A reservoir ecological regulation method and system considering dynamic response of water quality

Abstract

This invention relates to the field of water environment management technology, and more particularly to a reservoir ecological scheduling method and system considering dynamic water quality response. The method includes the following steps: acquiring reservoir water quality monitoring data; performing eutrophication analysis based on the reservoir water quality monitoring data to obtain eutrophication data; tracing pollution sources based on the eutrophication data to obtain pollution source data; assessing pollutant emission intensity based on the pollution source data; identifying high-pollution areas based on pollutant emission intensity; detecting algae density in high-pollution areas to obtain algae density data; analyzing algal toxins in the water based on the algae density data to obtain algal toxin data; acquiring reservoir hydrological data and extracting maximum rainfall characteristics to obtain maximum rainfall data. This invention improves the response rate and pollution risk identification rate of dynamic reservoir water quality scheduling based on water environment management technology.

Description

Technical Field

[0001] This invention relates to the field of water environment management technology, and in particular to a reservoir ecological scheduling method and system that takes into account the dynamic response of water quality. Background Technology

[0002] Traditional reservoir ecological regulation lacks the ability to identify fine-grained pollution sources, making it impossible to accurately identify point and non-point source pollution types based on spatiotemporal dynamic changes. It ignores the chain-like ecological response process caused by eutrophication and fails to fully establish the dynamic correlation between changes in nutrient concentration, algal growth, toxin release, and dissolved oxygen. It lacks collaborative modeling of hydrodynamic parameters and hydrological processes, making it difficult to accurately simulate the diffusion paths and dilution processes of pollutants under different cross-sections and different substrate roughness conditions. The ecological regulation mechanism is rigid, generally using fixed thresholds or preset time triggering methods, which cannot achieve intelligent response based on real-time water quality evolution trends. The internal modules of the system are loosely connected, and the monitoring, analysis, simulation, and control links do not form an efficient linkage, which cannot support rapid identification and coordinated scheduling control when reservoir water quality is abnormal, resulting in low overall water environment regulation efficiency, delayed response, and insufficient ecological restoration capacity. Summary of the Invention

[0003] Therefore, it is necessary for the present invention to provide a reservoir ecological scheduling method and system that takes into account the dynamic response of water quality, so as to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a reservoir ecological management method considering dynamic water quality response includes the following steps:

[0005] Step S1: Obtain reservoir water quality monitoring data; perform eutrophication analysis based on reservoir water quality monitoring data to obtain eutrophication data; trace pollution sources based on eutrophication data to obtain pollution source data;

[0006] Step S2: Assess pollutant emission intensity based on pollution source data; identify high-pollution areas based on pollutant emission intensity; detect algae density in high-pollution areas to obtain algae density data; analyze algal toxins in the water body based on algae density data to obtain algal toxin data.

[0007] Step S3: Obtain reservoir hydrological data and extract maximum rainfall characteristics to obtain maximum rainfall data; calculate flow rate based on maximum rainfall data to obtain flow rate data; simulate algal toxin diffusion based on flow rate data and algal toxin data to obtain algal toxin diffusion data.

[0008] Step S4: Identify high-concentration toxin retention areas based on algal toxin diffusion data; perform dissolved oxygen anomaly detection based on high-concentration toxin retention areas to obtain dissolved oxygen anomaly data; and conduct reservoir ecological water release scheduling based on dissolved oxygen anomaly data to obtain reservoir ecological water release data.

[0009] This invention, by acquiring reservoir water quality monitoring data and conducting eutrophication analysis, enables timely monitoring of nutrient levels exceeding standards in water bodies, providing a data foundation for subsequent pollution source tracing. The introduction of pollution source tracking enhances the ability to identify pollution input pathways, helps distinguish between point and non-point source pollution, and improves the accuracy of pollution type determination. Based on pollution source data, emission intensity assessment and high-pollution area identification further refine the granularity of pollution spatial distribution, enabling quantitative assessment of local water body risks. Combining algae density detection and toxin analysis in high-pollution areas allows for the comprehensive construction of the correlation chain between nutrient concentration changes, abnormal algal proliferation, and toxin release, enhancing the dynamic characterization of aquatic ecological evolution. Hydrological data acquisition and maximum rainfall feature extraction strengthen the dynamic perception of external hydraulic inputs, making the quantitative conversion between rainfall, runoff, and inflow more responsive to on-site conditions, ensuring accurate and reasonable initial boundary conditions for toxin diffusion simulation. Toxin diffusion simulation couples hydrodynamic parameters such as water flow velocity, water depth, and substrate roughness with rainfall inflow processes to construct a fine-scale, multi-dimensional diffusion path analysis mechanism, thereby improving the prediction accuracy of toxin retention locations and dilution capabilities. Based on abnormal dissolved oxygen changes identified in high-concentration toxin retention areas, real-time perception of the impact of abnormal algal toxin release on the ecosystem can be achieved, enhancing the intelligent early warning capability of ecological risks. Finally, through an ecological scheduling mechanism triggered by dissolved oxygen anomalies, the traditional static threshold-guided water release control method is broken, realizing a water quality-driven dynamic scheduling decision-making model. This improves the efficiency of reservoir water quality linkage management and the timeliness of ecological scheduling response, and overall enhances the comprehensive regulation capability of the water environment under complex pollution evolution processes.

[0010] Preferably, this specification also provides a reservoir ecological scheduling system that considers dynamic water quality response, used to execute the reservoir ecological scheduling method considering dynamic water quality response as described above. The reservoir ecological scheduling system considering dynamic water quality response includes:

[0011] The pollution source tracing module is used to acquire reservoir water quality monitoring data; perform eutrophication analysis based on the reservoir water quality monitoring data to obtain eutrophication data; and trace pollution sources based on the eutrophication data to obtain pollution source data.

[0012] The water body algal toxin analysis module is used to assess pollutant emission intensity based on pollution source data; identify high-pollution areas based on pollutant emission intensity; detect algal density in high-pollution areas to obtain algal density data; and analyze water body algal toxins based on algal density data to obtain water body algal toxin data.

[0013] The toxin diffusion simulation module is used to acquire reservoir hydrological data, extract maximum rainfall characteristics to obtain maximum rainfall data, perform flow conversion based on maximum rainfall data to obtain flow data, and perform toxin diffusion simulation based on flow data and algal toxin data to obtain algal toxin diffusion data.

[0014] The reservoir ecological water release scheduling module is used to identify areas with high concentrations of toxins based on algal toxin diffusion data; to detect dissolved oxygen anomalies based on these areas, thus obtaining dissolved oxygen anomaly data; and to schedule reservoir ecological water release based on this data, thus obtaining reservoir ecological water release data.

[0015] The present invention provides a reservoir ecological scheduling system that considers dynamic water quality response. This system can implement any of the reservoir ecological scheduling methods of the present invention that consider dynamic water quality response. It is used as a medium for coordinating the operation and signal transmission between various modules to complete the reservoir ecological scheduling method that considers dynamic water quality response. The modules within the system cooperate with each other to improve the response rate and pollution risk identification rate of the reservoir's dynamic water quality scheduling. Attached Figure Description

[0016] 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:

[0017] Figure 1 This is a schematic diagram of the steps of a reservoir ecological scheduling method that considers dynamic water quality response according to the present invention.

[0018] Figure 2 This is a detailed flowchart of step S1 in the present invention;

[0019] Figure 3 This is a detailed flowchart of step S16 in the present invention;

[0020] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] The technical method of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0022] 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.

[0023] 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.

[0024] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a reservoir ecological scheduling method considering dynamic water quality response, the method comprising the following steps:

[0025] Step S1: Obtain reservoir water quality monitoring data; perform eutrophication analysis based on reservoir water quality monitoring data to obtain eutrophication data; trace pollution sources based on eutrophication data to obtain pollution source data;

[0026] In this embodiment, a fixed online multi-parameter water quality monitoring buoy platform is used to set up at least nine monitoring points in the upstream, midstream, and downstream sections of the reservoir. Monitoring parameters include total nitrogen (TN), total phosphorus (TP), ammonia nitrogen (NH3-N), dissolved oxygen (DO), pH, conductivity (EC), and water temperature (T). The sampling frequency is set to once per hour, with a duration of no less than 30 days. Ammonia nitrogen is detected using ultraviolet-visible spectroscopy, total phosphorus concentration is detected using ammonium molybdate spectrophotometry, and total nitrogen content is detected using ultraviolet spectrophotometry. The monitoring data is input into a eutrophication assessment process based on the national Class V surface water quality limit. Nutrient thresholds are set as follows: TP > 0.2 mg / L and TN > 1.0 mg / L, indicating eutrophication. Within the eutrophication area, the location of potential pollution sources is determined by comparing the spatiotemporal concentration gradient of pollutants and combining this with spatial location data of marked sewage outlets, agricultural irrigation outlets, and livestock breeding areas in a Geographic Information System (GIS). On-site sampling and verification of pollution points were conducted to analyze the matching of nitrogen and phosphorus speciation characteristics with source data, thereby confirming the location of pollution sources and their emission attributes.

[0027] Step S2: Assess pollutant emission intensity based on pollution source data; identify high-pollution areas based on pollutant emission intensity; detect algae density in high-pollution areas to obtain algae density data; analyze algal toxins in the water body based on algae density data to obtain algal toxin data.

[0028] In this embodiment, identified pollution sources are classified into point sources (such as direct discharge outlets of domestic sewage and industrial discharge outlets) and area sources (such as agricultural area sources and exposed slopes). Emission intensity calculation methods are set for each: point sources are calculated based on the actual pollutant flow velocity (m³ / s). 3 The emission flux was calculated by multiplying the annual average fertilization intensity per unit area (kg / ha / yr) by the pollutant concentration (mg / L). For non-point source pollution, the annual emission volume was estimated by multiplying the annual average fertilization intensity per unit area (kg / ha / yr) by the runoff coefficient (set based on topographic slope, soil type, and rainfall data). Areas with an annual emission flux greater than 500 kgTP or 5000 kgTN were designated as high-pollution areas. A 3×3 grid sampling point was established within this area, and 1 L of water sample was collected. After 12 hours of natural settling, the upper water layer was extracted, and algae samples were concentrated using a filter membrane and analyzed using an inverted microscope. Cell counting was performed, and algal species were identified according to an algal morphology identification manual. A concentration of 1000 cells / mL was used as the criterion for identifying algal-rich areas. For areas with algal concentrations exceeding 3000 cells / mL, toxin data were further collected. Microcystin-releasing enzyme (MC-LR) concentrations were determined using enzyme-linked immunosorbent assay (ELISA). If the toxin concentration exceeded 1 μg / L, a combined temperature-pH analysis was performed on the toxin formation process (temperature at 0.5°C intervals and pH at 0.2-unit gradients). Acid-base regulation experiments were used to establish toxin release and degradation kinetic curves. After the degradation curves stabilized, the adsorption coefficient was calculated based on the initial and residual concentrations. This coefficient was then used to simulate toxin migration under different hydrodynamic conditions, ultimately generating aquatic algal toxin data.

[0029] Step S3: Obtain reservoir hydrological data and extract maximum rainfall characteristics to obtain maximum rainfall data; calculate flow rate based on maximum rainfall data to obtain flow rate data; simulate algal toxin diffusion based on flow rate data and algal toxin data to obtain algal toxin diffusion data.

[0030] In this embodiment, radar rainfall echo fusion with ground rain gauge data is used to acquire minute-by-minute rainfall data for the reservoir basin over 24 hours. The maximum cumulative values ​​for 1 hour, 3 hours, and 24 hours are selected, and the maximum value is taken as the maximum rainfall, expressed in mm / h. A threshold of >50 mm / h is set as heavy rainfall. This is combined with the reservoir's catchment area A (km²). 2 The surface runoff coefficient C (dimensionless, set according to soil type and vegetation cover, e.g., 0.25-0.45 for sandy loam) is used, and the runoff calculation formula Q = C·A·P (where Q is the runoff volume in m³) is applied. 3 P (rainfall in m³) is converted to rainfall runoff. This is then combined with the reservoir drainage area S (km²). 2 ), converting the inflow rate per unit time into flow rate (m³). 3 / s). This flow rate was coupled with an initial distribution grid of algal toxins in the water body, and a two-dimensional shallow water convection-diffusion equation was used to simulate toxin diffusion. The boundary conditions were set to fixed boundaries (no toxin input), and the initial conditions were imported based on the toxin distribution matrix. The simulation time was 48 hours, the time step was 300 seconds, and the spatial resolution was no higher than 50m × 50m. The Crank–Nicolson numerical scheme was used to solve for the temporal and spatial evolution of toxin concentration, and the toxin concentration distribution layer was output, which is the algal toxin diffusion data.

[0031] Step S4: Identify high-concentration toxin retention areas based on algal toxin diffusion data; perform dissolved oxygen anomaly detection based on high-concentration toxin retention areas to obtain dissolved oxygen anomaly data; and conduct reservoir ecological water release scheduling based on dissolved oxygen anomaly data to obtain reservoir ecological water release data.

[0032] In this embodiment, continuous water areas with toxin concentrations greater than 1.5 μg / L in the algal toxin diffusion simulation output data are designated as high-concentration toxin retention areas. Underwater DO probe monitoring points are set up based on the center point of these areas, and optical dissolved oxygen probes (measurement range 0-20 mg / L, accuracy 0.1 mg / L) are used for continuous recording for 72 hours, at a frequency of once every 5 minutes. Water temperature (°C) and water flow velocity (m / s) are measured simultaneously, and the rate of change of DO concentration per unit time is calculated using linear difference. If the DO concentration decreases by more than 2 mg / L or falls below 4 mg / L within 6 hours, it is identified as an abnormal dissolved oxygen area. For the identified abnormal dissolved oxygen areas, a drainage scheduling model is constructed based on the topographic elevation map and reservoir drainage structure parameters (such as bottom hole diameter, bottom hole elevation, and gate opening range). The scheduling strategy, based on the target area volume, current toxin and DO concentrations, and the total reservoir storage capacity, prioritizes opening nearby discharge structures and controls the discharge flow rate (0.2-1.5 m). 3 Strategies such as ( / s) were employed to maintain drainage for no less than 4 hours and no more than 24 hours. During this period, the dissolved oxygen (DO) concentration in the water was retested every hour, and the drainage volume was adjusted in a timely manner according to the increase in concentration. Finally, the data on the release time, flow velocity, gate number, and regional water quality changes during the scheduling period were recorded and compiled into reservoir ecological water release data.

[0033] Preferably, step S1 includes the following steps:

[0034] Step S11: Obtain reservoir water quality monitoring data and calculate nutrient concentration to obtain nutrient concentration data;

[0035] In this embodiment, no fewer than 10 online automatic water quality monitoring stations are deployed in the upstream, midstream, and downstream sections of the reservoir. Each station uses a multi-channel flow injection analyzer to monitor the total nitrogen (TN), total phosphorus (TP), and nitrate nitrogen (NO3) in the water sample. - -N), nitrite nitrogen (NO2) - -N), ammonium nitrogen (NH4) + Simultaneous monitoring of total nitrogen (TN) and soluble reactive phosphorus (SRP) was conducted. Total nitrogen was determined using alkaline potassium persulfate digestion ultraviolet spectrophotometry, with absorbance read at 220 nm. Total phosphorus was detected using ammonium molybdate ascorbic acid reduction, with absorbance measured at 700 nm. Both nitrogen and phosphorus concentrations are expressed in mg / L. Sampling frequency was set to once every 6 hours, with a continuous collection period of 15 days. The average of daily data collected from each station was used as the representative nutrient concentration for that station. The final data table uses the reservoir area grid number as the primary key, listing each parameter independently. Outliers (e.g., total nitrogen > 15 mg / L, TP > 2.5 mg / L) were manually re-examined to remove them, ensuring data reliability.

[0036] Step S12: Perform algal growth simulation based on nutrient concentration data to obtain algal growth data; calculate algal biomass based on algal growth data to obtain algal biomass data;

[0037] In this embodiment, the concentrations of total nitrogen, total phosphorus, nitrate nitrogen, and soluble phosphorus were used as input factors to construct a table of algal proliferation rate per unit nutrient salt. The significant algal growth response was defined as occurring within the total phosphorus range of 0.02-0.2 mg / L; the calculation was based on the assumption that 0.1 mg / L LTP in water could increase the daily algal growth rate by approximately 5%. Nutrient data were compared with the locally measured average daily light duration (measured using a photosynthetically active radiation (PAR) sensor, in mol / m²). 2 The light intensity was combined with the d) and water temperature (measured by a CTD sensor, in °C), and the light intensity was set to >1200 μmol / m². 2 The optimal growth range for algae is 15-28℃. Based on this, an algebraic iteration method was used to calculate cell proliferation daily over a 10-day period, generating the daily algal cell growth count (cells / mL) for each region. Algal biomass was calculated using the cell volume accumulation method per unit volume, with cell volumes set for dominant algal species (e.g., Microcystis, Scenedesmus, Oscillatoria) (e.g., 110 μm for Microcystis). 3 The biomass (μg / L) is calculated by multiplying the cell count by the daily cell count. The output data is presented in three columns: grid number, algal species, and biomass.

[0038] Step S13: Identify waters with high algal biomass based on algal biomass data;

[0039] In this embodiment, the daily maximum value is selected from the algal biomass data, and a high algal biomass judgment threshold is set. Areas with a biomass exceeding 1000 μg / L are designated as high algal biomass zones. Areas meeting the criteria are categorized by number and spatial clustering analysis is performed, with adjacent areas having a center-to-center distance of less than 100 meters considered as the same high-algae block. A GIS platform is used to render the layers, outputting a vector map of the high-algae water area (shapefile format), and its boundary coordinates are marked on an elevation map, providing a precise coordinate basis for subsequent water body sampling and positioning. For each block, independent fields such as identifier, area, center coordinates, maximum and minimum biomass values ​​are generated for later retrieval and overlay analysis.

[0040] Step S14: Conduct water turbidity testing on water bodies with high algal biomass to obtain water turbidity data;

[0041] In this embodiment, artificial sampling locations were set up at the center and edge of the high-algae water area. 1L of water sample was collected from each sampling point, and the samples were measured on-site using a portable turbidity meter (measurement range 0-1000 NTU, resolution 0.1 NTU). The measurement environment required no direct sunlight. Before measurement, the water sample was thoroughly shaken 5 times and allowed to stand for 15 seconds. Three consecutive measurements were taken, and the average value was used as the final turbidity value (unit: NTU). In addition, a laser particle size analyzer was used to detect the diameter distribution of suspended particles to analyze the turbidity composition. More than 30 sampling points were set up in the area to form a spatial isopleth map of turbidity in the water body. Areas with turbidity greater than 30 NTU were recorded as abnormal turbidity areas. The high-algae water area numbers corresponding to each sampling point were compiled, and a turbidity data table was output, including fields such as sampling number, location coordinates, corresponding algal biomass, turbidity value, and particle size range.

[0042] Of particular importance, step S14 includes the following steps:

[0043] Step S141: Identify the boundaries of high-algae waters based on high-algae biomass waters and obtain high-algae water boundary data;

[0044] In this embodiment, high-frequency water color remote sensing imagery (such as Sentinel-2MSI imagery, 10-meter resolution, acquired within the last 7 days on sunny days) is used. Based on pre-acquired high algae biomass distribution data, which is derived from historical water quality monitoring statistics and near-infrared vegetation index (NDVI) processing results, areas with an NDVI greater than 0.2 can be preliminarily identified as water bodies with high algae coverage. Spatial vectorization tools (such as Python vectorization scripts based on GDAL) are used to convert the binary masks of high NDVI areas into vector boundaries, resulting in high algae-covered water body boundary data stored in shapefile format (shp). The boundary coordinate system uniformly adopts WGS84 projection to ensure spatial consistency of data in subsequent operations. To ensure boundary closure and calculation accuracy, Douglas-Peucker compression with a boundary simplification threshold set to within 1 meter is performed during processing, and topological checks are conducted to eliminate geometric errors such as self-intersection and collinearity, outputting standardized high algae-covered water body boundary data.

[0045] Step S142: Perform spatial cropping of the remote sensing image based on the boundary data of the high-algae water area to obtain the remote sensing image of the high-algae water area;

[0046] In this embodiment, a portion of the remote sensing image of algae-rich water areas is extracted through spatial cropping. Using a remote sensing image cropping module, the input is set to the original Sentinel-2 remote sensing image (containing bands B2, B3, B4, and B8) in a geographic information processing platform (such as QGIS or ArcGIS), the output format is GeoTIFF, and the output resolution is set to 10 meters to ensure consistency with the original image. The cropping operation uses a vector boundary file for masking, with the boundary buffer distance set to 0 meters to ensure the output image completely covers the algae-rich water area and excludes land areas. If cloud interference is present, pixel removal is performed using cloud mask data (QA60 band). Pixels with a cloud coverage probability greater than 20% are removed. The cropped image is stored in an uncompressed format for subsequent reflectance calculation. The final output is a remote sensing image of the algae-rich water area, which serves as the basis for reflectance processing.

[0047] Step S143: Extract effective band reflectance values ​​based on remote sensing images of waters with high algae levels;

[0048] In this embodiment, the effective band reflectance values ​​are calculated using cropped remote sensing images of high-algae water areas. The required bands include visible blue light (B2, center wavelength 490nm), green light (B3, center wavelength 560nm), red light (B4, center wavelength 665nm), and near-infrared light (B8, center wavelength 842nm). All band reflectance calculations are based on Top-of-Atmosphere (TOA) correction. First, the calibration coefficients of the Sentinel-2 Level-1C image are used to convert the digital values ​​(DN) to apparent reflectance. The reflectance R is calculated as R = DN × 0.0001. After reflectance correction, the values ​​are stored separately for each band using Python NumPy arrays, and pixel values ​​with reflectance less than 0 or greater than 1 are removed as invalid values. The effective band reflectance extraction results are output in CSV format, recording the geographic location (latitude and longitude) of each pixel and its B2, B3, B4, and B8 reflectance values ​​for subsequent spectral ratio calculations.

[0049] Step S144: Calculate the water body spectral ratio based on the effective band reflectance value; estimate the suspended particulate matter concentration based on the water body spectral ratio to obtain the suspended particulate matter concentration in the water body;

[0050] In this embodiment, the spectral ratios of the water body are calculated using the red-green ratio (RBR) and red-blue ratio (RBBR), where RBR = B4 / B3 and RBBR = B4 / B2. The calculation is performed pixel-by-pixel using the reflectance data obtained in step S143. If the denominator is less than 0.01, the corresponding ratio result is set as invalid to avoid abnormal amplification. After calculation, the results are saved in raster data format (RBR and RBBR are recorded per pixel). Subsequently, particulate matter concentration is estimated based on existing empirical formulas for remote sensing water body suspended matter concentration, using the formula CSPM = 43.5 × RBR^1.12 (unit: mg / L). This empirical formula is derived from the fitting analysis of measured sample points (r... 2 >0.85, sample size >50). The final output is a raster map of suspended solids concentration in the water body, with a spatial resolution maintained at 10 meters and a value range controlled between 0 and 500 mg / L. Pixels with a concentration greater than 500 mg / L are marked as outliers and removed to prevent errors in subsequent calculations.

[0051] Step S145: Perform standard turbidity conversion based on the concentration of suspended solids in the water to obtain water turbidity data.

[0052] In this embodiment, the internationally accepted suspended solids to turbidity conversion function is adopted. The conversion formula is: NTU = 0.9 × CSPM^0.62. The CSPM value obtained in the previous step is directly used in the calculation, and the conversion is performed pixel by pixel. During the conversion process, the maximum NTU value must be controlled to not exceed 2000 NTU. Pixels exceeding this value are uniformly assigned a value of 2000 NTU to maintain data standardization. All calculation results are output in GeoTIFF format with a spatial resolution of 10 meters, and the projection coordinate system is maintained as WGS84. To facilitate subsequent ecological scheduling analysis, all turbidity data are resampled and aligned with the velocity raster. Bilinear interpolation is used for sampling. The final output is a standardized water turbidity data raster map, recording the average turbidity, maximum turbidity, the location of abnormally high-value pixels, and their latitude and longitude coordinates. The output is in both tabular and raster formats.

[0053] Step S15: Calculate chlorophyll content based on water turbidity data to obtain chlorophyll concentration data; conduct eutrophication assessment based on chlorophyll concentration data and nutrient concentration data to obtain water eutrophication data.

[0054] In this embodiment, based on the obtained turbidity data points, a 2L water sample was used as the standard measurement volume. Filtering was performed using a GF / C glass fiber membrane, with the filtration pressure controlled below 0.3 atm to prevent cell rupture. After extraction in 90% acetone at 4°C in the dark for 24 hours, absorbance values ​​were read at three wavelengths (664nm, 647nm, and 630nm) using a spectrophotometer. The chlorophyll a concentration (μg / L) was calculated using the Lorenzen formula. The concentration measured at each point was coupled with the corresponding nutrient data for analysis. According to the OECD eutrophication assessment system, water bodies were classified as eutrophic based on chlorophyll a > 25 μg / L, TN > 1.5 mg / L, and TP > 0.1 mg / L. A comprehensive eutrophication index was generated using a three-index weighted method (with weights of 0.4, 0.3, and 0.3 for each index). Grids with an index greater than 0.6 were recorded as severely eutrophic areas. Finally, the eutrophication level of the water body is output by grid number, along with the original values ​​of three parameters and the comprehensive index value, and a spatial distribution map of eutrophication is generated.

[0055] Step S16: Based on the eutrophication data of the water body, trace the pollution sources to obtain pollution source data.

[0056] In this embodiment, based on the spatial distribution map of eutrophication, topographic tracking analysis is performed upstream of each eutrophic block using a 1:10000 hydro-topographic map. A flow path inversion tool is used to construct a flow direction map based on DEM data to trace the catchment path of eutrophic areas, marking all possible inflow points along the path. Combined with vector data such as land use maps, river discharge outlet datasets, and farmland irrigation and drainage system layers, the source of eutrophic areas is identified. Water sampling points are deployed in the potential source areas, and comparative analysis is performed using the same nutrient analysis process. The correlation coefficient between the pollution concentration at the source and the concentration in the downstream eutrophic area is calculated; a correlation greater than 0.85 indicates a major pollution source. The final output is a pollution source data table, including pollution source type, geographic coordinates, pollution load (unit: kg / d), corresponding eutrophic area number, and a time-series concentration change trend map. All data is synchronously updated to the reservoir ecological scheduling database to provide data support for subsequent responsive scheduling.

[0057] Preferably, step S16 includes the following steps:

[0058] Step S161: Identify areas with high nutrient loads based on eutrophication data of water bodies;

[0059] In this embodiment, the reservoir area is divided into several grid cells using the spatial distribution of three indicators—chlorophyll a concentration (μg / L), total nitrogen (TN, mg / L), and total phosphorus (TP, mg / L)—found in the eutrophication data. Each grid cell has a side length of 50 meters. For each grid cell, eutrophication thresholds are set based on chlorophyll a exceeding 25 μg / L, TN exceeding 1.5 mg / L, and TP exceeding 0.1 mg / L. Each grid cell is then screened for these three indicators. Grid cells exceeding at least two of the three thresholds are designated as high trophic load areas. This process is achieved through GIS spatial overlay analysis. The "Raster Calculator" function in ArcGIS 10.8 software is used to perform logical operations on the three-parameter raster layers to obtain the vector boundary of the high trophic load area. To avoid interference from isolated points, areas with at least three consecutive grid cells exceeding the threshold and a total area greater than 0.05 square kilometers are considered valid high trophic load areas. Export the boundary polygons as Shapefile format to facilitate subsequent data extraction and on-site verification.

[0060] Step S162: Detect nitrogen and phosphorus content in areas with high nutrient load to obtain nitrogen and phosphorus content data;

[0061] In this embodiment, for areas with high nutrient loads, water sampling points were arranged in a uniform grid of 100 meters × 100 meters within each area. Each point collected 2 liters of water sample in nitrogen- and phosphorus-free polyethylene sampling bottles. A portable water quality analyzer was used on-site for preliminary rapid determination of nitrogen and phosphorus levels in the water samples. Specific quantitative determination was performed using a flow injection analyzer (FIA). Total phosphorus (TP) was determined using the ammonium molybdate spectrophotometric method, with a detection range of 0.005 mg / L to 10 mg / L and a resolution of 0.001 mg / L. Total nitrogen (TN) was detected using an alkaline potassium persulfate digestion method combined with an ultraviolet spectrophotometer, with a detection range of 0.01 mg / L to 20 mg / L and a resolution of 0.01 mg / L. The sampling process ensured that water samples from each sampling point were delivered to the laboratory immediately, transported in a 4°C refrigerated box, with transportation time controlled within 4 hours. After sampling, all measurement results were entered into the database, and a continuous concentration distribution map of nitrogen and phosphorus in the high nutrient load area was generated by spatial interpolation method (inverse distance weighting method IDW, with power parameter set to 2). The data was stored in grid form with a unit size of 50m×50m.

[0062] Step S163: Conduct wastewater outlet pollution detection based on nitrogen and phosphorus content data to obtain wastewater outlet pollution data;

[0063] In this embodiment, the coordinates and basic discharge parameters of all drainage outlets are first extracted based on the registration information of the drainage outlets in the reservoir basin. Water samples are collected within a 50-meter radius of these drainage outlets, with a sample volume of 2 liters, using the same sampling bottles and refrigerated transport specifications. The nitrogen and phosphorus content in the water samples is analyzed using a flow injection analyzer, and the results are used as baseline data for the pollution load of the drainage outlets. Simultaneously, flow rate data from the drainage outlets are collected using an ultrasonic flow meter with a measurement accuracy of ±1%. The measurement frequency is set to once every 15 minutes, with a continuous monitoring period of no less than 7 days. The collected nitrogen and phosphorus concentrations (mg / L) and corresponding flow rates (m³ / s) are then compared. 3 The emission load is calculated using the following formula: Pollution load (kg / d) = Nitrogen and phosphorus concentration (mg / L) × Flow rate (m³ / s). 3 / s)×86400(seconds / day) / 1000(mg / kg). Pollution load data from all drainage outlets are aggregated into the database. The drainage outlet pollution data table includes the outlet number, location coordinates, nitrogen concentration, phosphorus concentration, average flow rate, and corresponding pollution load. Outliers have been removed through time-series statistical analysis to ensure data robustness.

[0064] Step S164: Match pollution sources based on the pollution data of the drainage outlet to obtain pollution source data.

[0065] In this embodiment, pollution source matching analysis is implemented by combining the spatial concentration distribution map of nitrogen and phosphorus in the high trophic load area with the pollution load data of the drainage outlets. Based on the geographical coordinates of the drainage outlets, the influence range of each drainage outlet on the grid cells in the high trophic load area is calculated using digital elevation model (DEM) data according to the hydrological flow direction and topographic slope. Spatial buffer analysis is used, with a buffer radius of 1000 meters, and the correlation analysis is performed between the nitrogen and phosphorus concentrations of the high trophic load grid cells in the buffer and the pollution load of the corresponding drainage outlets. Specifically, the correlation is calculated using the Pearson correlation coefficient, and drainage outlets with a calculation result greater than 0.85 are identified as major pollution sources. For drainage outlets identified as major pollution sources, their emission time-series characteristics are further analyzed. Pollution load time-series curves are plotted using drainage outlet flow and nitrogen and phosphorus concentration change data, and time-lag analysis is performed with the nitrogen and phosphorus concentration change curves in the high trophic load area to verify the rationality of the pollution transmission path. Finally, pollution source data is output, including pollution source type (domestic sewage, industrial discharge, agricultural non-point source pollution, etc.), geographical location, pollution load, influence range, and time-series characteristics, which serve as the data basis for subsequent governance and response measures of reservoir ecological scheduling.

[0066] Preferably, step S2, which assesses pollutant emission intensity based on pollution source data, includes:

[0067] Pollution types are classified based on pollution source data to obtain point source pollution data and area source pollution data.

[0068] In this embodiment, pollution sources are classified according to their physical form and emission method based on the spatial coordinates, emission characteristics, and emission methods contained in the pollution source data. Point source pollution refers to a single, clearly defined discharge outlet, such as factory wastewater discharge outlets, sewage treatment plant effluent outlets, and centralized discharge wells. Non-point source pollution refers to non-centralized emission areas without obvious discharge points, such as agricultural runoff, dispersed emissions from residential areas, and urban stormwater runoff. First, the pollution source data is imported into a Geographic Information System (GIS) software (ArcGIS 10.8). Spatial buffer zones are set according to the pollution source coordinates. Point source pollution uses a buffer zone with a radius of 50 meters. Non-point source pollution areas are obtained by integrating land use data and watershed topography, with continuous agricultural land and residential areas with an area of ​​not less than 0.1 square kilometers as non-point source pollution areas. Combining surface slope information and land cover type data, buffer zone overlay analysis and spatial clustering analysis (DBSCAN algorithm, radius parameter set to 100 meters, minimum number of points to 3) are used to classify pollution sources, automatically distinguishing between point source and non-point source areas. Point source pollution data records include pollution source number, geographic coordinates, pollution type and emission intensity, while non-point source pollution data include the boundary of the pollution area, the area of ​​the polygon and its corresponding land use type.

[0069] Ammonia nitrogen pollution data was obtained by detecting ammonia nitrogen pollution based on point source pollution data;

[0070] In this embodiment, for point source pollution locations, on-site water samples were collected continuously for 7 days, with a sampling frequency of no less than 3 times per day to ensure data coverage of different discharge periods. Polyethylene sampling bottles were used, with a sample volume of 2 liters. A portable multi-parameter water quality analyzer (Hach HQ40d) was used on-site to predict ammonia nitrogen concentration. After the water samples were sent to the laboratory, ammonia nitrogen concentration was determined using Nessler's reagent spectrophotometry. The instrument model was a UV-Vis spectrophotometer (e.g., Shimadzu UV-2600), with a measurement wavelength of 420 nm, a detection range of 0.01 mg / L to 10 mg / L, and a resolution of 0.005 mg / L. The sample storage and processing temperature was strictly controlled at 4℃, and the measurement error was controlled within ±3%. The measurement results were recorded at specific time points to form a time series database. Outliers were removed using the 3σ principle. Finally, the ammonia nitrogen concentrations from multiple time periods were weighted and averaged to generate point source ammonia nitrogen concentration data, in mg / L.

[0071] Phosphorus nutrient concentration data were obtained by detecting phosphorus nutrient concentration based on non-point source pollution data.

[0072] In this embodiment, sampling points were set up in a 100m × 100m grid within the non-point source pollution area. Water samples were preferentially collected from drainage ditches, low-lying areas in fields, and the edges of water bodies based on land use type. Sampling covered at least three sampling cycles each during the rainy and dry seasons. Each sampling point collected 3 liters of samples, which were refrigerated on-site using polyethylene sampling bottles and transported to the laboratory for total phosphorus (TP) and dissolved phosphorus (DIP) determination. Total phosphorus was determined using the ammonium molybdate spectrophotometric method at a detection wavelength of 880nm, with a detection limit of 0.002mg / L. Dissolved phosphorus was measured after filtration through a 0.45μm filter membrane. The data were spatially interpolated (using the inverse distance weighting method with a power parameter of 2) to generate a distribution map of phosphorus nutrient concentration in the non-point source area. The data format was raster, with each grid cell measuring 50m × 50m, and the unit was mg / L. Statistical analysis was performed on the rainy and dry season sampling data to generate the spatiotemporal distribution characteristics of seasonal phosphorus nutrient concentrations.

[0073] Calculate ammonia nitrogen emission flux based on ammonia nitrogen pollution data;

[0074] In this embodiment, the ammonia nitrogen emission flux is calculated using point source polluted water outlets as the unit, combining ammonia nitrogen concentration (mg / L) and emission flow rate (m³ / s). 3 / s) data. Flow rate is measured using an online ultrasonic flow meter, with a measurement frequency of once every 5 minutes, and data storage duration is no less than 7 days. The hourly ammonia nitrogen emission load is calculated using the formula: Emission flux (kg / d) = Ammonia nitrogen concentration (mg / L) × Flow rate (m³ / d). 3 The emission flux value for each time period is obtained by multiplying the emission flux time series data by 86400 (seconds / day) / 1000 (mg / kg). The average daily emission flux is calculated by summing the emission flux time series data. For discrepancies between ammonia nitrogen concentration and flow rate data over time, linear interpolation is used to align the data. After calculation, the average daily emission flux is compared with the maximum emission flux, which is defined as the highest value within 7 days.

[0075] Total phosphorus flux was calculated based on phosphorus nutrient concentration data.

[0076] In this embodiment, runoff is estimated using rainfall data from standard rain gauges within the region and the surface runoff coefficient method. Rainfall data is collected hourly, and the runoff coefficient is calculated based on soil type, slope, and vegetation cover, with a value ranging from 0.1 to 0.5. The calculation formula is: Total phosphorus flux (kg / d) = Total phosphorus concentration (mg / L) × Runoff (m³ / d). 3 The formula ( / d) / 1000 multiplies the phosphorus concentration of each grid cell in the spatial area by the corresponding runoff volume, and sums the results to obtain the total phosphorus flux for the entire area source region. To ensure accuracy, the runoff volume is verified using the SWAT precipitation-runoff model, and the model parameters are repeatedly adjusted using historical flow observation data. The calculation results include the average daily total phosphorus flux and the maximum daily flux, in kg / d.

[0077] The intensity of pollutant emissions is assessed based on ammonia nitrogen emission flux and total phosphorus emission flux.

[0078] In this embodiment, the ammonia nitrogen emission flux (kg / d) and total phosphorus flux (kg / d) are divided by the corresponding environmental capacity thresholds for the pollutants. The environmental capacity threshold for ammonia nitrogen is set at 10 kg / d, and the environmental capacity threshold for total phosphorus is set at 1 kg / d, to obtain standardized emission intensity indices. A weighted average method is used to calculate the comprehensive emission intensity index, with a weight of 0.6 for ammonia nitrogen and 0.4 for total phosphorus. The calculation formula is: Comprehensive emission intensity = 0.6 × (ammonia nitrogen emission flux / 10) + 0.4 × (total phosphorus flux / 1). Emission intensity levels are distinguished based on the magnitude of the comprehensive index, with specific level thresholds as follows: low intensity < 0.5, medium intensity 0.5-1.0, and high intensity > 1.0. This assessment is implemented using Excel or MATLAB calculation tools, and the output results include the individual pollutant emission flux and the comprehensive emission intensity index, providing quantitative data support for subsequent ecological scheduling decisions.

[0079] Preferably, step S2, which involves detecting algae density in highly polluted areas, includes:

[0080] Algae samples were collected from highly polluted areas, and the algae samples were concentrated by sedimentation to obtain enriched algae suspension data.

[0081] In this embodiment, a high nutrient load area was used as the sampling location, and water samples were collected from the surface layer (0-0.5 meters) within this area. 5-liter transparent polyethylene sampling bottles were used, and disturbance of sediment was avoided during sampling. Three replicates were taken at each sampling point to ensure representative data. After sampling, the water samples were transported to the laboratory and stored at 4°C. The concentration process employed centrifugation sedimentation: the collected water sample was placed in a 500mL centrifuge bottle, the centrifuge speed was set to 3000 rpm, the centrifugation time to 20 minutes, and the centrifugation temperature was controlled at 20°C. A high-speed refrigerated centrifuge conforming to national standards (such as a Beckman Coulter Allegra X-15R) was selected. After centrifugation, the supernatant was discarded, and approximately 20mL of sediment at the bottom was retained, which is the enriched algae suspension. This suspension was concentrated approximately 25 times, and the volume was calculated by the ratio of the initial sampling volume to the concentrated volume. The concentrated volume and concentration factor were recorded in detail. To avoid damaging algal cells, no chemical fixatives are used before centrifugation, and subsequent processing is carried out promptly after centrifugation.

[0082] Based on the data of enriched algal suspensions, glass slide samples were prepared to obtain algal microscopic examination sample data.

[0083] In this embodiment, the enriched algal suspension was thoroughly mixed, and 10 μL of the suspension was evenly spread onto the center of a pre-cleaned, dust-free glass slide using a micropipette. After spreading, the sample was gently covered with a sterile coverslip to avoid air bubbles and ensure uniform sample thickness of approximately 0.1 mm. The prepared slide was immediately placed in a microscope sample holder for observation, or stored briefly at 4°C for no more than 1 hour. Sample preparation must be performed in a dust-free environment, and operators should wear clean gloves to reduce the risk of sample contamination. The temperature during sample preparation was controlled at room temperature (20±2°C), and the humidity was controlled at 40%-60% to maintain stable cell morphology.

[0084] Algae images were acquired based on algae microscopic sample data;

[0085] In this embodiment, a compound microscope with two magnification levels—100x (10x objective lens × 10x eyepiece) and 400x (40x objective lens × 10x eyepiece)—was selected, conforming to national standards (e.g., Olympus CX23). The microscope is equipped with a digital camera with a resolution of at least 2048×1536 pixels. At least 10 fields of view were randomly selected from each slide for image capture to ensure uniform distribution. Automatic exposure and white balance were used during image acquisition to avoid overexposure or color shift. Each image was saved in lossless TIFF format, with the file name including the sample number, acquisition date, and field of view number for easy data management. After image acquisition, basic quality checks were performed to ensure clear images, accurate focus, uniform background, and no significant interference from impurities.

[0086] Algal cell morphology is identified from algal images to obtain algal cell morphology data;

[0087] In this embodiment, digital image processing technology is used to identify cell morphology in acquired algal images. The images undergo grayscale conversion and background removal. An image thresholding method (with the threshold set to the top 5% of grayscale values ​​in the image's grayscale histogram) is used to separate algal cells from the background. Subsequently, morphological processing is performed, including dilation and erosion operations, with the structuring element size set to 3×3 pixels to remove image noise. For the segmented algal regions, cell contours are extracted, and their area, major axis, minor axis, and roundness (defined as 4π × area / perimeter²) are calculated. The cell morphology data table includes the cell number and area (μm²) of each cell. 2 The measurements included the major axis length (μm), minor axis length (μm), roundness value, and cell position coordinates. During the measurement process, the image scale was based on the microscope calibration scale, with 1 pixel corresponding to 0.2 μm. All operations were performed using the image processing toolbox based on Matlab, ensuring that the processing steps had fixed parameters and a consistent workflow.

[0088] The number of algal cells is calculated based on algal cell morphology data to obtain algal cell quantity data;

[0089] In this embodiment, based on cell contours, a connected component analysis algorithm is used to count the number of independent cells in each image. The statistical results for each image are summarized in a data table, including the image number and the corresponding total number of cells. Since the area of ​​the acquired field of view is known (e.g., 100 × 100 μm), the data is used in this method. 2 The average number of algal cells per unit field of view was calculated by averaging the cell counts across all fields of view. To improve statistical accuracy, at least 100 images were analyzed, and the overall mean and standard deviation were calculated. Cell count data were expressed as unitless counts, and the data storage format included sampling time, sample number, field of view number, cell count, and statistical results.

[0090] Algal concentration data is obtained by converting algal cell count data to unit volume.

[0091] In this embodiment, the algal cell concentration per unit volume is calculated by combining the algal suspension concentration factor and the sample slide field of view area. The specific conversion formula is: Algal concentration (cells / mL) = (Average number of cells per field of view × Slide sample area magnification × Concentration factor) / (Field of view volume). The field of view volume is calculated using the slide sample thickness (0.1 mm) and the field of view area (e.g., 100 μm × 100 μm), converted to 0.001 mm. 3 The concentration factor is calculated by the ratio of the sampling volume (e.g., 5L) to the concentration volume (20mL) in step 1. During this conversion, all length units are converted to meters, and the cell concentration results are stored in the database, along with the sampling time, sample number, and calculation parameters.

[0092] Algae density data is obtained by detecting algae density based on algae concentration data.

[0093] In this embodiment, algae density detection is based on algae concentration data, combined with water flow velocity and water volume data. Water flow velocity is measured using an acoustic Doppler current meter with an accuracy of ±0.01 m / s, a sampling frequency of once per minute, and measurement points are evenly distributed in high nutrient load areas. Water volume is calculated using reservoir depth profile and water surface area data; water depth is measured using a sonar depth sounder with an accuracy of ±0.05 m. Algae density is calculated using the formula: Algae Density (cells / m³) 3 = Algae concentration (cells / mL) × 1000 (mL / L) × Total water volume (m³) 3 ) / Monitored water volume (m 3 The data on algae density at monitoring time and location are complete and stored in a format that includes time, spatial coordinates, water depth, water volume, and calculation results. The data is used for dynamic response analysis of reservoir ecological scheduling.

[0094] Preferably, step S2, which involves analyzing algal toxins in the water based on algal density data, includes:

[0095] High-density algae areas were identified based on algae density data.

[0096] In this embodiment, algae density data (unit: cells / m) is used. 3 The data was imported into a Geographic Information System (GIS) platform (such as ArcGIS 10.8), and a spatial database was established based on the spatial coordinates of different sampling points in the reservoir and the corresponding algae density data. The Inverse Distance Weighted (IDW) spatial interpolation method was used, with an interpolation radius of 500 meters and a weighting index of 2, to generate a continuous spatial distribution map of algae density. A high-density algae threshold of 1 × 10^6 cells / m² was set. 3 Using this threshold as a criterion, continuous areas with a density greater than or equal to the threshold in the spatial distribution map are automatically extracted as high-density algae areas. The area boundaries are vectorized to an accuracy of 10 meters. The extracted area, boundary, and spatial location data are stored in a database for subsequent temperature and pH measurement and positioning. All coordinates conform to the WGS-84 standard.

[0097] Water temperature data were obtained by measuring water temperature in areas with high algae density.

[0098] In this embodiment, multiple water temperature measurement stations are deployed within the high-density algae area, with the distance between measurement points controlled within 100 meters to ensure coverage of the entire high-density area. A high-precision digital water quality multi-parameter instrument (such as the YSIProDSS) is used for water temperature measurement, with a measurement accuracy of ±0.05℃ and a resolution of 0.01℃. Water temperature is sampled along the water depth profile at the measurement points, including the surface (0-0.5m), middle layer (0.5-2m), and bottom layer (2-5m), with a sampling frequency of once per hour for 72 consecutive hours to ensure dynamic timeliness of the data. Data is uploaded to the data management platform in real time via a wireless transmission module. All temperature data includes a sampling timestamp and measurement point coordinates, and the data storage format is a CSV file. To eliminate outliers, a sliding window method is used, with a window width of 3 hours, to calculate the moving average temperature value as the final temperature data.

[0099] The pH of the water body was measured based on areas with high-density algae to obtain pH data.

[0100] In this embodiment, pH measurements correspond to temperature measurements, using a multi-parameter water quality analyzer equipped with pH electrodes (e.g., Hach HQ40d). Instrument calibration was strictly performed according to the instrument manual, using standard buffer solutions at pH 4.00, 7.00, and 10.00 to calibrate the electrodes. Calibration was performed before each measurement. The pH measurement range was 0-14, with a resolution of 0.01 and an accuracy of ±0.02 pH units. Sampling covered the surface, middle, and bottom layers at a frequency of once per hour for 72 consecutive hours. pH values, corresponding times, and coordinates were recorded in real-time on-site, and the data was uploaded and stored in the same format as the temperature data. To prevent instrument drift, measurements were compared daily with standard buffer solutions to ensure the measurement error did not exceed ±0.02; otherwise, the instrument was recalibrated. Abnormal pH data were removed using the 3σ method. The final data is provided in time-series format for subsequent toxin degradation rate analysis.

[0101] Toxin degradation rate was obtained by analyzing water temperature and pH data.

[0102] In this embodiment, the toxin degradation rate was determined using a laboratory water sample toxin degradation kinetics test method based on spatiotemporal data of water temperature and pH. Representative water samples were selected, and the temperature and pH conditions were kept consistent with the field monitoring data. The toxin degradation rate is commonly expressed using a first-order kinetic model: dC / dt=-kC, where C is the toxin concentration and k is the degradation rate constant. The degradation rate constant k is affected by temperature and pH and is adjusted using the Arrhenius equation: k=k_ref×exp[-Ea / R×(1 / T-1 / T_ref)], where T is the absolute temperature (K), Ea is the activation energy (J / mol, set to 50000J / mol), R is the gas constant 8.314J / (mol·K), and k_ref is the degradation rate constant at the reference temperature (T_ref=298K). The effect of pH on the k value was determined experimentally using a specific adjustment factor. The adjustment coefficient was calculated by linear interpolation within the pH range of 6.5 to 9.0. The final result of the toxin degradation rate k is expressed in units of 1 / d, generating dynamic degradation rate data with a time resolution of hours. In the experimental procedure, the toxin concentration was determined using high-performance liquid chromatography-mass spectrometry (HPLC-MS), with a measurement error of less than 5%.

[0103] Toxin adsorption was simulated based on the toxin degradation rate to obtain toxin adsorption data.

[0104] In this embodiment, the adsorption process is described using the Langmuir isotherm adsorption model, with the formula Q = (Q_max × K × C) / (1 + K × C), where Q is the amount of toxin adsorbed per unit mass of sediment (mg / kg), Q_max is the maximum adsorption capacity (set to 2.5 mg / g), K is the adsorption equilibrium constant (unit L / mg, value 0.3), and C is the toxin concentration in the water (mg / L). The toxin concentration is dynamically adjusted over time according to the degradation rate, with a simulation time step of 1 hour. The sediment adsorption capacity data were determined experimentally using batch adsorption experiments. The dry weight of sediment samples was determined using the drying method (dried at 105℃ to constant weight), and the toxin content was determined using HPLC-MS. The adsorption model calculation process was implemented using MATLAB programming, with input parameters including real-time toxin concentration, sediment content, and ambient temperature. The toxin adsorption data is output in time series form, in mg / kg, with timestamps.

[0105] Toxin migration data is obtained by performing toxin adsorption data;

[0106] In this embodiment, toxin migration analysis was performed using a hydrodynamic and diffusion model, considering water transport and toxin re-release from sediments. Hydrodynamic parameters were measured using ADCP (Acoustic Doppler Current Profiler) deployed within the reservoir, with a measurement time resolution of 10 minutes and a flow velocity accuracy of ±0.01 m / s. The diffusion coefficient was set based on the site temperature and flow regime, with a typical value of 1 × 10⁻⁵ m⁻¹. 2 The toxin migration process was numerically solved using a two-dimensional convection-diffusion equation, discretized using the finite difference method, with a time step of 600 seconds and a spatial grid size of 50 m × 50 m. Boundary conditions were set based on the toxin concentration input at the reservoir inlet and an open boundary at the outlet. The sediment re-release rate was calculated using first-order kinetics, with a release rate constant set to 0.1 / d. Model inputs included the toxin adsorption amount output from step 5, along with the on-site water flow velocity and diffusion parameters. The output was the spatiotemporal distribution concentration of the toxin in the water (mg / L). Data was stored using time and spatial coordinates and was formatted to support import from a GIS platform.

[0107] The algal toxin content in the water body is assessed based on toxin migration data, and the algal toxin data in the water body is obtained.

[0108] In this embodiment, the algal toxin content in each sampling unit is calculated by combining toxin migration concentration data and algal spatial distribution data through spatial overlay analysis. The formula for calculating algal toxin content is: Tox_content = C_toxin × B_mass, where C_toxin is the toxin concentration in the water body of that unit (mg / L), and B_mass is the algal biomass, which is calculated by multiplying the algal cell density by the mass of a single cell (set to 1 × 10^-10 mg / cell). The volume of the water body unit is calculated based on water depth and area, in L. The total toxin mass is estimated by combining the unit water body volume, in mg. The data from each unit are summarized to generate an overall spatial distribution map of algal toxins in the reservoir. The data results are labeled with time and space, and are used for ecological scheduling decision support.

[0109] Preferably, step S3 includes the following steps:

[0110] Step S31: Obtain reservoir hydrological data and extract the maximum rainfall characteristics to obtain the maximum rainfall data;

[0111] In this embodiment, rainfall data is collected using multiple automatic weather stations located within the reservoir basin. The station density is controlled at one station per 100 square kilometers to ensure uniform spatial coverage of the data. The rain gauges used are weighing automatic rain gauges with an accuracy of ±0.1 mm and a resolution of 0.2 mm. They automatically record rainfall time series, with a sampling frequency of once every 10 minutes. Rainfall data from the past year is collected, historical extreme events are filtered, and the maximum rainfall data is extracted using time series extreme value analysis methods (such as the annual maximum rainfall method). The maximum rainfall is expressed in millimeters (mm), and the time scale is set to 24-hour cumulative rainfall. To avoid the influence of outliers, the maximum rainfall value is subjected to a 3σ test to remove extreme outliers. The data recording time stamp uses the UTC standard, and the data storage format is a structured database, supporting subsequent automatic retrieval.

[0112] Step S32: Calculate rainfall runoff based on the maximum rainfall data;

[0113] In this embodiment, the rainfall runoff is calculated based on the empirical formula in hydrology: Q = P × A × C, where Q is the runoff (cubic meters, m³). 3 P is the maximum rainfall (meters, m), and A is the catchment area (square meters, m²). 2C is the runoff coefficient (dimensionless). The runoff coefficient C is determined based on land use type and soil permeability, ranging from 0.3 to 0.9. Specifically, a C value of 0.5 for a typical agricultural watershed is selected as the calculation basis. The maximum rainfall P is obtained from step S31, converted to meters (mm / 1000). In the calculation of runoff Q, the watershed area A is obtained from reservoir watershed measurements, converted to square meters. By substituting the above parameters into the formula, the runoff caused by rainfall is calculated, and the result is rounded to two decimal places in cubic meters. The calculation process is implemented through programming to ensure the correctness of each parameter input and unit conversion.

[0114] Step S33: Obtain reservoir drainage area data; perform flow conversion based on rainfall runoff and reservoir drainage area data to obtain flow data;

[0115] In this embodiment, the reservoir basin area data is derived from high-precision digital elevation model (DEM) analysis in a geographic information system (GIS). The basin boundary is extracted using digital hydrological analysis tools, achieving an area measurement accuracy of ±1%. The basin area is measured in square kilometers (km²). 2 ), converted to square meters (m 2 The rainfall runoff Q and watershed area A calculated in step S32 are used in subsequent calculations. The flow rate data per unit time (cubic meters per second, m³ / s) are further calculated. 3 / s), the calculation formula is: flow rate = rainfall runoff / time interval, where the time interval is determined to be 24 hours (86400 seconds) based on the time scale of the rainfall data. Rainfall runoff (m³) 3 Divide the average flow rate by the time in seconds to obtain the average flow rate, rounded to two decimal places. Flow rate data is recorded synchronously with reservoir water level observation data, with consistent time stamps, and stored in the database to ensure subsequent data retrieval and analysis.

[0116] Step S34: Obtain reservoir cross-sectional data; identify the substrate material based on the reservoir cross-sectional data to obtain substrate material data; determine the cross-sectional roughness based on the substrate material data;

[0117] In this embodiment, underwater sonar sounding technology (such as a multibeam echo sounder (MBES)) is used to acquire the topographic profile of the reservoir cross section, achieving a resolution of 0.5 meters. The cross section data includes water depth and substrate morphology information. Seabed sediment samples collected on-site are used, and a particle size analyzer is employed to detect the sediment particle size distribution, ranging from 0.001 to 2 mm, indicating a fine particle distribution. The substrate material is categorized into sand, gravel, and silt based on particle size distribution. The substrate material data corresponds to each sampling point in the cross section, and the data storage includes the substrate material category code. Cross section roughness is represented by the Manning roughness coefficient (n), with different materials corresponding to different Manning coefficients: 0.025 for sandy substrates, 0.035 for gravel, and 0.012 for silt. Based on the substrate material distribution in the cross section, the overall roughness is calculated using an area-weighted method, specifically by multiplying the Manning coefficient of each material by its corresponding area percentage and then summing the results. The roughness data is stored as dimensionless Manning coefficient values.

[0118] Step S35: Calculate the bottom undulation based on the reservoir cross-section data; calculate the flow resistance coefficient based on the bottom undulation and cross-sectional roughness; correct the flow rate data based on the flow resistance coefficient to obtain the corrected flow rate data;

[0119] In this embodiment, the undulation of the seabed at the cross-section is calculated using multibeam echo sounding data. The specific calculation formula is: undulation = (water depth at the highest point of the cross-section - water depth at the lowest point of the cross-section) / average water depth of the cross-section. The calculated result is a unitless ratio, typically ranging from 0.01 to 0.2. The flow resistance coefficient C_d is calculated by combining the Manning roughness coefficient n and the seabed undulation, using the empirical formula C_d = (g × n) / ( ... 2 ) / R^(1 / 3), where g is the gravitational acceleration 9.81 m / s². 2 R is the hydraulic radius (m), which is calculated from the cross-sectional area and wetted perimeter. Then, C_d is corrected for undulation using the formula C_d_corrected = C_d × (1 + 5 × undulation), ensuring that the influence of the substrate morphology on flow resistance is considered. The corrected flow rate Q_corrected is adjusted according to the flow-resistance relationship, calculated as Q_corrected = Q / (1 + C_d_corrected), where Q is the flow rate data obtained in step S33, and Q_corrected is the corrected flow rate data. All calculated values ​​are rounded to three decimal places, and the flow rate unit is cubic meters per second (m²). 3 The calculation results are stored in a database for subsequent water quality analysis.

[0120] Step S36: Perform toxin dilution analysis on the algal toxin data of the water body based on the corrected flow data to obtain toxin dilution data;

[0121] In this embodiment, the toxin dilution analysis is based on the dilution model C_diluted=C_initial×(Q_initial / Q_corrected), where C_diluted is the diluted toxin concentration (mg / L), C_initial is the initial toxin concentration (mg / L), and Q_initial is the original flow rate (m³ / L). 3 / s), Q_corrected is the flow rate (m³) after correction in step S35. 3 / s). The input parameter C_initial is provided by algal toxin concentration monitoring data, accurate to 0.001 mg / L. All flow rate data are averaged, with a time window of 24 hours. Consistent flow rate units are maintained during calculations, and dilution concentration results are retained to four decimal places. This calculation is performed in the MATLAB environment, employing vectorization to achieve rapid calculations across multiple time periods and spatial points. Toxin dilution data serves as the basis for dynamic changes in water toxins and is directly used in subsequent simulations.

[0122] Step S37: Perform toxin diffusion simulation based on toxin dilution data to obtain algal toxin diffusion data.

[0123] In this embodiment, a two-dimensional diffusion transport numerical model is used for calculation. First, toxin dilution data is input into the system as the initial concentration condition for the simulation, ensuring that the starting point accurately reflects the current distribution of toxins in the water. Water flow velocity data is obtained from an Acoustic Doppler Current Profiler (ADCP), which provides a flow velocity time resolution of 10 minutes and a spatial resolution of 50 meters, specifically including the horizontal velocity component of the water flow. The diffusion coefficient used in the diffusion process is set to 1 x 10^-5 square meters per second. The diffusion coefficient is dynamically adjusted based on the measured water temperature data, with the adjustment range limited to ±10% to reflect the influence of temperature on the diffusion rate. Numerical calculations employ an explicit finite difference method, dividing the space into 50-meter by 50-meter grids, with a time step of 300 seconds to ensure numerical stability and calculation accuracy. A flux-free boundary condition is used at the model boundary to prevent toxin diffusion from outside the boundary into the model. The simulation runs continuously for 48 hours, fully simulating the migration and diffusion changes of toxins in the water. The final output of toxin concentration distribution results is stored in the form of raster data, which supports subsequent visualization using geographic information systems and provides a basis for reservoir ecological scheduling.

[0124] Preferably, step S37 includes the following steps:

[0125] Step S371: Upload the toxin dilution data to the water environment diffusion simulation platform;

[0126] In this embodiment, the toxin dilution data storage format is a CSV file, with fields including timestamp (UTC time, format YYYY-MM-DD HH:MM:SS), latitude and longitude of the monitoring point (using WGS84 coordinate system), and toxin concentration (unit mg / L, accurate to four decimal places). The data is uploaded to the water environment diffusion simulation platform server via an encrypted FTP channel using the TCP / IP network transmission protocol. Before uploading, data integrity is verified using an MD5 hash algorithm to confirm that the file has not been tampered with, ensuring consistency between the uploaded and local data. The upload process is automatically segmented into 10MB blocks, supporting breakpoint resumption to ensure upload stability under network fluctuations. After upload, the server returns a confirmation message, confirming that the data has been completely received and stored in the specified database table. The upload interface uses a REST API call, with the transmission rate automatically adjusted according to network bandwidth to ensure real-time performance. The time series of the toxin dilution data is sorted during upload to ensure temporal continuity and facilitate subsequent simulation calls.

[0127] Step S372: Set the water flow velocity range to 0.05m / s-1.2m / s, the water depth range to 0.5m-10.0m, and the water temperature range to 5℃-35℃;

[0128] In this embodiment, in the parameter setting interface of the water environment diffusion simulation platform, the flow velocity parameter is set as an interval input format, with a minimum input value of 0.05 m / s and a maximum input value of 1.2 m / s, strictly limited to meters per second (m / s), and the value is retained to two decimal places. The water depth parameter is in meters (m), with a setting range of 0.5 m (shallowest) to 10.0 m (deepest), and the data source is the statistical range of measured data from cross-sectional water depth monitoring equipment (such as sonar depth sounders). The water temperature range is in degrees Celsius (°C), with a minimum input value of 5°C and a maximum input value of 35°C. Data is collected from automatic water quality monitoring stations using platinum resistance thermometers, with a temperature resolution of 0.1°C and an error of ±0.2°C. The above three parameter settings are used for initializing the simulation boundary conditions. The platform generates a parameter field through a parameter interpolation algorithm to ensure that the simulation process covers the dynamic changes of the entire set range. After the parameters are set, they are verified by the preprocessing module to ensure that the input values ​​comply with physical and engineering specifications and that there is no abnormal data exceeding the standard.

[0129] Step S373: Set the cross-sectional bottom roughness n value to 0.015-0.035, and the cross-sectional undulation variation to 0.3m-2.0m;

[0130] In this embodiment, the cross-sectional bed roughness is expressed in Manning roughness (n), using an interval numerical input method with a minimum value of 0.015 and a maximum value of 0.035, derived from the cross-sectional bed material analysis results, specifically corresponding to the roughness range of sandy to fine gravelly beds. The cross-sectional relief is expressed in meters (m), with a range of 0.3m to 2.0m, calculated using multibeam echo sounding data, and the formula is the difference between the maximum and minimum water depths at the cross-section. These parameters are input into the simulation platform as key parameters for flow resistance and turbulence characteristics, directly affecting the hydrodynamic conditions in the diffusion model. After parameter input, the system performs numerical stability testing to ensure that there are no divergences or numerical instabilities in the simulation calculation within the parameter range. The cross-sectional bed roughness and relief settings are transmitted as quantitative parameters in the simulation calculation, participating in the dynamic calculation of the flow resistance coefficient and diffusion coefficient.

[0131] Step S374: Set the simulation time step to 1 min-15 min, the total simulation duration to 6 h-72 h, and the diffusion scale to 50 m-500 m in the local area;

[0132] In this embodiment, the simulation time step is measured in seconds, with a range of 60 seconds (1 minute) to 900 seconds (15 minutes). The step size is determined based on the dynamic characteristics of the water body and the requirements for numerical stability. Smaller steps result in higher simulation accuracy but increase computational complexity. The total simulation duration is measured in hours, ranging from 6 to 72 hours to meet the simulation requirements for short- and medium-duration diffusion processes. The diffusion scale is defined as the simulation spatial range, set as a local area of ​​50 to 500 meters, divided into spatial units using a two-dimensional grid. The grid resolution is controlled between 10 and 50 meters, and the number of grid cells is automatically calculated based on the diffusion scale. After the above parameters are input through the simulation platform interface, the numerical solver pre-configuration module is invoked. Based on the time step and spatial grid, numerical stability conditions (e.g., CFL conditions) are set, and the maximum allowable time step is automatically calculated. During simulation initialization, the total number of steps is calculated based on the total duration and time step to ensure accurate simulation timing. The diffusion scale and spatial grid together determine the computational scale of the model and the accuracy of the output data.

[0133] Step S375: Run the water convection-diffusion numerical calculation module to obtain algal toxin diffusion data.

[0134] In this embodiment, when the numerical calculation module is started, the uploaded toxin dilution data is first loaded as the initial toxin concentration field input. Combined with the flow velocity, water depth, water temperature, roughness, undulation, time step, simulation duration, and diffusion scale parameters set in steps S372 to S374, the boundary conditions and initial conditions of the convection-diffusion equation are constructed. The calculation uses the finite difference method to discretize the two-dimensional convection-diffusion partial differential equation. Time propagation uses explicit or implicit time integration, and spatial discretization uses central difference or upwind difference schemes. The water velocity vector field is generated by linear interpolation of parameters within the input range. The influence of water temperature on the diffusion coefficient is adjusted using the empirical formula D_T = D_20 × θ^(T-20), where D_20 is the diffusion coefficient at 20℃, θ is taken as 1.03, and T is the current water temperature. Water depth and cross-sectional bed parameters participate in the hydrodynamic calculation, affecting the calculation of water flow velocity and turbulent diffusion coefficient. During the calculation process, an adaptive time step is used to monitor the residuals to ensure numerical convergence. The calculation results represent the spatiotemporal distribution of toxin concentration in each grid cell within a two-dimensional space. The output format is time-series raster data, with units of mg / L and values ​​rounded to four decimal places. After simulation, the results are stored in a database for subsequent analysis and application scheduling.

[0135] Preferably, step S4 includes the following steps:

[0136] Step S41: Identify areas with high concentrations of toxins based on algal toxin diffusion data;

[0137] In this embodiment, the input algal toxin diffusion data is first imported into a GIS spatial analysis system, where the toxin concentration data of the two-dimensional spatial distribution is imported. The data format is time-series grid point concentration values ​​in mg / L, with the time resolution consistent with the simulation time step. A concentration threshold of 0.05 mg / L is set, and the toxin concentration of grid cells is filtered to identify areas where the concentration exceeds this threshold within a continuous time period. A spatial clustering algorithm (based on proximity-based connected region identification) is used to spatially merge the grid points exceeding the threshold, forming polygonal boundaries of several high-concentration toxin retention areas. For each area, its area (in square meters), maximum concentration value, and average concentration value are calculated and stored as regional feature parameters. This process utilizes Python libraries such as GDAL and Shapely to perform data reading and spatial calculations, ensuring a spatial accuracy of 10-meter grid resolution.

[0138] Step S42: Calculate the water flow velocity based on the high concentration of toxin retention area to obtain water flow velocity data;

[0139] In this embodiment, based on the high-concentration toxin retention area, the data interface of the reservoir flow velocity measurement device is invoked to obtain measured flow velocity data within the retention area. The flow velocity unit is meters per second (m / s). At least three cross-sections spanning the area are set up using ADCP (Acoustic Doppler Current Profiler) to measure the vertical and horizontal distribution of the water flow velocity. Spatial interpolation processing is performed on the flow velocity data at the measurement points, using Kriging interpolation to generate a continuous flow velocity field within the retention area. During data processing, the interpolation radius is set to 50 meters to ensure spatial continuity of the data, and the time synchronization accuracy is controlled within 5 minutes. The flow velocity data is verified, and outliers are removed (data points with flow velocities exceeding 2 m / s or below 0.01 m / s are considered outliers) using a 3x standard deviation method. Finally, the mean, maximum, and distribution matrix of the flow velocity within the retention area are output as water flow velocity data, providing a basis for subsequent exchange capacity calculations.

[0140] Step S43: Evaluate the water exchange capacity based on the water flow velocity data to obtain water exchange capacity data; identify areas of hydrodynamic stagnation based on the water exchange capacity data;

[0141] In this embodiment, the water exchange capacity is evaluated using the water exchange rate, calculated as E = Q / V, where E is the water exchange rate (unit: d⁻¹) and Q is the flow rate within the retention area (m³). 3 / d), V is the volume of water in the retention area (m³) 3 The flow rate Q is calculated by integrating the product of flow velocity data and the cross-sectional area of ​​the retention area, with the area measured in square meters. The integration is performed using a numerical integration method. The water volume V is obtained by multiplying the area of ​​the retention area by the average water depth (obtained from a depth gauge, in meters, with an accuracy of 0.1 meters). The water exchange rate threshold is set to 0.1d⁻¹; areas below this value are defined as hydrodynamically stagnant areas. Spatial positioning is performed using a geographic information system to generate a vector map of the stagnant area boundaries. The characteristics of hydrodynamically stagnant areas include the area, average water exchange rate, and its spatial distribution. The data is saved in a database table format for reference in subsequent monitoring section layout.

[0142] Step S44: Set up dissolved oxygen monitoring sections according to the hydrodynamic stagnation area to obtain dissolved oxygen monitoring data; calculate the oxygen concentration change rate according to the dissolved oxygen monitoring data; determine the dissolved oxygen anomaly detection based on the oxygen concentration change rate to obtain dissolved oxygen anomaly data;

[0143] In this embodiment, based on the boundary of the hydrodynamic stagnation zone, at least three cross-sections are selected within the stagnation zone to set up dissolved oxygen monitoring points. The cross-sections are arranged to reference the uniformity of the flow velocity distribution in the reference area, with a cross-section length and a point spacing of 50 meters and 10 meters, respectively. Dissolved oxygen monitoring uses a portable electrochemical dissolved oxygen probe, with a measurement unit of mg / L, an accuracy of ±0.1 mg / L, and a sampling frequency of 15 minutes. The collected data is uploaded to the data processing system and organized according to the time series. The oxygen concentration change rate is calculated using the central difference method, with the formula ΔDO / Δt=(DO_t+1-DO_t-1) / (2×Δt), where Δt is the time interval (15 minutes) and DO is the dissolved oxygen concentration. An abnormal threshold for the oxygen concentration change rate is set at ±0.05 mg / (L·min), and data points exceeding this range are considered to have dissolved oxygen anomalies. Dissolved oxygen anomaly data is stored in the form of abnormal time, location, and abnormal amplitude to facilitate the location of potential hypoxia or hyperoxygenation in the water body.

[0144] Step S45: Based on the abnormal dissolved oxygen data, schedule the ecological water release of the reservoir to obtain the ecological water release data of the reservoir.

[0145] In this embodiment, based on dissolved oxygen anomaly data, combined with the current reservoir water level, inflow and outflow rates, and historical scheduling rules, the required ecological water release volume is calculated. The outflow rate (unit: m³) is measured using a control sluice gate flow meter. 3 / s), with a scheduling time resolution set to 1 hour. The minimum ecological water release flow threshold is set to 5m³ / s. 3 / s, the maximum permissible ecological water release flow is 50m³. 3 / s, the water release volume is allocated proportionally based on the size and severity of the abnormal area. The scheduling plan includes the water release flow rate, water release time period, and water release cross-section location. The data format includes time and flow rate (m³ / s). 3 The system displays the coordinates (latitude and longitude) of the water release section. Water release scheduling instructions are remotely issued and executed through the control system. Scheduling status and real-time flow data are synchronously transmitted back to the database for subsequent effect evaluation and water quality dynamic response analysis. All scheduling data is archived hourly to ensure historical tracking and a basis for subsequent scheduling adjustments.

[0146] Of particular importance, step S45 includes the following steps:

[0147] Step S451: Determine the water quality risk level based on the abnormal dissolved oxygen data to obtain water quality risk data;

[0148] In this embodiment, dissolved oxygen anomaly data were collected from multiple dissolved oxygen monitoring sections deployed in areas of slowed hydrodynamic flow. The section layout was determined based on the previous toxin migration paths and the slowed hydrodynamic flow zone. Dissolved oxygen electrode sensors were used, with a depth interval of 0.5m and a horizontal interval of 25m. The monitoring frequency was set to once every 10 minutes. The dissolved oxygen concentration values ​​collected at each monitoring point were compared with the National Surface Water Environmental Quality Standard GB3838-2002. The dissolved oxygen thresholds for Class I water bodies were set at 7.5mg / L, Class II at 6.0mg / L, and Class III at 5.0mg / L. Values ​​below 5.0mg / L were considered abnormal. For abnormal points, a risk scoring method was used to calculate the water quality risk value based on the magnitude of the oxygen concentration decrease, the duration, and the affected area. The scoring criteria were as follows: 4 points for a concentration below 4.0mg / L, 6 points for below 3.0mg / L; 3 points for a continuous abnormal duration exceeding 2 hours, 5 points for exceeding 6 hours; and 5 points for an affected area exceeding 2500m². 2 2 points, exceeding 5000m 2 A score of 4 points is awarded. Areas with a comprehensive score of over 10 points are marked as high-risk areas, those with a score of 6-10 points are medium-risk areas, and those with a score below 6 points are low-risk areas. Finally, a spatially distributed water quality risk level map is generated in GIS raster format, and water quality risk data is output.

[0149] Step S452: Based on water quality risk data, divide the target water body regulation zone to obtain ecological regulation zone data;

[0150] In this embodiment, the water quality risk level raster data output in step S451 is called, and high-risk and medium-risk areas are the main intervention targets. Spatial clustering algorithm is used to identify connected regions, and risk areas with a horizontal spacing of ≤50m and a vertical spacing of ≤0.5m between adjacent raster cells are merged, while areas with an area less than 400m² are removed. 2 Scattered areas were demarcated. After demarcation, control priorities were assigned according to risk level: high-risk areas had a priority of 1, medium-risk areas 2, and low-risk areas were not included in the current scheduling scope. Subsequently, a water boundary vector map of the control area was obtained, and the area volume was analyzed using DEM water depth data. A boundary closure algorithm was used to convert the control area into a closed water body spatial entity for subsequent water volume regulation calculations. All area boundary, priority, and spatial location information were uniformly output as ecological control area data, in Shapefile format with a corresponding DBF data table containing fields such as number, area, average depth, and priority.

[0151] Step S453: Calculate the dispatchable water volume based on the data of the ecological regulation area;

[0152] In this embodiment, the lower limit of the dispatched water level is set as the ecological water demand line, which is 1.2m above the dead water level; the upper limit of the dispatched water level is set as either the pre-flood limit water level or the real-time water level, whichever is lower. The current reservoir capacity is determined based on the water level-capacity curve; the water volume in the dispatched interval is calculated using the triangle approximation method: water surface area × (upper limit water level - lower limit water level) / 2. The current dispatching period is set within a 24-hour cycle, with the hour as the basic unit of calculation, combined with historical evaporation (daily average evaporation is approximately 4mm, calculated at 0.17m). 3 / m 2 The total dispatchable water volume is calculated by deducting evaporation losses (d). All dispatchable water volume values ​​are associated with the regional control objects, and a dispatchable water volume data table containing fields (region number, control priority, region volume, and dispatchable water volume) is output.

[0153] Step S454: Calculate the water release flow rate for the specified time period based on the available water volume;

[0154] In this embodiment, the scheduling cycle is set to 24 hours, divided into 8 water release periods, each lasting 3 hours. Based on the adjustable water volume data obtained in step S453, the total water release volume is allocated according to the priority of each control area, with high-priority areas receiving no less than 60% of the total water release volume. The water release demand in each area is calculated by back-calculating the required replacement volume based on the oxygen concentration change rate (the oxygen concentration recovery target is 6.0 mg / L, corresponding to a dilution ratio > 1.8). The area volume is divided by the dilution ratio to determine the target water release volume for that area. The target water release volume is evenly distributed across the 8 time periods, and divided by 3 hours to obtain the target flow rate per time period, in m³. 3 / s. The actual discharge outlet structural parameters are checked against the gate cross-sectional width (e.g., 3m), opening height (e.g., 1.5m), and head difference to ensure the flow rate does not exceed the maximum safe discharge capacity. Finally, the flow rate information for each time period is summarized into a time-period discharge flow rate data table, with fields including time period number, control area number, and target flow rate (m³ / s). 3 / s), scheduling duration (h), total scheduling water volume (m³) 3 ).

[0155] Step S455: Conduct ecological water release scheduling of the reservoir according to the water release flow rate during the time period to obtain ecological water release data of the reservoir.

[0156] In this embodiment, the scheduling command is issued from the reservoir scheduling automatic control system to the gate execution unit. The execution command is initiated 10 minutes before the start of each scheduling period, and the gate opening height is determined based on the flow data from step S454. Discharge verification is performed using the upstream water head and the gate opening area, employing the flow formula Q=μ·B·H·√(2gH), where μ is 0.6, B is the gate width, and H is the upstream water head. On-site water level gauges and flow meters record the actual discharge flow and water level fluctuations at 1-minute intervals to ensure the error with the planned value does not exceed ±5%. During the scheduling process, the scheduling console records the operation time, opening height, duration, and actual flow in real time, and stores this data in the scheduling log. After each period, scheduling feedback data is recorded and integrated into complete reservoir ecological water release data. The data includes fields such as the control area number, discharge period, target flow, actual flow, scheduled water volume, execution error, and execution status, and is exported in both Excel and CSV formats.

[0157] Preferably, this specification also provides a reservoir ecological scheduling system that considers dynamic water quality response, used to execute the reservoir ecological scheduling method considering dynamic water quality response as described above. The reservoir ecological scheduling system considering dynamic water quality response includes:

[0158] The pollution source tracing module is used to acquire reservoir water quality monitoring data; perform eutrophication analysis based on the reservoir water quality monitoring data to obtain eutrophication data; and trace pollution sources based on the eutrophication data to obtain pollution source data.

[0159] The water body algal toxin analysis module is used to assess pollutant emission intensity based on pollution source data; identify high-pollution areas based on pollutant emission intensity; detect algal density in high-pollution areas to obtain algal density data; and analyze water body algal toxins based on algal density data to obtain water body algal toxin data.

[0160] The toxin diffusion simulation module is used to acquire reservoir hydrological data, extract maximum rainfall characteristics to obtain maximum rainfall data, perform flow conversion based on maximum rainfall data to obtain flow data, and perform toxin diffusion simulation based on flow data and algal toxin data to obtain algal toxin diffusion data.

[0161] The reservoir ecological water release scheduling module is used to identify areas with high concentrations of toxins based on algal toxin diffusion data; to detect dissolved oxygen anomalies based on these areas, thus obtaining dissolved oxygen anomaly data; and to schedule reservoir ecological water release based on this data, thus obtaining reservoir ecological water release data.

[0162] 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.

[0163] 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 reservoir ecological scheduling method considering dynamic water quality response, characterized in that, Includes the following steps: Step S1: Obtain reservoir water quality monitoring data; Eutrophication analysis of the water body was conducted based on the reservoir water quality monitoring data to obtain eutrophication data. Pollution source data is obtained by tracing pollution sources based on water eutrophication data. Step S2: Assess pollutant emission intensity based on pollution source data; High-pollution areas are identified based on pollutant emission intensity; algae density is measured in these high-pollution areas to obtain algae density data; and algal toxins in the water are analyzed based on the algae density data to obtain algal toxin data. Step S3: Obtain reservoir hydrological data and extract the maximum rainfall characteristics to obtain the maximum rainfall data; Flow rate data is obtained by converting the maximum rainfall data; toxin diffusion data is obtained by simulating the diffusion of algal toxins based on the flow rate data and water algal toxin data; wherein, step S3 includes the following steps: Step S31: Obtain reservoir hydrological data and extract the maximum rainfall characteristics to obtain the maximum rainfall data; Step S32: Calculate rainfall runoff based on the maximum rainfall data; Step S33: Obtain reservoir drainage area data; perform flow conversion based on rainfall runoff and reservoir drainage area data to obtain flow data; Step S34: Obtain reservoir cross-sectional data; identify the substrate material based on the reservoir cross-sectional data to obtain substrate material data; determine the cross-sectional roughness based on the substrate material data; Step S35: Calculate the bottom undulation based on the reservoir cross-section data; calculate the flow resistance coefficient based on the bottom undulation and cross-sectional roughness; correct the flow rate data based on the flow resistance coefficient to obtain the corrected flow rate data; Step S36: Perform toxin dilution analysis on the algal toxin data of the water body based on the corrected flow data to obtain toxin dilution data; Step S37: Perform toxin diffusion simulation based on toxin dilution data to obtain algal toxin diffusion data; wherein, step S37 includes the following steps: Step S371: Upload the toxin dilution data to the water environment diffusion simulation platform; Step S372: Set the water flow velocity range to 0.05m / s-1.2m / s, the water depth range to 0.5m-10.0m, and the water temperature range to 5°C-35°C; Step S373: Set the cross-sectional bottom roughness n value to 0.015-0.035, and the cross-sectional undulation variation to 0.3m-2.0m; Step S374: Set the simulation time step to 1 min-15 min, the total simulation duration to 6 h-72 h, and the diffusion scale to 50 m-500 m in the local area; Step S375: Run the water convection-diffusion numerical calculation module to obtain algal toxin diffusion data; Step S4: Identify high-concentration toxin retention areas based on algal toxin diffusion data; perform dissolved oxygen anomaly detection based on high-concentration toxin retention areas to obtain dissolved oxygen anomaly data; and conduct reservoir ecological water release scheduling based on dissolved oxygen anomaly data to obtain reservoir ecological water release data.

2. The reservoir ecological scheduling method considering dynamic water quality response according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain reservoir water quality monitoring data and calculate nutrient concentration to obtain nutrient concentration data; Step S12: Perform algal growth simulation based on nutrient concentration data to obtain algal growth data; calculate algal biomass based on algal growth data to obtain algal biomass data; Step S13: Identify waters with high algal biomass based on algal biomass data; Step S14: Conduct water turbidity testing on water bodies with high algal biomass to obtain water turbidity data; Step S15: Calculate chlorophyll content based on water turbidity data to obtain chlorophyll concentration data; conduct eutrophication assessment based on chlorophyll concentration data and nutrient concentration data to obtain water eutrophication data. Step S16: Based on the eutrophication data of the water body, trace the pollution sources to obtain pollution source data.

3. The reservoir ecological scheduling method considering dynamic water quality response according to claim 2, characterized in that, Step S16 includes the following steps: Step S161: Identify areas with high nutrient loads based on eutrophication data of water bodies; Step S162: Detect nitrogen and phosphorus content in areas with high nutrient load to obtain nitrogen and phosphorus content data; Step S163: Conduct wastewater outlet pollution detection based on nitrogen and phosphorus content data to obtain wastewater outlet pollution data; Step S164: Match pollution sources based on the pollution data of the drainage outlet to obtain pollution source data.

4. The reservoir ecological scheduling method considering dynamic water quality response according to claim 1, characterized in that, Step S2, which assesses pollutant emission intensity based on pollution source data, includes: Pollution types are classified based on pollution source data to obtain point source pollution data and area source pollution data. Ammonia nitrogen pollution data was obtained by detecting ammonia nitrogen pollution based on point source pollution data; Phosphorus nutrient concentration data were obtained by detecting phosphorus nutrient concentration based on non-point source pollution data. Calculate ammonia nitrogen emission flux based on ammonia nitrogen pollution data; Total phosphorus flux was calculated based on phosphorus nutrient concentration data. The intensity of pollutant emissions is assessed based on ammonia nitrogen emission flux and total phosphorus emission flux.

5. The reservoir ecological scheduling method considering dynamic water quality response according to claim 1, characterized in that, Step S2, which involves detecting algae density in highly polluted areas, includes: Algae samples were collected from highly polluted areas, and the algae samples were concentrated by sedimentation to obtain enriched algae suspension data. Based on the data of enriched algal suspensions, glass slide samples were prepared to obtain algal microscopic examination sample data. Algae images were acquired based on algae microscopic sample data; Algal cell morphology is identified from algal images to obtain algal cell morphology data; The number of algal cells is calculated based on algal cell morphology data to obtain algal cell quantity data; Algal concentration data is obtained by converting algal cell count data to unit volume. Algae density data is obtained by detecting algae density based on algae concentration data.

6. The reservoir ecological scheduling method considering dynamic water quality response according to claim 1, characterized in that, Step S2 involves analyzing algal toxins in the water based on algal density data, including: High-density algae areas were identified based on algae density data. Water temperature data were obtained by measuring water temperature in areas with high algae density. The pH of the water body was measured based on areas with high-density algae to obtain pH data. Toxin degradation rate was obtained by analyzing water temperature and pH data. Toxin adsorption was simulated based on the toxin degradation rate to obtain toxin adsorption data. Toxin migration data is obtained by performing toxin adsorption data; The algal toxin content in the water body is assessed based on toxin migration data, and the algal toxin data in the water body is obtained.

7. The reservoir ecological scheduling method considering dynamic water quality response according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Identify areas with high concentrations of toxins based on algal toxin diffusion data; Step S42: Calculate the water flow velocity based on the high concentration of toxin retention area to obtain water flow velocity data; Step S43: Evaluate the water exchange capacity based on the water flow velocity data to obtain water exchange capacity data; identify areas of hydrodynamic stagnation based on the water exchange capacity data; Step S44: Set up dissolved oxygen monitoring sections according to the hydrodynamic stagnation area to obtain dissolved oxygen monitoring data; calculate the oxygen concentration change rate according to the dissolved oxygen monitoring data; determine the dissolved oxygen anomaly detection based on the oxygen concentration change rate to obtain dissolved oxygen anomaly data; Step S45: Based on the abnormal dissolved oxygen data, schedule the ecological water release of the reservoir to obtain the ecological water release data of the reservoir.

8. A reservoir ecological scheduling system considering dynamic water quality response, characterized in that, For implementing the reservoir ecological scheduling method considering dynamic water quality response as described in claim 1, the reservoir ecological scheduling system considering dynamic water quality response includes: The pollution source tracing module is used to acquire reservoir water quality monitoring data; perform eutrophication analysis based on the reservoir water quality monitoring data to obtain eutrophication data; and trace pollution sources based on the eutrophication data to obtain pollution source data. The water body algal toxin analysis module is used to assess pollutant emission intensity based on pollution source data; identify high-pollution areas based on pollutant emission intensity; detect algal density in high-pollution areas to obtain algal density data; and analyze water body algal toxins based on algal density data to obtain water body algal toxin data. The toxin diffusion simulation module is used to acquire reservoir hydrological data, extract maximum rainfall characteristics to obtain maximum rainfall data, perform flow conversion based on maximum rainfall data to obtain flow data, and perform toxin diffusion simulation based on flow data and algal toxin data to obtain algal toxin diffusion data. The reservoir ecological water release scheduling module is used to identify areas with high concentrations of toxins based on algal toxin diffusion data; to detect dissolved oxygen anomalies based on these areas, thus obtaining dissolved oxygen anomaly data; and to schedule reservoir ecological water release based on this data, thus obtaining reservoir ecological water release data.

Images