Water quantity and quality model visualization method based on digital twinning

Abstract

The application relates to the technical field of model visualization, in particular to a water quantity and water quality model visualization method based on digital twinning. The method comprises the following steps: acquiring grid geometry data, pollution concentration data of each time step and pollution source correlation data, and counting pollution concentration change indexes, pollution distribution dispersion indexes and regional diffusion dynamic indexes in different grid regions; taking the ratio of the pollution concentration change index and the pollution distribution dispersion index of the same grid region as a pollution active coefficient; taking the product of the pollution active coefficient and the regional diffusion dynamic index as a pollution diffusion representation coefficient; and calculating the time sequence difference degree of the pollution diffusion representation coefficient between different scheduling scenes in the same grid region as a scheduling sensitivity coefficient of the grid region. The application realizes quantitative visualization of the pollution situation by constructing the pollution diffusion representation coefficient and introducing multi-scheduling scene difference analysis.

Description

Technical Field

[0001] This invention relates to the field of model visualization technology, and in particular to a method for visualizing water quantity and quality models based on digital twins. Background Technology

[0002] In the refined management of water resources and the prevention and control of water environment risks in watersheds, the collaborative analysis and visualization of water quantity and quality have gradually become key supporting technologies in water conservancy scheduling, pollution prevention and control, and emergency management. With the widespread application of online monitoring equipment, remote sensing technology, and hydrodynamic-water quality numerical models, the types and scale of data that can be obtained at the watershed scale have increased significantly. However, traditional methods of water quantity and quality analysis and display are still mainly based on single-section monitoring, static statistical charts, or simple two-dimensional rendering, which are difficult to fully reflect the spatial diffusion patterns of pollutants and their dynamic evolution process over time.

[0003] On the one hand, existing water quality assessment methods mostly focus on judging concentration exceedances or mean analysis, lacking the characterization of dynamic characteristics such as pollution change rate, diffusion acceleration, and spatial dispersion. This leads to a lag in identifying the pollution diffusion process and makes it difficult to capture key stages of pollution evolution in a timely manner. On the other hand, while some digital twin or 3D visualization systems can graphically display the state of water bodies, their visualization results often rely on fixed thresholds or empirical rules, lacking systematic quantitative analysis of differences across multiple scheduling scenarios. This makes it difficult to accurately reflect the intrinsic relationship between scheduling behavior and pollution diffusion, thus limiting their practical value in scheduling decisions and emergency simulations. Summary of the Invention

[0004] Therefore, it is necessary for the present invention to provide a method for visualizing water quantity and quality models based on digital twins in order to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a method for visualizing water quantity and quality models based on digital twins includes the following steps: Step S1: Obtain grid geometric data, pollution concentration data at each time step, and pollution source correlation data, and statistically analyze pollution concentration change indicators, pollution distribution dispersion indicators, and regional diffusion dynamic indicators in different grid areas. Step S2: The ratio of the pollution concentration change index to the pollution distribution dispersion index in the same grid area is used as the pollution activity coefficient; the product of the pollution activity coefficient and the regional diffusion dynamic index is used as the pollution diffusion characterization coefficient. Step S3: Calculate the time series difference of the pollution diffusion characterization coefficient between different scheduling scenarios in the same grid area, and use it as the scheduling sensitivity coefficient of the grid area; construct a visual assessment quantity for pollution situation awareness based on the scheduling sensitivity coefficient and the pollution diffusion characterization coefficient. Step S4: Determine the visualization control parameters for each grid area based on the visualization evaluation, and send the visualization control parameters to the digital twin visualization platform controller to perform split-screen comparison dynamic rendering.

[0006] This invention achieves a refined characterization and dynamic perception of pollution diffusion by using a gridded model of the watershed system and conducting collaborative analysis of pollution concentration data, spatial distribution characteristics, and scheduling response characteristics over a continuous time dimension. This method not only captures the rate and acceleration characteristics of pollution concentration changes over time but also identifies the dispersion of pollution distribution, its expansion direction, and its coupling relationship with the mainstream hydrodynamic direction, thus overcoming the limitations of traditional pollution assessments that rely solely on single cross-sections or static indicators. By introducing multi-dimensional quantification methods such as first-order difference, second-order difference, and spatial displacement, abrupt changes, accelerated diffusion, and directional shifts in the pollution diffusion process can be effectively identified, improving the sensitivity and accuracy of identifying key stages of pollution evolution.

[0007] By constructing a pollution diffusion characterization coefficient and introducing a multi-scheduling scenario comparison mechanism, the impact of different scheduling schemes on pollution diffusion behavior can be quantitatively analyzed under a unified dimension. Through comprehensive calculation of the difference fluctuation amplitude and the proportion of significant periods, the response intensity and persistence of pollution diffusion to scheduling changes can be accurately reflected, providing a quantifiable basis for the selection of scheduling schemes and risk assessment. Simultaneously, by introducing the concentration of diffusion peaks and the characteristics of diffusion direction reversal, the assessment results take into account both the concentration of pollution intensity and the stability of its evolution, effectively avoiding misjudgments caused by short-term anomalies or frequent fluctuations.

[0008] At the visualization level, this method transforms the aforementioned multi-source, multi-scale analysis results into unified visual assessment metrics. Combined with regional priority ranking, split-screen comparison display, color mapping, and animation playback parameters, it achieves a hierarchical presentation and dynamic display of the pollution situation. This approach highlights key areas and high-risk periods, enabling managers to quickly grasp the key points of pollution spread and its changing trends in complex information environments. Simultaneously, leveraging the dynamic rendering capabilities of the digital twin platform, pollution situations across different scheduling scenarios and time stages can be intuitively compared, significantly improving the intuitiveness, foresight, and operability of pollution emergency response and scheduling decisions. Through the synergistic application of the above technical solutions, the overall method demonstrates significant improvements in pollution spread identification accuracy, scheduling response analysis depth, and the intuitiveness of situation presentation, making it suitable for pollution monitoring, emergency response, and intelligent scheduling scenarios in complex watershed environments. Attached Figure Description

[0009] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating the steps of the water quantity and quality model visualization method based on digital twins of the present invention. Figure 2 This is a split-screen visualization diagram of the watershed pollution diffusion situation according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a comprehensive visualization interface for pollution situation awareness and dispatch response according to an embodiment of the present invention. Detailed Implementation

[0010] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0011] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0012] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0013] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a method for visualizing water quantity and quality models based on digital twins, the method comprising the following steps: Step S1: Obtain grid geometric data, pollution concentration data at each time step, and pollution source correlation data, and statistically analyze pollution concentration change indicators, pollution distribution dispersion indicators, and regional diffusion dynamic indicators in different grid areas. Step S2: The ratio of the pollution concentration change index to the pollution distribution dispersion index in the same grid area is used as the pollution activity coefficient; the product of the pollution activity coefficient and the regional diffusion dynamic index is used as the pollution diffusion characterization coefficient. Step S3: Calculate the time series difference of the pollution diffusion characterization coefficient between different scheduling scenarios in the same grid area, and use it as the scheduling sensitivity coefficient of the grid area; construct a visual assessment quantity for pollution situation awareness based on the scheduling sensitivity coefficient and the pollution diffusion characterization coefficient. Step S4: Determine the visualization control parameters for each grid area based on the visualization evaluation, and send the visualization control parameters to the digital twin visualization platform controller to perform split-screen comparison dynamic rendering.

[0014] Furthermore, the indicators for statistically analyzing changes in pollution concentration in step S1 include: The watershed is divided into several grid regions based on grid geometric data. Pollution concentration data in continuous time step series is extracted for each grid region to calculate the average concentration value and form a time series concentration curve. In one embodiment, based on a pre-constructed digital twin hydrodynamic-water quality model, two-dimensional or three-dimensional grid geometric data of the watershed are read, and the entire watershed water system space is discretized into multiple regular or irregular grid units. Several adjacent grid units are then combined into the same grid region according to spatial connectivity. Subsequently, during model simulation or online prediction, for each grid region, pollutant concentration data corresponding to all grid nodes within that region are extracted over multiple consecutive time steps. The node concentrations within the same time step are then spatially averaged to obtain the average pollution concentration value for that grid region at each time step, thus forming a time-series concentration curve reflecting the change in pollution concentration over time.

[0015] For example, for a river section consisting of 25 computational grids, the ammonia nitrogen concentration values ​​of each grid node are obtained in each 10-minute time step, and the average of the 25 concentration values ​​is calculated. After obtaining the average for 24 consecutive time steps, the time series concentration curve of the region on a 4-hour scale is formed.

[0016] The concentration change between adjacent time steps in the time series concentration curve is calculated as a first-order difference sequence. In one embodiment, for the aforementioned time-series concentration curve, the average pollution concentration values ​​of adjacent time steps are differentially calculated according to the time step sequence. That is, the concentration value of the previous time step is subtracted from the concentration value of the later time step, thereby obtaining a first-order difference value reflecting the rate of change of pollution concentration. The difference results corresponding to all adjacent time steps are arranged in chronological order to form a first-order difference sequence. This first-order difference sequence is used to characterize the increasing or decreasing trend and magnitude of change of pollution concentration over time.

[0017] For example, if the average pollution concentration in a certain grid area is 1.2 mg / L, 1.5 mg / L, and 1.3 mg / L in three consecutive time steps, then the corresponding first-order difference sequence is +0.3 mg / L and 0.2 mg / L.

[0018] It should be noted that the first-order difference only reflects the change between adjacent time steps and does not involve trend judgment on longer time scales. The result can be positive, negative or zero.

[0019] The difference values ​​of the first-order difference sequence are used to obtain the second-order difference sequence; In one embodiment, after obtaining the first-order difference sequence, the adjacent first-order difference values ​​are further differentially processed in chronological order, that is, the first-order difference value at the later time is subtracted from the first-order difference value at the previous time, thereby obtaining the second-order difference value, and the value is serialized to form a second-order difference sequence. The second-order difference sequence is used to reflect the change in the rate of change of pollution concentration itself, that is, the acceleration or deceleration characteristics of the change in pollution concentration.

[0020] For example, if the first-order difference sequence is +0.3 mg / L and 0.2 mg / L, then the corresponding second-order difference value is The concentration was 0.5 mg / L, indicating that the rate of change in pollution concentration dropped significantly during this period.

[0021] It should be noted that the time length of the second-order difference sequence will be reduced compared to the original time series, but this change does not affect its use in extracting extreme value features.

[0022] The average absolute value of each time step in the first-order difference sequence is taken as the average rate of change, and the maximum value of the second-order difference sequence is taken as the peak acceleration. The product of the average rate of change and the peak acceleration is taken as the pollution concentration change index of the grid area.

[0023] In one embodiment, the absolute values ​​of the difference values ​​corresponding to each time step in the first-order difference sequence are taken to eliminate the influence of the direction of concentration increase or decrease on the judgment of the intensity of change. Then, all absolute values ​​are averaged to obtain the average rate of change, which characterizes the overall fluctuation level of pollution concentration in the grid region throughout the entire time series. Next, the largest difference value is extracted from the second-order difference sequence to represent the maximum acceleration that occurs during the pollution concentration change, i.e., the moment of most dramatic change rate. Finally, the average rate of change is multiplied by the peak acceleration to obtain a pollution concentration change index that comprehensively reflects the intensity and abrupt change characteristics of pollution concentration change.

[0024] For example, if the average absolute value of the first-order difference in a certain area is 0.25 mg / L and the maximum value of the second-order difference sequence is 0.6 mg / L, then the pollution concentration change index for that area is 0.15.

[0025] Of particular importance is that the pollution concentration data for each time step in step S1 are obtained as follows: Real-time water quality data, including dissolved oxygen, ammonia nitrogen, and oil levels, are obtained from the online monitoring equipment at the water intake section, the sections before and after the gate, and the water distribution gate section. In one embodiment, online monitoring equipment deployed at key locations upstream and downstream of the water conservancy project collects water quality monitoring data in real time at various time steps. The online monitoring equipment is installed at least at the intake section, the section upstream of the sluice gate, the section downstream of the sluice gate, and the diversion gate section to reflect the spatial distribution characteristics of pollutants before, during, and after the project's scheduling and distribution process. The collected water quality indicators include, but are not limited to, dissolved oxygen, ammonia nitrogen, and oil indicators. All indicators are sampled synchronously at a unified time step, thereby forming multi-section, multi-indicator time-series water quality data, providing a basic input for subsequent pollution concentration modeling.

[0026] For example, the system can collect dissolved oxygen and ammonia nitrogen concentration data from online sensors at each cross-section every 5 minutes and mark them as data records of the same time step.

[0027] Based on the pre-acquired outflow from the hub, gate opening, and real-time water quality data, water quantity and quality simulations are performed to obtain predicted pollution concentration values. In one embodiment, while acquiring real-time water quality data, pre-acquired scheduling parameters such as the outflow from the dam and the opening degree of each gate are invoked, and these scheduling parameters, along with the real-time water quality data, are input into the water quantity and quality simulation module. Through coupled water quantity and quality simulation, the transport, dilution, and attenuation processes of pollutants in the water body are calculated, thereby obtaining corresponding predicted pollution concentration values ​​at each continuous time step. These predicted values ​​are used to compensate for insufficient monitoring point coverage or time lag issues, enabling the system to form a continuous and complete pollution concentration time series.

[0028] For example, when the system detects that the gate opening increases and the discharge flow rate rises at a certain time step, the dilution effect can be considered simultaneously in the water quality simulation to predict and correct the ammonia nitrogen concentration in subsequent time steps.

[0029] When a sudden water pollution emergency is received, the location of the emergency pollution source, the amount of pollutants released, and the type of pollutants are extracted. A preset water quality simulation model is then run to generate pollution concentration data for each time step under the emergency scenario.

[0030] In one embodiment, upon receiving a signal triggering a sudden water pollution emergency, the system automatically switches to emergency scenario processing mode. The system first extracts key information corresponding to the emergency event, such as the location of the pollution source, the amount of pollutants released, and the type of pollutants, and loads this information as emergency input parameters into a preset water quality simulation model. By introducing a sudden pollution source term into the model, the system simulates the diffusion, migration, and evolution of pollutants under emergency conditions, thereby generating pollution concentration data at each time step in the emergency scenario to support subsequent emergency dispatch and visualization analysis.

[0031] For example, when receiving an event message that "an oil pollution leak occurred in a tributary at 10:00", the system can recalculate the changes in oil pollution concentration at each downstream section in subsequent time steps starting from 10:00. Of particular importance is that obtaining pollution source association data in step S1 includes: Collect watershed topography, gate location distribution, and intake and drainage outlet locations and elevations to form fixed basic data; In one embodiment, under non-emergency conditions, fixed basic data related to the watershed structure are pre-collected and stored. This fixed basic data includes static spatial information such as watershed topography, the location and distribution of each sluice gate, and the location and corresponding elevation of the intake and discharge outlets. This data is used to characterize water flow paths and pollutant propagation constraints, serving as the spatial basis for pollution source correlation analysis.

[0032] For example, a basic database containing gate coordinates and elevation information can be established using topographic survey data and engineering design data.

[0033] When a sudden water pollution emergency occurs, the spatial coordinates of the emergency pollution source, the amount of pollutants released, and the start time of the event are recorded to form emergency pollution source data. In one embodiment, when a sudden water pollution emergency occurs, the relevant information of the event is recorded in a structured manner to form emergency pollution source data. The emergency pollution source data includes at least the spatial coordinates of the emergency pollution source, the amount of pollutants released, and the event start time, which is used to clarify the spatiotemporal starting point of the pollution and serve as the core input for subsequent pollution diffusion simulation and dispatch response analysis.

[0034] For example, the information that "2 tons of ammonia-containing pollutants were released at a certain coordinate point at 9:30" can be recorded as an emergency pollution source data.

[0035] The real-time opening degree and discharge flow sequence of each gate are obtained from the hub dispatch center as dispatch parameter data; In one embodiment, a data interface is established with the hub dispatch center to obtain the opening status of each gate and the corresponding discharge flow sequence in real time, and this data is stored as dispatch parameter data. The dispatch parameter data reflects the operating conditions of the water body at different time steps and is an important basis for analyzing the controlled characteristics of pollution diffusion.

[0036] For example, the system can obtain the real-time opening degree of each gate from the dispatch center once per minute and form a continuous time series record.

[0037] Set early warning thresholds for pollutant concentrations at key cross-sections, and mark the early warning level when the predicted concentration exceeds the threshold; In one embodiment, corresponding pollutant concentration warning thresholds are pre-set for key sections such as the water intake and the sections before and after the gate. When the pollution concentration obtained from water quality simulation or prediction exceeds the threshold, the corresponding warning level is automatically marked to reflect the degree of pollution risk and to provide constraints for subsequent visual assessment and scheduling decisions.

[0038] For example, when the predicted concentration of ammonia nitrogen at a water intake exceeds a preset threshold, the section can be marked as a high-level warning state.

[0039] By combining fixed basic data, emergency pollution source data, scheduling parameter data, and early warning thresholds, pollution source correlation data is formed.

[0040] In one embodiment, the aforementioned fixed basic data, emergency pollution source data, scheduling parameter data, and early warning thresholds are uniformly integrated to construct a pollution source association data set. This pollution source association data is used to comprehensively describe the pollution occurrence conditions, diffusion environment, and scheduling influencing factors, providing complete data support for subsequent pollution diffusion characterization, scheduling sensitivity analysis, and visualization assessment.

[0041] For example, the pollution source information corresponding to a certain emergency pollution event, the gate operation status at that time, and the warning threshold of the relevant section can be uniformly linked to form a complete set of data records.

[0042] Furthermore, the statistical indicators of pollution distribution dispersion in step S1 include: For each grid region, at each time step, the pollution concentration data of all grid nodes in that region are extracted to form the spatial distribution of the concentration in that region at that time step; In one embodiment, during the calculation of the digital twin water quantity and quality model, for each pre-divided grid region, the pollutant concentration values ​​of all computational grid nodes contained in that region are read within a specified time step, and the spatial coordinate information corresponding to each node is retained, thereby forming the spatial distribution of pollutant concentration within that grid region at that time step. This spatial distribution of concentration exists in the form of a correspondence between "spatial location and concentration value," which is used to characterize the spatial non-uniformity of pollution within the region.

[0043] For example, for a region consisting of 16 two-dimensional grid nodes, the pollution concentration value of each node is obtained at a certain time step, and the node's position in the plane coordinate system is matched one-to-one to form the spatial distribution of concentration in that region at that time step.

[0044] Identify the spatial coordinates of concentration peak locations from the spatial distribution of concentration; In one embodiment, after obtaining the spatial distribution of concentration at a certain time step, the pollution concentration values ​​corresponding to all grid nodes in the region are compared, the node with the highest concentration value is identified, and the spatial coordinates corresponding to that node are determined as the concentration peak position for that time step. When multiple nodes have the same maximum concentration value, any one of them can be selected, or the peak position can be determined according to the priority rule of the nodes in the mainstream direction. This peak position is used to reflect the main accumulation location of pollution within the region.

[0045] For example, if the pollution concentration of a certain grid node in a region is 2.3 mg / L within a certain time step, and is higher than the concentration values ​​of other nodes in the region, then the planar coordinates of that node are determined as the concentration peak location of that time step.

[0046] Calculate the degree of deviation of the pollution concentration of all grid nodes in the region from the average concentration of the region at that time step, and take the sum of the squares of the deviations of each node as the variance of the concentration distribution at that time step; In one embodiment, the average pollution concentration value of the region is calculated based on the pollution concentration data of all grid nodes in the region within the time step. Subsequently, for each grid node, the difference between its pollution concentration and the average concentration of the region is calculated to characterize the degree of deviation of the pollution level of the node relative to the overall level of the region. Furthermore, the deviation is squared and the squared results corresponding to all nodes are accumulated to obtain the concentration distribution variance of the region at the time step, which is used to quantify the degree of dispersion of pollution within the region.

[0047] For example, if the average pollution concentration in a certain area is 1.5 mg / L at a certain time step, and the concentration at a certain node is 2.0 mg / L, then the deviation of that node is 0.5 mg / L, and its square value is 0.25. By summing the square values ​​of all nodes, the variance of the concentration distribution at that time step is obtained.

[0048] The spatial displacement distance of the concentration peak position between consecutive time steps is tracked, and the peak displacement distance is multiplied by the concentration distribution variance as the pollution distribution dispersion index for that grid area.

[0049] In one embodiment, within multiple consecutive time steps, the spatial coordinates of the concentration peak positions corresponding to each time step are sequentially acquired, and the spatial displacement distance between the peak positions of adjacent time steps is calculated. This displacement distance is used to reflect the spatial migration amplitude of the high-pollution area. Subsequently, the peak displacement distance corresponding to the time step is multiplied by the previously calculated concentration distribution variance to form a pollution distribution dispersion index that comprehensively considers the spatial dispersion of pollution and peak migration behavior.

[0050] For example, if a grid area experiences a spatial displacement of 30 meters between two adjacent time steps at the concentration peak position, and the concentration distribution variance of the corresponding time step is 0.4, then the pollution distribution dispersion index for that time step is 12.

[0051] Furthermore, the statistical regional diffusion dynamics indicators in step S1 include: The spatial range of grid nodes within the tracked grid area where the pollution concentration exceeds a preset threshold is recorded as the threshold-exceeding range. In one embodiment, during the calculation of each time step of the digital twin water quantity and quality model, for a given grid region, the pollution concentration values ​​corresponding to each grid node within that region are read one by one and compared with a pre-set pollution concentration threshold. When the pollution concentration of a grid node exceeds the preset threshold, the grid node is marked as an over-threshold node. Furthermore, all over-threshold nodes within the same time step are combined according to their spatial connectivity to form an over-threshold range reflecting the spatial distribution of pollution exceeding standards. This over-threshold range is used to characterize the actual impact range of pollution within the region.

[0052] For example, when the preset threshold is 1.5 mg / L, if the pollution concentration of 8 adjacent grid nodes in a region exceeds 1.5 mg / L in a certain time step, the connected region formed by these 8 nodes is identified as the threshold range for that time step.

[0053] Record the boundary expansion rate of the threshold range in the time series, and calculate the rate of change of the influence area of ​​the threshold range between consecutive time steps; In one embodiment, within consecutive time steps, the aforementioned identified threshold range is tracked hourly, the set of boundary grid nodes of the threshold range in each time step is extracted, and the average rate of outward expansion of the threshold range boundary is calculated based on the changes in the spatial position of the boundary nodes in adjacent time steps to characterize the expansion speed of the pollution impact range; at the same time, based on the number of grid nodes or the corresponding actual area contained in the threshold range, the rate of change of the impact area between consecutive time steps is calculated to reflect the growth or shrinkage trend of the pollution exceeding the standard area.

[0054] For example, if the area exceeding the threshold in a certain grid region expands from covering 10 grid nodes to covering 15 grid nodes in two adjacent time steps, the corresponding rate of change of the affected area can be calculated, and the boundary expansion rate can be calculated in combination with the outward movement distance of the boundary nodes.

[0055] Extract the mainstream direction information of the grid area from the pollution source association data, identify the main expansion direction of the exceeding threshold range, and determine whether the main expansion direction is consistent with the mainstream direction. In one embodiment, based on pre-constructed pollution source association data, the dominant water flow direction information corresponding to the grid area is extracted. The mainstream direction can be determined by a combination of gate discharge flow, river slope direction, or historical velocity field. At the same time, based on the overall migration trend of the boundary position of the threshold range within a continuous time step, the main expansion direction of the threshold range in the current time series is identified, and the main expansion direction is judged to be consistent with the mainstream direction to determine whether the pollution diffusion follows the dominant hydrodynamic direction.

[0056] For example, if the mainstream direction of a region is from west to east, and the range exceeding the threshold mainly expands eastward within a continuous time step, then the expansion direction is determined to be consistent with the mainstream direction; if it mainly expands northward, then it is determined to be inconsistent.

[0057] Multiply the boundary expansion rate by the rate of change of the affected area. When the expansion direction is consistent with the mainstream direction, the product is amplified by a preset amplification ratio. When they are inconsistent, the product is reduced by a preset reduction ratio to obtain the regional diffusion dynamic index.

[0058] In one embodiment, after obtaining the boundary expansion rate and the rate of change of the affected area, the two are multiplied to comprehensively reflect the dynamic characteristics of the pollution-exceeding area in terms of spatial expansion speed and scale changes. Subsequently, based on the aforementioned judgment result of the consistency between the expansion direction and the mainstream direction, the product value is directionally corrected. When the expansion direction is consistent with the mainstream direction, the product result is amplified according to a preset amplification ratio to emphasize the risk characteristics of downstream diffusion. When the expansion direction is inconsistent with the mainstream direction, the product result is reduced according to a preset reduction ratio to obtain the final regional diffusion dynamic index.

[0059] For example, if the product of the boundary expansion rate and the rate of change of the affected area within a certain time step is 5, and the expansion direction is consistent with the mainstream direction, and the preset amplification ratio is 1.5, then the regional diffusion dynamic index for that time step is 7.5; if the directions are inconsistent and the shrinkage ratio is 0.7, then the corresponding index is 3.5.

[0060] Furthermore, the main expansion directions of the identification exceeding the threshold range include: Extract the set of boundary grid nodes exceeding the threshold range at each time step; In one embodiment, after identifying the threshold range, for the current time step of a certain grid region, all grid nodes constituting the threshold range are traversed, and grid nodes that are adjacent to at least one node that is not at the threshold are selected. These nodes are then identified as boundary grid nodes, thereby forming a set of boundary grid nodes corresponding to that time step. This set of boundary grid nodes is used to describe the spatial outer contour structure of the threshold range.

[0061] For example, if the range exceeding the threshold within a certain time step consists of several adjacent grid nodes, where the outermost ring of nodes is directly adjacent to the nodes that have not exceeded the threshold, then the set of outermost nodes is identified as the set of boundary grid nodes.

[0062] For each boundary node in the boundary grid node set, calculate the spatial position difference between the node at the current time step and the previous time step to obtain the displacement vector of the node; In one embodiment, for each boundary grid node identified at the current time step, the spatial coordinates of the node at the current time step and the previous time step are obtained respectively. A difference operation is then performed on the coordinates of the two time steps to obtain a displacement vector characterizing the spatial migration direction and distance of the boundary node per unit time. This displacement vector reflects the expansion trend of the local boundary in the time dimension.

[0063] For example, if the spatial coordinates of a boundary node in the previous time step are (x1, y1) and its spatial coordinates in the current time step are (x2, y2), then the displacement vector of that node is (x2, y1). x1, y2 y1).

[0064] Based on the magnitude of the pollution concentration gradient corresponding to each boundary node, the displacement vector of each boundary node is assigned a weight, and the weighted displacement vectors of all boundary nodes are summed to obtain the composite displacement vector. In one embodiment, after obtaining the displacement vectors of each boundary node, the magnitude of the pollution concentration gradient corresponding to each boundary node at the current time step is further calculated. The pollution concentration gradient is used to characterize the steepness of the change in pollution concentration around the node. Subsequently, weights are assigned to the corresponding displacement vectors of each node according to the magnitude of the pollution concentration gradient, so that the boundary nodes with larger pollution gradients have a higher influence in the synthesis process. Finally, the weighted displacement vectors of all boundary nodes are superimposed to obtain a synthetic displacement vector that reflects the overall trend of spatial expansion beyond the threshold range.

[0065] For example, if the displacement vector of a boundary node is small but its pollution concentration gradient is large, then the node will still have a significant impact on the direction of the synthesized displacement vector during the vector synthesis process.

[0066] Calculate the direction angle of the composite displacement vector, which is the main expansion direction of the over-threshold range.

[0067] In one embodiment, after obtaining the synthetic displacement vector, its pointing angle in a preset spatial coordinate system is determined by calculating the direction angle of the vector. The direction of the synthetic displacement vector is then used as the main expansion direction of the exceedance range within the time series. This main expansion direction characterizes the overall spatial diffusion trend of the pollution-exceeding area and provides a basis for subsequent consistency judgments with the mainstream direction.

[0068] For example, if the synthesized displacement vector points southeast in the planar coordinate system, then that direction is determined to be the main expansion direction of the over-threshold range at that time step.

[0069] Furthermore, step S3 includes the following steps: Step S31: For the same grid area, extract the pollution diffusion characterization coefficients of the area under at least two different scheduling scenarios to form a time series of pollution diffusion characterization coefficients of the area under each scheduling scenario; In one embodiment, in the digital twin water quantity and quality model, water quantity and quality simulation or prediction calculations are run under different scheduling scenarios for the same grid area. The scheduling scenarios may include different gate opening combinations, different discharge flow configurations, or different scheduling strategy schemes. In each scheduling scenario, the pollution diffusion characterization coefficient of the grid area in continuous time steps is calculated according to the aforementioned method, and the pollution diffusion characterization coefficients under the same scheduling scenario are arranged in chronological order to form a time series of pollution diffusion characterization coefficients for the corresponding scheduling scenario.

[0070] For example, two different operating conditions, "routine dispatching scenario" and "emergency dispatching scenario", can be selected to calculate the pollution diffusion characterization coefficient of the same grid area at each time step, thereby forming two time series for subsequent comparative analysis.

[0071] Step S32: Align the time steps of the pollution diffusion characterization coefficient time series of different scheduling scenarios, calculate the difference of the pollution diffusion characterization coefficient of different scenarios at each time step, and form a difference sequence; In one embodiment, after obtaining the time series of pollution diffusion characterization coefficients under multiple scheduling scenarios, time step alignment is performed on each time series to ensure that the pollution diffusion characterization coefficients of different scheduling scenarios under the same time index are comparable; then, at each aligned time step, the difference between the pollution diffusion characterization coefficients corresponding to different scheduling scenarios is calculated, and the differences corresponding to all time steps are arranged in chronological order to form a difference sequence that reflects the degree of influence of scheduling scenario differences on pollution diffusion.

[0072] For example, at the same time step, if the pollution diffusion characterization coefficient for the regular scheduling scenario is 4.2 and that for the emergency scheduling scenario is 5.0, then the difference at that time step is 0.8.

[0073] Step S33: Calculate the standard deviation of the difference series as the amplitude of the difference fluctuation.

[0074] In one embodiment, after forming the difference sequence, statistical analysis is performed on the difference sequence to calculate its overall standard deviation, thereby quantifying the degree of fluctuation of the pollution diffusion characterization coefficients under different scheduling scenarios over time. The fluctuation amplitude of the difference is used to reflect the stability or sensitivity of the impact of scheduling changes on pollution diffusion; the larger the standard deviation, the more unstable the difference in the impact of different scheduling scenarios on pollution diffusion.

[0075] For example, if the difference sequence of a certain grid region fluctuates greatly at multiple time steps, its calculated standard deviation is relatively high, and the corresponding difference fluctuation amplitude is also large.

[0076] Furthermore, step S3 also includes the following steps: Step S34: Count the number of time steps in the difference sequence whose absolute value exceeds the preset difference threshold, and calculate the proportion of the number of time steps to the total number of time steps as the proportion of the period with significant difference. In one embodiment, based on the difference sequence, a difference threshold is pre-set to distinguish between significant and non-significant differences, and each time step in the difference sequence is judged; when the absolute value of the difference corresponding to a certain time step exceeds the difference threshold, the time step is marked as a significant difference time step; then the number of all significant difference time steps is counted, and the ratio of the number of significant difference time steps to the total number of time steps corresponding to the difference sequence is calculated, thereby obtaining the proportion of significant difference time periods that reflect the degree of persistence of significant differences.

[0077] For example, if the absolute value of the difference exceeds the preset difference threshold in 6 out of 20 time steps, the proportion of time steps with significant differences is 0.3.

[0078] Step S35: Multiply the amplitude of the difference fluctuation by the proportion of the period with significant difference to obtain the scheduling sensitivity coefficient of the grid area; In one embodiment, after obtaining the amplitude of the difference fluctuations and the proportion of periods with significant differences, the two are multiplied to form a scheduling sensitivity coefficient that comprehensively reflects the intensity and duration of the impact of scheduling changes on pollution diffusion. This scheduling sensitivity coefficient measures the responsiveness of the grid area to changes in scheduling strategies; a higher value indicates that the pollution diffusion in that area is more sensitive to scheduling adjustments.

[0079] For example, if the difference fluctuation range of a certain region is 1.2 and the proportion of periods with significant differences is 0.25, then the scheduling sensitivity coefficient of that region is 0.3.

[0080] Step S36: Construct a visual assessment quantity for pollution situation awareness based on the scheduling sensitivity coefficient and the pollution diffusion characterization coefficient.

[0081] In one embodiment, after obtaining the scheduling sensitivity coefficient, it is combined with the pollution diffusion characterization coefficient under the corresponding time series to simultaneously reflect the objective intensity of pollution diffusion and its response characteristics to scheduling changes, thereby constructing a pollution situation awareness assessment quantity for digital twin visualization. This visualization assessment quantity serves as the basic indicator for subsequent split-screen comparison display, color mapping, and animation control, highlighting key areas where scheduling adjustments significantly impact the pollution diffusion situation.

[0082] For example, when a grid area has a high pollution diffusion characterization coefficient and a large scheduling sensitivity coefficient within a certain time period, its corresponding visualization evaluation quantity will be significantly higher than that of other areas, and thus it will be highlighted in the visualization interface.

[0083] Furthermore, step S36 includes the following steps: Step S361: Identify local maxima time steps in the pollution diffusion characterization coefficient time series to form a diffusion peak time set; In one embodiment, after obtaining the time series of pollution diffusion characterization coefficients for a certain grid area in continuous time steps, the time series is scanned step by step, and the pollution diffusion characterization coefficients corresponding to the current time step are compared with the values ​​of the previous and next time steps. When the pollution diffusion characterization coefficients of the current time step are greater than the values ​​of the two adjacent time steps at the same time, the time step is determined to be a local maximum time step and is included in the diffusion peak time set to characterize the high-intensity concentrated period of pollution diffusion in the time dimension.

[0084] For example, if the pollution diffusion characterization coefficient of a time series is higher at the 5th time step than at the 4th and 6th time steps, then the 5th time step is identified as the diffusion peak time step.

[0085] Step S362: Calculate the proportion of the pollution diffusion characterization coefficient corresponding to each time step in the diffusion peak time set to the total of the time series, and obtain the diffusion concentration. In one embodiment, after forming a set of diffusion peak times, the pollution diffusion characterization coefficient values ​​corresponding to each time step in the set are extracted and accumulated; at the same time, the values ​​of all time steps in the complete pollution diffusion characterization coefficient time series are summed; then, the ratio of the accumulated value in the set of diffusion peak times to the sum of the time series is calculated to obtain the diffusion concentration degree, which reflects the degree of concentration of pollution diffusion intensity during the peak period.

[0086] For example, if the sum of the time series of pollution diffusion characterization coefficients in a certain area is 50, and the sum of the coefficients corresponding to the diffusion peak time set is 20, then the diffusion concentration is 0.4.

[0087] Step S363: Using diffusion concentration as the weight, the scheduling sensitivity coefficient is weighted to obtain the weighted value of the scheduling response; In one embodiment, after obtaining the scheduling sensitivity coefficient and diffusion concentration, the diffusion concentration is used as a weighting factor and multiplied by the scheduling sensitivity coefficient of the corresponding grid area to obtain a weighted scheduling response value. This weighted scheduling response value is used to highlight areas that exhibit significant response characteristics to scheduling changes during periods of high-intensity pollution diffusion.

[0088] For example, if the scheduling sensitivity coefficient of a certain region is 0.6 and the diffusion concentration is 0.5, then the scheduling response weighting value of that region is 0.3.

[0089] Step S364: Calculate the number of sign changes in the pollution diffusion characterization coefficient time series in continuous time steps to obtain the number of diffusion direction reversals; In one embodiment, for the time series of pollution diffusion characterization coefficients, the direction of numerical change between adjacent time steps is calculated. When the direction of change between a certain time step and the previous time step switches from increasing to decreasing or from decreasing to increasing, the change is determined as a sign change. The number of sign changes in the entire time series is accumulated and statistically analyzed to obtain the number of diffusion direction reversals, which is used to reflect the repeated fluctuation characteristics of pollution diffusion intensity in the time dimension.

[0090] For example, if a time series shows a trend of "rising - rising - falling - falling - rising", then there are two reversals in the direction of diffusion.

[0091] Step S365: Divide the weighted value of the scheduling response by the number of diffusion direction reversals plus one to obtain the direction correction value; multiply the direction correction value by the diffusion concentration to obtain the visual evaluation value of the grid region.

[0092] In one embodiment, after obtaining the scheduling response weighted value and the number of diffusion direction reversals, the number of diffusion direction reversals is incremented by one as the denominator of the correction factor, and the scheduling response weighted value is divided by the correction factor to weaken the impact of scheduling response under frequent diffusion direction reversals, thereby obtaining the direction correction value; subsequently, the direction correction value is multiplied by the diffusion concentration to form a visual evaluation quantity that comprehensively considers the diffusion intensity concentration, the degree of scheduling response, and the stability of diffusion trend.

[0093] For example, if the scheduling response weighting value of a certain region is 0.3 and the diffusion direction reversal number is 2, then the direction correction value is 0.1; after multiplying by the diffusion concentration of 0.4, the visual evaluation value of the region is 0.04.

[0094] Furthermore, step S4 includes the following steps: Step S41: Summarize the visual evaluation values ​​of all grid areas to form a regional evaluation sequence. Sort the regional evaluation sequence according to the numerical value to obtain the regional priority order. In one embodiment, after calculating the visual assessment values ​​for each grid area, the visual assessment values ​​corresponding to all grid areas are centrally summarized and formed into a one-dimensional regional assessment sequence according to the time step or the current moment. In this regional assessment sequence, each sequence element uniquely corresponds to a grid area, and the element value is the visual assessment value of that grid area at the current time state. Subsequently, the values ​​in the regional assessment sequence are sorted according to a preset sorting rule, preferably sorted from largest to smallest, thereby obtaining the regional priority order of each grid area, which is used to characterize the urgency of concern for each area under the current pollution diffusion situation.

[0095] For example, when the system contains 25 grid regions, the 25 visual evaluation values ​​can be grouped into a sequence of length 25 and sorted by numerical value. The grid region ranked first in the sorting result is determined as the current highest priority region.

[0096] Step S42: According to the priority order of the regions, the grid regions with higher priority are assigned to the main display area of ​​the split-screen comparison display, and the remaining grid regions are assigned to the auxiliary display area. The corresponding split-screen comparison weight parameters are calculated based on the sorting position of each region's visualization evaluation quantity in the region evaluation sequence. In one embodiment, the grid areas are divided into display layers based on region priority. Specifically, grid areas that rank higher in the region priority order and whose quantity meets the display capacity requirements of the main display area are allocated to the main display area for split-screen comparison display, for focused display; the remaining grid areas are allocated to the secondary display area to provide background or auxiliary information. Simultaneously, the system calculates the corresponding split-screen comparison weight parameter based on the ranking position of each grid area in the region evaluation sequence. This split-screen comparison weight parameter is used to quantify the display ratio, resolution proportion, or display prominence of different areas in the split-screen display.

[0097] For example, when the main display area can display up to 4 grid areas at the same time, the system can allocate the top 4 grid areas in terms of priority to the main display area, allocate the remaining areas to the secondary display area, and assign different weight coefficients to the grid areas in the main display area according to their ranking order.

[0098] Step S43: Calculate the maximum and minimum values ​​of the region evaluation sequence, and determine the color level range in the color mapping parameters based on the maximum and minimum values.

[0099] In one embodiment, statistical analysis is performed on all values ​​in the regional evaluation sequence to calculate the maximum and minimum values ​​of the sequence. Based on these maximum and minimum values, the system further determines the color gradation range in the color mapping parameters to establish a mapping relationship between the visual evaluation quantity and the color display. This method ensures that visual evaluation quantities in different numerical ranges correspond to different color levels when displayed, thereby enhancing the intuitive expression of differences between regions.

[0100] For example, when the minimum value of the regional evaluation sequence is 0.15 and the maximum value is 0.92, the system can set the color gradation range to a continuous color gradation from cool to warm colors to correspond to the change in evaluation quantity from low to high.

[0101] Furthermore, step S4 also includes the following steps: Step S44: Calculate the rate of change of the visual evaluation quantity of each grid region between consecutive time steps, and determine the gradient mode in the color mapping parameters based on the magnitude of the rate of change; In one embodiment, for each grid region, the rate of change of its visual evaluation value over consecutive time steps is calculated. This rate of change characterizes the degree of dynamic change of the evaluation value of that region over time. Subsequently, the system determines the gradient mode in the color mapping parameters based on the magnitude of the rate of change for each region, so that regions with a larger rate of change exhibit a more obvious gradient effect in color display, while regions with a smaller rate of change exhibit a relatively gentle color change.

[0102] For example, when the rate of change of the visual evaluation value of a certain grid area is significantly higher than that of other areas in two adjacent time steps, the system can set a faster or stronger color gradient mode for that area.

[0103] Step S45: Statistically analyze the fluctuation frequency of the visualization evaluation quantity of each grid area in the time series, determine the time step interval for animation playback based on the fluctuation frequency, and determine the update frequency for animation playback based on the correspondence between the time step interval and the fluctuation frequency, thereby forming the animation playback parameters; In one embodiment, statistical analysis is performed on the changes in the visual evaluation values ​​of each grid region over a complete time series to calculate the corresponding fluctuation frequency, which reflects the stability or severity of the evaluation value over time. Based on the fluctuation frequency, the time step interval for animation playback is determined, and further, according to the correspondence between the time step interval and the fluctuation frequency, the update frequency for animation playback is determined, thereby forming animation playback parameters so that high-fluctuation areas are displayed at a higher update frequency, while low-fluctuation areas are displayed at a lower update frequency.

[0104] For example, when the visual evaluation value of a certain grid area fluctuates significantly multiple times in a short period of time, the system can shorten its animation time step interval to improve the real-time performance of dynamic presentation.

[0105] Step S46: Combine the split-screen comparison weight parameters, color mapping parameters, and animation playback parameters to form visualization control parameters, and send them to the digital twin visualization platform controller to execute split-screen comparison dynamic rendering.

[0106] In one embodiment, the split-screen comparison weight parameters, color mapping parameters, and animation playback parameters are integrated to form a complete set of visualization control parameters. Subsequently, these visualization control parameters are sent to the controller of the digital twin visualization platform, which then performs dynamic rendering operations for split-screen comparison based on these parameters, achieving coordinated display of different grid areas in terms of spatial layout, color representation, and dynamic effects.

[0107] For example, the digital twin visualization platform can highlight the dynamic changes of high-priority grid areas in the main display area based on the visualization control parameters, while presenting the status of other areas in a simplified manner in the auxiliary display area.

[0108] See Figure 2 Visualize control parameters and display the pollution diffusion situation of the same watershed under different scheduling scenarios or at different times in a split-screen comparison.

[0109] The left and right screens in the diagram correspond to different pollution diffusion characterization results. The river area uses a color mapping method to display the pollution diffusion intensity, with the color intensity changing according to the numerical value of the visualized assessment, to intuitively reflect the spatial distribution differences and evolution trends of pollution. By comparing the location, range, and expansion direction of high-pollution areas in different screens, the impact of changes in scheduling conditions on the pollution diffusion path and scope can be intuitively identified, thereby assisting in scheduling decisions and emergency response analysis.

[0110] See Figure 3 The visualized assessment quantities, scheduling sensitivity coefficients, and pollution diffusion characterization coefficients of each grid area are uniformly mapped to the digital twin visualization interface for comprehensive display.

[0111] The upper part of the map displays pollution status information for key areas within the watershed in map form. Different areas are represented by colors and labels to reflect the magnitude of their corresponding visual assessment quantities. The middle part displays parameter information related to scheduling, which is used to reflect the differences in the response of pollution diffusion under different scheduling schemes. The lower part shows the changes of the pollution diffusion characterization coefficient over time in the form of a time series curve, and is linked with the animation playback parameters to depict the dynamic evolution of the pollution status.

[0112] This comprehensive visualization interface enables the interconnected perception of pollution diffusion status, dispatch response characteristics, and time evolution trends within the same view, thereby improving the intuitiveness and accuracy of pollution situation assessment and dispatch decisions.

[0113] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0114] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A visualization method for water quantity and quality models based on digital twins, characterized in that, Includes the following steps: Step S1: Obtain grid geometric data, pollution concentration data at each time step, and pollution source correlation data, and statistically analyze pollution concentration change indicators, pollution distribution dispersion indicators, and regional diffusion dynamic indicators in different grid areas. Step S2: The ratio of the pollution concentration change index to the pollution distribution dispersion index in the same grid area is used as the pollution activity coefficient; the product of the pollution activity coefficient and the regional diffusion dynamic index is used as the pollution diffusion characterization coefficient. Step S3: Calculate the time series difference of the pollution diffusion characterization coefficient between different scheduling scenarios in the same grid area, and use it as the scheduling sensitivity coefficient of the grid area; A visual assessment metric for pollution situation awareness is constructed based on the scheduling sensitivity coefficient and the pollution diffusion characterization coefficient. Step S4: Determine the visualization control parameters for each grid area based on the visualization evaluation, and send the visualization control parameters to the digital twin visualization platform controller to perform split-screen comparison dynamic rendering.

2. The method for visualizing water quantity and quality models based on digital twins according to claim 1, characterized in that, The indicators for statistically analyzing changes in pollution concentration in step S1 include: The watershed is divided into several grid regions based on grid geometric data. Pollution concentration data in continuous time step series is extracted for each grid region to calculate the average concentration value and form a time series concentration curve. The concentration change between adjacent time steps in the time series concentration curve is calculated as a first-order difference sequence. The difference values ​​of the first-order difference sequence are used to obtain the second-order difference sequence; The average absolute value of each time step in the first-order difference sequence is taken as the average rate of change, and the maximum value of the second-order difference sequence is taken as the peak acceleration. The product of the average rate of change and the peak acceleration is taken as the pollution concentration change index of the grid area.

3. The method for visualizing water quantity and quality models based on digital twins according to claim 2, characterized in that, The statistical indicators of pollution distribution dispersion in step S1 include: For each grid region, at each time step, the pollution concentration data of all grid nodes in that region are extracted to form the spatial distribution of the concentration in that region at that time step; Identify the spatial coordinates of concentration peak locations from the spatial distribution of concentration; Calculate the degree of deviation of the pollution concentration of all grid nodes in the region from the average concentration of the region at that time step, and take the sum of the squares of the deviations of each node as the variance of the concentration distribution at that time step; The spatial displacement distance of the concentration peak position between consecutive time steps is tracked, and the peak displacement distance is multiplied by the concentration distribution variance as the pollution distribution dispersion index for that grid area.

4. The method for visualizing water quantity and quality models based on digital twins according to claim 3, characterized in that, The statistical regional diffusion dynamics indicators in step S1 include: The spatial range of grid nodes within the tracked grid area where the pollution concentration exceeds a preset threshold is recorded as the threshold-exceeding range. Record the boundary expansion rate of the threshold range in the time series, and calculate the rate of change of the influence area of ​​the threshold range between consecutive time steps; Extract the mainstream direction information of the grid area from the pollution source association data, identify the main expansion direction of the exceeding threshold range, and determine whether the main expansion direction is consistent with the mainstream direction. Multiply the boundary expansion rate by the rate of change of the affected area. When the expansion direction is consistent with the mainstream direction, the product is amplified by a preset amplification ratio. When they are inconsistent, the product is reduced by a preset reduction ratio to obtain the regional diffusion dynamic index.

5. The method for visualizing water quantity and quality models based on digital twins according to claim 4, characterized in that, The main expansion directions of the identification exceeding the threshold range include: Extract the set of boundary grid nodes exceeding the threshold range at each time step; For each boundary node in the boundary grid node set, calculate the spatial position difference between the node at the current time step and the previous time step to obtain the displacement vector of the node; Based on the magnitude of the pollution concentration gradient corresponding to each boundary node, the displacement vector of each boundary node is assigned a weight, and the weighted displacement vectors of all boundary nodes are summed to obtain the composite displacement vector. Calculate the direction angle of the composite displacement vector, which is the main expansion direction of the over-threshold range.

6. The method for visualizing water quantity and quality models based on digital twins according to claim 5, characterized in that, Step S3 includes the following steps: Step S31: For the same grid area, extract the pollution diffusion characterization coefficients of the area under at least two different scheduling scenarios to form a time series of pollution diffusion characterization coefficients of the area under each scheduling scenario; Step S32: Align the time steps of the pollution diffusion characterization coefficient time series of different scheduling scenarios, calculate the difference of the pollution diffusion characterization coefficient of different scenarios at each time step, and form a difference sequence; Step S33: Calculate the standard deviation of the difference series as the amplitude of the difference fluctuation.

7. The method for visualizing water quantity and quality models based on digital twins according to claim 6, characterized in that, Step S3 also includes the following steps: Step S34: Count the number of time steps in the difference sequence whose absolute value exceeds the preset difference threshold, and calculate the proportion of the number of time steps to the total number of time steps as the proportion of the period with significant difference. Step S35: Multiply the amplitude of the difference fluctuation by the proportion of the period with significant difference to obtain the scheduling sensitivity coefficient of the grid area; Step S36: Construct a visual assessment quantity for pollution situation awareness based on the scheduling sensitivity coefficient and the pollution diffusion characterization coefficient.

8. The method for visualizing water quantity and quality models based on digital twins according to claim 7, characterized in that, Step S36 includes the following steps: Step S361: Identify local maxima time steps in the pollution diffusion characterization coefficient time series to form a diffusion peak time set; Step S362: Calculate the proportion of the pollution diffusion characterization coefficient corresponding to each time step in the diffusion peak time set to the total of the time series, and obtain the diffusion concentration. Step S363: Using diffusion concentration as the weight, the scheduling sensitivity coefficient is weighted to obtain the weighted value of the scheduling response; Step S364: Calculate the number of sign changes in the pollution diffusion characterization coefficient time series in continuous time steps to obtain the number of diffusion direction reversals; Step S365: Divide the weighted value of the scheduling response by the number of diffusion direction reversals plus one to obtain the direction correction value; multiply the direction correction value by the diffusion concentration to obtain the visual evaluation value of the grid region.

9. The method for visualizing water quantity and quality models based on digital twins according to claim 8, characterized in that, Step S4 includes the following steps: Step S41: Summarize the visual evaluation values ​​of all grid areas to form a regional evaluation sequence. Sort the regional evaluation sequence according to the numerical value to obtain the regional priority order. Step S42: According to the priority order of the regions, the grid regions with higher priority are assigned to the main display area of ​​the split-screen comparison display, and the remaining grid regions are assigned to the auxiliary display area. The corresponding split-screen comparison weight parameters are calculated based on the sorting position of each region's visualization evaluation quantity in the region evaluation sequence. Step S43: Calculate the maximum and minimum values ​​of the region evaluation sequence, and determine the color level range in the color mapping parameters based on the maximum and minimum values.

10. The method for visualizing water quantity and quality models based on digital twins according to claim 9, characterized in that, Step S4 also includes the following steps: Step S44: Calculate the rate of change of the visual evaluation quantity of each grid region between consecutive time steps, and determine the gradient mode in the color mapping parameters based on the magnitude of the rate of change; Step S45: Statistically analyze the fluctuation frequency of the visualization evaluation quantity of each grid area in the time series, determine the time step interval for animation playback based on the fluctuation frequency, and determine the update frequency for animation playback based on the correspondence between the time step interval and the fluctuation frequency, thereby forming the animation playback parameters; Step S46: Combine the split-screen comparison weight parameters, color mapping parameters, and animation playback parameters to form visualization control parameters, and send them to the digital twin visualization platform controller to execute split-screen comparison dynamic rendering.

Images