Drainage basin water resource dynamic scheduling optimization method and system based on digital twinning

Abstract

The invention relates to the technical field of water resource scheduling, in particular to a drainage basin water resource dynamic scheduling optimization method and system based on digital twinning, and the method comprises the steps: obtaining drainage basin static geographic data and real-time monitoring data, and constructing a digital twinning model; identifying a target water supply node according to the real-time monitoring data and a preset scheduling target; simulating in the digital twinborn model to obtain water quality evolution time sequence data of new and old water sources; determining first prediction time and second prediction time according to the water quality evolution time sequence data of the new and old water sources, and determining a dynamic switching window; calculating a time sequence consistency evaluation index; and outputting the dynamic switching window and the time sequence consistency evaluation index as a decision basis for optimizing a water source scheduling scheme. According to the method, accurate quantification of the water source switching time sequence decoupling risk is realized, a reliable basis is provided for scheduling decision, and the risks of water supply quality reduction and scheduling failure caused by time sequence asynchronization are effectively avoided.

Description

Technical Field

[0001] This invention relates to the field of water resource scheduling technology, specifically to a method and system for dynamic scheduling and optimization of water resources in a basin based on digital twins. Background Technology

[0002] With the increasing demand for water resource allocation in river basins, digital twin technology has been applied to construct virtual-real river basin simulation systems to achieve dynamic monitoring and prediction of water flow processes. However, existing methods mainly focus on water allocation and balance, assisting personnel in formulating water supply plans by synchronizing water source supply and pipeline distribution status in real time, and judging the completion of switching mainly relies on changes in water status.

[0003] However, in actual scheduling, water supply nodes often need to switch between different water sources. During the switching process, water volume scheduling often changes rapidly, while water quality replacement has a lag, and the old and new water sources will advance and mix in the water transmission path. If the timing of the switch completion is still judged based on the water volume scheduling results, it is easy for scheduling records to show that water is being supplied by the new water source, while the water quality at the water supply node is still mainly determined by the old water source, resulting in distorted scheduling status and unstable water supply quality.

[0004] Therefore, there is an urgent need for a method and system for dynamic scheduling and optimization of water resources in a basin based on digital twins, which can combine digital twin technology to simulate the time-series evolution of water quality during the water source switching process, and provide a basis for judging the merits of different scheduling schemes. Summary of the Invention

[0005] (1) Technical problems to be solved

[0006] The purpose of this invention is to provide a method and system for dynamic scheduling optimization of water resources in a basin based on digital twins, so as to make up for the shortcomings of existing scheduling that rely only on water quantity criteria and lack water quality time-series consistency evaluation.

[0007] (2) Technical solution

[0008] To achieve the above objectives, on the one hand, the present invention provides a method for dynamic scheduling and optimization of watershed water resources based on digital twins, the method comprising:

[0009] Step S1: Obtain static geographic data of the watershed, including water conveyance path topology, water conservancy project nodes and water supply nodes; construct a digital twin model based on the static geographic data of the watershed; obtain real-time monitoring data of each water source and the water supply nodes in the watershed, the real-time monitoring data including water quantity, water level and water quality indicators.

[0010] Step S2: Based on the real-time monitoring data and the preset scheduling target, identify the target water supply node with water source switching requirements through the digital twin model; simulate the evolution of the water quality characteristics of the old and new water sources of the target water supply node over time under the water source switching condition in the digital twin model to obtain the time series data of the water quality evolution of the old and new water sources of the target water supply node; determine the first predicted time when the water quality front of the new water source reaches the target water supply node and the second predicted time when the water quality tail of the old water source leaves the target water supply node based on the time series data of the water quality evolution of the old and new water sources.

[0011] Step S3: Determine the dynamic switching window of the target water supply node based on the first prediction time and the second prediction time; calculate the time consistency evaluation index of the target water supply node based on the time series deviation between the water quality front and the water quality tail; output the dynamic switching window and the time series consistency evaluation index as the decision basis for optimizing the water source scheduling scheme.

[0012] Furthermore, the method for constructing a digital twin model based on the static geographic data of the watershed includes:

[0013] The topology of the water conveyance path is discretized to obtain a set of nodes and a set of pipe segments. A hydraulic calculation network is established based on the set of nodes and the set of pipe segments. Water quality characteristics are set as state variables in the set of nodes and the set of pipe segments, and a mass conservation relationship for the transmission of water quality characteristics along the flow direction in the water conveyance path is established based on the convection-diffusion equation. The hydraulic calculation network and the mass conservation relationship are integrated to form a digital twin model.

[0014] Furthermore, the method for simulating the evolution of water quality characteristics of the old and new water sources of the target water supply node over time under the water source switching condition in the digital twin model, and obtaining the time series data of water quality evolution of the old and new water sources of the target water supply node, includes:

[0015] Based on real-time monitoring data before water source switching, an initial field of water quality characteristics for each node and pipe segment is set in the hydraulic calculation network; when the digital twin model simulation starts, the boundary conditions of the water quality characteristics of the new water source inlet node to be switched are updated from a first constant value representing the concentration of the old water source to a second constant value representing the concentration of the new water source.

[0016] Based on the updated boundary conditions, a real-time hydraulic gradient provided by the hydraulic calculation network is applied to the mass conservation relationship of the water quality characteristics. The simulation time step is gradually advanced using a numerical discrete solution method to calculate the concentration values ​​of water quality characteristics of each node and pipe segment in the hydraulic calculation network.

[0017] Within each simulation time step, the concentration values ​​of water quality characteristics of the new and old water sources of the target water supply node and the corresponding simulation time are recorded, and the data are arranged in chronological order to obtain the time series data of water quality evolution of the new and old water sources, respectively representing the concentration change process of the new and old water sources.

[0018] Furthermore, the method for updating the boundary conditions of the water quality characteristic quantity of the new water source inlet node to be switched from a first constant value representing the concentration of the old water source to a second constant value representing the concentration of the new water source includes:

[0019] The nominal water quality characteristic concentration of the new water source is obtained as a second constant value; the sequence of operating instructions for the hydraulic control equipment controlling the inflow of the new water source during the switching period is obtained; and a water source switching progress function characterizing the water source switching progress is determined based on the sequence of operating instructions. ,in, Monotonically increasing from 0 to 1, This indicates the simulation time of the digital twin model.

[0020] During the digital twin model simulation, the water quality characteristic boundary conditions of the new water source inlet node are applied during the water source switching period. Set as ;in, The first constant value, The boundary condition is set to a second constant value; before the water source switching period, the boundary condition is maintained at a first constant value, and after the water source switching period, the boundary condition is maintained at a second constant value.

[0021] Furthermore, the method for determining the first predicted time when the water quality front of the new water source reaches the target water supply node and the second predicted time when the water quality tail of the old water source leaves the target water supply node based on the time series data of the evolution of water quality from the new and old water sources includes:

[0022] Extract the time series of the new water source quality at the target water supply node from the time series data of the evolution of water quality between the old and new water sources. Time series of water quality of old water sources ; Obtain the preset qualified concentration threshold for new water sources .

[0023] The new water source water quality time series In the middle, when from a point in time A continuous period of time begins to exist. This makes it possible for any time All And the length of the continuous time period is not less than the preset minimum stable duration. At that time, the time point mentioned As the first prediction time.

[0024] The old water source water quality time series In the middle, when from a point in time A continuous period of time begins to exist. For any time , All conditions meet the preset influence concentration requirements, and the length of the continuous period is not less than the preset minimum stability duration. At that time, the time point mentioned As the second prediction time.

[0025] Furthermore, the preset new water source qualified concentration threshold The preset minimum stable duration is adaptively determined based on the water usage attributes of the target water supply node.

[0026] Furthermore, the method for calculating the temporal consistency evaluation index of the target water supply node based on the temporal deviation between the water quality front and the water quality tail includes:

[0027] According to the first predicted time Second prediction time Calculate the absolute timing deviation ; Obtain the baseline switching time corresponding to the water usage attribute of the target water supply node. ; Calculate the water quality sensitivity weighting coefficient based on water quality differences ;

[0028] According to the absolute timing deviation Reference switching time Water quality sensitivity weighting coefficient The time series consistency evaluation index was calculated. .

[0029] Furthermore, the calculation of water quality sensitivity weighting coefficients based on water quality differences... The methods include:

[0030] Identify a set of key water quality difference indicators between the old and new water sources, including turbidity, pH value, and total dissolved solids content; assign a corresponding basic weight value to each water quality difference indicator in the set of key water quality difference indicators based on the water use attribute of the target water supply node.

[0031] Based on the concentration difference between the old and new water sources for each water quality difference indicator, the difference contribution is calculated using a preset piecewise function. The basic weight value of each water quality difference indicator is multiplied by its corresponding difference contribution, and then summed to obtain a comprehensive water quality difference score. Finally, the comprehensive water quality difference score is mapped to the [0,1] interval using an S-shaped function to obtain the water quality sensitivity weight coefficient. .

[0032] Based on the same inventive concept, the present invention also provides a watershed water resources dynamic scheduling and optimization system based on digital twins, the system being used to execute the watershed water resources dynamic scheduling and optimization method based on digital twins.

[0033] (3) Beneficial effects

[0034] Compared with existing technologies, the beneficial effect of this invention is that by constructing a digital twin model that couples hydrodynamic and water quality transport mechanisms, it simulates and predicts the spatiotemporal evolution of the water quality front and the water quality tail of the old water source during water source switching, thus achieving precise quantification of the risk of temporal decoupling. By outputting dynamic switching windows and temporal consistency evaluation indicators, it provides dispatchers with intuitive and reliable decision-making basis, thereby avoiding risks such as local water supply interruptions, water quality degradation, and distortion of ecological water replenishment effects caused by temporal asynchrony. Attached Figure Description

[0035] Figure 1 This is a flowchart of the dynamic scheduling and optimization method for watershed water resources based on digital twins according to Embodiment 1 of the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Before providing examples, it's necessary to describe the application scenarios of this invention. In multi-source joint scheduling scenarios, scheduling strategies typically require dynamic flow allocation among different water sources to simultaneously meet water supply security, cost, and ecological constraints. However, the execution of water volume regulation actions at the control center does not necessarily mean that the water quality characteristics at the water supply nodes will change synchronously immediately. Due to the complexity of water transmission and distribution paths, long water ages, and the continuous characteristics of multi-source mixing, the gradual replacement of old water bodies by new water bodies along the transmission path inevitably involves a process of advancement and mixing, leading to a time-dimensional deviation between water volume scheduling response and water quality replacement response. During the stage of accumulated deviation, although the water supply at different nodes conforms to the scheduling command, the water quality attributes are still affected by the residual old water, resulting in a temporal misalignment state inconsistent with the scheduling objectives. When the scheduling strategy is executed at the wrong time or deemed complete, it may cause local areas to remain in a suboptimal state for a long time, such as the mixed water bodies staying at the water supply nodes for too long, delayed replacement completion leading to reduced resource utilization efficiency, and even causing a decline in process performance and a deterioration in user experience. Therefore, given the availability of multiple scheduling strategies, there is an urgent need to construct a mechanism that can predict the temporal consistency between different scheduling actions and water body replacement processes before execution, identify time periods that do not meet the replacement conditions, avoid generating scheduling schemes that are physically impossible to implement or lack operational advantages, and provide quantifiable indicators for screening and optimization.

[0038] Example 1: As Figure 1 As shown in the figure, this embodiment provides a method for dynamic scheduling and optimization of watershed water resources based on digital twins. The method includes:

[0039] Step S1: Obtain static geographic data of the watershed, including water conveyance path topology, water conservancy project nodes and water supply nodes; construct a digital twin model based on the static geographic data of the watershed; obtain real-time monitoring data of each water source and the water supply nodes in the watershed, the real-time monitoring data including water quantity, water level and water quality indicators.

[0040] Step S2: Based on the real-time monitoring data and the preset scheduling target, identify the target water supply node with water source switching needs through the digital twin model; simulate the evolution of water quality characteristics of the old and new water sources of the target water supply node over time under the water source switching condition in the digital twin model to obtain the time series data of water quality evolution of the old and new water sources of the target water supply node; determine the first predicted time when the water quality front of the new water source reaches the target water supply node and the second predicted time when the water quality tail of the old water source leaves the target water supply node based on the time series data of water quality evolution of the old and new water sources; the water quality front specifically refers to the leading edge of the propagation of water quality characteristics of the new water source in the water transmission path during the water source switching process, and the water quality tail specifically refers to the trailing edge where the influence of water quality characteristics of the old water source fades.

[0041] Step S3: Determine the dynamic switching window of the target water supply node based on the first prediction time and the second prediction time; calculate the time consistency evaluation index of the target water supply node based on the time series deviation between the water quality front and the water quality tail; output the dynamic switching window and the time series consistency evaluation index as the decision basis for optimizing the water source scheduling scheme.

[0042] For example, taking a certain watershed A as an example, its water sources include surface water from the upstream hilly area, groundwater from the plain area, and a reclaimed water source. Watershed A is equipped with several pumping stations, gates, and regulating reservoirs, supplying water to multiple water supply nodes, including urban domestic water users, industrial cooling water users, and agricultural irrigation water users. The water conveyance path topology, water conservancy project nodes, and water supply nodes of watershed A are obtained. The water conveyance path topology describes the direction and connection relationships of the water conveyance paths from each water source to each water conservancy project node, and from each water conservancy project node to each water supply node, including the start and end node relationships and direction information of each segment of the water conveyance path. Water conservancy project nodes include pumping station nodes, pressure regulating gate nodes, distribution gate chamber nodes, and regulating reservoir nodes deployed along the water conveyance path, used to indicate the control and regulation positions during the water conveyance process. Water supply nodes include urban domestic water nodes, industrial cooling water nodes, and agricultural irrigation water nodes, used to indicate the water intake locations for different water uses.

[0043] A digital twin model is constructed, including a digital representation of the water conveyance path topology. Corresponding model nodes are established for each hydraulic engineering node and water supply node, and corresponding model pipe segments are established for each water conveyance path segment. Static parameters such as length, elevation, control method, and design flow range are configured for each node and pipe segment. The digital twin model achieves a virtual mapping of the actual water conveyance state of basin A through hydraulic calculations and water quality transport calculations. Then, the water quantity, water level, and water quality indicators at each water source and water supply node within the basin are acquired. Monitoring data for each water source includes intake flow rate, intake water level, and water quality indicators, including turbidity, total dissolved solids, nitrate concentration, and residual chlorine. Monitoring data for each water supply node includes node inflow rate, node pressure or water level, and similar water quality indicators. The monitoring data comes from online monitoring equipment and is provided to the digital twin model as real-time input.

[0044] The preset scheduling objective is to achieve a rational allocation of various water sources while ensuring the water quality requirements corresponding to different water uses. This avoids the long-term use of water sources with better water quality for purposes with lower water quality requirements, and also avoids the misuse of water sources with specific water quality risks for purposes sensitive to water quality. In particular, for urban domestic water supply nodes, it is desirable to use water sources with more suitable drinking water quality as much as possible, while prioritizing the allocation of water sources with larger quality fluctuations or higher salinity to industrial cooling or agricultural irrigation. Therefore, the scheduling objective not only involves water quantity balance but also requires ensuring that the water quality status of new and old water sources is not mixed for the same purpose over a long period. The digital twin model analyzes the supply and demand matching and water quality suitability of each water supply node based on current water quantity, water level, and water quality indicator monitoring data, combined with the preset scheduling objective. When the model predicts that within a certain time window, if water supply node X continues to use groundwater, there will be insufficient water supply or the water quality will no longer be suitable for its drinking water use, while the surface water source can meet the needs of water supply node X in terms of both quantity and quality, the model will identify that water supply node X has a need to switch from groundwater source to surface water source, and will designate water supply node X as the target water supply node.

[0045] For the water source switching at the target water supply node X, an operational plan was set up to switch from groundwater to surface water, and the switching conditions were simulated in a digital twin model. During the simulation, the model tracked the water quality characteristics of the groundwater and surface water sources along the water conveyance path over time and the mixing process, obtaining the time-varying results of the water quality characteristics of the new water source (surface water source) and the old water source (groundwater source) at water supply node X, thus forming the time-series data of the water quality evolution of the new and old water sources at the target water supply node. Based on the time-series data of the water quality evolution of the new and old water sources, the time when the water quality characteristics of the new water source first reach and continuously meet the drinking water requirements of water supply node X was identified as the first prediction time, and the time when the water quality characteristics of the old water source decayed to the point where their impact on the drinking water use of water supply node X could be ignored and remained stable was identified as the second prediction time. In the simulation, the water quality characteristics of the new and old water sources can be tracked separately as conservative tracers.

[0046] After obtaining the first and second predicted times, directly using the assumption that the water source before a certain point is the old source and the water source after that point is the new source as the basis for switching decisions will ignore factors such as pressure regulation, valve action delays, and water mixing buffers during actual operation. This can easily lead to inconsistencies between scheduling records and actual user perception. Therefore, a more robust time interval, determined through comprehensive evaluation, is selected between the first and second predicted times as the dynamic switching window. This window indicates that issuing water source switching-related scheduling commands within this time interval is more likely to achieve results highly consistent with the water quality replacement process. For example, if the model calculates the first predicted time as 10:30 on a certain day and the second predicted time as 13:00 on the same day, the dynamic switching window can be determined as the time interval from 11:00 to 12:30 on the same day. Performing the switching operation within the dynamic switching window can avoid the problem of water quality not being stable when switching too early, and also avoid the problem of prolonged mixing due to switching too late.

[0047] While determining the dynamic switching window, the temporal consistency evaluation index of water supply node X is calculated to be 0.26 based on the temporal deviation between the water quality front and the water quality tail (an evaluation index closer to 1 indicates good consistency, and closer to 0 indicates poor consistency). During the scheduling decision-making process, the scheduling system can perform water source switching simulations in the digital twin model for different candidate scheduling schemes, obtaining their respective first prediction time, second prediction time, dynamic switching window, and temporal consistency evaluation index. For a candidate scheme, if the evaluation index is low and the dynamic switching window is wide, it indicates that the water quality is in a transitional state for a relatively long period of time, and the scheme should be adjusted or discarded; for schemes with higher evaluation indices and more concentrated dynamic switching windows, they can be given priority as the actual implementation scheme. The finally selected scheduling scheme will simultaneously record its corresponding dynamic switching window, and the switching command will be controlled within the dynamic switching window during operation, thereby improving the temporal consistency of the water source switching and water quality replacement process while ensuring water quality safety.

[0048] The method for constructing a digital twin model based on the static geographic data of the watershed includes:

[0049] The topology of the water conveyance path is discretized to obtain a set of nodes and a set of pipe segments. A hydraulic calculation network is established based on the set of nodes and the set of pipe segments. Water quality characteristics are set as state variables in the set of nodes and the set of pipe segments, and a mass conservation relationship for the transmission of water quality characteristics along the flow direction in the water conveyance path is established based on the transport equation based on the convection-diffusion mechanism. The hydraulic calculation network and the mass conservation relationship are integrated to form a digital twin model.

[0050] For example, when the water conveyance path topology is discretized, the path is divided into several connected model pipe segments according to its physical structure. Each turning point, control point, or functional location is defined as a model node, allowing the water conveyance system to be represented as a discrete network consisting of a set of nodes and a set of pipe segments. Each straight pipe segment in the water conveyance path is defined as a unit segment, while pumping stations, gates, regulating reservoirs, and water supply nodes are included as independent nodes in the node set, retaining connection, direction, and geometric information consistent with the actual facilities. Based on the discretized set of nodes and pipe segments, a hydraulic calculation network is established. In this network, static water level or pressure state is set for each node, and hydraulic characteristic parameters such as pipe diameter, length, roughness, and elevation difference are set for each pipe segment. The hydraulic relationships between adjacent nodes are constrained using continuity and energy equations, ensuring that for any given change in water inflow, the time-varying responses of the water level at each node and the flow rate in each pipe segment can be calculated.

[0051] For each node and each pipe segment, multiple water quality characteristic state variables, including turbidity, total dissolved solids, nitrate concentration, and residual chlorine, are set, and these state variables are used to characterize the mixing and propulsion process of new and old water sources in the water conveyance path. The flow-diffusion equation is used to discretize the water quality characteristics along each pipe segment, ensuring that the spatiotemporal changes of these water quality characteristics simultaneously satisfy both propulsion and diffusion mixing along the flow direction. That is, the mass flux is driven by the flow rate of the pipe segment, and at the nodes, instantaneous mixing occurs according to the proportion of the flow rates of the connecting pipe segments.

[0052] The method for simulating the evolution of water quality characteristics of the old and new water sources of the target water supply node over time under the water source switching condition in the digital twin model, and obtaining the time series data of water quality evolution of the old and new water sources of the target water supply node, includes:

[0053] Based on real-time monitoring data before water source switching, an initial field of water quality characteristics for each node and pipe segment is set in the hydraulic calculation network; when the digital twin model simulation starts, the boundary conditions of the water quality characteristics of the new water source inlet node to be switched are updated from a first constant value representing the concentration of the old water source to a second constant value representing the concentration of the new water source.

[0054] The method for updating the boundary conditions of the water quality characteristic quantities of the new water source inlet node to be switched from a first constant value representing the concentration of the old water source to a second constant value representing the concentration of the new water source includes:

[0055] The nominal water quality characteristic concentration of the new water source is obtained as a second constant value; the sequence of operating instructions for the hydraulic control equipment controlling the inflow of the new water source during the switching period is obtained; and a water source switching progress function characterizing the water source switching progress is determined based on the sequence of operating instructions. ,in, Monotonically increasing from 0 to 1, This indicates the simulation time of the digital twin model.

[0056] During the digital twin model simulation, the water quality characteristic boundary conditions of the new water source inlet node are applied during the water source switching period. Set as ;in, The first constant value, The boundary condition is set to a second constant value; before the water source switching period, the boundary condition is maintained at a first constant value, and after the water source switching period, the boundary condition is maintained at a second constant value.

[0057] Based on the updated boundary conditions, a real-time hydraulic gradient provided by the hydraulic calculation network is applied to the mass conservation relationship of the water quality characteristics. The simulation time step is gradually advanced using a numerical discrete solution method to calculate the concentration values ​​of water quality characteristics of each node and pipe segment in the hydraulic calculation network.

[0058] Within each simulation time step, the concentration values ​​of water quality characteristics of the new and old water sources of the target water supply node and the corresponding simulation time are recorded, and the data are arranged in chronological order to obtain the time series data of water quality evolution of the new and old water sources, respectively representing the concentration change process of the new and old water sources.

[0059] For example, before the water source switchover was initiated, groundwater continuously supplied water to water supply node X. Real-time monitoring data showed that turbidity remained at approximately 0.6 NTU, total dissolved solids concentration was approximately 850 mg / L, and residual chlorine level was approximately 0.05 mg / L. These monitoring values ​​were directly used as the initial field for water quality characteristics corresponding to all nodes and pipe sections along the water transmission path, ensuring that the water quality status in the digital twin model remained consistent with the actual site conditions. After the scheduling plan was implemented at 10:00, the inflow valve of the surface water source entered a phased opening process. On-site monitoring commands showed that the valve opening degree was progressively increased in four actions at 10:00, 10:05, 10:10, and 10:15, with the surface water inflow increasing proportionally. Monitoring data of the surface water at the intake side was sent to the digital twin model in real time, showing water quality characteristics of approximately turbidity of 1.2 NTU, total dissolved solids of approximately 450 mg / L, and residual chlorine of approximately 0.25 mg / L. As the valve opening degree increases step by step according to the scheduling instructions, the water source switching progress function s(t) is determined by linear interpolation based on the valve opening degree sequence. For example, if the valve opening degree is 0% at 10:00, the corresponding... At 10:15, the valve opening was 100%, corresponding to This indicates that the proportion of new water inflow gradually increases from 0 to 1 during the period from 10:00 to 10:15.

[0060] The specific process of advancing the numerical discrete solution method in the digital twin model is as follows: the digital twin model continuously advances the simulation with a time step of 10 seconds. Once the digital twin model obtains the node pressure and pipe flow rate from the real-time hydraulic calculation network, it directly applies the corresponding hydraulic gradient to the convection-diffusion relationship, updates water quality characteristics in each water delivery pipe segment, and performs mixing calculations at the nodes according to the flow rate ratio. The surface water quality front continues to advance in the pipeline. The turbidity at water supply node X no longer remains at 0.6 NTU, but rises to 0.9 NTU at 10:30 and 1.4 NTU at 10:45. During the same period, the total dissolved solids concentration decreases from 850 mg / L to 650 mg / L, and the residual chlorine gradually increases to 0.15 mg / L. As time continues, the characteristics of the old water source gradually fade, and the characteristics of the new water source gradually become dominant, forming a clear replacement process in the water delivery path. During the simulation, the concentration values ​​of water quality characteristics of the new and old water sources are state variables obtained by the digital twin model based on their respective initial states and boundary conditions, through solving the mass conservation relationship, and are used to track their independent evolution processes. The calculation results at each time step record the respective values ​​of the concentrations of the new and old water sources at the water supply node X, and are accumulated in the order of the simulation time, thereby forming complete time-series data of the evolution of water quality characteristics of the new and old water sources over time.

[0061] The method for determining the first predicted time when the water quality front of the new water source reaches the target water supply node and the second predicted time when the water quality tail of the old water source leaves the target water supply node based on the water quality evolution time series data of the new and old water sources includes:

[0062] Extract the time series of the new water source quality at the target water supply node from the time series data of the evolution of water quality between the old and new water sources. Time series of water quality of old water sources ; Obtain the preset qualified concentration threshold for new water sources .

[0063] The new water source water quality time series In the middle, when from a point in time A continuous period of time begins to exist. This makes it possible for any time All And the length of the continuous time period is not less than the preset minimum stable duration. At that time, the time point mentioned As the first prediction time.

[0064] The old water source water quality time series In the middle, when from a point in time A continuous period of time begins to exist. For any time , All conditions meet the preset influence concentration requirements, and the length of the continuous period is not less than the preset minimum stability duration. At that time, the time point mentioned As the second prediction time. The preset new water source qualified concentration threshold. The preset minimum stable duration is adaptively determined based on the water usage attributes of the target water supply node.

[0065] For example, the time series of the new water source quality at the target water supply node X is extracted from the time series data of the evolution of water quality between the old and new water sources. Time series of water quality of old water sources In this example, water supply node X is used for urban domestic water supply, and the corresponding new water source qualified concentration threshold is... The water quality requirements include: turbidity ≤ 1.5 NTU, total dissolved solids ≤ 650 mg / L, and residual chlorine ≥ 0.10 mg / L. Based on relevant operating standards, a minimum stabilization time is preset. Set to 10 minutes.

[0066] New water source water quality time series In the analysis, from 11:00 onwards, turbidity, total dissolved solids, and residual chlorine all consistently met the acceptable concentration thresholds for the new water source, and remained stable for 10 minutes from 11:00 to 11:10. Therefore, 11:00 was determined as the first prediction time. (This is in the context of the old water source water quality time series.) From 12:20 onwards, the contribution of the old water source gradually decreased, meeting the following preset influence concentration conditions: the turbidity increase caused by the old water source ≤ 0.2 NTU, and the residual chlorine concentration of the old water source ≤ 0.02 mg / L. During the subsequent 20-minute period from 12:20 to 12:40, the preset influence concentration conditions were continuously met; therefore, 12:20 was designated as the second prediction time. The preset influence concentration conditions are used to determine whether the water quality characteristics of the old water source have lost their adverse influence capacity under the target use, and the determination criteria are related to the use attributes of the water supply node. For urban domestic water use, the focus is on assessing the contribution ratio of turbidity and residual chlorine; for industrial cooling water, priority is given to changes in residual chlorine and hardness; and for agricultural irrigation water, the main focus is on the accumulation characteristics of total dissolved solids and salt.

[0067] The method for calculating the time-series consistency evaluation index of the target water supply node based on the time-series deviation between the water quality front and the water quality tail includes:

[0068] According to the first predicted time Second prediction time Calculate the absolute timing deviation ; Obtain the baseline switching time corresponding to the water usage attribute of the target water supply node. ; Calculate the water quality sensitivity weighting coefficient based on water quality differences .

[0069] According to the absolute timing deviation Reference switching time Water quality sensitivity weighting coefficient The time series consistency evaluation index was calculated. .in, The range of values ​​is , The closer it is to 1, the better the temporal consistency during the switching process between the old and new water sources, and the smoother the water quality transition. The closer to 0, the higher the risk of temporal decoupling and the longer the period of water quality instability during the switching process. (Set to...) Excellent consistency. Good consistency However, the consistency is poor, and the scheduling scheme needs to be optimized.

[0070] The water quality sensitivity weighting coefficient is calculated based on water quality differences. The methods include:

[0071] Identify a set of key water quality difference indicators between the old and new water sources, including turbidity, pH value, and total dissolved solids content; assign a corresponding basic weight value to each water quality difference indicator in the set of key water quality difference indicators based on the water use attribute of the target water supply node.

[0072] Based on the concentration difference between the old and new water sources for each water quality difference indicator, the difference contribution is calculated using a preset piecewise function. The basic weight value of each water quality difference indicator is multiplied by its corresponding difference contribution, and then summed to obtain a comprehensive water quality difference score. Finally, the comprehensive water quality difference score is mapped to the [0,1] interval using an S-shaped function to obtain the water quality sensitivity weight coefficient. .

[0073] For example, the dynamic switching window has been determined from the first predicted time of 11:00 to the second predicted time of 12:20, but the width of this window is 80 minutes, which cannot directly reflect the reliability of the switching effect. Therefore, it is necessary to construct a quantifiable evaluation index based on the time series deviation between the water quality front and the water quality tail to analyze the consistency of this water source switching in the time dimension. The purpose attribute of water supply node X is urban domestic water use, which is highly sensitive to the water quality replacement process. Its corresponding baseline switching time is 60 minutes according to the table. That is, if the difference between the first predicted time and the second predicted time does not exceed approximately 60 minutes, the switching process generally will not produce user-perceived risks. Absolute time series deviation Minutes. After identifying turbidity and residual chlorine as key water quality differences in this water source switch, basic weight values ​​are assigned to them based on their intended use. For example, for urban domestic water use, turbidity changes have a significant impact on sensory perception, and its basic weight is assigned 0.6; residual chlorine concentration directly affects disinfection capacity, and its basic weight is 0.4. The concentration difference between the old and new water sources is 2.5 NTU - 0.6 NTU = 1.9 NTU in terms of turbidity, and 0.25 mg / L - 0.05 mg / L = 0.20 mg / L in terms of residual chlorine. Based on a preset piecewise function... The calculated contribution factor for turbidity difference was 0.78, and the contribution factor for residual chlorine difference was 0.80; among which, This corresponds to the concentration difference between the old and new water sources for water quality indicators. and This is the threshold value for the graded difference of this water quality difference index. To determine the maximum expected difference in this water quality difference index based on engineering experience, turbidity is taken as... , , Take residual chlorine , , The overall water quality difference score S = 0.6 × 0.78 + 0.4 × 0.80 ≈ 0.79. This is calculated using the S-shaped mapping function. Mapping 0.79 to the [0,1] interval yields the water quality sensitivity weighting coefficient. ,in This is a scaling parameter used to adjust the steepness of the mapping curve. Its value typically ranges from 3 to 10, determined based on the actual water quality sensitivity distribution; here, a value of 4 is used. (Time-series consistency evaluation index) This indicates that the current switching scheme has low time consistency. If implemented, it may result in the switching scheduling record being updated before the actual water quality at water supply node X still exhibits characteristics of the old water source. Therefore, when facing highly sensitive urban domestic water demand scenarios, the time interval between the departure of the old water source's tailwater and the arrival of the new water source's frontwater should be shortened as much as possible. This can be achieved, for example, by increasing surface water flow in advance or optimizing valve opening timing, to bring the time consistency evaluation index close to 1.0, thereby establishing better time consistency between scheduling execution and user perception.

[0074] Example 2: Based on the same inventive concept, this example also provides a watershed water resources dynamic scheduling and optimization system based on digital twins, the system being used to execute the watershed water resources dynamic scheduling and optimization method based on digital twins.

[0075] It should be noted that the specific methods of performing operations in the systems described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0076] Finally, it should be noted that although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for dynamic scheduling and optimization of watershed water resources based on digital twins, characterized in that, The method includes: Acquire static geographic data of the watershed, including water conveyance path topology, water conservancy project nodes, and water supply nodes; construct a digital twin model based on the static geographic data of the watershed; acquire real-time monitoring data of each water source and the water supply nodes within the watershed, the real-time monitoring data including water quantity, water level, and water quality indicators; Based on the real-time monitoring data and preset scheduling targets, the target water supply node with water source switching needs is identified through the digital twin model; the evolution of water quality characteristics of the old and new water sources of the target water supply node over time under water source switching conditions is simulated in the digital twin model to obtain the time series data of water quality evolution of the old and new water sources of the target water supply node; based on the time series data of water quality evolution of the old and new water sources, the first predicted time of the arrival of the water quality front of the new water source at the target water supply node and the second predicted time of the departure of the water quality tail of the old water source from the target water supply node are determined; The dynamic switching window of the target water supply node is determined based on the first and second prediction times; the temporal consistency evaluation index of the target water supply node is calculated based on the temporal deviation between the water quality front and the water quality tail; the dynamic switching window and the temporal consistency evaluation index are output as the decision basis for optimizing the water source scheduling scheme.

2. The method for dynamic scheduling and optimization of watershed water resources based on digital twins according to claim 1, characterized in that, The method for constructing a digital twin model based on the static geographic data of the watershed includes: The topology of the water conveyance path is discretized to obtain a set of nodes and a set of pipe segments. A hydraulic calculation network is established based on the set of nodes and the set of pipe segments. Water quality characteristics are set as state variables in the set of nodes and the set of pipe segments, and a mass conservation relationship for the transmission of water quality characteristics along the flow direction in the water conveyance path is established based on the convection-diffusion equation. The hydraulic calculation network and the mass conservation relationship are integrated to form a digital twin model.

3. The method for dynamic scheduling and optimization of watershed water resources based on digital twins according to claim 2, characterized in that, The method for simulating the evolution of water quality characteristics of the old and new water sources of the target water supply node over time under the water source switching condition in the digital twin model, and obtaining the time series data of water quality evolution of the old and new water sources of the target water supply node, includes: Based on real-time monitoring data before water source switching, an initial field of water quality characteristics for each node and pipe segment is set in the hydraulic calculation network; when the digital twin model simulation starts, the boundary conditions of the water quality characteristics of the new water source inlet node to be switched are updated from a first constant value representing the concentration of the old water source to a second constant value representing the concentration of the new water source. Based on the updated boundary conditions, a real-time hydraulic gradient provided by the hydraulic calculation network is applied to the mass conservation relationship of the water quality characteristics. The simulation time step is gradually advanced using a numerical discrete solution method to calculate the concentration values ​​of water quality characteristics of each node and pipe segment in the hydraulic calculation network. Within each simulation time step, the concentration values ​​of water quality characteristics of the new and old water sources of the target water supply node and the corresponding simulation time are recorded, and the data are arranged in chronological order to obtain the time series data of water quality evolution of the new and old water sources, respectively representing the concentration change process of the new and old water sources.

4. The method for dynamic scheduling and optimization of watershed water resources based on digital twins according to claim 3, characterized in that, The method for updating the boundary conditions of the water quality characteristic quantities of the new water source inlet node to be switched from a first constant value representing the concentration of the old water source to a second constant value representing the concentration of the new water source includes: The nominal water quality characteristic concentration of the new water source is obtained as a second constant value; the sequence of operating instructions for the hydraulic control equipment controlling the inflow of the new water source during the switching period is obtained; and a water source switching progress function characterizing the water source switching progress is determined based on the sequence of operating instructions. ,in, Monotonically increasing from 0 to 1, Indicates the simulation time of the digital twin model; During the digital twin model simulation, the water quality characteristic boundary conditions of the new water source inlet node are applied during the water source switching period. Set as ;in, The first constant value, The boundary condition is set to a second constant value; before the water source switching period, the boundary condition is maintained at a first constant value, and after the water source switching period, the boundary condition is maintained at a second constant value.

5. The method for dynamic scheduling and optimization of watershed water resources based on digital twins according to claim 4, characterized in that, The method for determining the first predicted time when the water quality front of the new water source reaches the target water supply node and the second predicted time when the water quality tail of the old water source leaves the target water supply node based on the water quality evolution time series data of the new and old water sources includes: Extract the time series of the new water source quality at the target water supply node from the time series data of the evolution of water quality between the old and new water sources. Time series of water quality of old water sources ; Obtain the preset qualified concentration threshold for new water sources ; The new water source water quality time series In the middle, when from a point in time A continuous period of time begins to exist. This makes it possible for any time All And the length of the continuous time period is not less than the preset minimum stable duration. At that time, the time point mentioned As the first prediction time; The old water source water quality time series In the middle, when from a point in time A continuous period of time begins to exist. For any time , All conditions meet the preset influence concentration requirements, and the length of the continuous period is not less than the preset minimum stability duration. At that time, the time point mentioned As the second prediction time.

6. The method for dynamic scheduling and optimization of watershed water resources based on digital twins according to claim 5, characterized in that, The preset qualified concentration threshold for new water sources The preset minimum stable duration is adaptively determined based on the water usage attributes of the target water supply node.

7. The method for dynamic scheduling and optimization of watershed water resources based on digital twins according to claim 6, characterized in that, The method for calculating the time-series consistency evaluation index of the target water supply node based on the time-series deviation between the water quality front and the water quality tail includes: According to the first predicted time Second prediction time Calculate the absolute timing deviation ; Obtain the baseline switching time corresponding to the water usage attribute of the target water supply node. ; Calculate the water quality sensitivity weighting coefficient based on water quality differences ; According to the absolute timing deviation Reference switching time Water quality sensitivity weighting coefficient The time series consistency evaluation index was calculated. .

8. The method for dynamic scheduling and optimization of watershed water resources based on digital twins according to claim 7, characterized in that, The water quality sensitivity weighting coefficient is calculated based on water quality differences. The methods include: Identify a set of key water quality difference indicators between the old and new water sources, including turbidity, pH value, and total dissolved solids content; assign a corresponding basic weight value to each water quality difference indicator in the set of key water quality difference indicators based on the water use attribute of the target water supply node. Based on the concentration difference between the old and new water sources for each water quality difference indicator, the difference contribution is calculated using a preset piecewise function. The basic weight value of each water quality difference indicator is multiplied by its corresponding difference contribution, and then summed to obtain a comprehensive water quality difference score. Finally, the comprehensive water quality difference score is mapped to the [0,1] interval using an S-shaped function to obtain the water quality sensitivity weight coefficient. .

9. A watershed water resources dynamic scheduling and optimization system based on digital twins, characterized in that, The system is used to execute the watershed water resources dynamic scheduling optimization method based on digital twins as described in claims 1-8.

Images