and update the pump performance data in the model. If the error originates from valve characteristic deviation, the valve flow coefficient curve of the model is corrected based on the actual flow rate and simulated flow rate under different valve openings. If the error originates from the field water consumption model, the crop coefficient or soil hydrodynamic parameters of the corresponding field are adjusted using measured soil moisture changes and model-calculated water consumption.

[0041] In practice, the strategy execution consistency report showed that during three consecutive control cycles, the actual monitored pressure drop of the L3 pipe section on the east side of the irrigation area was consistently higher than the simulated predicted value, with an average deviation of 0.028 MPa, continuously exceeding the allowable tolerance of ±0.015 MPa. Based on this, it was determined that the digital mirror model had model errors. In practice, real-time sensing data was extracted for the time periods of difference, including pressure sensors recording 0.52 MPa and 0.48 MPa upstream and downstream of the L3 pipe section, flow meters recording 32 cubic meters per hour, and simulation calculation parameters of the digital mirror model within the corresponding time periods, including calculated pressure values ​​of 0.535 MPa and 0.51 MPa, and flow rate of 33 cubic meters per hour. In practice, the source of model error was analyzed: comparing the actual and simulated pressure drop differences ruled out the influence of pump and valve characteristics. Since the pressure consistency of other pipe sections was within the allowable range, and no abnormalities were found in soil moisture monitoring, the error was identified as originating from inaccurate pipeline resistance parameters of the L3 pipe section.

[0042] In practical implementation, to address the errors caused by pipeline resistance parameters, the difference between the actual monitored pressure drop of 0.04 MPa and the simulated pressure drop of 0.025 MPa in the L3 pipe section is used to employ an iterative optimization algorithm to invert and update the friction coefficient and local resistance coefficient of the L3 pipe section in the digital mirror model. The update process is based on the pressure drop balance formula:

[0043] in: For the pressure drop of the pipeline section, This is the friction coefficient. The pipe section is 150 meters long. The pipe diameter is 0.2 meters. The flow velocity is 2.83 meters per second. The acceleration due to gravity is 9.8 meters per second squared. This represents the local drag coefficient. Substituting the actual pressure drop of 0.04 MPa into the formula, and adjusting it iteratively... and The value of was adjusted to make the calculated pressure drop closer to the actual value, and the friction coefficient was updated from 0.018 to 0.023, and the local resistance coefficient was updated from 0.8 to 1.2.

[0044] In some embodiments, if the error originates from pump characteristic curve drift, for example, if the actual pump head of 38 meters at a frequency of 45Hz is lower than the sample curve head of 42 meters, then the actual operating point (flow rate of 50 cubic meters per hour, head of 38 meters) is fitted with multiple points of the sample curve to generate the current actual pump head-flow characteristic curve, replacing the original performance curve data in the digital mirror model. In specific implementations, if the error originates from valve characteristic deviation, for example, if the deviation between the actual flow rate of 120 cubic meters per hour and the simulated flow rate of 130 cubic meters per hour when the DN300 valve is open at 50% persists, then the flow coefficient curve of the valve in the digital mirror model is corrected based on the actual flow test data at the five valve opening positions (30%, 40%, 50%, 60%, 70%).

[0045] It is understandable that the error calibration of the field water consumption model can be performed using soil moisture data at different crop growth stages. For example, the difference between the measured daily average decrease of 0.8% in soil moisture during the wheat jointing stage and the model's calculated decrease of 1.0% can be used to adjust the crop coefficient of the corresponding field in the digital mirror model, rather than being limited to a single calibration method. It is also understandable that the frequency of reverse calibration of model parameters can be adjusted according to the persistence of the error. For example, short-term deviations caused by occasional sensor noise may not trigger calibration, while systematic deviations for more than three consecutive periods will immediately initiate the update process.

[0046] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A simulation method for irrigation district water resource management based on digital twin technology, characterized in that, include: Establish a digital mirror model of the physical entity of the irrigation area, the digital mirror model including the pipeline network topology, equipment attribute parameters and spatial distribution information; Data is synchronously and in real time sensed from the integrated skid-mounted equipment at the irrigation area site. The real-time sensed data includes the current and voltage of the water pump, the valve opening feedback, the inlet and outlet pressure difference of the filter, the injection flow of the fertilizer applicator, and the soil moisture and meteorological parameters distributed in the irrigation area. The real-time sensing data is injected into the digital mirror model, which drives the digital mirror model to simulate the dynamic hydraulic response of the irrigation area water network under the current equipment state, generating a full system simulation state including pressure field, flow distribution, and water quality concentration field. Based on the simulation state of the entire system, the actual operating efficiency index of the current irrigation district water network is calculated and compared with the preset operating efficiency benchmark target to identify the deviation. Based on the identified deviations, multiple rounds of virtual control simulation with preset strategies are performed in the digital mirror model, and the optimal set of control strategies that meet the operational performance benchmark target is selected from the results of the multiple rounds of virtual control simulation. The optimal control strategy set is distributed to the integrated skid-mounted equipment at the irrigation area site for execution, completing a closed-loop management cycle.

2. The simulation method for irrigation district water resource management based on digital twin technology according to claim 1, characterized in that, The data synchronously and in real-time sensed from the integrated skid-mounted equipment at the irrigation area site includes: By using data acquisition interfaces deployed on water pump motor drivers, electric regulating valve controllers, differential pressure sensor transmitters, fertilizer metering pump controllers, soil moisture sensor nodes, and weather stations, raw equipment signals are acquired through a combination of periodic polling and event triggering. The original device signal is subjected to signal quality assessment, which includes checking for data jumps, out-of-bounds values, and communication timeouts. The original equipment signals that have passed the quality assessment are standardized and converted, converting current and voltage into pump operating power and efficiency, valve opening feedback into equivalent flow area, filter inlet and outlet pressure difference into clogging coefficient, fertilizer injector flow rate into instantaneous fertilizer concentration, and soil moisture and meteorological parameters into regional evapotranspiration water demand. The standardized data is aligned and packaged according to preset spatiotemporal labels to form a real-time sensing data package with timestamps and geographic coordinates.

3. The simulation method for irrigation district water resource management based on digital twin technology according to claim 2, characterized in that, The driving digital mirror model simulates the dynamic hydraulic response of the irrigation network under the current equipment state, generating a full-system simulation state including pressure field, flow distribution, and water quality concentration field, including: The operating power and efficiency of the water pump, the equivalent flow area of ​​the valve, and the clogging coefficient of the filter in the real-time sensing data packet are used as boundary conditions and internal parameters to update the attribute parameters of the corresponding devices in the digital mirror model. Based on the updated digital mirror model, the physical control equations describing water flow, mass conservation and solute transport in the irrigation network are solved, and the coupled simulation calculations of transient hydraulics and water quality are carried out. In the simulation calculation process, the regional evapotranspiration water demand in the real-time sensing data packet is taken as a dynamic field water consumption term, and the instantaneous value of fertilizer solution concentration is taken as a solute source term of water source, and substituted into the physical control equation. After completing the transient simulation calculation for a preset duration, the pressure, flow, and water concentration values ​​of all nodes and pipe sections in the digital mirror model are extracted to form a pressure field distribution map, flow distribution map, and water concentration field distribution map covering the entire irrigation area, which together constitute the simulation state of the entire system.

4. The simulation method for irrigation district water resource management based on digital twin technology according to claim 1, characterized in that, Based on the simulation of the entire system, the actual operational efficiency indicators of the current irrigation district water network are calculated and compared with the preset operational efficiency benchmark target to identify deviations, including: From the pressure field distribution map of the entire system simulation state, the pressure values ​​of key water-using nodes are extracted, and the water supply pressure guarantee rate and pressure uniformity index are calculated as sub-indicators of water supply efficiency. From the flow distribution map of the entire system simulation state, extract the flow value of each field inlet, compare it with the crop water requirement of the corresponding field, and calculate the irrigation water satisfaction and distribution uniformity index as the irrigation efficiency sub-indicators. From the water quality concentration field distribution map of the entire system simulation state, extract the fertilizer concentration value at the inlet of each field, compare it with the preset fertilizer formula concentration, and calculate the fertilizer uniformity and accuracy index as a sub-index of fertilizer efficiency. The water supply efficiency sub-indicators, irrigation efficiency sub-indicators, and fertilization efficiency sub-indicators are integrated according to preset weights to calculate the comprehensive operational efficiency index value of the current irrigation district water network. The comprehensive operational efficiency index value is compared with the preset operational efficiency benchmark target value, the difference is calculated, and the difference is used as the deviation.

5. The simulation method for irrigation district water resource management based on digital twin technology according to claim 4, characterized in that, Based on the identified deviations, multiple rounds of virtual control simulations with preset strategies are performed in the digital mirror model. From the results of these multiple rounds of virtual control simulations, the optimal set of control strategies that meet the operational performance benchmark is selected, including: Analyzing the composition of the aforementioned deviations, it was determined that the main factors leading to the failure of the comprehensive operational efficiency index to meet the standards were insufficient water supply efficiency, insufficient irrigation efficiency, or insufficient fertilization efficiency. Based on the main factors, one or more candidate control strategies are selected from the preset strategy knowledge base. The candidate control strategies include adjusting the pump speed combination, adjusting the valve opening combination, starting the filter backwashing program, modifying the fertilizer applicator injection ratio, and switching the field irrigation group. In the digital mirror model, each group of candidate control strategies is loaded sequentially, and the state changes of the integrated skid-mounted device after executing the candidate control strategies are simulated. Based on the equipment state after the simulation execution, the digital mirror model is re-driven to perform coupled simulation calculations of transient hydraulics and water quality, to obtain the new full-system simulation state corresponding to each group of candidate control strategies, and to calculate its new comprehensive operating efficiency index value. The new comprehensive operational efficiency index values ​​corresponding to all candidate control strategies are compared with the operational efficiency benchmark target. Strategies that enable the new comprehensive operational efficiency index values ​​to reach or exceed the benchmark target are selected, and these strategies are combined to form the optimal control strategy set.

6. The simulation method for irrigation district water resource management based on digital twin technology according to claim 5, characterized in that, The step of selecting one or more candidate control strategies from a preset strategy knowledge base includes: The strategy knowledge base is constructed from historical successful control cases, domain expert rules, and equipment operation constraints. When the main factor is insufficient water supply efficiency, the selected candidate control strategies focus on increasing the pressure of the pipeline network, including activating standby pumps, increasing the frequency of operating pumps, and closing downstream non-critical branch valves. When the main factor is insufficient irrigation efficiency, the selected candidate control strategies focus on optimizing flow distribution, including extending the rotation irrigation time in high water demand areas, adjusting the opening sequence of the rotation irrigation group, and increasing the opening degree of the branch pipe valves supplying water to water-scarce areas. When the main factor is insufficient fertilization efficiency, the selected candidate control strategies focus on stabilizing fertilizer solution concentration, including calibrating the closed-loop control parameters of the fertilizer applicator's injection volume, switching to the backup fertilization channel, and adding a pressure stabilizing device to the upstream pipeline of the fertilizer applicator.

7. The simulation method for irrigation district water resource management based on digital twin technology according to claim 1, characterized in that, The optimal control strategy set is distributed to the integrated skid-mounted equipment at the irrigation area for execution, including: Each strategy instruction in the optimal control strategy set is translated into a standardized control command that can be executed by specific equipment. The standardized control command includes the target frequency and start / stop of the water pump, the target opening degree of the valve, the backwash trigger signal of the filter, the target injection ratio and start / stop of the fertilizer applicator, and the group switching sequence of the field solenoid valve. Through a secure industrial communication protocol, the standardized control commands are sequentially sent to the pump control unit, valve control unit, filter control unit, fertilizer control unit, and field valve controller at the irrigation area according to a preset execution sequence. After the command is issued, the real-time sensing data returned from the irrigation area is continuously monitored to verify whether the standardized control command is executed correctly and whether the equipment status meets the strategy expectations. The verified actual execution results are compared with the virtual control simulation prediction results in the digital mirror model to generate a strategy execution consistency report.

8. The simulation method for irrigation district water resource management based on digital twin technology according to claim 7, characterized in that, The method also includes correcting the digital mirror model based on the policy execution consistency report, including: When the strategy execution consistency report shows that the actual pressure, flow rate or concentration changes differ from the simulation prediction values ​​in a continuous manner and exceed the allowable range, it is determined that the digital mirror model has model errors. Extract the real-time sensing data during the period in which the difference occurs, and the internal simulation calculation parameters of the digital mirror model during the corresponding period; The source of the model error was analyzed to determine whether it stemmed from inaccurate pipeline resistance parameters, drift of pump characteristic curves, deviation of valve flow characteristics, or inaccurate soil infiltration parameters. Based on the error analysis results, the corresponding equipment attribute parameters, pipeline topology parameters, or field water consumption model parameters in the digital mirror model are reverse-calibrated and updated to make the model output closer to the physical entity response.

9. The simulation method for irrigation district water resource management based on digital twin technology according to claim 8, characterized in that, Reverse calibration and updating of the corresponding equipment attribute parameters, pipeline topology parameters, or field water consumption model parameters in the digital mirror model include: If the error originates from the pipeline resistance parameters, the difference between the actual monitored pipeline pressure drop and the simulated pressure drop is used to invert and update the friction coefficient and local resistance coefficient of the corresponding pipe segment in the digital mirror model using an iterative optimization algorithm. If the error originates from the drift of the pump characteristic curve, the actual operating point of the pump is fitted with the sample curve to generate the current actual pump head-flow characteristic curve, and the pump performance curve data in the digital mirror model is updated accordingly. If the error originates from valve characteristic deviation, then the flow coefficient curve of the valve in the digital mirror model is corrected based on the actual flow rate and simulated flow rate at different valve openings. If the error originates from the field water consumption model, the crop coefficient or soil hydrodynamic parameters of the corresponding field in the digital mirror model are adjusted by using the measured changes in soil moisture and the soil moisture consumption calculated by the model.

10. A simulation system for irrigation district water resource management based on digital twin technology, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the irrigation district water resource management simulation method based on digital twin technology as described in any one of claims 1 to 9.

Images