A watershed ecological flow regulation method and system based on hydrophilic biological protection

Abstract

The application discloses a kind of based on the hydrophilic biological protection watershed ecological flow scheduling method and system, by comprehensively considering the final water flow velocity in target watershed and final flow direction as the first influencing factor, water conservancy engineering facilities of tributary where target watershed is as the second influencing factor, the activity range and migration route of target species are set as seed cell and organization cell, seed cell and organization cell are evolved using cellular automaton model control, and the evolution result of the target watershed is obtained;Subsequently, the ecological situation of the target species in the target watershed is determined according to the evolution result, and the corresponding scheduling strategy is selected in the resource library, and the intelligent scheduling of the target watershed is completed;To solve the technical problem that the existing technology cannot realize comprehensive scheduling in the face of whole watershed ecology, the present scheme uses cellular automaton model to comprehensively deduce the multidimensional data of whole watershed ecology, realizes the intelligent scheduling and ecological restoration of whole watershed.

Description

Technical Field

[0001] This invention relates to the field of watershed big data technology, and in particular to a watershed ecological flow scheduling method and system based on the protection of hydrophilic organisms. Background Technology

[0002] "Aquatic organisms" refer to groups of organisms that rely on water bodies (including freshwater, seawater, and wetlands) to complete part or all of their life cycle. This includes fish, amphibians, aquatic mammals (such as whales and dolphins), aquatic invertebrates (such as shellfish, shrimp, and crabs), aquatic plants (such as reeds and Vallisneria natans), and birds that depend on water bodies (such as wading birds and waterfowl). "Aquatic organism conservation" is a systematic action aimed at maintaining the integrity of aquatic ecosystems and ensuring biodiversity. It involves scientific intervention and management to mitigate the threats to the survival of aquatic organisms caused by human activities and environmental changes. It is a key area of ​​global biodiversity conservation and ecological governance. In the aforementioned watershed ecological governance models, the core challenge lies in how to dynamically adjust the state or allocation of controllable natural resources and engineering facilities within the watershed to achieve synergistic optimization of ecological protection and resource utilization. This includes water resource allocation, engineering facility allocation, ecological restoration resource allocation, and data and computing power allocation.

[0003] Traditional governance models (such as single-project restoration and static scheduling management) are no longer sufficient to cope with the dual pressures of climate change and human activities. Engineers utilize data collected from the watershed and their own experience to schedule resources in a single dimension. While this can temporarily address the rationality of that single resource, it cannot simultaneously address the rationality of scheduling in other dimensions (such as water resources, ecological restoration, and data and computing power). Considering the overall ecology of the entire watershed, its positive effect is far from adequate. Furthermore, relying solely on experience to select scheduling strategies and manually judging and extrapolating watershed data for a future period based on personal experience is inaccurate and inefficient. Against this backdrop, facing the comprehensive scheduling challenges of watershed ecological restoration and intelligent scheduling, there is an urgent need for an artificial intelligence strategy for watershed ecological flow scheduling based on the protection of aquatic organisms.

[0004] Cellular automata (CA) models are discrete, rule-based dynamical systems in which a large number of simple units (cells) interact locally on a spatial grid to form complex global behaviors. In recent years, they have become a mainstream model in ecological simulation studies; utilizing grid dynamics where spatial interactions and temporal causality are both local, they have the ability to simulate the spatiotemporal evolution of complex systems.

[0005] Therefore, how to use cellular automata models to comprehensively extrapolate multi-dimensional data on the entire watershed's ecology and achieve ecological restoration and intelligent scheduling across the entire watershed is a challenge that this solution will address. Summary of the Invention

[0006] This invention provides a watershed ecological flow scheduling method and system based on hydrophilic organism protection, in order to solve the technical problem that existing technologies cannot achieve comprehensive scheduling of the entire watershed ecology. This solution uses a cellular automata model to comprehensively extrapolate multi-dimensional data of the entire watershed ecology, thereby realizing the ecological restoration and intelligent scheduling of the entire watershed.

[0007] To address the aforementioned technical problems, embodiments of the present invention provide a watershed ecological flow scheduling method based on hydrophilic organism protection, comprising:

[0008] The initial water flow velocity and initial flow direction in the target watershed are collected. The target watershed is then acquired using a drone within a preset time period using remote sensing images. The initial water flow velocity and initial flow direction are then optimized using the remote sensing images to obtain the final water flow velocity and final flow direction of the target watershed.

[0009] Obtain the maximum operating efficiency and location information of water conservancy facilities in the tributaries of the target watershed;

[0010] Identify the target species to be studied, locate the target species in the target watershed using the remote sensing images, and determine the activity range and migration route of the target species;

[0011] The target watershed is gridded and a spatial coordinate system is established. At the same time, a cellular automata model is invoked, and the final water flow velocity and final flow direction of the target watershed are used as the first influencing factor, and the maximum operating efficiency of the water conservancy facilities is used as the second influencing factor. The grid cells occupied by the target species are determined in the spatial coordinate system according to the activity range and migration route of the target species.

[0012] The grid cells occupied by the target species in the activity range are set as seed cells, and the grid cells occupied by the target species in the migration route are set as tissue cells. At the same time, the first influencing factor and the second influencing factor are input into the cellular automata model, and the seed cells and the tissue cells are controlled to evolve to obtain the evolution results of the target watershed.

[0013] Based on the evolution results, the ecological status of the target species in the target watershed is determined, and a corresponding scheduling strategy is selected from the resource pool based on the ecological status to complete the intelligent scheduling of the target watershed.

[0014] As a preferred embodiment, the step of optimizing the initial water flow velocity and initial flow direction using the remote sensing image specifically includes:

[0015] The remote sensing image is decomposed frame by frame. The remote sensing images of two adjacent frames are highlighted. The highlighted feature points in the remote sensing images of the two adjacent frames are determined respectively. The highlighted feature points in the earlier remote sensing image are set as the reference points, and the highlighted feature points in the later remote sensing image are set as the comparison points.

[0016] Calculate the spatial distance value between the comparison points corresponding to each reference point, and at the same time, determine the original position of the reference point in the remote sensing image, and modify the original position according to the spatial distance value to obtain the behavioral trajectory of the reference point;

[0017] The initial water flow velocity and initial flow direction are corrected based on the spatial distance value and the behavioral trajectory to obtain the final water flow velocity and final flow direction.

[0018] As a preferred embodiment, the step of locating the target species in the target watershed using the remote sensing image specifically includes:

[0019] The biological characteristics information of the target species are obtained, and a target recognition model is constructed. The biological characteristics information is input into the target recognition model for training to generate a target species capture model.

[0020] The remote sensing image is input into the target species capture model for processing, and the tagging information of the target species in the remote sensing image is output.

[0021] The marker information in the remote sensing image is located, and those marker information appearing within a preset spatial range that is less than a preset threshold are identified as noise and deleted. The remaining marker information in the filtered remote sensing image is located to determine the activity range and migration route of the target species.

[0022] As a preferred embodiment, the step of using the final flow velocity and final flow direction of the target watershed as the first influencing factor specifically includes:

[0023] The final water flow velocity of the target watershed is set as the first water velocity influence factor, and the final flow direction of the target watershed is set as the first flow direction influence factor.

[0024] Set time units, segment the first water velocity influence factor according to the time units to obtain a discrete set of water velocity, and segment the first flow direction influence factor according to the time units to obtain a discrete set of flow direction;

[0025] The discrete sets of water velocity and flow direction at the same time unit are mapped to generate a discrete set that simultaneously has water velocity data and flow direction data at the same time unit, which is used as the first influencing factor.

[0026] As a preferred embodiment, the second influencing factor further includes the location information of the water conservancy engineering facilities; specifically, the step of using the maximum operating efficiency of the water conservancy engineering facilities as the second influencing factor involves:

[0027] Based on the activity range of the target species and the location information of the water conservancy facilities, the original influence distance between the target species and different water conservancy facilities is calculated;

[0028] Based on the migration route of the target species and the location information of the water conservancy facilities, the migration distance of the target species in each time unit is calculated.

[0029] Using the migration distance value as a weight, the original influence distance at each time unit is multiplied by the corresponding weight and the maximum operating efficiency of the water conservancy facility to obtain the migration influence value of the target species under the influence of the water conservancy facility at different time units, which is used as the second influence factor.

[0030] As a preferred embodiment, the step of controlling the evolution of the seed cells and the tissue cells specifically includes:

[0031] The seed cells and tissue cells are controlled to grow separately, and the growth probabilities of the seed cells and tissue cells are corrected in real time according to the first influence factor and the second influence factor.

[0032] The seed cells and tissue cells are monitored within a preset evolution time. When it is determined that the growth area of ​​the seed cells reaches a preset area, or the tissue cells stop growing, or the evolution time reaches the preset evolution time, the seed cells and tissue cells are controlled to stop growing, and the evolution ends.

[0033] As a preferred embodiment, prior to controlling the evolution of the seed cells and the tissue cells, the method further includes:

[0034] Calculate the maximum processing value based on the current amount of all acquired data; simultaneously, obtain the maximum processing value for each processing node in the current server.

[0035] Determine the load capacity of each computing node in different time units, and correct the maximum computing value of each computing node according to the load capacity. Match the corrected maximum computing value with the maximum processing value to determine the computing node with operational risk in a certain time unit.

[0036] The computational node that exhibits operational risk at a certain time unit is used as a third influencing factor and input into the cellular automata model.

[0037] Accordingly, another embodiment of the present invention also provides a watershed ecological flow scheduling system based on hydrophilic organism protection, including: a water flow data module, an engineering data module, a species data module, an influencing factor module, a model evolution module, and an intelligent scheduling module;

[0038] The water flow data module is used to collect the initial water flow velocity and initial flow direction in the target watershed, use a drone to acquire continuous remote sensing images of the target watershed within a preset time period, and optimize the initial water flow velocity and initial flow direction using the remote sensing images to obtain the final water flow velocity and final flow direction of the target watershed.

[0039] The engineering data module is used to obtain the maximum operating efficiency and location information of the water conservancy engineering facilities in the tributary of the target watershed;

[0040] The species data module is used to identify the target species to be studied, locate the target species in the target watershed through the remote sensing images, and determine the activity range and migration route of the target species.

[0041] The influencing factor module is used to perform gridding processing on the target watershed and establish a spatial coordinate system. At the same time, it calls the cellular automata model, takes the final water flow velocity and final flow direction of the target watershed as the first influencing factor, takes the maximum operating efficiency of the water conservancy facilities as the second influencing factor, and determines the grid cell occupied by the target species in the spatial coordinate system according to the activity range and migration route of the target species.

[0042] The model evolution module is used to set the grid cells occupied by the target species in the activity range as seed cells and the grid cells occupied by the target species in the migration route as tissue cells. At the same time, the first influencing factor and the second influencing factor are input into the cellular automaton model, and the seed cells and tissue cells are controlled to evolve to obtain the evolution results of the target watershed.

[0043] The intelligent scheduling module is used to determine the ecological status of the target species in the target watershed based on the evolution results, select the corresponding scheduling strategy from the resource pool based on the ecological status, and complete the intelligent scheduling of the target watershed.

[0044] As a preferred embodiment, the water flow data module is used to optimize the initial water flow velocity and initial flow direction using the remote sensing image. Specifically, this includes: decomposing the remote sensing image frame by frame; highlighting adjacent frames; determining the highlighted feature points in the adjacent frames; setting the highlighted feature points in the earlier remote sensing image as reference points; setting the highlighted feature points in the later remote sensing image as comparison points; calculating the spatial distance between the comparison points corresponding to each reference point; determining the original position of the reference point in the remote sensing image; modifying the original position based on the spatial distance value to obtain the behavioral trajectory of the reference point; and correcting the initial water flow velocity and initial flow direction based on the spatial distance value and the behavioral trajectory to obtain the final water flow velocity and final flow direction.

[0045] As a preferred embodiment, the species data module is used to locate the target species in the target watershed using the remote sensing image. Specifically, this includes: acquiring the biological characteristic information of the target species; simultaneously, constructing a target recognition model; inputting the biological characteristic information into the target recognition model for training to generate a target species capture model; inputting the remote sensing image into the target species capture model for processing; outputting the marker information of the target species in the remote sensing image; locating the marker information in the remote sensing image; identifying and deleting marker information with a total number of markers appearing within a preset spatial range that is less than a preset threshold, and locating the remaining marker information in the filtered remote sensing image to determine the activity range and migration route of the target species.

[0046] As a preferred embodiment, the step of using the final water flow velocity and final flow direction of the target watershed as the first influencing factor specifically includes: setting the final water flow velocity of the target watershed as the first water velocity influencing factor, and setting the final flow direction of the target watershed as the first flow direction influencing factor; setting a time unit, dividing the first water velocity influencing factor according to the time unit to obtain a discrete set of water velocity, dividing the first flow direction influencing factor according to the time unit to obtain a discrete set of flow direction; mapping the discrete sets of water velocity and flow direction in the same time unit to generate a discrete set that simultaneously has water velocity data and flow direction data in the same time unit, which is used as the first influencing factor.

[0047] As a preferred embodiment, the second influencing factor further includes the location information of the water conservancy engineering facilities. Specifically, the step of using the maximum operating efficiency of the water conservancy engineering facilities as the second influencing factor in the influencing factor module involves: calculating the original influence distance between the target species and different water conservancy engineering facilities based on the activity range of the target species and the location information of the water conservancy engineering facilities; calculating the migration distance value of the target species in each time unit based on the migration route of the target species and the location information of the water conservancy engineering facilities; using the migration distance value as a weight, multiplying the original influence distance in each time unit by the corresponding weight and the maximum operating efficiency of the water conservancy engineering facility, to obtain the migration influence value of the target species under the influence of the water conservancy engineering facilities in different time units, which is then used as the second influencing factor.

[0048] As a preferred embodiment, the model evolution module is used to control the evolution of the seed cell and the tissue cell, specifically including: controlling the growth of the seed cell and the tissue cell respectively, and simultaneously correcting the growth probability of the seed cell and the tissue cell in real time according to the first influencing factor and the second influencing factor; monitoring the seed cell and the tissue cell within a preset evolution time, and controlling the seed cell and the tissue cell to stop growing and ending the evolution when it is determined that the growth area of ​​the seed cell reaches a preset area, or the tissue cell stops growing, or the evolution time reaches the preset evolution time.

[0049] As a preferred embodiment, the system further includes: a computing power impact module; the computing power impact module is used to calculate the current amount of all acquired data to obtain the maximum processing value before the control of the seed cell and the tissue cell evolves; simultaneously, it obtains the maximum processing value of each computing node in the current server; determines the load of each computing node in different time units, and corrects the maximum processing value of each computing node according to the load, matches the corrected maximum processing value with the maximum processing value, and determines the computing node with operational risk in a certain time unit; the computing node with operational risk in a certain time unit is used as a third impact factor and input into the cellular automata model.

[0050] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0051] This invention comprehensively considers the final flow velocity and final flow direction in the target watershed as the first influencing factor, the hydraulic engineering facilities of the tributaries where the target watershed is located as the second influencing factor, and sets the activity range and migration route of the target species as seed cells and tissue cells. A cellular automata model is used to control the evolution of the seed cells and tissue cells to obtain the evolutionary results of the target watershed. Subsequently, based on the evolutionary results, the ecological status of the target species in the target watershed is determined, and a corresponding scheduling strategy is selected from the resource pool to complete the intelligent scheduling of the target watershed. This addresses the technical problem of existing technologies being unable to achieve comprehensive scheduling of the entire watershed ecosystem. This solution utilizes a cellular automata model to comprehensively extrapolate multi-dimensional data of the entire watershed ecosystem, achieving ecological restoration and intelligent scheduling of the entire watershed. Attached Figure Description

[0052] Figure 1 This is a schematic flowchart of a watershed ecological flow scheduling method based on hydrophilic organism protection according to an embodiment of the present invention.

[0053] Figure 2 This is a schematic diagram of a watershed ecological flow scheduling system based on hydrophilic organism protection, according to an embodiment of the present invention. Detailed Implementation

[0054] 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.

[0055] Example 1

[0056] Please refer to Figure 1 The flowchart of a watershed ecological flow scheduling method based on hydrophilic organism protection provided by an embodiment of the present invention includes steps 101 to 106, each step as follows:

[0057] Step 101: Collect the initial water flow velocity and initial flow direction in the target watershed, use a drone to acquire continuous remote sensing images of the target watershed within a preset time period, and optimize the initial water flow velocity and initial flow direction using the remote sensing images to obtain the final water flow velocity and final flow direction of the target watershed.

[0058] In the first aspect of this embodiment, the step of optimizing the initial water flow velocity and initial flow direction using the remote sensing image specifically includes: decomposing the remote sensing image frame by frame, highlighting two adjacent frames of the remote sensing image, determining the highlighted feature points in the two adjacent frames of the remote sensing image respectively, setting the highlighted feature points in the earlier remote sensing image as reference points, and setting the highlighted feature points in the later remote sensing image as comparison points; calculating the spatial distance value between the comparison points corresponding to each reference point, and simultaneously determining the original position of the reference point in the remote sensing image, and modifying the original position according to the spatial distance value to obtain the behavioral trajectory of the reference point; correcting the initial water flow velocity and initial flow direction according to the spatial distance value and the behavioral trajectory to obtain the final water flow velocity and final flow direction.

[0059] Specifically, data acquisition and preprocessing: Initial hydrological data acquisition: Acquire initial water flow velocity (e.g., ADCP velocity data) and initial flow direction (e.g., compass azimuth) of the target watershed through hydrological monitoring stations or buoy sensors. The data format must be uniformly in a geographic coordinate system (e.g., WGS84). Example: If the initial data are discrete point measurements, a continuous velocity / flow direction field within the watershed needs to be generated through Kriging interpolation. UAV remote sensing image acquisition: Use a multispectral UAV (e.g., DJIP4 Multispectral) to capture visible light and near-infrared band images at fixed intervals (e.g., one frame every 30 seconds) during a preset time period (e.g., 10:00-14:00 daily during flood season). The resolution must be ≥10cm / pixel. The recommended flight altitude is 100-300 meters, and POS data (position, attitude, timestamp) should be recorded simultaneously.

[0060] Image preprocessing: Radiometric correction (eliminating illumination changes), geometric correction (based on DEM registration), and denoising processing (such as non-local mean filtering) are performed on remote sensing images to ensure spatial alignment of temporal images. Highlight feature point extraction and matching: Feature enhancement processing: The following two methods are used in parallel processing: (1) Optical flow method: The Farneback dense optical flow algorithm is applied to adjacent frame images to extract the water flow motion vector field; (2) Feature point method: Highlight feature points (such as water surface ripples and floating objects) are detected using SIFT / SURF operators, and mismatched points are removed by RANSAC algorithm. Optimization strategy: Near-infrared band can enhance the contrast between water and land boundaries and is given priority for feature extraction. Spatiotemporal correlation modeling: The displacement matrix of the reference point (previous frame) and the comparison point (later frame) is established:

[0061] ;in The spatial distance value is improved through sub-pixel level interpolation; Δx and Δy are the displacement components in the x and y directions. Let be the grayscale value of the image at coordinates (x, y) in frame t; This is to minimize the error function.

[0062] Trajectory Reconstruction and Hydrological Parameter Optimization: Trajectory Generation Algorithm: Kalman filtering is applied to the displacements of feature points in N consecutive frames (e.g., 10 frames) to smooth the data and generate spatiotemporal trajectories. Watershed DEM data is introduced to correct projection errors caused by topography. Velocity / Flow Direction Correction Model:

[0063] Construct the optimization function: ;

[0064] ;

[0065] Where α and β are weighting coefficients (default 0.5), which are dynamically adjusted by fitting the measured data using the least squares method; To ultimately optimize the water flow velocity; To ultimately optimize the water flow direction; The time interval between adjacent images; Calculate the displacement direction angle.

[0066] Dynamic verification mechanism: Verification points are set up at river bifurcation points to compare the speed retrieved by UAVs with the data from radar velocimeters. When the relative error is greater than 15%, re-optimization is triggered.

[0067] This implementation method improves the accuracy of traditional hydrological monitoring by more than 40% through multi-source data fusion and spatiotemporal sequence analysis. It is particularly suitable for remote watersheds without ground monitoring stations and can be used for flash flood early warning. During heavy rain, it updates the flow field every 5 minutes and identifies areas of abnormal acceleration (such as...). A red alert is triggered when the speed is greater than 3 m / s. It is also used for sediment transport research, calculating sediment transport rates using long-term trajectory data, with a correlation coefficient R² reaching 0.82.

[0068] Step 102: Obtain the maximum operating efficiency and location information of the water conservancy facilities in the tributary where the target watershed is located.

[0069] Specifically, facility data is dynamically acquired: Data source: The maximum operating efficiency of reservoirs / dams on tributaries (such as maximum flood discharge Q_max, in m³ / s) is obtained in real time through the water resources department's API; facility location information (latitude and longitude coordinates, control basin area A_control) is extracted from the engineering GIS database.

[0070] Spatial correlation modeling: The location of the facility is marked on the UAV remote sensing image, and the radius of influence R of the facility is established. The facility location is overlaid with the water flow trajectory for analysis to identify the water area directly affected by the project (e.g., when the distance of the trajectory point from the facility is ≤R, it is marked as the control zone).

[0071] Operational efficiency coupling correction: When water flow is detected entering the control zone, the velocity is corrected according to the facility's operating status. If the facility is operating at full load (Q_current / Q_max≥0.9), the velocity correction is ignored.

[0072] Step 103: Identify the target species to be studied, locate the target species in the target watershed using the remote sensing images, and determine the activity range and migration route of the target species.

[0073] In a first aspect of this embodiment, the step of locating the target species in the target watershed using the remote sensing image specifically includes: acquiring the biological characteristic information of the target species; simultaneously, constructing a target recognition model; inputting the biological characteristic information into the target recognition model for training to generate a target species capture model; inputting the remote sensing image into the target species capture model for processing to output the marker information of the target species in the remote sensing image; locating the marker information in the remote sensing image; identifying and deleting marker information with a total number of markers appearing within a preset spatial range that is less than a preset threshold, and locating the remaining marker information in the filtered remote sensing image to determine the activity range and migration route of the target species.

[0074] Specifically, in the implementation process, the target species (such as various migratory organisms) are first identified based on ecological survey data, and their biological characteristics are extracted (such as the spindle-shaped body shape and gray-brown spectral reflectance curve of the back of fish species; the high reflectance characteristics of white feathers of birds in the near-infrared band). A target recognition model is built using a deep learning framework (such as YOLOv7 or MaskR-CNN), and trained on more than 5,000 UAV remote sensing images containing the target species. Data augmentation techniques (such as random rotation and illumination changes) are used during training to improve the model's generalization ability. The trained target species capture model is deployed to an edge computing device to process the real-time acquired remote sensing images frame by frame, and output bounding box information with confidence scores (threshold set above 0.8). The marked points are aggregated and analyzed using a spatial clustering algorithm (such as DBSCAN), and an activity range threshold is set (such as the number of marked points ≥ 5 and the spatial density ≥ 1 per hectare in 10 consecutive frames) to remove isolated noise points (such as misidentified floating objects). For migration route tracking, a time-series trajectory fusion technology is used to overlay daily location results onto the watershed digital elevation model (DEM). Combined with water flow velocity data (from the aforementioned hydrological model), the correlation between species movement direction and hydraulic factors is analyzed. Finally, a heat map of activity range (spatial resolution 1km×1km) and migration path vector lines (accuracy ±50m) are output. The spatiotemporal overlap area between species distribution and water conservancy facilities is displayed in real time through a GIS platform, providing a basis for ecological scheduling decisions (such as controlling the gate opening time to avoid fish migration peaks).

[0075] Step 104: The target watershed is gridded and a spatial coordinate system is established. At the same time, the cellular automata model is invoked, and the final water flow velocity and final flow direction of the target watershed are used as the first influencing factor, and the maximum operating efficiency of the water conservancy facilities is used as the second influencing factor. The grid cells occupied by the target species are determined in the spatial coordinate system according to the activity range and migration route of the target species.

[0076] In a first aspect of this embodiment, the step of using the final flow velocity and final flow direction of the target watershed as the first influencing factor specifically includes: setting the final flow velocity of the target watershed as the first flow velocity influencing factor, and setting the final flow direction of the target watershed as the first flow direction influencing factor; setting a time unit, dividing the first flow velocity influencing factor according to the time unit to obtain a discrete set of flow velocity, dividing the first flow direction influencing factor according to the time unit to obtain a discrete set of flow direction; mapping the discrete set of flow velocity and the discrete set of flow direction in the same time unit to generate a discrete set that simultaneously has flow velocity data and flow direction data in the same time unit, which is used as the first influencing factor.

[0077] Specifically, in the implementation process, the target watershed is first gridded, using regular rectangular grids (e.g., 100m×100m) or irregular triangular meshes (TINs), and a UTM spatial coordinate system is established to ensure geometric accuracy. A cellular automata (CA) model is then invoked, defining each grid cell in the watershed as a cell, whose state is driven by three types of parameters: hydrological, engineering, and ecological. The first influencing factor (hydrodynamic factor): the optimized final flow velocity and direction are discretized by time units (e.g., 1 hour), generating a spatiotemporal mapping matrix. For each grid cell in a time slice, its hydrodynamic propagation intensity is calculated. The second influencing factor (engineering control factor): for grid cells within the influence radius R of the hydraulic facilities, an operational efficiency attenuation coefficient is introduced, applying a 15%~30% attenuation to the final flow velocity of that grid (the specific value is determined through Manning's formula inversion). Species distribution constraints: Based on the activity range output by the target species capture model, occupied grids are marked (e.g., habitat grids are assigned a value of 1, others are assigned 0). Ecological protection priorities are set in the CA evolution rules: if a grid simultaneously meets the threshold and is a species-occupied area, an ecological safety limit on water flow velocity is triggered (e.g., forcing the final water flow velocity ≤ 1.2 m / s). Through three-factor coupled iterative calculation (time step Δt = 10 minutes), the watershed grid state evolution sequence is dynamically output.

[0078] Furthermore, in another embodiment, the second influencing factor further includes the location information of the water conservancy engineering facilities; in the step of using the maximum operating efficiency of the water conservancy engineering facilities as the second influencing factor, specifically: based on the activity range of the target species and the location information of the water conservancy engineering facilities, the original influence distance between the target species and different water conservancy engineering facilities is calculated; based on the migration route of the target species and the location information of the water conservancy engineering facilities, the migration distance value of the target species in each time unit is calculated; using the migration distance value as a weight, the original influence distance in each time unit is multiplied by the corresponding weight and the maximum operating efficiency of the water conservancy engineering facility to obtain the migration influence value of the target species under the influence of the water conservancy engineering facilities in different time units, which is used as the second influencing factor.

[0079] Specifically, in the implementation process, the system comprehensively considers the location information of water conservancy facilities and the spatiotemporal distribution characteristics of target species to dynamically assess the impact of project operation on species migration. First, based on the precise coordinates of the water conservancy facilities (such as the center point of dams and reservoirs) and the activity range boundaries of the target species, the actual water flow distance between them is calculated. This distance is a network distance measured along the river centerline, rather than a simple straight-line distance, to more accurately reflect the hydraulic connection. For target species with migratory habits (such as migratory fish), the system continuously tracks their migration routes and dynamically generates weighting coefficients by analyzing the species' movement distances in different time periods. This coefficient reflects the species' sensitivity to the influence of water conservancy facilities at a specific time period—the weighting coefficient increases when a species rapidly approaches a facility, indicating that the facility's current impact on the species is more significant. Combining the maximum operating efficiency of the water conservancy facilities (such as maximum flood discharge capacity), the system calculates the comprehensive impact value of the facilities on species migration in each time period. This impact value considers not only the distance between facilities and species, but also the operational intensity of facilities (such as the current discharge volume relative to maximum capacity) and the migration status of species (such as the different sensitivities during the breeding season or non-breeding season). When the impact value exceeds a preset threshold, the system automatically adjusts the water flow velocity in the relevant water area and highlights the "engineering-ecological conflict zone" on the electronic map, providing intuitive decision support for ecological scheduling. Practical applications show that this dynamic assessment method can significantly improve the accuracy of water conservancy project regulation. Simultaneously, through a visual interface, managers can clearly understand the degree of ecological impact of different facilities at different times, achieving more scientific integrated watershed management.

[0080] Step 105: Set the grid cells occupied by the target species in the activity range as seed cells, and set the grid cells occupied by the target species in the migration route as tissue cells. At the same time, input the first influencing factor and the second influencing factor into the cellular automata model, and control the seed cells and tissue cells to evolve to obtain the evolution result of the target watershed.

[0081] In a first aspect of this embodiment, the step of controlling the evolution of the seed cell and the tissue cell specifically includes: controlling the seed cell and the tissue cell to grow respectively, and simultaneously correcting the growth probability of the seed cell and the tissue cell in real time according to the first influencing factor and the second influencing factor; monitoring the seed cell and the tissue cell within a preset evolution time, and controlling the seed cell and the tissue cell to stop growing and ending the evolution when it is determined that the growth area of ​​the seed cell reaches a preset area, or the tissue cell stops growing, or the evolution time reaches the preset evolution time.

[0082] Specifically, this system employs cellular automata (CA) technology to construct a dynamic evolutionary model, performing a spatiotemporal coupling analysis of the target species' habitat distribution and the impact of water conservancy projects. During implementation, the watershed grid is first labeled with ecological attributes: grid cells occupied by the target species' stable activity areas (such as breeding areas and foraging areas) are marked as "seed cells," which possess self-sustaining and expansion characteristics; simultaneously, grid cells traversed by the species' migration path are marked as "tissue cells," which primarily undertake connection and transmission functions. During the model initialization phase, the system sets differentiated growth rules for each type of cell: seed cells determine their initial growth probability based on the species' habitat suitability index, while tissue cells are configured with transmission weights based on the connectivity of the migration route.

[0083] During the evolution process, the system receives two types of dynamic influencing factors in real time: hydrodynamic factors (flow velocity and flow direction data) adjust growth conduction efficiency by changing the strength of hydraulic connections between cells. Specifically, when the flow velocity exceeds the species tolerance threshold, the growth probability of the corresponding cell will decrease according to an S-shaped curve. The hydraulic engineering influencing factors are analyzed by spatial superposition. Cells within the influence radius of the engineering facilities will receive operational efficiency modulation signals. These signals will dynamically correct the cell state according to the intensity of engineering scheduling (such as the proportion of flood discharge). For example, when the flood discharge of the dam reaches 80% of the design capacity, the conduction efficiency of tissue cells within a 5km downstream range will decrease by 30%.

[0084] The model employs an adaptive time-step progression mechanism, detecting three types of termination conditions in each iteration: first, the expansion area of ​​the seed cell cluster reaches the preset ecological protection zone zoning requirements (e.g., core habitat area ≥ 50 km²); second, the connectivity interruption of tissue cells persists for more than 3 time steps (indicating migration corridor disruption); and third, the maximum simulation duration is reached (e.g., 72-hour flood process simulation). When any condition is triggered, the system automatically terminates the evolution and outputs the final cell state distribution map. This result can intuitively display the potential change trend of species habitats under the operation of water conservancy projects, including the displacement of core habitats and the identification of bottleneck sections in migration corridors.

[0085] Furthermore, in another embodiment, before controlling the evolution of the seed cell and the tissue cell, the method further includes: calculating the current amount of all acquired data to obtain the maximum processing value; simultaneously, obtaining the maximum processing value of each computing node in the current server; determining the load of each computing node in different time units, and correcting the maximum processing value of each computing node according to the load, matching the corrected maximum processing value with the maximum processing value, and determining the computing node with operational risk in a certain time unit; and inputting the computing node with operational risk in a certain time unit as a third influencing factor into the cellular automaton model.

[0086] Specifically, during implementation, the system employs a distributed computing architecture to parallelize the cellular automata model, addressing the high-performance computing demands of large-scale watershed grids. Before initiating the evolutionary simulation, the system assesses the computational load in real time: firstly, it calculates the total amount of data to be processed, including the number of watershed grids (e.g., over 1 million cells), the total number of time units (e.g., a 72-hour simulation divided into 432 steps at 10-minute intervals), and the update frequency of dynamic influencing factors, comprehensively calculating the maximum processing value (peak computational load); simultaneously, it monitors the real-time status of each computing node in the server cluster, collecting metrics such as CPU core utilization, memory usage, and network throughput, dynamically adjusting the maximum computing capacity of each node. Through a load balancing algorithm, the system divides the computational task into multiple sub-task packages and dynamically schedules them based on the actual workload of each node in each time unit (e.g., the percentage of completed tasks and the length of the pending task queue). When a computing node is detected to be at risk of overload in a specific time unit (e.g., CPU utilization exceeding 90% for 3 consecutive minutes), the system automatically marks the node as a "risk node" and converts its processing latency, task backlog, and other parameters into a third influencing factor. This factor is input into the cellular automata model in the form of negative feedback, triggering three adaptive adjustments: first, reducing the computational precision of the affected time unit (e.g., temporarily reducing the grid resolution from 100m to 200m); second, dynamically extending the computation time limit of that time step (e.g., increasing it from the default 10 seconds to 30 seconds); and third, activating a backup node to take over some computational tasks. Simultaneously, the system records the risk events of each node to optimize subsequent task allocation strategies, forming a dynamic balance between computational resources, model precision, and timeliness requirements.

[0087] Step 106: Determine the ecological status of the target species in the target watershed based on the evolution results, and select the corresponding scheduling strategy from the resource pool based on the ecological status to complete the intelligent scheduling of the target watershed.

[0088] Specifically, after simulating watershed evolution using a cellular automata model, this system generates multi-dimensional ecological assessment results, including core indicators such as the target species' habitat integrity index, migration corridor connectivity, and engineering impact heatmaps. The system's built-in intelligent decision-making engine performs multi-level matching of these quantitative results with contingency plans in the resource database: First, it triggers tiered responses based on the habitat reduction ratio; when the core habitat area decreases by more than 15%, a level-one response is initiated, automatically matching a reservoir ecological water replenishment plan. Second, it analyzes the distribution of migration obstacle points and intelligently recommends fishway optimization or temporary channel opening schemes for identified key obstruction sections (such as three or more consecutive tissue cell breaks). Finally, combining the spatiotemporal characteristics of engineering impacts, it generates dam and gate scheduling suggestions (such as reducing flood discharge in different time periods) while ensuring flood control safety. The system features a dynamic feedback mechanism; after each scheduling implementation, real-time ecological response data is collected via IoT devices (such as fish sonar monitoring and bird drone observation), automatically correcting the strategy parameters in the resource database. All scheduling strategies are stored on the blockchain to ensure traceability of the decision-making process, providing a closed-loop solution for smart watershed management.

[0089] This invention comprehensively considers the final flow velocity and final flow direction in the target watershed as the first influencing factor, the hydraulic engineering facilities of the tributaries where the target watershed is located as the second influencing factor, and sets the activity range and migration route of the target species as seed cells and tissue cells. A cellular automata model is used to control the evolution of the seed cells and tissue cells to obtain the evolutionary results of the target watershed. Subsequently, based on the evolutionary results, the ecological status of the target species in the target watershed is determined, and a corresponding scheduling strategy is selected from the resource pool to complete the intelligent scheduling of the target watershed. This addresses the technical problem of existing technologies being unable to achieve comprehensive scheduling of the entire watershed ecosystem. This solution utilizes a cellular automata model to comprehensively extrapolate multi-dimensional data of the entire watershed ecosystem, achieving ecological restoration and intelligent scheduling of the entire watershed.

[0090] Example 2

[0091] Please refer to Figure 2 The diagram below illustrates the structure of a watershed ecological flow scheduling system based on hydrophilic organism protection, as provided in another embodiment of the present invention. The system includes: a water flow data module, an engineering data module, a species data module, an influencing factor module, a model evolution module, and an intelligent scheduling module. The specific details of each module are as follows:

[0092] The water flow data module is used to collect the initial water flow velocity and initial flow direction in the target watershed, use a drone to acquire continuous remote sensing images of the target watershed within a preset time period, and optimize the initial water flow velocity and initial flow direction using the remote sensing images to obtain the final water flow velocity and final flow direction of the target watershed.

[0093] In this embodiment, the water flow data module is used to optimize the initial water flow velocity and initial flow direction using the remote sensing image. Specifically, this includes: decomposing the remote sensing image frame by frame; highlighting adjacent frames; determining the highlighted feature points in the adjacent frames; setting the highlighted feature points in the earlier remote sensing image as reference points; setting the highlighted feature points in the later remote sensing image as comparison points; calculating the spatial distance between the comparison points corresponding to each reference point; determining the original position of the reference point in the remote sensing image; modifying the original position based on the spatial distance value to obtain the behavioral trajectory of the reference point; and correcting the initial water flow velocity and initial flow direction based on the spatial distance value and the behavioral trajectory to obtain the final water flow velocity and final flow direction.

[0094] The engineering data module is used to obtain the maximum operating efficiency and location information of the water conservancy facilities in the tributary where the target watershed is located.

[0095] The species data module is used to identify the target species to be studied, locate the target species in the target watershed using the remote sensing images, and determine the activity range and migration route of the target species.

[0096] In this embodiment, the species data module is used to locate the target species in the target watershed using the remote sensing image. Specifically, this includes: acquiring the biological characteristic information of the target species; simultaneously, constructing a target recognition model; inputting the biological characteristic information into the target recognition model for training to generate a target species capture model; inputting the remote sensing image into the target species capture model for processing; outputting the marker information of the target species in the remote sensing image; locating the marker information in the remote sensing image; identifying and deleting marker information with a total number of markers appearing within a preset spatial range that is less than a preset threshold, and locating the remaining marker information in the filtered remote sensing image to determine the activity range and migration route of the target species.

[0097] The influencing factor module is used to perform gridding processing on the target watershed and establish a spatial coordinate system. At the same time, it calls the cellular automata model, takes the final water flow velocity and final flow direction of the target watershed as the first influencing factor, takes the maximum operating efficiency of the water conservancy facilities as the second influencing factor, and determines the grid cell occupied by the target species in the spatial coordinate system according to the activity range and migration route of the target species.

[0098] In a first aspect of this embodiment, the step of using the final water flow velocity and final flow direction of the target watershed as first influencing factors by the influencing factor module specifically includes: setting the final water flow velocity of the target watershed as a first water velocity influencing factor, and setting the final flow direction of the target watershed as a first flow direction influencing factor; setting a time unit, dividing the first water velocity influencing factor according to the time unit to obtain a discrete set of water velocity, dividing the first flow direction influencing factor according to the time unit to obtain a discrete set of flow direction; mapping the discrete set of water velocity and the discrete set of flow direction in the same time unit to generate a discrete set that simultaneously has water velocity data and flow direction data in the same time unit, which is used as the first influencing factor.

[0099] In another embodiment, the second influencing factor further includes the location information of the water conservancy engineering facilities. Specifically, the step of using the maximum operating efficiency of the water conservancy engineering facilities as the second influencing factor in the influencing factor module involves: calculating the original influence distance between the target species and different water conservancy engineering facilities based on the activity range of the target species and the location information of the water conservancy engineering facilities; calculating the migration distance value of the target species in each time unit based on the migration route of the target species and the location information of the water conservancy engineering facilities; using the migration distance value as a weight, multiplying the original influence distance in each time unit by the corresponding weight and the maximum operating efficiency of the water conservancy engineering facility, to obtain the migration influence value of the target species under the influence of the water conservancy engineering facilities in different time units, which is then used as the second influencing factor.

[0100] The model evolution module is used to set the grid cells occupied by the target species in the activity range as seed cells and the grid cells occupied by the target species in the migration route as tissue cells. At the same time, the first influencing factor and the second influencing factor are input into the cellular automaton model, and the seed cells and tissue cells are controlled to evolve to obtain the evolution result of the target watershed.

[0101] In this embodiment, the model evolution module is used to control the evolution of the seed cell and the tissue cell, specifically including: controlling the growth of the seed cell and the tissue cell respectively, and simultaneously correcting the growth probability of the seed cell and the tissue cell in real time according to the first influence factor and the second influence factor; monitoring the seed cell and the tissue cell within a preset evolution time, and controlling the seed cell and the tissue cell to stop growing and ending the evolution when it is determined that the growth area of ​​the seed cell reaches a preset area, or the tissue cell stops growing, or the evolution time reaches the preset evolution time.

[0102] The intelligent scheduling module is used to determine the ecological status of the target species in the target watershed based on the evolution results, select the corresponding scheduling strategy from the resource pool based on the ecological status, and complete the intelligent scheduling of the target watershed.

[0103] In another embodiment, it further includes: a computing power influence module; the computing power influence module is used to calculate the current amount of all acquired data before the control of the seed cell and the tissue cell evolves, to obtain the maximum processing value; at the same time, to obtain the maximum processing value of each computing node in the current server; to determine the load of each computing node in different time units, and to correct the maximum processing value of each computing node according to the load, to match the corrected maximum processing value with the maximum processing value, to determine the computing node with operational risk in a certain time unit; and to input the computing node with operational risk in a certain time unit as a third influence factor into the cellular automaton model.

[0104] This invention comprehensively considers the final flow velocity and final flow direction in the target watershed as the first influencing factor, the hydraulic engineering facilities of the tributaries where the target watershed is located as the second influencing factor, and sets the activity range and migration route of the target species as seed cells and tissue cells. A cellular automata model is used to control the evolution of the seed cells and tissue cells to obtain the evolutionary results of the target watershed. Subsequently, based on the evolutionary results, the ecological status of the target species in the target watershed is determined, and a corresponding scheduling strategy is selected from the resource pool to complete the intelligent scheduling of the target watershed. This addresses the technical problem of existing technologies being unable to achieve comprehensive scheduling of the entire watershed ecosystem. This solution utilizes a cellular automata model to comprehensively extrapolate multi-dimensional data of the entire watershed ecosystem, achieving ecological restoration and intelligent scheduling of the entire watershed.

[0105] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A watershed ecological flow scheduling method based on hydrophilic organism protection, characterized in that, include: The initial water flow velocity and initial flow direction in the target watershed are collected. The target watershed is then acquired using a drone within a preset time period using remote sensing images. The initial water flow velocity and initial flow direction are then optimized using the remote sensing images to obtain the final water flow velocity and final flow direction of the target watershed. Obtain the maximum operating efficiency and location information of water conservancy facilities in the tributaries of the target watershed; Identify the target species to be studied, locate the target species in the target watershed using the remote sensing images, and determine the activity range and migration route of the target species; The target watershed is gridded and a spatial coordinate system is established. At the same time, a cellular automata model is invoked, and the final water flow velocity and final flow direction of the target watershed are used as the first influencing factor, and the maximum operating efficiency of the water conservancy facilities is used as the second influencing factor. The grid cells occupied by the target species are determined in the spatial coordinate system according to the activity range and migration route of the target species. The grid cells occupied by the target species in the activity range are set as seed cells, and the grid cells occupied by the target species in the migration route are set as tissue cells. At the same time, the first influencing factor and the second influencing factor are input into the cellular automata model, and the seed cells and the tissue cells are controlled to evolve to obtain the evolution results of the target watershed. Based on the evolution results, the ecological status of the target species in the target watershed is determined, and a corresponding scheduling strategy is selected from the resource pool based on the ecological status to complete the intelligent scheduling of the target watershed. The step of optimizing the initial water flow velocity and initial flow direction using the remote sensing image specifically includes: breaking down the remote sensing image frame by frame, highlighting adjacent frames, identifying the highlighted feature points in the adjacent frames, setting the highlighted feature points in the earlier remote sensing image as reference points, and setting the highlighted feature points in the later remote sensing image as comparison points; calculating the spatial distance between the comparison points corresponding to each reference point, determining the original position of the reference point in the remote sensing image, and modifying the original position according to the spatial distance value to obtain the behavioral trajectory of the reference point; and correcting the initial water flow velocity and initial flow direction according to the spatial distance value and the behavioral trajectory to obtain the final water flow velocity and final flow direction.

2. The watershed ecological flow scheduling method based on hydrophilic organism protection as described in claim 1, characterized in that, The step of locating the target species in the target watershed using the remote sensing image specifically includes: The biological characteristics information of the target species are obtained, and a target recognition model is constructed. The biological characteristics information is input into the target recognition model for training to generate a target species capture model. The remote sensing image is input into the target species capture model for processing, and the tagging information of the target species in the remote sensing image is output. The marker information in the remote sensing image is located, and those marker information appearing within a preset spatial range that is less than a preset threshold are identified as noise and deleted. The remaining marker information in the filtered remote sensing image is located to determine the activity range and migration route of the target species.

3. The watershed ecological flow scheduling method based on hydrophilic organism protection as described in claim 1, characterized in that, The step of using the final flow velocity and final flow direction of the target watershed as the first influencing factor specifically includes: The final water flow velocity of the target watershed is set as the first water velocity influence factor, and the final flow direction of the target watershed is set as the first flow direction influence factor. Set time units, segment the first water velocity influence factor according to the time units to obtain a discrete set of water velocity, and segment the first flow direction influence factor according to the time units to obtain a discrete set of flow direction; The discrete sets of water velocity and flow direction at the same time unit are mapped to generate a discrete set that simultaneously has water velocity data and flow direction data at the same time unit, which is used as the first influencing factor.

4. The watershed ecological flow scheduling method based on hydrophilic organism protection as described in claim 1, characterized in that, The second influencing factor also includes the location information of the water conservancy engineering facilities; specifically, the step of using the maximum operating efficiency of the water conservancy engineering facilities as the second influencing factor is as follows: Based on the activity range of the target species and the location information of the water conservancy facilities, the original influence distance between the target species and different water conservancy facilities is calculated; Based on the migration route of the target species and the location information of the water conservancy facilities, the migration distance of the target species in each time unit is calculated. Using the migration distance value as a weight, the original influence distance at each time unit is multiplied by the corresponding weight and the maximum operating efficiency of the water conservancy facility to obtain the migration influence value of the target species under the influence of the water conservancy facility at different time units, which is used as the second influence factor.

5. The watershed ecological flow scheduling method based on hydrophilic organism protection as described in claim 1, characterized in that, The steps for controlling the evolution of the seed cells and the tissue cells specifically include: The seed cells and tissue cells are controlled to grow separately, and the growth probabilities of the seed cells and tissue cells are corrected in real time according to the first influence factor and the second influence factor. The seed cells and tissue cells are monitored within a preset evolution time. When it is determined that the growth area of ​​the seed cells reaches a preset area, or the tissue cells stop growing, or the evolution time reaches the preset evolution time, the seed cells and tissue cells are controlled to stop growing, and the evolution ends.

6. The watershed ecological flow scheduling method based on hydrophilic organism protection as described in claim 1, characterized in that, Prior to controlling the evolution of the seed cells and the tissue cells, the method further includes: Calculate the maximum processing value based on the current amount of all acquired data; simultaneously, obtain the maximum processing value for each processing node in the current server. Determine the load capacity of each computing node in different time units, and correct the maximum computing value of each computing node according to the load capacity. Match the corrected maximum computing value with the maximum processing value to determine the computing node with operational risk in a certain time unit. The computational node that exhibits operational risk at a certain time unit is used as a third influencing factor and input into the cellular automata model.

7. A watershed ecological flow scheduling system based on the protection of hydrophilic organisms, characterized in that, include: The system includes modules for water flow data, engineering data, species data, influencing factors, model evolution, and intelligent scheduling. The water flow data module is used to collect the initial water flow velocity and initial flow direction in the target watershed, use a drone to acquire continuous remote sensing images of the target watershed within a preset time period, and optimize the initial water flow velocity and initial flow direction using the remote sensing images to obtain the final water flow velocity and final flow direction of the target watershed. The engineering data module is used to obtain the maximum operating efficiency and location information of the water conservancy engineering facilities in the tributary of the target watershed; The species data module is used to identify the target species to be studied, locate the target species in the target watershed through the remote sensing images, and determine the activity range and migration route of the target species. The influencing factor module is used to perform gridding processing on the target watershed and establish a spatial coordinate system. At the same time, it calls the cellular automata model, takes the final water flow velocity and final flow direction of the target watershed as the first influencing factor, takes the maximum operating efficiency of the water conservancy facilities as the second influencing factor, and determines the grid cell occupied by the target species in the spatial coordinate system according to the activity range and migration route of the target species. The model evolution module is used to set the grid cells occupied by the target species in the activity range as seed cells and the grid cells occupied by the target species in the migration route as tissue cells. At the same time, the first influencing factor and the second influencing factor are input into the cellular automaton model, and the seed cells and tissue cells are controlled to evolve to obtain the evolution results of the target watershed. The intelligent scheduling module is used to determine the ecological status of the target species in the target watershed based on the evolution results, and select the corresponding scheduling strategy from the resource pool based on the ecological status to complete the intelligent scheduling of the target watershed. The step of optimizing the initial water flow velocity and initial flow direction using the remote sensing image specifically includes: breaking down the remote sensing image frame by frame, highlighting adjacent frames, identifying the highlighted feature points in the adjacent frames, setting the highlighted feature points in the earlier remote sensing image as reference points, and setting the highlighted feature points in the later remote sensing image as comparison points; calculating the spatial distance between the comparison points corresponding to each reference point, determining the original position of the reference point in the remote sensing image, and modifying the original position according to the spatial distance value to obtain the behavioral trajectory of the reference point; and correcting the initial water flow velocity and initial flow direction according to the spatial distance value and the behavioral trajectory to obtain the final water flow velocity and final flow direction.

8. The watershed ecological flow scheduling system based on hydrophilic organism protection as described in claim 7, characterized in that, The model evolution module is used to control the evolution steps of the seed cells and the tissue cells, specifically including: The seed cells and tissue cells are controlled to grow separately, and the growth probabilities of the seed cells and tissue cells are corrected in real time according to the first influence factor and the second influence factor. The seed cells and tissue cells are monitored within a preset evolution time. When it is determined that the growth area of ​​the seed cells reaches a preset area, or the tissue cells stop growing, or the evolution time reaches the preset evolution time, the seed cells and tissue cells are controlled to stop growing, and the evolution ends.

9. The watershed ecological flow scheduling system based on hydrophilic organism protection as described in claim 7, characterized in that, Also includes: The module that affects computing power; The computing power impact module is used to calculate the current amount of all acquired data before the control of the seed cell and the tissue cell evolves, to obtain the maximum processing value; at the same time, it obtains the maximum computing value of each computing node in the current server; determines the load of each computing node in different time units, and corrects the maximum computing value of each computing node according to the load, and matches the corrected maximum computing value with the maximum processing value to determine the computing node with operational risk in a certain time unit; The computational node that exhibits operational risk at a certain time unit is used as a third influencing factor and input into the cellular automata model.

Images