A stripping method after the non - linear superposition effect of the warm wastewater discharge adjacent to a nuclear power plant

Abstract

The present invention discloses a stripping method after the non-linear superposition effect of adjacent nuclear power plant's warm wastewater discharge, including: collecting the water intake and discharge information, hydro-meteorological and topographic data of the research power plant to form a basic data set; establishing and calibrating the hydrodynamic and thermal mathematical model of the research water area to obtain a calibrated model; using the calibrated model to carry out the simulation of warm wastewater discharge during the operation of all units to obtain the full-field temperature rise field data; after shutting down the concerned unit, carrying out the simulation based on the same calibrated model to obtain the half-field temperature rise field data; performing non-linear stripping processing on the full-field temperature rise field data and the half-field temperature rise field data to obtain the temperature rise field data of the concerned unit; and conducting statistical analysis on the temperature rise field data of the concerned unit to obtain different temperature rise ranges and corresponding areas. The present invention solves the problems of dealing with the non-linear superposition effect and thermal cycle cumulative effect of warm wastewater discharge, and improves the stripping accuracy and result rationality of warm wastewater discharge.

Description

Technical Field

[0001] The present invention belongs to the field of environmental science and engineering, and in particular, to a stripping method after the non-linear superposition effect of warm wastewater discharged from adjacent nuclear and thermal power plants. Background Art

[0002] Nuclear power and thermal power are important components of China's energy system. With the advancement of policies, the number of adjacent nuclear and thermal power plant sites is gradually increasing, and the distance between adjacent sites is getting shorter. After cooling by taking water from natural water bodies through the water intake, nuclear and thermal power plants discharge "warm wastewater" at a temperature higher than that of the natural water body, and this warm wastewater will have an impact on the water ecosystem and environment. The impact range of warm wastewater is an important indicator that the ecological environment management department and the natural resources management department focus on. The specification requires that the prediction deviation of the temperature rise range shall not exceed 5%. How to accurately strip the impact range of warm wastewater from each unit under the condition of adjacent nuclear and thermal power plants is of great significance for the preliminary demonstration and operation supervision of the project.

[0003] At present, there is a relatively mature research foundation for warm wastewater simulation and evaluation technologies. The traditional hydrodynamic numerical model method can better simulate the diffusion process of warm wastewater under single or multiple drainage outlet conditions. This type of method usually uses the control equations in the Euler coordinate system to describe the transport process of warm wastewater, and uses the finite difference or finite volume method for numerical solution. In terms of time step processing, most adopt the fixed time step strategy, and grid encryption is used near the drainage outlet to improve the accuracy. For the change of boundary conditions, linear interpolation or step change methods are commonly used. For the multi-drainage outlet scenario, researchers usually use the linear superposition principle to evaluate the comprehensive impact of warm wastewater from multiple outlets, or independently simulate the impact range of a single outlet.

[0004] However, the existing technical methods have obvious deficiencies in dealing with the stripping problem after the non-linear superposition of warm wastewater discharged from adjacent nuclear and thermal power plants. First of all, the traditional linear stripping method ignores the non-linear superposition effect of warm wastewater in the complex flow field. Especially in the high-temperature gradient region, the buoyancy effect caused by temperature will change the local flow field structure, resulting in the impact of warm wastewater from multiple units not being a simple linear superposition. Secondly, the conventional fixed time step strategy cannot simultaneously meet the warm wastewater simulation accuracy requirements in different stages of the tidal cycle. Especially at critical moments such as the tide conversion period, numerical oscillations or insufficient accuracy often occur. In addition, when the water intake is within the impact range of warm wastewater, a thermal cycle effect of drainage - water intake - re-drainage will be formed, and this heat accumulation process follows complex tide control laws. The existing methods lack effective means to accurately strip it. These technical deficiencies lead to significant deviations in the stripping of the non-linear superposition effect of warm wastewater discharged from adjacent nuclear and thermal power plants, and it is difficult to meet the requirements of refined management. Summary of the Invention

[0005] Objective of the Invention: To provide a method for stripping the non - linear superposition effect of the warm water discharge from an adjacent nuclear - thermal power plant, aiming to solve at least one technical problem existing in the prior art.

[0006] Technical Solution: A method for stripping the non - linear superposition effect of the warm water discharge from an adjacent nuclear - thermal power plant includes:

[0007] Collect the intake and discharge information, hydro - meteorological and topographic data of the power plant under study to form a basic data set;

[0008] Based on the basic data set, establish and calibrate the hydrodynamic and thermal mathematical model of the study water area to obtain a calibrated model;

[0009] Use the calibrated model to conduct the simulation of the warm water discharge during the operation of all units to obtain the full - field temperature rise data;

[0010] After shutting down the unit of interest, conduct the simulation based on the same calibrated model to obtain the half - field temperature rise data;

[0011] Conduct non - linear stripping processing on the full - field temperature rise data and the half - field temperature rise data to obtain the temperature rise data of the unit of interest;

[0012] Conduct statistical analysis on the temperature rise data of the unit of interest to obtain the different temperature rise ranges and corresponding areas, as well as the temperature rise at the intake water temperature rise or sensitive positions of the unit.

[0013] Beneficial Effects: The present invention solves the non - linear superposition effect of the warm water discharge, avoids the numerical oscillation of the conventional interpolation method in the violently changing area, accurately strips the direct warm water discharge effect and the thermal cycle cumulative effect, and improves the accuracy of warm water discharge stripping. Description of the Drawings

[0014] Figure 1 It is a step - flow chart of a method for stripping the non - linear superposition effect of the warm water discharge from an adjacent nuclear - thermal power plant provided by an embodiment of the present application.

[0015] Figure 2 It is a step - flow chart of conducting non - linear stripping processing provided by an embodiment of the present application.

[0016] Figure 3 It is a step - flow chart of constructing a streamline coordinate system provided by an embodiment of the present application.

[0017] Figure 4 It is a step - flow chart of performing non - linear stripping in the streamline coordinate system provided by an embodiment of the present application. Detailed Embodiment

[0018] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0019] It should be particularly noted that, for clearly showing the step flow of the present application, serial numbers are marked for each step in the specification. These serial numbers are only for the convenience of description and do not limit the execution order of the steps. In actual operation, according to the technical requirements of the specific implementation scenario, the steps can be executed in an order different from that shown in the specification, and in some cases, parallel processing between steps can also be achieved.

[0020] As Figure 1 shown, a stripping method after the non-linear superposition effect of adjacent nuclear power plant cooling water discharge includes the following steps:

[0021] S1. Collect the intake and discharge water information, hydrometeorological and topographic data of the research power plant to form a basic data set;

[0022] Specifically, the intake and discharge water information includes the structure and location information of the intake and discharge structures and the discharge flow rate and temperature rise data. Among them, the structure and location information of the intake and discharge structures can be parameters such as the coordinates of the intake, the coordinates of the discharge, the shape and size of the intake and discharge structures, and the water depth; the discharge flow rate and temperature rise data can be parameters such as the maximum discharge flow rate under the design condition, the average discharge flow rate under the normal condition, and the discharge temperature rise. The hydrometeorological data includes the hydrological observation data and meteorological data of the research water area. The hydrological observation data can be parameters such as the tide level, flow velocity, flow direction, and water temperature; the meteorological data can be parameters such as the air temperature, wind speed, wind direction, precipitation, and radiation intensity. The topographic data includes the isobath map, water depth point data, etc.

[0023] S2. Establish and calibrate the hydrodynamic and thermal mathematical model of the research water area based on the basic data set to obtain a calibrated model;

[0024] Specifically, to establish a mathematical model, equations are usually used to describe the laws of hydrodynamic (such as the velocity and direction of water flow) and thermal (such as water temperature change) in the water area. By comparing the predicted results of the model with the actual measurement data, the parameters of the model are optimized so that it can more accurately reflect the actual situation of the water area. Finally, a calibrated mathematical model is obtained, so that researchers can use it to simulate and analyze the dynamic changes in the water area, such as predicting the future trends of water flow and temperature.

[0025] S3. Use the calibrated model to carry out the cooling water discharge simulation of all units in operation to obtain the full-field temperature rise field data;

[0026] Specifically, when simulating the operation of the unit, how the warm discharged water diffuses and its influence range and degree can be understood, so as to understand the potential impact of the unit operation on the environment. Through simulation calculation, the temperature distribution data of the entire water area can be obtained, and it can be known which areas will have temperature rise and the degree of temperature rise.

[0027] S4. After shutting down the concerned units, simulate based on the same calibrated model to obtain the half-field temperature rise field data;

[0028] Specifically, stop the operation of some units to reduce the influence of their warm discharged water on the water temperature distribution. Use the calibrated model to calculate the temperature change of the water area in this case, and obtain the temperature rise distribution of a certain part of the water area (rather than the entire field) through simulation.

[0029] S5. Perform non-linear stripping processing on the full-field temperature rise field data and the half-field temperature rise field data to obtain the temperature rise field data of the concerned units;

[0030] Specifically, extract the difference between the full-field data and the half-field data through a non-linear algorithm to obtain the temperature rise field data of the concerned units, that is, analyze the contribution of these units to the water temperature rise separately.

[0031] S6. Perform statistical analysis on the temperature rise field data of the concerned units to obtain different temperature rise ranges and the corresponding areas, as well as the temperature rise at the water intake temperature rise or sensitive positions of the units.

[0032] Specifically, use mathematical and statistical methods to analyze the temperature rise field data caused by the concerned units, and obtain different temperature rise ranges: divide the temperature rise data into several ranges, such as a lower temperature rise range (such as 0.5 °C), a medium temperature rise range (such as 1 - 3 °C), and a higher temperature rise range (such as above 4 °C). Analyze the water area corresponding to each temperature rise range, that is, see how large an area is affected by these temperature changes.

[0033] This embodiment can accurately simulate the temperature influence of the warm discharged water of the nuclear power plant on the water area; through non-linear stripping processing, the non-linear superposition effect of the warm discharged water of multiple units is decomposed, and the independent temperature rise influence of a single or multiple concerned units is clearly extracted; through statistical analysis of the temperature rise field of the concerned units, the influence area of different temperature rise ranges can be quantified, which helps to identify which areas are more affected, so as to formulate more accurate environmental protection or operation adjustment strategies. It solves the problems of dealing with the non-linear superposition effect and the cumulative effect of the thermal cycle of the warm discharged water, and improves the stripping accuracy of the warm discharged water.

[0034] According to one aspect of the present application, the steps of forming the basic data set include:

[0035] S11. Collect and analyze the structure and location information of the water intake and drainage structures of the power plant and surrounding existing power plant units, including parameters such as the water intake coordinates, outlet coordinates, shape and size of the water intake and drainage structures, and water depth, to form a water intake and drainage structure information dataset; collect and analyze the drainage flow and temperature rise data of the power plant and surrounding existing power plant units, including parameters such as the maximum drainage flow under design conditions, the average drainage flow under normal conditions, and the drainage temperature rise, to form a water intake and drainage operating condition dataset;

[0036] S12. Collect multi-year hydrological observation data for the study waters, including parameters such as tide level, flow velocity, flow direction, and water temperature, to form a hydrological observation dataset; collect meteorological data for the study waters, including parameters such as temperature, wind speed, wind direction, precipitation, and radiation intensity, to form a meteorological observation dataset;

[0037] S13. Collect underwater topographic survey data of the study waters, including bathymetric maps and water depth point data, to form an original topographic dataset; digitize the original topographic dataset to create a digitized water depth topographic dataset; inspect the digitized water depth topographic dataset, identify and correct abnormal data points, and obtain a corrected water depth topographic dataset;

[0038] S14. Collect the measured temperature rise field data of existing power plants in operation in the study waters, including the temperature rise distribution in different seasons and under different tidal conditions, to form a measured temperature rise field data set; perform quality control on the measured temperature rise field data set, eliminate abnormal data, and form a corrected temperature rise field data set; extract key features from the corrected temperature rise field data set, including the maximum temperature rise position, temperature rise contour line distribution features, etc., to obtain a temperature rise feature data set.

[0039] According to one aspect of the present application, the step of obtaining a calibration model includes:

[0040] S21. Based on the corrected water depth terrain dataset, determine the scope of the computational domain and generate a computational domain boundary dataset; mesh the computational domain to form an initial grid dataset; locally encrypt the initial grid dataset, set the grid size near the drainage outlet to 20m, and generate an encrypted grid dataset; perform grid quality inspection and optimization on the encrypted grid dataset to generate an optimized grid dataset; interpolate the corrected water depth terrain dataset onto the optimized grid dataset to obtain a grid terrain dataset.

[0041] S22. Based on the hydrological observation dataset, produce the hydrodynamic time series file for the open boundary of the model to form the open boundary condition dataset; based on the meteorological observation dataset, produce the time series files of wind speed and wind direction to form the meteorological boundary condition dataset; set the positions of the intake and drainage outlets of each unit, the intake and drainage volumes, and the drainage temperature rise conditions to generate the intake and drainage boundary condition dataset; set the initial calculation conditions, including parameters such as initial water level, flow velocity, and water temperature, to form the initial condition dataset; construct the boundary condition of the feature mode retention type to generate the smooth boundary condition dataset.

[0042] S23. Set the initial bottom roughness coefficient, hydraulic viscosity coefficient, temperature drainage diffusion coefficient, and overall heat dissipation coefficient of the water surface to form the initial parameter set; based on the initial parameter set, optimize the grid dataset, the smooth boundary condition dataset, and the initial condition dataset, and conduct a preliminary run of the model to generate the preliminary simulation results; compare the preliminary simulation results with the corrected temperature rise field dataset, calculate the correlation coefficient and the root mean square error to form the model evaluation index. Based on the model evaluation index, adjust the model parameters to form the adjusted parameter set; repeat running the model until the model evaluation index meets the requirements to obtain the calibrated parameter set and the calibrated model.

[0043] According to one aspect of the present application, the steps of generating the smooth boundary condition dataset include:

[0044] Based on the basic dataset, analyze the boundary condition data (such as applying the non-linear feature extraction algorithm), extract the key feature modes of the boundary conditions to form the boundary feature mode library;

[0045] Based on the boundary feature mode library, construct the boundary condition reconstruction model to generate the boundary condition reconstruction mode;

[0046] Construct the recurrence relationship of the weight function, optimize the reconstruction coefficient of the boundary condition reconstruction model to ensure the continuity of the change of the boundary conditions;

[0047] Apply the optimized boundary condition reconstruction model to achieve the smooth transition of the boundary conditions and generate the smooth boundary condition data that meets the physical constraints;

[0048] Based on the smooth boundary condition data, establish and calibrate the hydrodynamic and thermal mathematical model of the research water area to obtain the calibrated model.

[0049] Specifically, based on the open boundary condition dataset and the meteorological boundary condition dataset, a non-linear feature extraction algorithm is applied to extract the key feature patterns of the boundary conditions, generating a boundary feature pattern library. The calculation formula is: Mi = Φ(Bi, τi), where Bi is the boundary condition vector, τi is the time scale parameter, and Φ is the non-linear mapping function. The boundary feature pattern library contains k feature patterns {M1, M2,..., Mk}, and each pattern represents a typical boundary condition change pattern. Based on the boundary feature pattern library, a boundary condition reconstruction model is constructed to obtain the boundary condition reconstruction model. The reconstruction formula is: B(t) = Σαi(t)·Mi, where αi(t) is the time-varying weight coefficient. A recurrence relation of the weight function is designed: αi(t+Δt) = αi(t) + Δαi(t), where Δαi(t) is determined by the physical constraint conditions. The boundary condition reconstruction model is applied to achieve the smooth transition of the boundary conditions and generate a smooth boundary condition dataset.

[0050] In an embodiment of the present application, the open boundary condition dataset and the meteorological boundary condition dataset are read, and multi-time scale decomposition is performed using the wavelet transform method to respectively obtain the short-time scale boundary component (T<1h), the medium-time scale boundary component (1h<T<12h), and the long-time scale boundary component (T>12h), forming a multi-scale boundary condition dataset. The calculation formula is: B_multi(t, x) = WT{B(t, x)}, where WT is the wavelet transform operator, and B(t, x) is the original boundary condition. The statistical features of the boundary conditions at each time scale are extracted, including the mean, standard deviation, maximum value, minimum value, etc., forming a boundary statistical feature dataset.

[0051] The multi-scale boundary condition dataset is read, a high-dimensional feature space is constructed, the boundary conditions are mapped into the feature space, and non-linear features are extracted using kernel principal component analysis (KPCA). The calculation formula is: Mi = KPCA(Bi, K), where Bi is the boundary condition vector and K is the kernel function. The kernel function selects the radial basis function (RBF): K(x, y) = exp(-γ||x-y|| 2 ), where γ is the kernel parameter, and the optimal value is determined through cross-validation. The non-linear feature extraction is performed on the boundary conditions of each time scale respectively to obtain the short-time scale feature set, the medium-time scale feature set, and the long-time scale feature set.

[0052] Read the short - term scale feature set, medium - term scale feature set, and long - term scale feature set, and calculate the proportion of the explained variance of each feature; select the feature subset with the cumulative explained variance reaching 95%, and form the optimized short - term feature set, optimized medium - term feature set, and optimized long - term feature set respectively. Conduct a physical meaning analysis on the selected features to ensure that each feature pattern has a clear physical interpretation, and form the feature physical interpretation data set. Integrate the optimized short - term feature set, optimized medium - term feature set, and optimized long - term feature set into a unified boundary feature pattern library, which contains boundary feature patterns {M1, M2,..., Mk} of different time scales.

[0053] Read the boundary feature pattern library, construct the basic framework of the boundary condition reconstruction model, and design the reconstruction formula: B(t, x)= Σαi(t)·Mi(x), where αi(t) is the time - varying weight coefficient. Read the historical boundary condition data, and use the least - squares method to calculate the optimal reconstruction coefficient αi(t) to form the initial reconstruction coefficient set. Introduce physical constraint conditions, including energy conservation, momentum conservation, etc., to construct a constrained optimization problem. Solve to obtain the optimized reconstruction coefficient set that satisfies the physical constraints. Read the optimized reconstruction coefficient set, analyze the variation law of the reconstruction coefficient with time, and design the recurrence relationship of the weight function: αi(t + Δt)= αi(t)+ Δαi(t), where Δαi(t) is determined by the physical constraint conditions, and construct the weight adjustment model: Δαi(t)= f(B(t), ▽B(t), F_ext(t)), where B(t) is the current boundary condition, ▽B(t) is the boundary condition gradient, and F_ext(t) is the external forcing factor; determine the parameters of the weight adjustment model through historical data training to form the weight adjustment model.

[0054] Read the boundary feature pattern library and the weight adjustment model, and construct a complete boundary condition reconstruction model; design a smooth transition function to ensure the continuity of the boundary condition change: B_smooth(t)=(1 - β(t))·B_current(t)+ β(t)·B_target(t); where β(t) is the smoothing factor, which smoothly transitions from 0 to 1, β(t)=3t 2 - 2t 3 , t∈[0, 1]. Apply the boundary condition reconstruction model to smoothly reconstruct the boundary conditions, generate the smooth boundary condition data set, and verify the physical rationality of the reconstructed boundary conditions to ensure that they satisfy basic physical laws such as mass conservation and energy conservation.

[0055] According to one aspect of the present application, the steps of obtaining the full - field temperature rise field data include:

[0056] S31. Based on the hydrometeorological data, analyze the tide level time series, decompose the tidal cycle into the flood tide period, high - flat period, ebb tide period, and low - flat period to form typical tidal phase classification data;

[0057] S32. Calculate the optimal time step for different typical tidal phases based on the typical tidal phase classification data to form a typical phase time step strategy.

[0058] S33. Apply the phase time step strategy to the calibrated model for warm water discharge simulation to ensure high-precision calculations at critical moments of tidal current changes, and generate more accurate full-field temperature rise field data accordingly.

[0059] Specifically, based on the water intake and discharge condition data set, set the water intake and discharge conditions for all units to operate simultaneously to form a full-field water intake and discharge condition data set; load the full-field water intake and discharge condition data set into the calibrated model to form an initial full-field simulation model. Conduct multi-time scale feature-preserving tidal phase decomposition simulation. Based on the hydrological observation data set, decompose the tidal cycle into key phases: flood acceleration period, flood peak period, flood deceleration period, slack period, ebb acceleration period, ebb peak period, ebb deceleration period, and generate a tidal phase classification data set. Develop a tidal phase recognition algorithm to automatically identify the tidal phase characteristics at different positions within the computational domain and generate a tidal phase spatial distribution data set.

[0060] For each key tidal phase in the tidal phase classification data set, design an optimal time step strategy to generate a phase time step strategy. The calculation formula is: Δt(φ) = Δt_base * W(φ), where φ is the tidal phase, W(φ) is the phase weight function, W(φ)≈0.2 during the tidal current transition period, W(φ)≈1.0 during the tidal current stable period, and Δt_base is the base time step. Apply the phase time step strategy to the initial full-field simulation model to form an optimized time step full-field model.

[0061] Run the optimized time step full-field model to generate the original full-field simulation results. Conduct phase feature-preserving interpolation processing on the original full-field simulation results to ensure that the reconstructed temperature rise field maintains physical properties; save the processed full-field simulation results to form a full-field temperature rise field time series data set. Conduct full-field temperature rise feature analysis. Based on the full-field temperature rise field time series data set, calculate the spatio-temporal statistical features of the temperature rise field, including the position of the maximum temperature rise, the range of different temperature rise isotherms, etc., to form a full-field temperature rise feature data set. Analyze the temperature rise distribution characteristics under different tidal phases to generate a full-field tidal phase temperature rise feature data set.

[0062] In an embodiment of the present application, tidal cycle identification and segmentation are performed. The hydrological observation data set is read, and the tidal level time series data is extracted; the Hilbert-Huang transform (HHT) is applied to perform time-frequency analysis on the tidal level series to identify the main tidal wave components and form a tidal wave feature data set. Based on the tidal level derivative and extreme points, the key phases of the tidal cycle are divided: rising tide acceleration period (RA), rising tide extreme period (RP), rising tide deceleration period (RD), slack period (S), falling tide acceleration period (FA), falling tide extreme period (FP), falling tide deceleration period (FD), and the calculation formulas are: φ(t) = RA, if dh / dt > 0 and d 2 h / dt 2 > 0; RP, if dh / dt> 0 and |d 2 h / dt 2 | < ε; RD, if dh / dt > 0 and d 2 h / dt 2 < 0; S, if |dh / dt| < Δ; FA, if dh / dt < 0 and d 2 h / dt 2 < 0; FP, if dh / dt < 0 and |d 2 h / dt 2 | < ε; FD, if dh / dt <0 and d 2 h / dt 2 > 0; where h is the tidal level, ε and Δ are threshold parameters, t is the time parameter, and d is the differential symbol. A tidal phase classification data set is generated, including the time intervals and characteristic parameters of each tidal phase.

[0063] The tidal level data of multiple sites in the tidal phase classification data set and the hydrological observation data set are read, the propagation law of the tidal phase in space is analyzed, the tidal phase propagation speed and direction are calculated, and a tidal phase propagation feature data set is formed; a tidal phase spatial propagation model is constructed: φ(x, t) = φ(x0, t - τ(x, x0)), where τ(x, x0) is the time required for the tidal phase to propagate from position x0 to position x; based on the tidal phase propagation feature data set and the grid terrain data set, the spatio-temporal distribution of the tidal phase at each grid point in the calculation domain is calculated, and a tidal phase spatial distribution data set is formed.

[0064] Read the flow velocity data in the hydrological observation dataset and the tidal phase classification dataset, analyze the flow field characteristics under different tidal phases, including the magnitude of flow velocity, flow direction, shear strength, etc., and form a tidal current characteristic dataset. Identify key flow field structures, such as vortices, fronts, convergence / divergence zones, etc., and generate a key flow field structure dataset. Based on the tidal current characteristic dataset and the key flow field structure dataset, construct typical flow field patterns for different tidal phases and form a tidal phase flow field pattern dataset.

[0065] According to one aspect of the present application, the steps of calculating the optimal time step for different tidal phases to form a phase time step strategy include:

[0066] Based on the tidal phase classification data, analyze the transport characteristics of the thermal discharge water under different tidal phases, calculate the characteristic time scales of each tidal phase, and form phase characteristic time scale data;

[0067] Based on the phase characteristic time scale data, construct a time step calculation formula, determine the optimal time step for different tidal phases and spatial positions, and generate a phase-space time step distribution; apply spatial smoothing processing to the phase-space time step distribution to ensure continuous change of the time step in adjacent grid cells and form a smoothed time step distribution;

[0068] Integrate the smoothed time step distribution to generate a complete phase time step strategy for use in the thermal discharge water simulation process.

[0069] Specifically, read the tidal phase classification dataset and the tidal current characteristic dataset, and analyze the transport characteristics of the thermal discharge water under different tidal phases; calculate the characteristic time scales of the thermal discharge water transport under different tidal phases: τ_phase(φ) = min{L / U(φ), H 2 / D, T_tide / η(φ)}, where L is the characteristic length, U(φ) is the flow velocity at phase φ, H is the water depth, D is the diffusion coefficient, T_tide is the tidal period, and η(φ) is the local topographic change rate at phase φ; generate a dataset of phase characteristic time scales to characterize the influence degree of different tidal phases on the transport of warm water discharge. Read the dataset of phase characteristic time scales and design a time step optimization strategy based on tidal phases. The calculation formula is: Δt(φ, x) = Δt_base * W(φ, x), where W(φ, x) is the weight function of tidal phase φ at position x and is designed as follows: W(φ, x) = 0.2, if φ ∈ {S} or x ∈ high_gradient_area; 0.4, if φ ∈ {RA, FA} or x ∈ medium_gradient_area; 0.7, if φ ∈ {RD, FD} or x ∈ low_gradient_area; 1.0, if φ ∈ {RP, FP} and x ∈ stable_area; where high_gradient_area is the area with large temperature gradient or flow velocity gradient, usually including the area near the drainage outlet; medium_gradient_area is the medium gradient area; low_gradient_area is the low gradient area; stable_area is the stable area; based on the above formula, calculate the optimal time step for each grid point and each tidal phase in the computational domain to form a phase time step strategy. Read the phase time step strategy and detect the spatial discontinuity of the time step; design a spatial smoothing algorithm to ensure that the change of the time step between adjacent grid cells does not exceed a preset threshold: |Δt(x_i) - Δt(x_j)| ≤ γ·min{Δt(x_i), Δt(x_j)}, for adjacent cells x_i and x_j, where γ is the smoothing coefficient, generally taken as 0.2 - 0.3; apply the iterative smoothing algorithm to generate a dataset of spatially continuous smoothed time step distributions. Integrate the dataset of smoothed time step distributions into the phase time step strategy to form the final time step optimization strategy.

[0070] In an embodiment of the present application, the full-field simulation initial model and the phase time-step strategy are read, and the simulation control parameters are set; a phase adaptive simulation controller is developed to dynamically adjust the time step and the numerical solution parameters according to the tidal phase at the current simulation time. The working process of the simulation controller: determine the tidal phase φ(t, x) at the current simulation time t; obtain the corresponding time step Δt(φ, x) according to the phase time-step strategy; set the numerical solver parameters, including the iteration accuracy, relaxation factor, etc.; perform the simulation calculation for one time step; update the model state and prepare for the next time step calculation. Generate a phase adaptive control parameter dataset to guide the simulation process.

[0071] Perform the full-field warm water discharge simulation and dynamically adjust the time step according to the phase adaptive control parameter dataset. For each tidal phase, store the corresponding warm water discharge simulation results respectively to form a tidal phase warm water discharge result set; the storage format is designed as: R(φ, t) = {T(x, y, t) | φ(t) = φ}, for φ ∈ {RA, RP, RD, S, FA, FP, FD}, where T(x, y, t) is the temperature field at the position (x, y) and time t; for the results of each tidal phase, extract the key features to form a tidal phase feature summary dataset.

[0072] Read the tidal phase warm water discharge result set and design a feature-preserving interpolation algorithm. For any time t, determine the tidal phase interval [t_a, t_b] it is in and the corresponding tidal phases φ_a and φ_b; apply the tidal phase feature-preserving interpolation function: T(x, y, t) = (1 - α(t))·T(x, y, t_a, φ_a) + α(t)·T(x, y, t_b, φ_b) + C(x, y, t), where α(t) is the interpolation weight and C(x, y, t) is the feature-preserving correction term; design of the feature-preserving correction term: C(x, y, t) = β(t)·[▽T(x, y, t_a)·(t - t_a) - ▽T(x, y, t_b)·(t_b - t)], where β(t) is the correction coefficient and ▽T is the temperature gradient. Through feature-preserving interpolation, reconstruct the complete warm water discharge time series to form a full-field temperature rise field time series dataset. Verify the physical rationality of the reconstruction result to ensure the continuity and energy conservation of the temperature field.

[0073] According to one aspect of the present application, the steps of obtaining the half-field temperature rise field data include:

[0074] S41. Based on the intake and discharge water information before and after shutting down the concerned unit, combine the calibrated model to identify the changes in the intake and discharge water conditions during the conversion process from the full-field model to the half-field model, and form the intake and discharge water condition change data;

[0075] S42. Perform a feature pattern-preserving transition on the data of the changes in the water intake and discharge conditions, and combine with a smooth transition function to generate smooth transition conditions; establish an energy balance equation for the changing conditions, apply energy balance constraints to optimize the smooth transition conditions, and generate energy balance conditions.

[0076] S43. Apply the energy balance conditions to the half-field warm water discharge simulation to ensure the continuity and physical rationality of the condition changes, and generate half-field temperature rise field data.

[0077] Specifically, the half-field simulation conditions are set as follows: based on the full-field water intake and discharge condition data set, close the water intake and discharge ports of the concerned unit to form a half-field water intake and discharge condition data set; load the half-field water intake and discharge condition data set into the calibrated model to form an initial half-field simulation model. Perform a feature pattern-preserving boundary transition process. During the transition from the full-field simulation initial model to the half-field simulation initial model, identify the boundary condition change area to form a boundary change area data set; apply the feature pattern-preserving boundary transition method to the boundary change area data set to generate a boundary condition transition model, and the calculation formula is: B_trans(t) = B_initial + (B_final - B_initial)·f(t), where f(t) is the smooth transition function, satisfying f(0)=0, f(T)=1, and T is the transition period; B_initial is the initial boundary condition; B_final is the final boundary condition; generate smooth transition boundary conditions to ensure the continuity and physical rationality of the boundary condition changes. Establish an energy balance equation for the boundary conditions: E[B_trans(t)] = E[B_initial] + ∫F(t)dt, where E is the boundary energy evaluation function and F is the boundary energy change rate; design an energy balance feedback controller: Δαi(t) = K·[E_target - E(t)]·▽E_i, apply energy balance constraints to optimize the smooth transition boundary conditions, and generate energy balance boundary conditions. Apply the energy balance boundary conditions and the phase time step strategy to the initial half-field simulation model to form an optimized half-field model; run the optimized half-field model to generate the original half-field simulation results; perform phase feature-preserving interpolation processing on the original half-field simulation results to form a time series data set of the half-field temperature rise field. Based on the time series data set of the half-field temperature rise field, calculate the spatio-temporal statistical features of the temperature rise field to form a half-field temperature rise feature data set; analyze the temperature rise distribution features under different tidal phases to generate a half-field tidal phase temperature rise feature data set.

[0078] In an embodiment of the present application, the full-field water intake and drainage condition dataset and the half-field water intake and drainage condition dataset are read, and the difference between the two is calculated; the boundary condition change region is identified, mainly including the regions near the water intake and drainage outlets of the concerned unit, and the calculation formula is: ΔB(x) = ||B_full(x) - B_half(x)||; B_full(x) is the boundary condition function value under the full-field water intake and drainage conditions; B_half(x) is the boundary condition function value under the half-field water intake and drainage conditions; a change significance threshold ε_B is set, and the boundary change region is defined as: Ω_B = {x | ΔB(x) > ε_B}; a boundary change region dataset is generated, including the spatial range and change intensity of the change region.

[0079] The boundary change region dataset and the tidal phase flow field pattern dataset are read, the spatial propagation characteristics of the boundary change are analyzed, and the transmission effect of the flow field on the boundary condition change is considered; a boundary change propagation model is constructed: ΔB(x, t+Δt) = ΔB(x, t) + ∫_ΩK(x, y, t)·ΔB(y, t)·dy, where K(x, y, t) is the propagation kernel function, which characterizes the boundary change propagation intensity from position y to position x. Based on the tidal flow field characteristics, the propagation kernel function is designed as: K(x, y, t) = exp(-||x-y|| 2 / σ 2 )·exp(-(t_travel(y→x) / τ) 2 ), where t_travel(y→x) is the water particle transport time from position y to position x, τ is the characteristic time scale, and σ is the characteristic spatial scale; a boundary change propagation characteristic dataset is generated to describe the spatio-temporal propagation law of the boundary condition change.

[0080] The boundary change region dataset and the boundary change propagation characteristic dataset are read, and for the boundary condition change region, a smooth transition function is designed; for the boundary condition at position x, the smooth transition function is designed as: B_trans(x, t)= B_initial(x) + (B_final(x) - B_initial(x))·f(x, t), where f(x, t) is the smooth transition function, satisfying f(x,0)=0, f(x, T)=1. The core smooth function is designed as: f(x, t) = (10·(t / T) 3 - 15·(t / T) 4 +6·(t / T) 5)·g(x), where g(x) is a spatial modulation function that considers the flow field characteristics at position x: g(x) = exp(-d(x, Ω_c) / λ), where d(x, Ω_c) is the distance from position x to the core change region Ω_c, and λ is the spatial modulation coefficient; generate a smooth transition boundary condition to ensure the continuity and physical rationality of the boundary condition change.

[0081] Read the smooth transition boundary condition and design a boundary energy evaluation function. The energy evaluation function considers three aspects: kinetic energy, potential energy, and thermal energy: E[B] = ∫_Ω (α·ρv 2 / 2 + β·ρgh + γ·ρcT)·dΩ; where ρ is the density, v is the flow velocity, g is the acceleration due to gravity, h is the water level, T is the temperature, and α, β, γ are weight coefficients; calculate the energy E[B_initial] of the initial boundary condition and the energy E[B_final] of the target boundary condition; construct an energy change rate function: F(t) = (E[B_final] - E[B_initial])·df(t) / dt, where df(t) / dt is the derivative of the smooth transition function; generate a boundary energy feature dataset to describe the characteristics of the boundary condition energy change.

[0082] Read the boundary energy feature dataset and design an energy balance feedback controller. The controller objective is to ensure that the energy change during the boundary condition change conforms to physical laws; establish a boundary condition energy balance equation: E[B_trans(t)] = E[B_initial] + ∫0 t F(τ)·dτ; where ∫0 t represents the integral from time 0 to the current time t, and F(τ) is the energy change rate function; design an energy balance feedback controller: Δα_i(t) = K·[E_target(t) - E[B_trans(t)]]·▽α_i E[B_trans(t)], where K is the control gain, E_target(t) is the target energy, and ▽α_i E is the gradient of the energy with respect to the reconstruction coefficient α_i. The controller workflow is as follows: calculate the energy E[B_trans(t)] of the current boundary condition; calculate the deviation ΔE from the target energy = E_target(t) - E[B_trans(t)]; calculate the gradient ▽α_i E of the energy with respect to the reconstruction coefficient; update the reconstruction coefficient: α_i(t+Δt) = α_i(t) + Δα_i(t); reconstruct the boundary condition B_trans(t+Δt); generate an energy control parameter dataset to guide the energy balance control of the boundary condition.

[0083] Read the dataset of smooth transition boundary conditions and energy control parameters, apply the energy balance feedback controller, and iteratively optimize the boundary conditions. The optimization algorithm is as follows: for t = 0 to T: Calculate the current boundary condition B_trans(t); Evaluate the energy balance state E[B_trans(t)]; Apply the energy controller to calculate the correction Δα_i(t); Update the reconstruction coefficient α_i(t); Reconstruct the boundary condition B_trans(t). Optimization objective: Minimize the cumulative amount of energy balance deviation: min ∫0 T (E[B_trans(t)] - E_target(t)) 2 ·dt; Apply the Lagrange multiplier method to solve the constrained optimization problem, obtain the optimal boundary condition that satisfies the energy balance, generate the final energy balance boundary condition, and use it for the half-field warm water discharge simulation.

[0084] As Figure 2 shown, according to one aspect of the present application, the steps of performing non-linear peeling to obtain the data of the temperature rise field of the concerned unit include:

[0085] S51. Based on the full-field temperature rise field data, extract the flow velocity field and construct a streamline coordinate system;

[0086] S52. Convert the full-field temperature rise field data and the half-field temperature rise field data to the streamline coordinate system to obtain the streamline temperature rise field data;

[0087] S53. Based on the streamline temperature rise field data, calculate the heat flux along the streamline and analyze the influence of temperature on fluid density to generate a non-linear correction term;

[0088] S54. Combine the heat flux along the streamline and the non-linear correction term to perform non-linear peeling in the streamline coordinate system, obtain the peeling result and map it back to the original coordinate system to form the data of the temperature rise field of the concerned unit.

[0089] As Figure 3 shown, according to one aspect of the present application, the steps of constructing a streamline coordinate system include:

[0090] Based on the flow velocity field, calculate the streamline trajectory; Based on the streamline trajectory, establish a streamline orthogonal curvilinear coordinate system, including the coordinate along the streamline direction and the coordinate perpendicular to the streamline direction;

[0091] Establish a two-way mapping relationship between the Cartesian coordinate and the streamline orthogonal curvilinear coordinate system to form a streamline coordinate system;

[0092] Derive the heat transfer equation in the streamline coordinate system, combine the temperature-dependent diffusion coefficient and the density flow correction term to form a streamline heat transfer model for performing non-linear peeling.

[0093] As Figure 4As shown, according to one aspect of the present application, the steps of performing non-linear stripping in the streamline coordinate system to obtain the stripping result include:

[0094] Based on the streamline temperature rise field data, calculate the heat fluxes along the streamline for the whole field and the half field respectively to obtain the heat flux difference;

[0095] According to the relationship between temperature and density, calculate the non-linear correction term of the fluid density change caused by temperature to the heat flux along the streamline;

[0096] Apply the streamline segment non-linear stripping formula to combine the heat flux difference and the non-linear correction term to obtain the stripped temperature rise field data in the streamline coordinate system, that is, the stripping result, for mapping back to the original coordinate system.

[0097] Specifically, construct the streamline coordinate system. Based on the full-field temperature rise field time series data set, extract the velocity fields at each moment to form the full-field velocity field time series data set; based on the full-field velocity field time series data set, calculate the main streamline distribution to generate the streamline distribution data set; establish the streamline orthogonal curvilinear coordinate system (s, n), where s is the direction along the streamline and n is the direction perpendicular to the streamline, to form the streamline coordinate system data set; develop the coordinate mapping algorithm: (x, y) ↔ (s, n) to form the coordinate mapping model.

[0098] Reconstruct the heat flux conservation equation. In the streamline coordinate system, reconstruct the heat transfer equation: ΨT / Ψt + UΨT / Ψs = Ψ / Ψs(DsΨT / Ψs) + Ψ / Ψn(DnΨT / Ψn) + S; where T is the temperature field, U is the flow velocity, Ds and Dn are the diffusion coefficients in the streamline direction and the direction perpendicular to the streamline respectively, S is the heat source term, and Ψ is the partial derivative; generate the streamline heat transfer model for the calculation and analysis of the heat flux. Perform streamline segment non-linear stripping. Using the coordinate mapping model, convert the full-field temperature rise field time series data set and the half-field temperature rise field time series data set to the streamline coordinate system to obtain the streamline temperature rise field data set and the streamline half-field temperature rise field data set; decompose the heat flux of the warm water discharge along each streamline, considering the influence of the temperature gradient along the streamline direction on the fluid density; non-linear stripping formula: ΔT concerned unit(s, n) = ΔT full field(s, n) - ΔT half field(s, n) + ∫Φ(s, n, T)ds, where Φ is the non-linear correction term of the fluid density change caused by temperature to the heat flux; perform the non-linear stripping calculation to obtain the streamline stripped temperature rise field data set in the streamline coordinate system.

[0099] Perform streamline temperature rise restoration. Using the coordinate mapping model, map the streamline stripped temperature rise field data set back to the original grid coordinate system, and use the interpolation method that maintains heat conservation to ensure the restoration accuracy to generate the temperature rise field data of the concerned unit.

[0100] According to one aspect of the present application, the steps of performing non-linear stripping processing further include:

[0101] Based on the flow velocity field and tidal phase classification data, identify the thermal circulation paths from the drainage outlets to the water intake outlets under different tidal phases to form a set of thermal circulation paths; use the set of thermal circulation paths to construct a tidal-phase-related thermal circulation matrix to characterize the heat transfer intensity from each drainage outlet to each water intake outlet under different tidal phases;

[0102] Apply the thermal circulation matrix to the stripped temperature rise field data to separate the direct warm water discharge influence and the thermal circulation cumulative effect, and obtain the temperature rise field data of the concerned units after thermal circulation decoupling.

[0103] According to one aspect of the present application, the steps of forming the set of thermal circulation paths and constructing the thermal circulation matrix include:

[0104] Calculate the connectivity index between each drainage outlet and each water intake outlet under different tidal phases to form drainage outlet - water intake outlet connectivity data;

[0105] Based on the connectivity data, calculate the paths of water masses from the drainage outlets to the water intake outlets (such as applying the Lagrangian particle tracking method) to form a set of thermal circulation paths;

[0106] Evaluate the heat transfer intensity of each path in the set of thermal circulation paths to generate thermal circulation path intensity data;

[0107] Based on the thermal circulation path intensity data, construct a tidal-phase-related thermal circulation matrix; apply singular value decomposition to the thermal circulation matrix to extract the main thermal circulation modes to form thermal circulation main mode data for the thermal circulation decoupling algorithm.

[0108] Specifically, based on the full-field flow velocity field time series data set and the tidal phase classification data set, develop a tidal-phase-associated thermal circulation path identification algorithm to identify the heat transfer paths from the drainage outlets to the water intake outlets under different tidal phases to form a tidal-phase thermal circulation path set; calculation formula: P(φ) = {p1(φ), p2(φ),..., pn(φ)}, where φ is the tidal phase and pi(φ) is the i-th thermal circulation path; introduce a path intensity evaluation index to quantify the heat transfer ability of each path to generate a thermal circulation path intensity data set. Construct a tidal-phase-related thermal circulation matrix: C(φ) = [cij(φ)], where cij(φ) represents the thermal circulation coefficient from drainage outlet i to water intake outlet j under tidal phase φ; apply singular value decomposition to extract the main thermal circulation modes: C(φ) = U·Σ·V T ; where U is an orthogonal matrix, Σ is a diagonal matrix, V is another orthogonal matrix, T is the transpose; retain the thermal circulation modes corresponding to the main eigenvalues, filter the noise modes to form a thermal circulation main mode data set. Establish a recursive model based on the thermal circulation main mode data set: ΔT = ΔT_direct + C(φ)·ΔT; transform it into a decoupling equation: ΔT_direct = [I - C(φ)]-1 · ΔT; Develop a stable solution algorithm to avoid ill-conditioned matrix problems; Use the initial data set of the temperature rise field of the concerned unit as the preliminary stripped temperature rise field data set and apply the decoupling algorithm to generate the decoupled temperature rise field data set for thermal cycles. Based on the decoupled temperature rise field data set for thermal cycles, perform the final reconstruction of the temperature rise field. Apply physical constraint conditions to ensure that the temperature rise field satisfies the requirements of energy conservation and continuity; Generate the final time series data set of the temperature rise field of the concerned unit as the final result of the stripping of the warm water discharge.

[0109] In an embodiment of the present application, perform flow field feature analysis and streamline calculation. Read the time series data set of the full-field temperature rise field, extract the velocity fields at each moment, and form the time series data set of the full-field velocity field; Perform feature analysis on the time series data set of the full-field velocity field to identify characteristic flow field structures such as the main flow area, vortex area, and shear layer, and generate the data set of flow field feature structures; Apply the streamline tracking algorithm to calculate the key streamlines in the characteristic flow field: dx / dt = u(x, y, t); dy / dt = v(x, y, t); where u(x, y, t) is the velocity component of the flow field in the x direction at the position (x, y) at time t; v(x, y, t) is the velocity component of the flow field in the y direction at the position (x, y) at time t; Use the fourth-order Runge-Kutta method to solve the above ordinary differential equations to obtain the streamline trajectory. Streamline density control strategy: Increase the streamline density in the high flow velocity gradient region to ensure that the streamline distribution can fully characterize the flow field features; Generate the initial streamline distribution data set, which contains the main streamline trajectories within the computational domain.

[0110] Perform streamline optimization and layering. Read the initial streamline distribution data set and the flow field feature structure data set to perform streamline optimization; Apply the streamline smoothing algorithm to eliminate local irregular fluctuations: x'(s) = (1 - ω)·x(s) + ω·(x(s - Δs) + x(s + Δs)) / 2; y'(s) = (1 - ω)·y(s) + ω·(y(s - Δs) + y(s + Δs)) / 2; where ω is the smoothing factor, s is the streamline parameterization variable; x(s) and y(s) are the abscissa and ordinate of the streamline under the parameterization variable s respectively; Δs is the increment of the streamline parameterization variable. Detect and process streamline intersections and special streamline topological structures (such as saddle points, foci) to ensure the topological correctness of the streamline system; Classify the streamlines into two categories: core streamlines and auxiliary streamlines: Core streamlines: Consistent with the main flow direction, relatively few in number, used to define the skeleton of the streamline coordinate system; Auxiliary streamlines: Fill the areas between the core streamlines to improve the accuracy of the coordinate system; Generate the optimized streamline distribution data set, which contains the optimized core streamlines and auxiliary streamlines.

[0111] Construct orthogonal curves. Read the optimized streamline distribution dataset and construct a curve system orthogonal to the streamlines; orthogonal curve calculation method: dx / dt = -v(x, y, t); dy / dt = u(x, y, t); use the fourth-order Runge-Kutta method to solve the above ordinary differential equations to obtain the orthogonal curve trajectory; select appropriate starting points on each core streamline to generate curves orthogonal to the streamlines; apply the orthogonality optimization algorithm to ensure the orthogonality of the orthogonal curves and the streamlines: min ∑_i ∑_j (τ_i·n_j) 2 , where τ_i is the tangent vector of streamline i and n_j is the normal vector of orthogonal curve j; generate an orthogonal curve dataset containing the curve system orthogonal to the streamlines.

[0112] Perform streamline coordinate system parameterization. Read the optimized streamline distribution dataset and the orthogonal curve dataset, establish a streamline orthogonal curve coordinate system, parameterize the streamlines, and define the streamline coordinate s: s(x, y) = ∫0 c sqrt(dx 2 + dy 2 ), where c is the path from the starting point of the streamline to the point (x, y); parameterize the orthogonal curves and define the orthogonal coordinate n: n(x, y) = ±∫0 c sqrt(dx 2 + dy 2 ), and the sign depends on whether the point (x, y) is on the left or right side of the reference streamline; establish a two-way mapping relationship: (x, y) ↔ (s, n) to form a set of coordinate mapping functions. Calculate the metric tensor gij and the Christoffel symbols Γijk of the coordinate transformation to form a coordinate system metric dataset. Integrate the set of coordinate mapping functions and the coordinate system metric dataset to generate a complete streamline coordinate system dataset and a coordinate mapping model.

[0113] Derive the heat transfer equation in the streamline coordinate system. Read the streamline coordinate system dataset and the coordinate system metric dataset; in the streamline coordinate system (s, n), derive the heat transfer equation: ΨT / Ψt + UΨT / Ψs = (1 / sqrt(g))·Ψ / Ψs(sqrt(g)·Ds·ΨT / Ψs) + (1 / sqrt(g))·Ψ / Ψn(sqrt(g)·Dn·ΨT / Ψn) + S, where T is the temperature field, U is the flow velocity along the streamline direction, g is the determinant of the metric tensor, Ds and Dn are the diffusion coefficients in the streamline direction and the direction perpendicular to the streamline respectively, and S is the heat source term; analyze the physical meaning of each term in the equation: the first term is the time rate of change of temperature, the second term is convective heat transfer, the third term is diffusion heat transfer along the streamline direction, the fourth term is diffusion heat transfer perpendicular to the streamline direction, and the fifth term is the heat source term; generate a streamline heat transfer equation dataset containing the mathematical expression and physical interpretation of the heat transfer equation.

[0114] Construct a temperature-dependent diffusion coefficient model. Read the time-series data set of the full-field temperature rise field, analyze the relationship between temperature and diffusion coefficient; construct a temperature-dependent diffusion coefficient model: Ds(T) = Ds0·[1 + αs·(T-T0) + βs·(T-T0) 2 ; Dn(T) = Dn0·[1 + αn·(T-T0) + βn·(T-T0) 2 ; where Ds0 and Dn0 are the diffusion coefficients at the reference temperature T0, and αs, αn, βs, βn are temperature-dependent coefficients; use historical warm water discharge simulation data and measured data, and apply the nonlinear regression method to determine the model parameters; generate a temperature-dependent diffusion model to describe the influence relationship of temperature on the diffusion coefficient.

[0115] Construct a density-dependent flow correction model. Read the time-series data set of the full-field temperature rise field, analyze the influence of temperature on the water body density; construct a temperature-density relationship model: ρ(T) = ρ0·[1 - α·(T-T0) + β·(T-T0) 2 - γ·(T-T0) 3 , where ρ0 is the water body density at the reference temperature T0, and α, β, γ are temperature-density relationship coefficients; analyze the influence of density change on the flow field, and construct a density flow correction term: ΔU(s, n) = g·H 2 / (8ν)·Ψρ / Ψs·(1-(2n / H) 2 ), where g is the acceleration due to gravity, H is the water depth, ν is the kinematic viscosity coefficient, and Ψρ / Ψs is the density gradient along the streamline direction; generate a density flow correction model to describe the influence of density change caused by temperature on the flow field.

[0116] Integrate the streamline heat transfer equation data set, the temperature-dependent diffusion model and the density flow correction model to construct a complete streamline heat transfer model: ΨT / Ψt + (U+ΔU)ΨT / Ψs = (1 / sqrt(g))·Ψ / Ψs(sqrt(g)·Ds(T)·ΨT / Ψs) + (1 / sqrt(g))·Ψ / Ψn(sqrt(g)·Dn(T)·ΨT / Ψn) + S; design a numerical solution scheme, including a time discretization method (implicit difference) and a space discretization method (finite volume method); formulate stability conditions and convergence criteria to ensure the accuracy of the numerical solution; generate a complete streamline heat transfer model for subsequent heat flux calculation and peeling analysis.

[0117] According to one aspect of the present application, the steps of statistically analyzing the data of the temperature rise field of the concerned unit include:

[0118] S61. Extract the temperature rise field data for multiple time periods from the concerned unit temperature rise field data, set the standard temperature rise isotherm level, extract the isotherms of each temperature level, and form the multi-time period temperature rise isotherm data;

[0119] S62. Analyze the spatio-temporal variation characteristics of the multi-time period temperature rise isotherm data, count the maximum influence range and area of each temperature level within the calculation period, and generate the temperature rise extreme value statistical data;

[0120] S63. Process the multi-time period temperature rise isotherm data using the α-shape method, generate the precise envelope lines of each temperature level, and form the temperature rise envelope line data;

[0121] S64. Calculate the area enclosed by the temperature rise envelope line data, apply geographical projection correction, and generate the temperature rise envelope area data for evaluating the influence range of the warm water discharge.

[0122] According to one aspect of the present application, the steps for statistically analyzing the concerned unit temperature rise field data further include:

[0123] Extract the temperature rise time series at each water intake location from the concerned unit temperature rise field data to form the water intake temperature rise time series data;

[0124] Analyze the statistical characteristics of the water intake temperature rise time series data, calculate indicators such as the average temperature rise, maximum temperature rise, and temperature rise exceeding standard rate, and generate the water intake temperature rise statistical data;

[0125] Based on the water intake temperature rise statistical data, evaluate the temperature rise exceeding standard rate and risk level, set the risk rating standard, and generate the water intake risk assessment data;

[0126] Integrate the water intake risk assessment data, design the evaluation system for the influence level of the water intake temperature rise, and form the water intake influence assessment data for evaluating the influence degree of the warm water discharge on the water intake.

[0127] Specifically, perform the extraction of multi-time period temperature rise field data. Extract the temperature rise field data for different time periods from the concerned unit temperature rise field time series dataset, including typical tidal phases, representative moments of flood tide / ebb tide, representative moments of seasons, etc.; generate the multi-time period temperature rise field dataset for subsequent statistical analysis. Based on the multi-time period temperature rise field dataset, calculate the coverage range of different temperature rise isotherms (usually 0.5°C, 1°C, 2°C, 3°C, 4°C, etc.); for each temperature rise isotherm, calculate the area within its envelope range to form the temperature rise area statistical dataset.

[0128] Based on the multi-period temperature rise field data set, calculate the maximum influence range of each temperature rise isotherm during the entire simulation period to form a temperature rise envelope line data set; calculate the area within the envelope range of each temperature rise isotherm to form a temperature rise envelope area data set. Based on the time series data set of the temperature rise field of the concerned unit, extract the temperature rise time series at each water intake location to form a temperature rise time series data set of the water intake; statistically analyze the temperature rise time series data set of the water intake, and calculate indicators such as the average temperature rise, maximum temperature rise, and temperature rise exceeding standard rate at the water intake to form a water intake impact assessment data set.

[0129] Integrate the temperature rise area statistical data set, the temperature rise envelope area data set, and the water intake impact assessment data set to generate a comprehensive assessment report on the impact of the warm water discharge of the concerned unit, including content such as the temperature rise range map, the temperature rise area statistical table, and the water intake impact analysis; form the final warm water discharge impact assessment report as the final output result.

[0130] In an embodiment of the present application, a multi-period temperature rise field data set is read. For the temperature rise field data of each period, a set of standard temperature rise isothermal line levels L = {0.5°C, 1°C, 2°C, 3°C, 4°C} is set; an isothermal line extraction algorithm is applied to extract isothermal lines for each temperature rise level l∈L and each period t: C(l, t) = {(x, y) | T(x, y, t) = l}; the bilinear interpolation method is used to determine the exact position of the isothermal line between grids, improving the smoothness and accuracy of the isothermal line. The Douglas-Peucker simplification algorithm is applied to the extracted isothermal line to reduce redundant points while maintaining the shape characteristics: C'(l, t) = Douglas-Peucker(C(l, t), ε_DP), where ε_DP is the simplification threshold, usually set to 1 / 10 of the grid size; a multi-period temperature rise isothermal line data set is generated, including the isothermal line data of all periods and all temperature rise levels. The multi-period temperature rise isothermal line data set is read, and the change characteristics of each temperature rise isothermal line over time are analyzed; for each temperature rise level l∈L, its maximum influence range within the calculation period is statistically calculated: max_x(l) = max_t{max{x | (x, y)∈C(l, t)}}; min_x(l) = min_t{min{x | (x, y)∈C(l, t)}}; max_y(l) = max_t{max{y | (x, y)∈C(l, t)}}; min_y(l) = min_t{min{y | (x, y)∈C(l, t)}}; a rectangular envelope box is generated: B_rect(l) = [min_x(l), max_x(l)] × [min_y(l), max_y(l)]; the maximum influence area and average influence area of each temperature rise level are calculated: A_max(l) = max_t{Area(C(l, t))}; A_avg(l) = (1 / T)·∑_t Area(C(l, t)); a temperature rise extreme value statistical data set is generated, including the extreme value statistical results of each temperature rise level.

[0131] Read the multi - period temperature rise isotherm data set and the temperature rise extreme value statistical data set, and use the AlphaShape method to generate the precise envelope for each temperature rise level: E(l) = AlphaShape(∪_t C(l, t), α), where α is the shape parameter determined by an adaptive algorithm: α = β·sqrt(A_max(l) / π) / N_points, where β is the adjustment coefficient (usually 2 - 3) and N_points is the total number of isotherm point sets; apply the envelope smoothing algorithm to eliminate irregular concavities and convexities and improve the aesthetics of the envelope: E'(l) = SmoothingSpline(E(l), λ), where λ is the smoothing parameter and the optimal value is determined by cross - validation; apply topological checks to the envelope to ensure topological legality such as no self - crossing and no holes; generate the final temperature rise envelope data set, which contains the optimized envelopes for each temperature rise level. Read the temperature rise envelope data set, calculate the area enclosed by the envelopes for each temperature rise level, and the area calculation uses a high - precision polygon area calculation method: A(l) = 0.5·|∑_i (x_i·y_{i + 1}-x_{i + 1}·y_i)|; where (x_i, y_i) are the vertex coordinates on the envelope; apply geographical projection correction to consider the projection deformation from spherical coordinates to planar coordinates: A_corr(l) = A(l)·cos(φ_avg), where φ_avg is the average latitude of the study area; calculate the area difference between adjacent temperature rise levels: ΔA(l) = A_corr(l)-A_corr(l + Δl), and generate the temperature rise envelope area data set, which contains the envelope areas and related statistical results for each temperature rise level.

[0132] Read the time - series data set of the temperature rise field of the concerned unit and the data set of intake and drainage structure information, and extract the set of location coordinates of all water intakes I = {i1, i2,..., i m} from the data set of intake and drainage structure information; for each water intake location i j , extract the temperature rise time series from the time - series data set of the temperature rise field of the concerned unit: T_j(t) = T(x_j, y_j, t), for(x_j, y_j) = i j; Considering the actual size of the water intake, taking the water intake location as the center, calculate the average temperature rise in the surrounding area (usually a 3×3 grid): T'_j(t) = (1 / N)·∑_{||x - x_j||≤R, ||y - y_j||≤R} T(x, y, t), where R is the influence radius of the water intake (usually 1 - 1.5 times the width of the water intake), and N is the number of grids within the calculation range; Apply temporal smoothing filtering to eliminate short-term fluctuations: T''_j(t) = (1 / W)·∑_{|τ - t|≤w} w(τ - t)·T'_j(τ), where w(τ - t) is the temporal weight function, W is the sum of weights, and w is the half-width of the temporal window; Generate a time series dataset of the water intake temperature rise, containing the temperature rise time series at all water intake locations.

[0133] Read the time series dataset of the water intake temperature rise and calculate the statistical characteristics of the temperature rise at each water intake; Calculate basic statistics: Average temperature rise: μ_j = (1 / T)·∑_t T''_j(t); Maximum temperature rise: max_j = max_t{T''_j(t)}; Standard deviation of temperature rise: σ_j = sqrt[(1 / T)·∑_t (T''_j(t) - μ_j 2 ; 95th percentile temperature rise: P95_j = Percentile(T''_j, 95%). Evaluate the persistence of the temperature rise by calculating the exceedance duration at different temperature rise thresholds l: D_j(l) = the maximum duration of the continuous time period {t | T''_j(t) > l}; Analyze the periodicity of the temperature rise and identify the main periodic components using power spectrum analysis: PS_j(f) = |FFT(T''_j(t))| 2 , identify the peak frequency f_peak and the corresponding period P_j = 1 / f_peak; Generate a statistical dataset of the water intake temperature rise, containing the statistical characteristics of the temperature rise at each water intake.

[0134] Read the statistical dataset of the temperature rise at the water intake, evaluate the potential impact of the temperature rise on the water intake, and set the temperature rise evaluation threshold set according to the ecological safety threshold and the engineering design limit: L_thresh = {l_eco, l_eng}, usually l_eco = 1°C, l_eng = 3°C; calculate the temperature rise exceedance rate: R_j(l) = (number of time periods exceeding threshold l / total number of time periods) × 100%; evaluate the temperature rise risk level, and set the risk level standard: Risk_j = low risk, if max_j < l_eco; medium-low risk, if l_eco ≤ max_j < l_eng and R_j(l_eco) < 20%; medium risk, if l_eco ≤ max_j < l_eng and 20% ≤ R_j(l_eco) < 50%; medium-high risk, if max_j ≥ l_eng and R_j(l_eng) < 10%; high risk, if max_j ≥ l_eng and R_j(l_eng) ≥ 10%. Calculate the potential economic loss index of the temperature rise impact: EL_j = k_j·(μ_j / l_eng)·(R_j(l_eco) / 100%)·P_j, where k_j is the importance coefficient of the water intake, and P_j is the power or flow rate of the water intake, and generate the water intake risk assessment dataset, including the risk ratings and impact indicators of each water intake.

[0135] Read the statistical dataset of the temperature rise at the water intake and the water intake risk assessment dataset, comprehensively evaluate the impact degree of the warm water discharge of the concerned unit on each water intake, and generate the impact assessment summary table; design the evaluation system for the impact level of the water intake temperature rise, considering multiple factors: temperature rise amplitude factor: F_T = max_j / l_eng; duration factor: F_D = D_j(l_eco) / D_crit; exceedance frequency factor: F_R = R_j(l_eco) / R_crit; comprehensive impact index: II_j = w_T·F_T + w_D·F_D + w_R·F_R; where w_T, w_D, w_R are weight coefficients, and D_crit and R_crit are the critical duration and critical exceedance rate. Sort each water intake according to the II_j value, identify the key sensitive water intakes, and generate visual charts, including the temperature rise time series chart of the water intake, the temperature rise distribution histogram, the impact level bubble chart, etc. Integrate the above data and analysis results to form the water intake impact assessment dataset.

[0136] In a specific embodiment of this application, a method for stripping the nonlinear superposition effect of thermal discharge from an adjacent nuclear power plant is described as follows:

[0137] Step 1: Data collection and pre-processing.

[0138] 1.1. Collect information on water intake and drainage structures.

[0139] The water intake of units 1 to 4 of the power station is located on the shore; the water intake of units 5 to 6 is shared with units 1 to 4, and the outlet is located 2.2 km offshore; the water depth at the outlet is about 14.5 m, and the water discharge capacity of a single unit is about 65 m 3 / s, the temperature rise of the discharged water is about 8℃. The thermal power plant is located about 9km northeast of the power plant, and the discharge capacity of a single unit is about 35m 3 / s, the drainage temperature rises by about 5°C. This generates the drainage structure information dataset D1 and the drainage condition dataset D2.

[0140] 1.2. Collect hydrological and meteorological data.

[0141] Hydrological observation data from the study area over the past five years were collected. The area is primarily governed by the M2 semi-diurnal tide, with an average tidal range of approximately 3.2 meters and a maximum current velocity of approximately 1.2 meters per second. This data was used to create the hydrological observation dataset D3, which includes time series of tide levels, current velocities, flow directions, and water temperatures at multiple measurement points. Meteorological data from the same period were collected to create the meteorological observation dataset D4, which includes parameters such as temperature, wind speed, and wind direction.

[0142] 1.3. Underwater terrain data collection and processing.

[0143] Underwater topographic data from the study area were collected to form the original topographic dataset D5. The water depth in the study area is approximately 14.5 meters at the outfall, gradually decreasing to 2-3 meters along the coast. The original data was digitized to form the digitized bathymetric topographic dataset D6. Abnormal data correction was then performed to obtain the corrected bathymetric topographic dataset D7.

[0144] 1.4. Data collection and preprocessing of measured temperature rise field.

[0145] The measured temperature rise field data of the existing units during operation are collected. After data quality control, a corrected temperature rise field dataset is formed, and the temperature rise characteristic dataset D10 is extracted for subsequent model verification.

[0146] Step 2: Construction and calibration of high-precision mathematical model.

[0147] 2.1. Build a computing grid.

[0148] In this embodiment, a certain coastal power station is taken as the research object, and the calculation domain ranges from a rectangular area centered on the power plant, with a length of 30 km in the east-west direction and 25 km in the north-south direction. The Triangle grid generation software is used to perform unstructured triangular grid meshing on the calculation domain, generating 89,532 initial grid cells. According to the characteristics of warm water discharge diffusion, the grid is locally refined as follows: within a range of 500 m around the discharge outlet, the grid size is 20 m; within a range of 500 m - 2 km around the discharge outlet, the grid size is 150 m; within a range of 2 km - 3 km around the discharge outlet, the grid size is 250 m; within a range of 200 m around the intake, the grid size is 20 m; in other nearshore areas (water depth less than 10 m), the grid size is 50 m; in the far-field area, the grid size is 500 - 2000 m. After grid refinement, the total number of cells increases to 125,753. The grid quality inspection shows that the minimum interior angle is 30° and the maximum interior angle is 116.3°, meeting the requirements for numerical calculation stability. The water depth data is interpolated onto the grid nodes to generate the grid terrain dataset D15.

[0149] 2.2. Boundary conditions and initial conditions setting.

[0150] 2.2.1. Extraction of boundary condition characteristic modes.

[0151] The wavelet transform analysis is performed on the measured tidal level data. The Morlet wavelet is selected as the mother wavelet function, with the scale ranging from 1 to 128 and the time step of 10 minutes. The multi-scale boundary condition dataset 1 obtained after the transformation contains the following characteristic components: short-time scale component (T < 1 h): mainly manifested as high-frequency perturbations, with an amplitude of about 0.05 - 0.1 m; medium-time scale component (1 h < T < 12 h): mainly contains semi-diurnal tide components such as M2 (12.42 h), S2 (12 h), N2 (12.66 h), etc., with an amplitude of about 1.5 - 2.0 m; long-time scale component (T > 12 h): mainly contains diurnal tide components such as K1 (23.93 h), O1 (25.82 h), etc. and meteorological tide components, with an amplitude of about 0.3 - 0.5 m. The kernel principal component analysis is applied to extract the non-linear characteristics, and the radial basis function K(x, y) = exp(-0.05||x - y|| 2)。The optimal kernel parameter γ = 0.05 is determined through cross-validation. Under this parameter, the training error is 0.032 m and the validation error is 0.045 m. After feature extraction, a total of 38 feature patterns are obtained. Among them, the first 12 feature subsets with the cumulative explained variance reaching 95% are selected to form the boundary feature pattern library: M1: representing the M2 principal semidiurnal tide feature, with an explained variance of 55.7%; M2: representing the S2 semidiurnal tide feature, with an explained variance of 17.3%; M3: representing the K1 principal diurnal tide feature, with an explained variance of 9.5%; M4: representing the O1 diurnal tide feature, with an explained variance of 4.8%; M5 - M12: representing the interaction and nonlinear features of various tidal waves, with a cumulative explained variance of 7.7%.

[0152] 2.2.2. Reconstruction of the boundary conditions of the feature patterns.

[0153] Based on the feature pattern library, a boundary condition reconstruction model is constructed to implement feature pattern reconstruction for the east, south, and north open boundaries of the model. Taking the east boundary as an example, the calculation process of the reconstruction formula is as follows: The initial reconstruction coefficients are determined by the least squares method: α1(t0) = 1.73, α2(t0) = 0.68, α3(t0) = 0.45, α4(t0) = 0.29, α5(t0) = 0.18,..., α12(t0) = 0.05; The reconstruction coefficients are optimized by applying physical constraints and solved by the Lagrange multiplier method: Constraint 1: Tide level continuity, |B(t+Δt) - B(t)| ≤ 0.1 m / 10 min; Constraint 2: Energy conservation, ∫B 2 (t)dt = ∫B' 2 (t)dt, where B' is the measured tide level; The optimized reconstruction coefficients are: α1(t0) = 1.69, α2(t0) = 0.71,.... A recurrence relation is designed to calculate the time-varying weight coefficients: Δαi(t) = 0.02·(BT(t) - B(t))·Mi + 0.005·(dBT / dt - dB / dt)·Mi; where BT(t) is the target boundary condition and B(t) is the current reconstructed boundary condition. For example, calculate the reconstruction coefficients at t = t0 + 1 h: α1(t0 + 1 h) = 1.69 + 0.09 = 1.78; α2(t0 + 1 h) = 0.71 - 0.03 = 0.68;.... When applying the boundary condition reconstruction model, when converting from the full-field model to the half-field model, for the boundary conditions in the drainage outlet area of units 5 - 6, a smooth transition function is designed: f(t) = 10·(t / T) 3 - 15·(t / T) 4 + 6·(t / T) 5, where T = 1 h; when t = 0 min, f(0) = 0, maintaining the original boundary conditions; when t = 30 min, f(0.5) = 0.5, and the boundary conditions have transitioned by 50%; when t = 60 min, f(1.0) = 1.0, fully transitioning to the new boundary conditions. After smooth transition processing, in the drainage outlet area of Unit 5, the maximum temperature boundary condition change rate has decreased from 1.2 °C / 10 min to 0.25 °C / 10 min, improving numerical stability. The generated smooth boundary condition dataset is applied to the half-field simulation.

[0154] 2.3. Presetting and calibrating model parameters.

[0155] The initial parameter settings are based on empirical values and literature recommendations: bottom roughness: n = 0.025 (corresponding to sandy seabed in open sea areas); horizontal eddy viscosity coefficient: Eh = 0.1 m 2 / s; vertical eddy viscosity coefficient: Ev = 0.01 m 2 / s; horizontal temperature diffusion coefficient: Dh = 0.5 m 2 / s; vertical temperature diffusion coefficient: Dv = 0.01 m 2 / s. Using this set of initial parameters to run a 30-day simulation and comparing with the measured data of 6 tide level stations and 4 flow velocity stations: average correlation coefficient of tide level: 0.91; root mean square error of tide level: 0.22 m; average correlation coefficient of flow velocity: 0.83; root mean square error of flow velocity: 0.19 m / s. For the model performance, a parameter sensitivity test is conducted, with a total of 24 groups of parameter combinations tested. Finally, the optimal parameter set is determined: bottom roughness: n = 0.022; horizontal eddy viscosity coefficient: Eh = 0.12 m 2 / s; vertical eddy viscosity coefficient: Ev = 0.008 m 2 / s; horizontal temperature diffusion coefficient: Dh = 0.65 m 2 / s; vertical temperature diffusion coefficient: Dv = 0.007 m 2 / s. The simulation results under the optimal parameters are significantly improved: average correlation coefficient of tide level: 0.97; root mean square error of tide level: 0.12 m; average correlation coefficient of flow velocity: 0.91; root mean square error of flow velocity: 0.11 m / s. Existing temperature drainage monitoring data shows that the measured average temperature rise at 500 m around the drainage outlets of Units 1 - 4 is 2.3 °C, and the model calculated value is 2.1 °C, with an error of approximately 8.7%; the measured average temperature rise at 1000 m is 1.1 °C, and the model calculated value is 1.05 °C, with an error of approximately 4.5%, meeting the accuracy requirements of the temperature drainage model.

[0156] Step 3. Conduct full-field temperature drainage simulation calculations.

[0157] 3.1. Full-field simulation condition settings. Set the boundary conditions for all units to operate simultaneously: For Units 1 - 4 of the power plant: the water intake flow rate for each unit is 65 m 3 / s, and the drainage temperature rise is 8 °C; for Units 5 - 6 of the power plant: the water intake flow rate for each unit is 65 m 3 / s, and the drainage temperature rise is 8 °C; for 4 units of the thermal power plant: the water intake flow rate for each unit is 35 m 3 / s, and the drainage temperature rise is 5 °C. The momentum source term method is used for setting the discharge boundary conditions, and momentum and heat source terms are added to the boundary cells of the drainage outlet: Momentum source term: MD = ρQU, where ρ = 1025 kg / m 3 , Q is the flow rate, and U is the discharge velocity; Heat source term: MT = ρcpQΔT, where cp = 4182 J / (kg·°C), and ΔT is the temperature rise. For the drainage outlet of Unit 5, the calculated values are: Momentum source term: MD = 1025×65×2.5 = 166,562.5 kg·m / s 2 ; Heat source term: MT = 1025×4182×65×8 = 2,228,716,000 J / s.

[0158] 3.2. Multi - time - scale feature - preserving tidal phase decomposition simulation.

[0159] 3.2.1. Tidal phase decomposition.

[0160] Perform Hilbert - Huang transform analysis on the tidal level data of the study area for one month to identify the main tidal wave components: M2 semi - diurnal tide: amplitude 1.53 m, phase 54.3°; S2 semi - diurnal tide: amplitude 0.62 m, phase 81.5°; K1 diurnal tide: amplitude 0.37 m, phase 322.6°; O1 diurnal tide: amplitude 0.26 m, phase 270.1°. Calculate the characteristics of each tidal phase based on the tidal level derivative. Taking the main measurement station in the study area as an example: Average flood - tide duration: 5 hours and 42 minutes; Average ebb - tide duration: 6 hours and 40 minutes; Maximum flood - tide change rate: 0.73 m / h, occurring in the flood - tide acceleration period; Maximum ebb - tide change rate: - 0.61 m / h, occurring in the ebb - tide acceleration period. Divide the tidal phase according to the tidal level change rate and acceleration. Taking July 15 as an example: 00:00 - 02:10: Ebb - tide deceleration period (FD), dh / dt = - 0.31 m / h, d 2 h / dt 2 = 0.12 m / h 2 ; 02:10 - 02:40: Slack period (S), dh / dt = - 0.04 m / h; 02:40 - 05:20: Flood - tide acceleration period (RA), dh / dt = 0.45 m / h, d 2 h / dt 2 = 0.15 m / h 2; 05:20 - 06:40: Rising tide extreme period (RP), dh / dt = 0.72 m / h, |d 2 h / dt 2 | = 0.009 m / h 2 ; 06:40 - 08:30: Rising tide deceleration period (RD), dh / dt = 0.38 m / h, d 2 h / dt 2 = -0.18 m / h 2 ;... Analyze the spatial propagation characteristics of the tidal phase. Taking the M2 tidal wave as an example, the time delay from the eastern boundary to the center of the study area is about 25 minutes, and the propagation speed is about 20 km / h.

[0161] 3.2.2. Phase - maintaining time - step optimization.

[0162] Calculate the characteristic time scales for different regions and tidal phases: In the high - temperature gradient region near the drainage outlet: Rising tide extreme period (RP): U = 0.75 m / s, H = 14.5 m, D = 0.65 m 2 / s, characteristic time scale τ(RP) = 133 s; slack period (S): U = 0.08 m / s, characteristic time scale τ(S) = 1250 s; Calculate the optimal time steps: Δt(RP) = 60×0.7 = 42 s, Δt(S) = 60×0.2 = 12 s. In the main diffusion region of the drainage outlet thermal plume (range of 500 m - 2 km): Rising tide extreme period: τ(RP) = 267 s, Δt(RP) = 60×0.9 = 54 s; slack period: τ(S) = 2000 s, Δt(S) = 60×0.4 = 24 s. In the far - field region (beyond 2 km): Rising tide extreme period: τ(RP) = 400 s, Δt(RP) = 60×1.0 = 60 s; slack period: τ(S) = 3000 s, Δt(S) = 60×0.7 = 42 s. Organize the time steps of each region and each tidal phase obtained from the calculation into a phase - time - step strategy D33 table for guiding the simulation calculation.

[0163] 3.2.3. Full - field warm - water discharge simulation execution.

[0164] Based on the phase time-step strategy D33, a full-field simulation was performed to calculate the thermal discharge diffusion process for 31 days from July 1st to July 31st, 2023. The simulation results were output every 10 minutes, generating a total of 4464 time-step results. Taking 13:00 on July 15th (mid-tide) as an example, the feature-preserving interpolation algorithm was applied to process the data. The adjacent output times were 12:50 (t_a) and 13:00 (t_b), and the interpolation weight α(12:55) = 0.5 was calculated. The preliminary temperature field at 12:55 was obtained by linear interpolation: T_init = 0.5×T(12:50) + 0.5×T(13:00). The feature-preserving correction term was calculated: temperature gradients: ▽T(12:50) = (0.12, 0.08) °C / 100m, ▽T(13:00) = (0.14, 0.09) °C / 100m; the correction coefficient β = 0.3; C(x, y, 12:55) = 0.3×[0.15×5min - 0.05×5min] = 0.15 °C. The temperature field after feature-preserving interpolation: T(12:55) = T_init + C = T_init + 0.15 °C. After feature-preserving interpolation processing, a complete time-series dataset D35 of the full-field temperature rise was generated for subsequent stripping calculations.

[0165] Step 4: Half-field thermal discharge simulation calculation.

[0166] 4.1. Half-field simulation condition setting.

[0167] Construct half-field simulation conditions, that is, shut down units 5-6: For power plant units 1-4: The water intake flow rate of each unit is 65 m 3 / s, and the drainage temperature rise is 8 °C; for power plant units 5-6: Shut down, and the water intake and drainage flow rates are 0; for 4 units of the thermal power plant: The water intake flow rate of each unit is 35 m 3 / s, and the drainage temperature rise is 5 °C; Remove the momentum and heat source terms of units 5-6 to form a half-field water intake and drainage condition dataset D38.

[0168] 4.2. Feature-mode-preserving boundary transition processing.

[0169] 4.2.1. Smooth transition of boundary conditions.

[0170] Calculate the boundary condition differences between the full-field model and the half-field model: Drain outlet position of Unit 5: Δh = 0m, ΔT = 8°C, ΔU = 2.5m / s; within a 100m range around the drain outlet of Unit 5: average ΔT = 3.2°C, average ΔU = 0.8m / s; within a 500m range around the drain outlet of Unit 5: average ΔT = 0.9°C, average ΔU = 0.2m / s. Define the boundary change region Ω_B according to the magnitude of the change: Core change region Ω_c: within a 100m range around the drain outlet of Unit 5; Transition change region: within a range of 100m - 1000m around the drain outlet; Far-field region: outside 1000m from the drain outlet. Design a smooth transition function to achieve a smooth transition from the full-field to the half-field: Time transition function f(t): f(0) = 0, t = 0min; f(0.2) = 0.01, t = 12min, at this time the boundary condition only changes by 1%; f(0.5) = 0.5, t = 30min, the boundary condition changes by 50%; f(0.8) = 0.99, t = 48min, the boundary condition changes by 99%; f(1.0) = 1.0, t = 60min, the boundary condition completely changes. Spatial modulation function g(x): g(x) = 1.0, when x is within the core change region Ω_c; g(x) = exp(-d(x, Ω_c) / 500), when x is within the transition region, d is the distance to the core region (m); for example, at a distance of 200m from the core region, g = exp(-200 / 500) = 0.67; at a distance of 1000m from the core region, g = exp(-1000 / 500) = 0.14. Example of comprehensive smooth function calculation: Drain outlet position, t = 30min: B_trans = B_initial + (B_final - B_initial)×0.5×1.0 = B_initial + 0.5×ΔB; at a distance of 500m from the drain outlet, t = 30min: B_trans = B_initial + (B_final - B_initial)×0.5×0.37 = B_initial + 0.18×ΔB.

[0171] 4.2.2. Boundary energy balance constraint.

[0172] Calculate each component of the boundary condition energy evaluation function: Kinetic energy component: E_k = ∫0.4×1025×v 2 / 2×dΩ = 3.76×10 6 J (full-field condition); Potential energy component: E_p = ∫0.3×1025×9.8×h×dΩ = 2.14×10 8 J (full-field condition); Thermal energy component: E_t = ∫0.3×1025×4182×T×dΩ = 7.93×109 J (Full-field conditions). The energy change from full-field to half-field: ΔE_k = -9.12×10 5 J (due to the removal of the momentum source of units 5 - 6); ΔE_p = -7.83×10 6 J (due to the water level change); ΔE_t = -7.58×10 8 J (due to the removal of the heat source). Design the parameters of the energy balance controller: control gain K = 0.2; energy gradient direction weights: w_k = 0.35, w_p = 0.25, w_t = 0.4. Apply the energy balance controller to adjust the boundary condition reconstruction coefficient: Calculate the current energy deviation: ΔE = E_target - E_current = -1.83×10 7 J (at t = 15 min); Calculate the gradient of energy with respect to the reconstruction coefficient: ▽α_i E = [ΨE / Ψα1, ΨE / Ψα2, ..., ΨE / Ψα12]; Update the reconstruction coefficient: Δα_i = 0.2 × (-1.83×10 7 ) × ▽α_i E; Δα1 = -0.022, Δα2 = -0.018, ..., Δα12 = -0.003; Reconstruct the boundary condition: B_bal = Σ(α_i + Δα_i)×M_i. Through 50 iterative steps, the energy deviation during the boundary condition change process is reduced from the initial -7.67×10 8 J to -3.21×10 6 J, and the relative error is reduced to 0.4%, meeting the energy balance requirement.

[0173] 4.3. Execution of half-field warm water discharge simulation.

[0174] Apply the energy balance boundary condition D42 and the phase time step strategy D33 to the half-field simulation, and perform the simulation calculation for the same time period (from July 1st to July 31st, 2023) as the full-field simulation. The processing method is the same as that of the full-field simulation, and a time series dataset D44 of the half-field temperature rise field is generated.

[0175] 4.4. Analysis of half-field temperature rise characteristics.

[0176] Analyze the characteristics of the half - field temperature rise field and compare with the full - field simulation results: Average temperature rise at 500 m from the drainage outlets of Units 1 - 4: Half - field 2.1°C vs Full - field 2.1°C (no change); Temperature rise at the proposed drainage outlet location of Unit 5: Half - field 0.2°C vs Full - field 8.0°C (a decrease of 7.8°C); Temperature rise at the power plant water intake: Half - field 0.32°C vs Full - field 0.62°C (a decrease of 0.3°C); Temperature rise at the thermal power plant water intake: Half - field 0.18°C vs Full - field 0.21°C (a decrease of 0.03°C). Analyze the temperature rise characteristics of different tidal phases to form the half - field tidal - phase temperature rise characteristic dataset D46 for subsequent non - linear stripping analysis.

[0177] Step Five: Non - linear temperature rise stripping calculation.

[0178] 5.1. Heat flux conservation stripping with streamline coordinate transformation.

[0179] 5.1.1. Construction of the streamline coordinate system.

[0180] Extract the velocity field from the full - field temperature rise field time - series dataset D35 to form the full - field velocity field time - series dataset D47. Select a typical moment t = 168 h (mid - flood tide), with a velocity range of 0.2 - 0.8 m / s and a main flow direction of 45°. Starting from the drainage outlet of Unit 5, track the water particle trajectories along the main flow direction, calculate 21 main streamlines, with a streamline spacing of 50 m at the drainage outlet. Establish an orthogonal curvilinear coordinate system of the streamlines to form the streamline coordinate system dataset D49 and the coordinate mapping model D50.

[0181] 5.1.2. Reconstruction of the heat flux conservation equation.

[0182] Derive the heat transfer equation in the streamline coordinate system: ΨT / Ψt + UΨT / Ψs = (1 / sqrt(g))·Ψ / Ψs(sqrt(g)·Ds·ΨT / Ψs) + (1 / sqrt(g))·Ψ / Ψn(sqrt(g)·Dn·ΨT / Ψn) + S, where T is the temperature field, U is the flow velocity, g is the Jacobian determinant of the coordinate transformation, Ds and Dn are the diffusion coefficients, and S is the heat source term. Establish the temperature - dependent diffusion coefficient models Ds(T) = Ds0·[1 + αs·(T - T0) + βs·(T - T0) 2 , Dn(T) = Dn0·[1 + αn·(T - T0) + βn·(T - T0) 2 , where Ds0 = 0.65 m 2 / s, Dn0 = 0.4 m 2 / s, αs = 0.02 / °C, αn = 0.01 / °C, βs = βn = 0.001 / °C 2, T0 = 20 °C. Establish a temperature-density relationship model ρ(T) = ρ0·[1 - α·(T - T0)], where ρ0 = 1025 kg / m 3 , α = 2×10⁻ 4 / °C. Calculate the density flow correction term ΔU(s, n) = g·H 2 / (8ν)·Ψρ / Ψs·(1 - (2n / H) 2 ), where g = 9.8 m / s 2 , H = 14.5 m, ν = 10⁻ 6 m 2 / s. Form a streamline heat transfer model D51.

[0183] 5.1.3. Streamline segmented non-linear stripping

[0184] Convert the full-field temperature rise field time series dataset D35 and the half-field temperature rise field time series dataset D44 to the streamline coordinate system to form a streamline temperature rise field dataset D52 and a streamline half-field temperature rise field dataset D53. Calculate the streamline heat fluxes of the full-field and half-field temperature rise fields Q_full(s, n) = ∫0 s [(U + ΔU_full)·T_full - Ds·ΨT_full / Ψs]·sqrt(g)·ds', Q_half(s, n) = ∫0 s [(U + ΔU_half)·T_half - Ds·ΨT_half / Ψs]·sqrt(g)·ds'. Calculate the heat flux difference ΔQ(s, n) = Q_full(s, n) - Q_half(s, n) and the non-linear correction term Φ(s, n, T) = ΔU_full·ΨT_full / Ψs - ΔU_half·ΨT_half / Ψs + (Ds(T_full)·ΨT_full / Ψs - Ds(T_half)·ΨT_half / Ψs), and apply the stripping formula ΔTconcern(s, n) = T_full(s, n) - T_half(s, n) + ∫0 s Φ(s', n, T)·sqrt(g)·ds' / (U·sqrt(g)) to form a streamline stripped temperature rise field dataset D54. Where ΔU_full is the density flow correction term of the full-field temperature rise field, T_full is the temperature distribution in the full-field temperature rise field, s' is the integration variable, ΔU_half is the density flow correction term of the half-field temperature rise field, and T_half is the temperature distribution in the half-field temperature rise field.

[0185] 5.1.4. Streamline temperature rise restoration.

[0186] Apply the coordinate mapping model D50 to map the streamline stripping temperature rise field dataset D54 back to the Cartesian coordinate system. Use the interpolation method that conserves heat, ∫∫ T(s, n)·sqrt(g)·dsdn = ∫∫ T(x, y)·dxdy, to ensure the restoration accuracy and form the preliminary stripped temperature rise field dataset D55.

[0187] 5.2. Decoupling of the tide-controlled thermal cycle path.

[0188] 5.2.1. Identification of the tide-controlled thermal cycle path.

[0189] Calculate the connectivity index of the intake and discharge ports C(o, i, φ) = exp(-τ(o→i, φ) / τ0)·exp(-d(o→i, φ) / d0), where τ0 = 12h and d0 = 10km. For the flood tide phase (φ = RP), the connectivity C from the discharge port of Unit 5 to the power plant intake is 0.21, and the connectivity C to the intake of the thermal power plant is 0.08. Apply the Lagrangian particle tracking method to release 100 tracer particles and obtain the thermal cycle path set D56, where 32% reach the power plant intake and 7% reach the intake of the thermal power plant. Calculate the path intensity S(p, φ) = Q·ΔT·P(p, φ)·exp(-τ(p, φ) / τ_d), where Q = 65m 3 / s, ΔT = 8℃, τ_d = 24h, and obtain S(p1, RP) = 35.8m 3 ·℃ / s, S(p2, RP) = 5.2m 3 ·℃ / s. Form the thermal cycle path intensity dataset D57.

[0190] 5.2.2. Thermal cycle matrix decomposition.

[0191] Construct the thermal cycle matrix C(φ) = [cij(φ)]. For the discharge port of Unit 5 to the power plant intake, c51(RP) = 0.16; to the intake of the thermal power plant, c52(RP) = 0.02. Apply the singular value decomposition C(RP) = U·Σ·V T , and obtain the main eigenvalue σ1 = 0.22 and the corresponding eigenvector (0.85, 0.15). Form the thermal cycle main mode dataset D58.

[0192] 5.2.3. Decoupling of the thermal cycle effect.

[0193] Apply the decoupling equation ΔTdirect = [I - C(φ)] -1 ·ΔT to process the preliminary stripped temperature rise field dataset D55. Taking the power plant intake as an example, the preliminary stripped temperature rise ΔT = 0.7℃, and the direct temperature rise contribution after decoupling is ΔTdirect = 0.6℃. Form the thermal cycle decoupled temperature rise field dataset D59.

[0194] 5.3. Comprehensive Reconstruction of Temperature Rise Field

[0195] Apply physical constraint conditions to the temperature rise field data set D59 after thermal cycle decoupling to ensure that the temperature rise field meets the requirements of energy conservation and continuity, and generate the final time-series data set D60 of the temperature rise field of the concerned unit.

[0196] Step Six: Statistics and Evaluation of Temperature Rise Influence Range

[0197] 6.1. Extraction of Temperature Rise Field Data in Multiple Time Periods

[0198] Extract the temperature rise field data in different time periods from the time-series data set D60 of the temperature rise field of the concerned unit, including typical moments such as the mid-tide and mid-ebb, to form the multi-time-period temperature rise field data set D61.

[0199] 6.2. Calculation of the Area of Temperature Rise Isolines

[0200] Based on the multi-time-period temperature rise field data set D61, calculate the coverage and area of different temperature rise isolines (0.5°C, 1°C, 2°C, 3°C, 4°C) to form the temperature rise area statistics data set D62.

[0201] 6.3. Calculation of the Envelope of Temperature Rise Influence Range

[0202] 6.3.1. Extraction of Multi-Time-Period Temperature Rise Isolines

[0203] Extract each temperature rise isoline C(l, t) = {(x, y) | T(x, y, t) = l} from the multi-time-period temperature rise field data set D61, use bilinear interpolation to determine the exact position of the isoline between grids, and apply the Douglas-Peucker simplification algorithm to reduce redundant points to form the multi-time-period temperature rise isoline data set D631.

[0204] 6.3.2. Statistics of the Extreme Values of Temperature Rise Influence Range

[0205] Statistically analyze the maximum influence range of each temperature rise isoline during the entire simulation period. Taking the 2°C temperature rise isoline as an example, the maximum range during the mid-tide is an elliptical area centered on the drainage outlet, with a major axis direction of 45°, a major axis length of about 1.8 km, and a minor axis length of about 0.7 km; the maximum range during the mid-ebb is an elliptical area with a major axis direction of 225°, a major axis length of about 2.2 km, and a minor axis length of about 0.8 km. Form the temperature rise extreme value statistics data set D632.

[0206] 6.3.3. Optimization and Generation of Envelope Curves

[0207] Apply the α - shape method to generate the exact envelope of each temperature rise isotherm E(l)=AlphaShape(∪_t C(l,t),α), where α = β·sqrt(A_max(l) / π) / N_points and β = 2.5. Taking the 2℃ temperature rise isotherm as an example, merge all the point sets at different times (a total of 56172 points), and calculate α = 4.3m. Form the temperature rise envelope data set D63.

[0208] 6.3.4. Calculation and correction of the envelope area.

[0209] Calculate the area enclosed by the temperature rise envelope A(l)=0.5·|∑_i (x_i·y_{i + 1}-x_{i + 1}·y_i)|, and apply geographical projection correction A_corr(l)=A(l)·cos(φ_avg), where φ_avg = 38.4°N is the average latitude of the study area. Form the temperature rise envelope area data set D64, where the 2℃ temperature rise envelope area is 3.85 km 2 .

[0210] 6.4. Assessment of the impact of temperature rise at the water intake.

[0211] 6.4.1. Extraction of the temperature rise time series at the water intake location.

[0212] Extract the temperature rise time series at the power plant water intake and the thermal power plant water intake locations to form the water intake temperature rise time series data set D65.

[0213] 6.4.2. Analysis of the statistical characteristics of the water intake temperature rise.

[0214] Calculate the statistical characteristics of the power plant water intake: average temperature rise μ = 0.62℃, maximum temperature rise max = 0.91℃, standard deviation of temperature rise σ = 0.14℃, 95% percentile temperature rise P95 = 0.85℃. Statistical characteristics of the thermal power plant water intake: average temperature rise μ = 0.21℃, maximum temperature rise max = 0.38℃. Form the water intake temperature rise statistical data set D651.

[0215] 6.4.3. Assessment of the temperature rise exceedance rate and risk level.

[0216] Set the temperature rise assessment threshold set L_thresh={l_eco, l_eng}, where l_eco = 1℃ and l_eng = 3℃. Calculate the temperature rise exceedance rates of the power plant water intake R(l_eco)=0% and R(l_eng)=0%, and assess the risk level as "low risk". Form the water intake risk assessment data set D652.

[0217] 6.4.4. Generation of a comprehensive impact assessment report.

[0218] Integrate the analysis results to generate the warm drainage impact assessment report D67, including a temperature rise range map, a temperature rise area statistics table, and a water intake impact analysis.

[0219] The results show that compared with the traditional linear stripping method, the envelope area of this embodiment in the low temperature rise region (0.5°C) is 31.2 km 2 , while the traditional method is 56.3km 2 , the deviation is as high as 80.3%; in the high temperature rise area (4℃), the area of this embodiment is 0.43km 2 , the traditional method is 0.45km 2 , with a deviation of only 4.7%. This demonstrates that this embodiment has advantages in assessing the far-field impact of warm discharge and predicting temperature rise at the water intake, meeting the regulatory requirement of "temperature rise range prediction deviation no more than 5%."

[0220] The present invention provides a method for stripping the nonlinear superposition effects of thermal discharge water adjacent to a nuclear power plant, including collecting basic data; establishing and calibrating a hydrodynamic and thermodynamic mathematical model; simulating full-field temperature rise; simulating half-field temperature rise; nonlinear stripping; and statistical analysis. The method employs a heat flux conservation stripping method using streamline coordinate transformation to convert the temperature rise field to a streamline coordinate system, taking into account the nonlinear correction term formed by the effect of temperature on fluid density. Combined with a tide-controlled thermal cycle path decoupling algorithm, the method identifies thermal cycle paths under different tidal phases and constructs a thermal cycle matrix to separate the direct thermal discharge effect and the cumulative effect of the thermal cycle. Through streamline coordinate transformation, the present invention accurately captures the plume characteristics of the outlet, resolving the problem that traditional methods cannot accurately describe heat transport in high-temperature gradient regions. The heat flux conservation equation is reconstructed to account for the effect of temperature on fluid density, essentially resolving the nonlinear superposition effect. The method addresses the insufficient accuracy of traditional fixed time step methods during tidal changes and avoids numerical oscillation of conventional interpolation methods in areas of drastic changes. The phase-differentiation stripping strategy considers the physical characteristics of different tidal phases, improving stripping accuracy, particularly in the highly dynamic regions near the outlet. This approach overcomes the limitations of traditional warm water mass tracking methods in complex tidal environments. The thermal cycle matrix decomposition method extracts key thermal cycle patterns, reducing computational complexity. The decoupling algorithm accurately separates the direct effects of thermal drainage from the cumulative effects of thermal cycles. Nonlinear characteristic pattern extraction overcomes the limitations of traditional Fourier analysis in dealing with nonlinear boundary conditions. The pattern-preserving boundary reconstruction method ensures the physical rationality of boundary condition changes. Boundary energy balance constraints avoid unphysical oscillations caused by sudden changes in boundary conditions. This approach offers significant advantages, particularly when the boundary conditions vary widely.

[0221] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A stripping method after the non-linear superposition effect of the warm wastewater discharged adjacent to a nuclear power plant, characterized in that including: Collect the intake and drainage information, hydro-meteorological and topographic data of the power plant under study to form a basic data set; Establish and calibrate a hydrodynamic and thermal mathematical model for the study water area based on the basic data set to obtain a calibrated model; Use the calibrated model to conduct warm water discharge simulations for all unit operations to obtain the full-field temperature rise data; After shutting down the unit of concern, conduct simulations based on the same calibrated model to obtain the half-field temperature rise data; Perform non-linear stripping on the full-field temperature rise data and the half-field temperature rise data to obtain the temperature rise data of the unit of concern; Conduct statistical analysis on the temperature rise data of the unit of concern to obtain the different temperature rise ranges and corresponding areas, as well as the temperature rise at the intake water temperature or sensitive locations of the unit; The steps of performing non-linear stripping to obtain the temperature rise data of the unit of concern include: Based on the full-field temperature rise data, extract the velocity field and construct a streamline coordinate system; Convert the full-field temperature rise data and the half-field temperature rise data to the streamline coordinate system to obtain the streamline temperature rise data; Based on the streamline temperature rise data, calculate the heat flux along the streamline and analyze the influence of temperature on fluid density to generate a non-linear correction term; Combine the heat flux along the streamline and the non-linear correction term to perform non-linear stripping in the streamline coordinate system, obtain the stripping result and map it back to the original coordinate system to form the temperature rise data of the unit of concern; The steps of performing non-linear stripping in the streamline coordinate system to obtain the stripping result include: Based on the streamline temperature rise data, calculate the heat flux along the streamline for the full field and the half field respectively to obtain the heat flux difference; Calculate the non-linear correction term for the change in fluid density caused by temperature on the heat flux along the streamline according to the relationship between temperature and density; Apply the non-linear stripping formula to combine the heat flux difference and the non-linear correction term to obtain the stripped temperature rise field data in the streamline coordinate system, which is the stripping result.

2. The method according to claim 1, characterized in that, The steps of constructing a streamline coordinate system include: Based on the velocity field, calculate the streamline trajectory; Based on the streamline trajectory, establish a streamline orthogonal curvilinear coordinate system, including the coordinate along the streamline direction and the coordinate perpendicular to the streamline direction; Establish a two-way mapping relationship between the Cartesian coordinate and the streamline orthogonal curvilinear coordinate system to form the streamline coordinate system; Derive the heat transfer equation in the streamline coordinate system, and combine the temperature-dependent diffusion coefficient and the density flow correction term to form a streamline heat transfer model.

3. The method according to claim 1, wherein The steps of conducting warm water discharge simulations for all unit operations to obtain the full-field temperature rise data include: Based on the hydro-meteorological data, analyze the tidal level time series, decompose the tidal cycle into flood tide period, high tide flat period, ebb tide period and low tide flat period to form typical tidal phase classification data; Based on the typical tidal phase classification data, calculate the optimal time step for different typical tidal phases to form a typical phase time step strategy; Apply the phase time step strategy to the calibrated model for warm water discharge simulations to ensure high-precision calculations at critical moments of tidal current changes, and generate more accurate full-field temperature rise data accordingly.

4. The method according to claim 3, wherein The steps of calculating the optimal time step for different tidal phases to form a phase time step strategy include: Based on the tidal phase classification data, analyze the transport characteristics of warm water discharge under different tidal phases, calculate the characteristic time scales of each tidal phase to form phase characteristic time scale data; Based on the phase feature time-scale data, construct a time step calculation formula, determine the optimal time step for different tidal phases and spatial positions, and generate the phase space time step distribution; Apply spatial smoothing processing to the phase space time step distribution to ensure continuous variation of the time step for adjacent grid cells, forming a smooth time step distribution; Integrate the smooth time step distribution to generate a complete phase time step strategy.

5. The method according to claim 3, characterized in that The steps for non-linear stripping processing also include: Based on the flow velocity field and tidal phase classification data, identify the heat cycle paths from the drainage outlets to the water intake points under different tidal phases, forming a set of heat cycle paths; Use the set of heat cycle paths to construct a heat cycle matrix related to the tidal phase, characterizing the heat transfer intensity from each drainage outlet to each water intake point under different tidal phases; Apply the heat cycle matrix to strip the temperature rise field data, separate the direct warm water discharge impact and the heat cycle cumulative effect, and obtain the temperature rise field data of the concerned units after heat cycle decoupling.

6. The method according to claim 5, wherein The steps for forming the set of heat cycle paths and constructing the heat cycle matrix include: Calculate the connectivity index between each drainage outlet and each water intake point under different tidal phases, forming the drainage outlet - water intake point connectivity data; Based on the connectivity data, calculate the paths of water masses from the drainage outlets to the water intake points, forming a set of heat cycle paths; Evaluate the heat transfer intensity of each path in the set of heat cycle paths, generating heat cycle path intensity data; Based on the heat cycle path intensity data, construct a heat cycle matrix related to the tidal phase; Apply singular value decomposition to the heat cycle matrix, extract the main heat cycle modes, and form the main heat cycle mode data.

7. The method according to claim 1, wherein After shutting down the concerned units, the steps for obtaining the half-field temperature rise field data include: Based on the water intake and drainage information before and after shutting down the concerned units, combine the calibration model to identify the changes in water intake and drainage conditions during the conversion from the full-field model to the half-field model, forming the water intake and drainage condition change data; Perform a characteristic mode-preserving transition on the water intake and drainage condition change data, combine with a smooth transition function to generate a smooth transition condition; the calculation formula is: B_trans(t) = B_initial + (B_final - B_initial)·f(t), where f(t) is the smooth transition function, satisfying f(0)=0, f(T)=1, T is the transition period; B_initial is the initial boundary condition; B_final is the final boundary condition; Establish a boundary condition energy balance equation, apply energy balance constraints to optimize the smooth transition condition, and generate an energy balance condition; where the boundary condition energy balance equation is E[B_trans(t)] = E[B_initial] + ∫F(t)dt, E is the boundary energy evaluation function, F is the boundary energy change rate; t is the time parameter, d is the differential symbol; Apply the energy balance condition to the half-field warm water discharge simulation to ensure the continuity and physical rationality of the condition changes, and generate the half-field temperature rise field data.

8. The method according to claim 1, wherein The steps for obtaining the calibration model include: Based on the basic data set, analyze the boundary condition data, extract the key characteristic modes of the boundary conditions, and form a boundary characteristic mode library; Based on the boundary characteristic mode library, construct a boundary condition reconstruction model; Construct the recurrence relation of the weight function, optimize the reconstruction coefficient of the boundary condition reconstruction model, and ensure the continuity of the change of the boundary condition; the formula for constructing the recurrence relation of the weight function is: αi(t+Δt) = αi(t) + Δαi(t), where αi(t) is the time-varying weight coefficient, and Δαi(t) is determined by the physical constraint conditions; t is the time parameter; Apply the optimized boundary condition reconstruction model to achieve a smooth transition of the boundary condition and generate smooth boundary condition data that satisfy the physical constraints; Based on the smooth boundary condition data, establish and calibrate the hydrodynamic and thermal mathematical model of the research water area to obtain the calibrated model.

Images