Nuclear power temperature drainage plankton abnormal aggregation rolling prediction method and system

Abstract

The application discloses a nuclear power temperature drainage plankton abnormal gathering rolling prediction method and system, and belongs to the technical field of marine environment monitoring and prediction. The method comprises the following steps: automatically acquiring multi-source data; inputting the data into a dynamic ecological tightly coupled model for online coupling calculation, and outputting a plurality of plankton functional group biomass original prediction fields; using real-time observation and remote sensing data to correct the original prediction field through optimal interpolation; calculating a plankton community anomaly index based on the corrected analysis field, and generating a risk grade spatial distribution early warning map; automatically publishing and archiving the early warning map and the corrected prediction field as an initial field for the next prediction cycle, and realizing rolling iteration prediction. The application realizes online tight coupling of the heat discharge of the temperature drainage and the marine dynamic ecological process, takes a plurality of plankton functional groups as prediction objects, improves the early warning precision and the operational ability through rolling correction and automatic publishing, and can be used for normal management of nuclear power station cold source safety guarantee.

Description

Technical Field

[0001] This invention relates to the field of marine environmental monitoring and forecasting technology, and in particular to a rolling forecasting method and system for abnormal aggregation of planktonic organisms in the thermal discharge of nuclear power plants. Background Technology

[0002] During nuclear power plant operation, large quantities of seawater are used as cooling water. After heat exchange, the temperature rises, and the water is discharged into adjacent sea areas as warm wastewater. Continuous discharge of warm wastewater can significantly raise the water temperature in localized areas, altering the hydrodynamic environment. This thermal disturbance may induce abnormal proliferation or aggregation of specific planktonic functional groups, such as short-term explosive growth of diatoms and dinoflagellates, or cause small and medium-sized zooplankton to accumulate at the warm wastewater front. These abnormal changes in planktonic community structure may clog the intake filters of the nuclear power plant, affecting the normal use of cooling water and threatening the safety of the cold source. Accurately predicting abnormal changes in planktonic community structure under the influence of warm wastewater is crucial for ensuring the safe operation of nuclear power plants.

[0003] Currently, there are relevant technologies for biological monitoring and forecasting of nuclear power plant cold sources. For example, the Chinese patent "A Method and System for Predicting Marine Species Outbreaks Caused by Nuclear Power Plants" (application number 202211499491.5) predicts the risk of marine species outbreaks caused by nuclear power plants by constructing a hydrodynamic model, combining a food web simulation model and a maximum entropy algorithm. However, the biological model part of this method focuses on the distribution prediction of single species, making it difficult to simulate the different responses of different phytoplankton functional groups (such as diatoms, dinoflagellates, etc.) to thermal and thermal inputs and their interspecific interactions.

[0004] Existing forecasting methods generally have the following shortcomings: (1) Most hydrodynamic models do not include real-time simulation of thermal discharge from nuclear power plant thermal discharge, making it difficult to accurately characterize the temperature anomaly field caused by thermal discharge and its key driving effect on the dynamic changes of phytoplankton functional groups. (2) Even if an ecological model is introduced, its open boundary conditions mostly adopt the static field of climate state, lacking the forecast of the dynamic ecological element boundary in the future period, thus limiting the forecast accuracy. (3) Data transmission between the hydrodynamic module and the ecological module is mostly unidirectional and offline, which cannot simulate the real-time feedback of changes in ecological processes on the physical environment (such as underwater light field and heat budget), resulting in a disconnect between the physical field and the ecological response; (4) Existing solutions are mostly limited to scientific research simulation or single forecast stage, and do not have the rolling iteration capabilities of 24 / 7 unattended data acquisition, automatic model operation, intelligent result correction and automatic product release, which makes it difficult to support the normalized business management of nuclear power plant cold source safety.

[0005] In summary, there is an urgent need for a forecasting method and system that can tightly couple real-time thermal discharge of warm wastewater with marine dynamic ecological models online, use multiple planktonic functional groups as forecasting targets, and integrate observation correction and rolling operational capabilities. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a rolling forecasting method and system for abnormal phytoplankton aggregation in nuclear power plant thermal discharge, aiming to achieve online tight coupling of real-time thermal discharge with marine dynamic ecological models, dynamic forecasting of biomass of multiple phytoplankton functional groups, and improved early warning accuracy and operational capabilities through rolling corrections and automatic releases.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows: A rolling forecasting method for abnormal phytoplankton aggregation in thermal wastewater from nuclear power plants includes the following steps: S1: Automatically acquire and preprocess multi-source data, including meteorological forcing fields, nuclear power plant thermal discharge source strength, large-scale marine physical background boundary, large-scale marine ecological background boundary, previous day's forecast initial field, and real-time observation and remote sensing data; S2: Input the multi-source data obtained in S1 into the pre-constructed dynamic-ecological tightly coupled model. The model performs real-time bidirectional data exchange between the dynamic module and the ecological module in each calculation time step to simulate the dynamic driving of thermal discharge and heat discharge on the biomass of multiple planktonic functional groups and output the original forecast field containing the biomass and nutrient concentration of multiple planktonic functional groups. S3: Using the real-time observation and remote sensing data obtained in S1, perform optimal interpolation correction on the original forecast field output by S2 to obtain the corrected analysis field. S4: Based on the analysis field corrected in S3, the biomass forecast field of phytoplankton functional groups is recalculated through the coupling model. The phytoplankton community anomaly index is calculated based on the degree of deviation of the biomass of each phytoplankton functional group from the historical baseline value, and a spatial distribution early warning map of the risk level of abnormal phytoplankton aggregation induced by warm drainage is generated. S5: Automatically publish the warning map and the corrected forecast field for the day, and archive the corrected forecast field as the initial field for the next forecast cycle to achieve rolling iterative forecasting.

[0008] In the above scheme, the power module solves the three-dimensional primitive equations based on the static approximation and the Boussinesq approximation, and outputs the three-dimensional temperature field, salinity field, velocity field and vertical turbulent diffusion coefficient at each time step.

[0009] In the above scheme, the ecological module solves the biogeochemical processes including phytoplankton photosynthetic growth, zooplankton predation, detritus sedimentation and remineralization, and nutrient regeneration based on the hydrodynamic elements output by the power module, and outputs the biomass and nutrient concentration of each phytoplankton functional group.

[0010] In the above scheme, the power module and the ecological module exchange data in real time through an inter-module coupler: the power module transmits temperature, salinity, flow velocity and turbulence diffusion coefficient to the ecological module, and the ecological module feeds back phytoplankton biomass to the power module. Based on this, the power module dynamically updates the optical parameters of the water body, corrects the vertical attenuation profile of shortwave radiation and heat budget, and forms a closed-loop feedback.

[0011] In the above scheme, the large-scale ocean physical background boundary is provided by the HYCOM global forecast system, and the large-scale ocean ecological background boundary is provided by the CMEMS global biogeochemical analysis and forecast product. The two serve as the open boundary conditions for the dynamic module and the ecological module, respectively.

[0012] In the above scheme, the nuclear power plant temperature and wastewater discharge source strength includes discharge flow rate, discharge temperature and discharge outlet location. In each time step calculation of the power module, it is added to the control equation of the grid where the discharge outlet is located in the form of momentum source term and temperature and salinity source term.

[0013] In the above scheme, the optimal interpolation correction is performed according to the following formula: ; in, This provides a forecast field for temperature, salinity, nutrients, dissolved oxygen, and chlorophyll output from the model. The corrected analysis field, Number of effective observation stations For the k-th observation, It is a bilinear interpolation operator. This represents the global damping gain coefficient. It is an isotropic Gaussian spatial correlation function, and the correction range is limited to the influence radius around the observation point.

[0014] In the above scheme, the ecological module adopts a temperature-dependent growth function. As a temperature correction factor for various ecological elements; ; in, This refers to the current temperature transmitted by the power module at each time step.

[0015] In the above scheme, the ecological module uses the following light attenuation parameterization scheme to calculate the underwater photosynthetically effective radiation when calculating the photosynthetic growth of phytoplankton. : ; in, Solar radiation incident on the sea surface. The percentage of photosynthetically active radiation. Vertical depth coordinates The attenuation coefficient of pure water is... This is a salinity-related attenuation term. The chlorophyll ratio attenuation coefficient is... , , The values ​​represent the chlorophyll concentrations of different phytoplankton functional groups. This refers to the local water depth. To characterize the light attenuation contribution term of nearshore water turbidity as a function of water depth, it is smoothly constructed using three sigmoid functions: .

[0016] A rolling forecasting system for abnormal phytoplankton aggregation in thermal drainage from nuclear power plants includes: The multi-source data aggregation module is used to automatically acquire and preprocess multi-source data, which includes meteorological forcing fields, nuclear power plant thermal wastewater discharge source strength, large-scale marine physical background boundaries, large-scale marine ecological background boundaries, the previous day's forecast initial field, and real-time observation and remote sensing data. The dynamic-ecological tightly coupled model operation module is used to input the multi-source data into the pre-constructed dynamic-ecological tightly coupled model. The model performs real-time bidirectional data exchange between the dynamic module and the ecological module in each calculation time step, simulates the dynamic driving of thermal discharge and heat discharge on the biomass of multiple planktonic functional groups, and outputs the original forecast field containing the biomass and nutrient concentration of multiple planktonic functional groups. The optimal interpolation correction module is used to perform optimal interpolation correction on the original forecast field using the real-time observation and remote sensing data to obtain the corrected analysis field. The abnormal aggregation early warning level generation module is used to recalculate the biomass forecast field of phytoplankton functional groups based on the corrected analysis field and the coupling model, calculate the phytoplankton community anomaly index based on the degree of deviation of the biomass of each phytoplankton functional group from the historical baseline value, and generate a spatial distribution early warning map of the risk level of abnormal aggregation of phytoplankton induced by warm drainage. The rolling optimization and product release module is used to automatically release the warning map and the corrected forecast field for the day, and archive the corrected forecast field as the initial field for the next forecast period, so as to realize rolling iterative forecasting.

[0017] Through the above technical solution, the present invention provides a method and system for predicting the abnormal aggregation and rolling of planktonic organisms in the thermal drainage of nuclear power plants, which has the following beneficial effects: 1. Achieved close online coupling of thermal discharge and marine dynamic-ecological processes. This invention enables real-time bidirectional data exchange between the power module and the ecological module in each computation time step through an inter-module coupler: the power module transmits temperature, salinity, flow velocity, and turbulence diffusion coefficient to the ecological module to drive ecological calculations, and the ecological module feeds back phytoplankton biomass to the power module to dynamically update water optical parameters and correct heat balance, forming a complete closed-loop feedback, which solves the technical problem of the disconnect between physical environment and ecological response.

[0018] 2. By using multiple planktonic functional groups as forecast targets, it overcomes the limitations of single-species forecasting. The forecast targets of this invention cover multiple phytoplankton functional groups such as diatoms, dinoflagellates, and coccolithophores, as well as small and medium-sized zooplankton functional groups. It can more comprehensively capture the complex impact of warm water discharge on the phytoplankton community structure, and the scientific connotation of the early warning is richer.

[0019] 3. The introduction of dynamic ecological boundaries overcomes the accuracy limitations of static boundary conditions. This invention uses large-scale ocean physical background boundary (HYCOM global forecast system) and large-scale ocean ecological background boundary (CMEMS global biogeochemical analysis and forecast product) as dynamic open boundary conditions to reflect the physical and ecological change trends of large-scale sea areas in real time, significantly improving the accuracy of regional forecasts.

[0020] 4. A full-chain operational forecasting system was constructed, consisting of "tightly coupled physical mechanisms + optimal interpolation correction + rolling iteration". This invention forms a complete closed loop of "forecast-correction-archiving-re-forecasting" by automatically acquiring multi-source data, automatically running the model, performing optimal interpolation correction based on measured data, automatically releasing early warning products, and archiving and rolling iterative forecasting fields. This can continuously provide decision support for the safety of cold sources in nuclear power plants, and the forecast accuracy is continuously improved with operational iterations.

[0021] 5. An ecological parameterization scheme customized for nuclear power plant thermal wastewater scenarios was adopted, improving forecast accuracy. This invention introduces a temperature-dependent growth function into the ecological module to accurately characterize the nonlinear promoting effect of warm water discharge on phytoplankton growth. At the same time, it adopts a refined light attenuation parameterization scheme, and uses three sigmoid functions to smoothly construct the attenuation contribution of nearshore water turbidity as water depth changes, so as to more realistically simulate the underwater light field distribution and its impact on phytoplankton photosynthetic growth. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0023] Figure 1 This is a schematic diagram of a method for predicting the abnormal aggregation and rolling of planktonic organisms in the thermal drainage of nuclear power plants, as disclosed in an embodiment of the present invention.

[0024] Figure 2 This is a schematic diagram of the composition of the nuclear power plant thermal drainage plankton abnormal aggregation and rolling forecasting system disclosed in the embodiments of the present invention. Detailed Implementation

[0025] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0026] This invention provides a method and system for predicting abnormal aggregation and rolling of planktonic organisms in the thermal drainage of nuclear power plants. Specific embodiments are as follows: This embodiment uses a coastal nuclear power plant as an application scenario, with the intake and discharge outlet approximately 1.8 kilometers apart, and a designed discharge flow rate of approximately 75 m³ / h. 3 / s, temperature rise of approximately 7-10℃. The target forecast lead time is 96 hours.

[0027] The phytoplankton functional groups are set as follows: the phytoplankton functional group includes diatoms, dinoflagellates, and coccoliths; the zooplankton functional group includes small and medium-sized zooplankton. These functional groups are all common categories in marine ecological models.

[0028] The forecasting system in this embodiment is deployed on a high-performance computing cluster, with Linux as the operating system and Slurm job scheduling system. The entire process is automatically triggered at 00:00 UTC every day.

[0029] Model Architecture This forecasting system employs a two-layer, unidirectional nested architecture: the large-area model has a horizontal resolution of approximately 5 km, covering the coastal waters of China; the inner-area refined model has a horizontal resolution of approximately 100 m, covering an area of ​​approximately 10 km × 10 km around the nuclear power plant. The large-area model and the inner-area refined model are the same model (the dynamic and ecological modules have identical structures), differing only in resolution and coverage. The large-area model's boundary is driven by global products from HYCOM and CMEMS, while the inner-area model uses dynamic elements such as temperature, salinity, and current field, as well as ecological elements such as nutrients and phytoplankton, output from the large-area model, as its boundary field.

[0030] Method implementation, for example Figure 1 As shown, it includes the following steps: S1: Automatic Acquisition and Preprocessing of Multi-Source Data Write an automated script to retrieve the following data in parallel: (1) Coastline topographic data: The coastline and topography of the target sea area were extracted from electronic nautical charts and measured water depth data, and the model grid file was generated by interpolation, with 20 layers in the vertical direction; (2) Meteorological forcing field in the future period: Automatically download the data of the next 96 hours from the GFS global forecast system (10m wind field at sea surface, 2m air temperature, air pressure, relative humidity, shortwave radiation, longwave radiation, precipitation rate) and process it into NetCDF format; (3) Nuclear power plant thermal discharge source strength data: The total cooling water discharge flow rate Q (m³) is obtained from the nuclear power plant operation database. 3 / s), drainage temperature T (°C), and precise coordinates of the drainage outlet; (4) Large-scale ocean temperature-salinity current background boundary field: Download the three-dimensional daily temperature-salinity current field from HYCOM and interpolate to generate the physical open boundary conditions of the large-area model; (5) Large-scale marine ecological background boundary field: Download global biogeochemical analysis and forecast products from CMEMS, extract the three-dimensional analysis field and 10-day forecast field of variables such as nitrate, phosphate, silicate, dissolved oxygen, and chlorophyll, and interpolate to generate the ecological open boundary conditions of the large-area model; (6) Real-time observation and remote sensing data: Access data from ship monitoring stations around the nuclear power plant, including measured water temperature, salinity, nitrate, phosphate, dissolved oxygen, and phytoplankton biomass, as well as sea surface temperature (SST) products retrieved from satellites. Real-time observation and remote sensing data are used for corrections in subsequent S3 steps; (7) Initial field file: Read the initial field file generated by the previous day's forecast archive (the climatological initial field is replaced during the first run), and input it into the dynamic module and the ecological module respectively to provide the three-dimensional initial state of physical and ecological variables for the model hot start.

[0031] S2: Automated operation of the dynamic-ecological tightly coupled model Automatic triggering at 00:00 UTC daily. Multi-source data acquired by S1 is modularly allocated: meteorological forcing field and nuclear power plant thermal discharge source strength data are input into the dynamics module as the forcing field; large-scale ocean thermal-salinity-current background boundary field is input into the dynamics module as the physical boundary condition; large-scale ocean ecological background boundary field is input into the ecological module as the ecological boundary condition; initial field files are input into the dynamics and ecological modules respectively, providing the three-dimensional initial states of physical and ecological variables for model hot start.

[0032] S2.1: Solution of hydrodynamic processes in the power module Based on the input meteorological forcing field, nuclear power plant thermal wastewater discharge source strength, physical boundary conditions, and initial field, the dynamic module uses the three-dimensional primitive equations of the free surface and terrain-following coordinate system under the static approximation and Boussinesq approximation to solve the Reynolds-averaged Navier-Stokes equations. In each time step, it solves the spatiotemporal evolution of the three-dimensional flow field, temperature field, and salinity field, and outputs the three-dimensional temperature field, salinity field, velocity field, and vertical turbulent diffusion coefficient for each time step.

[0033] S2.2: Solving the biogeochemical processes in the ecological module The ecological module, based on the hydrodynamic element field input from the dynamic module, the large-scale marine ecological background boundary conditions input from S1, and the initial ecological field, calculates the biogeochemical cycle processes of ecological state variables such as nitrate, ammonium salt, phosphate, silicate, dissolved oxygen, and multiple phytoplankton functional groups, zooplankton, and detritus under given physical environmental constraints. The processes include at least the photosynthetic growth, respiratory metabolism, and natural death of phytoplankton, predation and feeding of zooplankton, detritus sedimentation and remineralization, and nutrient regeneration cycle of phytoplankton. The output is a preliminary ecological element field containing the biomass and nutrient concentrations of multiple phytoplankton functional groups at each time step.

[0034] The ecological dynamics governing equations for the ecological module are as follows: ; In the formula: Let i be the i-th ecological variable; , , These are the eastward, northward, and vertical velocity components, respectively; The horizontal diffusion coefficient is... This represents the vertical turbulent diffusion coefficient. The left-hand side of the equation represents the changes of various ecological variables over time. The terms on the right-hand side correspond to the following in order: the first term represents advection transport in the physical process, the second term represents turbulent diffusion in the horizontal direction, the third term represents turbulent diffusion in the vertical direction, and the last term... This represents the source-sink biological processes between ecological variables.

[0035] The ecological module adopts a temperature-dependent growth function. As a temperature correction factor for each ecological element, T is the current temperature (°C) transmitted by the power module at each time step.

[0036] The underwater photosynthetically active radiation I is calculated using the following light attenuation parameterization scheme: ; in, Solar radiation incident on the sea surface. The value represents the percentage of photosynthetically active radiation, and z represents the vertical depth coordinate. The attenuation coefficient of pure water is... This is a salinity-related attenuation term. The chlorophyll ratio attenuation coefficient is... , , , representing the chlorophyll concentrations of diatoms, dinoflagellates, and coccolithophores, respectively, and h represents the local water depth (negative value). To characterize the light attenuation contribution term of nearshore water turbidity as a function of water depth, it is smoothly constructed using three sigmoid functions: ; ; ; .

[0037] S2.3: The tight coupling mechanism between the power module and the ecological module The power module and the ecological module exchange data in real time and bidirectionally at each computation time step through an inter-module coupler, forming a tight coupling: Data transfer from the dynamics module to the ecological module: After completing the hydrodynamic solution at each time step, the dynamics module directly transmits the current three-dimensional temperature field, salinity field, velocity field, and vertical turbulent diffusion coefficient to the ecological module as the environmental driving conditions for ecological process calculation.

[0038] Data feedback from the ecological module to the dynamic module: After completing the ecological state solution, the ecological module feeds back the calculated phytoplankton biomass to the dynamic module; the dynamic module dynamically updates the inherent optical parameters of the water body, such as the light absorption coefficient and light scattering coefficient, based on the fed-back phytoplankton carbon biomass, corrects the vertical attenuation profile and heat budget of shortwave radiation, and thus affects the vertical temperature structure and density circulation of the water body, realizing real-time closed-loop feedback of ecological processes to hydrodynamic processes.

[0039] Coupler Synchronization: The bidirectional data exchange described above is executed serially within each coupling time step, ensuring strict synchronization of physical field and ecological state variables over time integration.

[0040] S2.4: Source term treatment of nuclear power plant thermal wastewater discharge In each time step of the power module, the source strength data of nuclear power plant thermal discharge is directly added to the right-hand side of the control equation corresponding to the spatial grid where the discharge outlet is located in the form of momentum source terms and temperature and salinity source terms. This enables synchronous coupling calculation of thermal discharge momentum injection, heat injection, and freshwater injection with the hydrodynamic processes of the surrounding water body.

[0041] The model starts from 00:00 UTC every day, uses 128 CPU cores for parallel computing, and completes a 96-hour forecast in about 2.5 hours. It outputs hourly three-dimensional temperature, salinity, flow field, and raw forecast fields of nitrate, phosphate, ammonium, silicate, as well as biomass of diatoms, dinoflagellates, coccolithophores, small zooplankton, and medium-sized zooplankton.

[0042] S3: Initial Field Correction After S2 completes the daily forecast, the optimal interpolation correction is triggered immediately.

[0043] The corrected equation is: ; in, It is a forecast field (3D grid) of temperature, salinity, nutrients, dissolved oxygen and chlorophyll output by the model. This is the corrected analysis field; This refers to the number of effective observation stations; It is the kth observation; It is a bilinear interpolation operator that maps the observation location to the background field. This is the interpolation of the background field at the observation point; It is the global damping gain coefficient, used to avoid excessive correction of the entire field by sparse observations. The spatial correlation function is characterized by isotropic Gaussian decay. In the formula L represents the horizontal distance (in km) from the grid point to the observation station, where L is the relevant scale. This applies only to distances smaller than the radius of influence. Corrections are applied to the grid points.

[0044] S4: Bioaccumulation Warning Level Generated Based on the analysis field corrected by S3, the biomass forecast field of planktonic functional groups is calculated by the S2 coupling model. The planktonic community anomaly index is calculated based on the degree of deviation of the biomass of each planktonic functional group from the historical baseline value, and a spatial distribution early warning map of the risk level of abnormal planktonic aggregation induced by nuclear power plant thermal discharge is generated accordingly.

[0045] Phytoplankton biomass (the sum of planktonic flora and fauna) risk threshold: Based on historical ecological baseline survey data of a nuclear power plant area, a threshold for phytoplankton biomass within five kilometers of the intake and discharge outlet is set. When it deviates from this baseline range, different levels of anomaly warnings are triggered. The risk levels are divided into at least three categories: blue (low risk), yellow (medium risk), and red (high risk), as shown in Table 1. Table 1. Classification and Countermeasures

[0046] S5: Product Launch and Rolling Optimization The system automatically publishes the daily warning map and the corrected forecast field to the nuclear power plant's early warning platform, and stores the corrected 3D ocean forecast field as the initial field for the next day. The system forms a closed loop of "forecast-correction-archiving-re-forecasting," achieving a 99.3% success rate without human intervention after 90 days of continuous operation. During the trial operation in the summer of 2024, it successfully issued warnings for two rapid diatom proliferation events, providing a window of more than 24 hours for the nuclear power plant to take filter protection measures in advance.

[0047] System Implementation Examples Corresponding to the above method embodiments, the present invention also provides a nuclear power plant thermal drainage plankton anomalous aggregation and rolling forecasting system, such as... Figure 2 As shown, it includes: The multi-source data aggregation module is used to perform automatic data acquisition and preprocessing for S1. Specifically, it includes a shoreline topography data acquisition submodule, a meteorological forcing field acquisition submodule, a thermal drainage source strength data acquisition submodule, a physical background boundary acquisition submodule, an ecological background boundary acquisition submodule, and a real-time observation and remote sensing data acquisition submodule. Each submodule runs in parallel, and the preprocessed data is stored in a shared storage space for downstream modules to access.

[0048] The dynamic-ecological tightly coupled model operation module is used to perform online dynamic-ecological coupled calculations driven by multi-source data in S2, and output the forecast field in the initial field file format. This module specifically includes: a dynamic submodule for solving the hydrodynamic processes in S2.1; an ecological submodule for solving the biogeochemical processes in S2.2; a coupler submodule for performing real-time bidirectional data exchange between the dynamic and ecological modules in S2.3; and a source term processing submodule for processing the nuclear power plant thermal wastewater discharge source terms in S2.4.

[0049] The optimal interpolation correction module is used to perform optimal interpolation correction of temperature, salinity, nutrients, dissolved oxygen, and chlorophyll in the output initial field file by S3.

[0050] The abnormal clustering early warning level generation module is used to perform risk level calculation and early warning map generation for S4.

[0051] The Rolling Optimization and Product Release module is used to perform S5 forecast field archiving and automatic product release.

[0052] System workflow: At 00:00 UTC daily, the multi-source data aggregation module starts, acquiring all input data in parallel and preprocessing it; subsequently, the dynamic-ecological tightly coupled model operation module starts, executing 96-hour online coupled forecasts and outputting the raw forecast field; the optimal interpolation correction module starts, correcting the raw forecast field using measured data; the anomaly clustering early warning level generation module starts, generating a risk level early warning map; and the rolling optimization and product release module starts, automatically releasing and archiving the early warning map and the corrected forecast field, preparing for the next day's forecast. The entire process requires no manual intervention.

[0053] The above description of the disclosed embodiments enables those skilled in the art to make or use 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 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 disclosed herein.

Claims

1. A method for predicting the abnormal aggregation and rolling of planktonic organisms in the thermal drainage of nuclear power plants, characterized in that, Includes the following steps: S1: Automatically acquire and preprocess multi-source data, including meteorological forcing fields, nuclear power plant thermal discharge source strength, large-scale marine physical background boundaries, large-scale marine ecological background boundaries, the previous day's forecast initial field, and real-time observation and remote sensing data; S2: Input the multi-source data obtained in S1 into the pre-constructed dynamic-ecological tightly coupled model. The model performs real-time bidirectional data exchange between the dynamic module and the ecological module in each calculation time step to simulate the dynamic driving of thermal discharge and heat discharge on the biomass of multiple planktonic functional groups and output the original forecast field containing the biomass and nutrient concentration of multiple planktonic functional groups. S3: Using the real-time observation and remote sensing data obtained in S1, perform optimal interpolation correction on the original forecast field output by S2 to obtain the corrected analysis field. S4: Based on the analysis field corrected by S3, the biomass forecast field of phytoplankton functional groups is recalculated through the coupling model. The phytoplankton community anomaly index is calculated based on the degree of deviation of the biomass of each phytoplankton functional group from the historical baseline value, and a spatial distribution early warning map of the risk level of abnormal phytoplankton aggregation induced by warm drainage is generated. S5: Automatically publish the warning map and the corrected forecast field for the day, and archive the corrected forecast field as the initial field for the next forecast cycle to achieve rolling iterative forecasting.

2. The method according to claim 1, characterized in that, The power module solves the three-dimensional primitive equations based on the static approximation and the Boussinesq approximation, and outputs the three-dimensional temperature field, salinity field, velocity field and vertical turbulent diffusion coefficient at each time step.

3. The method according to claim 1, characterized in that, The ecological module, based on the hydrodynamic elements output by the power module, solves the biogeochemical processes including phytoplankton photosynthetic growth, zooplankton predation, detritus sedimentation and remineralization, and nutrient regeneration, and outputs the biomass and nutrient concentration of each phytoplankton functional group.

4. The method according to claim 1, characterized in that, The power module and the ecological module exchange data in real time through an inter-module coupler: the power module transmits temperature, salinity, flow velocity and turbulence diffusion coefficient to the ecological module, and the ecological module feeds back phytoplankton biomass to the power module. Based on this, the power module dynamically updates the optical parameters of the water body, corrects the vertical attenuation profile of shortwave radiation and heat balance, and forms a closed-loop feedback.

5. The method according to claim 1, characterized in that, The large-scale ocean physical background boundary is provided by the HYCOM global forecast system, and the large-scale ocean ecological background boundary is provided by the CMEMS global biogeochemical analysis and forecast product. The two serve as the open boundary conditions for the dynamic module and the ecological module, respectively.

6. The method according to claim 1, characterized in that, The nuclear power plant's thermal wastewater discharge source strength includes discharge flow rate, discharge temperature, and discharge outlet location. In each time step calculation of the power module, it is added to the control equation of the grid where the discharge outlet is located in the form of momentum source term and temperature-salinity source term.

7. The method according to claim 1, characterized in that, The optimal interpolation correction is performed according to the following formula: ； in, This provides a forecast field for temperature, salinity, nutrients, dissolved oxygen, and chlorophyll output from the model. The corrected analysis field, Number of effective observation stations For the k-th observation, It is a bilinear interpolation operator. This represents the global damping gain coefficient. It is an isotropic Gaussian spatial correlation function, and the correction range is limited to the influence radius around the observation point.

8. The method according to claim 1, characterized in that, The ecological module adopts a temperature-dependent growth function. As a temperature correction factor for various ecological elements; ； in, This refers to the current temperature transmitted by the power module at each time step.

9. The method according to claim 3, characterized in that, When calculating the photosynthetic growth of phytoplankton, the ecological module uses the following light attenuation parameterization scheme to calculate the underwater photosynthetically effective radiation. : ； in, Solar radiation incident on the sea surface. The percentage of photosynthetically active radiation. Vertical depth coordinates The attenuation coefficient of pure water is... This is a salinity-related attenuation term. The chlorophyll ratio attenuation coefficient is... , , The values ​​represent the chlorophyll concentrations of different phytoplankton functional groups. This refers to the local water depth. To characterize the light attenuation contribution term of nearshore water turbidity as a function of water depth, it is smoothly constructed using three sigmoid functions: 。 10. A rolling forecasting system for abnormal phytoplankton aggregation in nuclear power plant thermal drainage, characterized in that, include: The multi-source data aggregation module is used to automatically acquire and preprocess multi-source data, which includes meteorological forcing fields, nuclear power plant thermal wastewater discharge source strength, large-scale marine physical background boundaries, large-scale marine ecological background boundaries, the previous day's forecast initial field, and real-time observation and remote sensing data. The dynamic-ecological tightly coupled model operation module is used to input the multi-source data into the pre-constructed dynamic-ecological tightly coupled model. The model performs real-time bidirectional data exchange between the dynamic module and the ecological module in each calculation time step, simulates the dynamic driving of thermal discharge and heat discharge on the biomass of multiple planktonic functional groups, and outputs the original forecast field containing the biomass and nutrient concentration of multiple planktonic functional groups. The optimal interpolation correction module is used to perform optimal interpolation correction on the original forecast field using the real-time observation and remote sensing data to obtain the corrected analysis field. The abnormal aggregation early warning level generation module is used to recalculate the biomass forecast field of phytoplankton functional groups based on the corrected analysis field and the coupling model, calculate the phytoplankton community anomaly index based on the degree of deviation of the biomass of each phytoplankton functional group from the historical baseline value, and generate a spatial distribution early warning map of the risk level of abnormal aggregation of phytoplankton induced by warm drainage. The rolling optimization and product release module is used to automatically release the warning map and the corrected forecast field for the day, and archive the corrected forecast field as the initial field for the next forecast period, so as to realize rolling iterative forecasting.

Images