A method and device for calibrating a lake water ecological management model based on mid-universe experimental data
By deploying open experimental units in the lake and applying differentiated intervention measures, a mesocosmic-scale model was constructed and converted into a whole-lake-scale model. This solved the problems of insufficient data and non-unique parameters in existing technologies, and achieved high efficiency and high reliability calibration of the whole-lake water ecological governance model.
Patent Information
- Application Number
- CN202610868384.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-16
- Publication Date
- 2026-08-25
AI Technical Summary
Existing lake water ecological models suffer from limited data and single operating conditions during parameter calibration, making it difficult for the models to capture the dynamic response characteristics of the system, resulting in non-unique parameter identification and high uncertainty in prediction results, making it difficult to adapt to changes in actual operating conditions.
Using a method based on mesocosmic experimental data, multiple open experimental units were deployed in the target lake, differentiated engineering interventions were applied, water quality and meteorological data were collected, a mesocosmic-scale aquatic ecosystem dynamics model was constructed, parameters were optimized, the conversion relationship between mesocosmic and whole-lake-scale parameters was established, a whole-lake aquatic ecosystem dynamics model was constructed and iteratively corrected.
By increasing the amount of data and conducting multi-condition simulations, the uniqueness of parameter identification and the reliability of calibration results were improved, the scale effect problem was solved, and the efficient and accurate calibration of the whole lake water ecological governance model was achieved.
Smart Images
Figure CN122634918A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of lake management model technology, specifically relating to a calibration method and apparatus for a lake water ecological management model based on data from the Middle Cosmic Experiment. Background Technology
[0002] Lake eutrophication is a global water environment problem, and scientific aquatic dynamics models are an important support for the effective management of lake aquatic ecosystems. Currently used lake aquatic ecosystem models include a large number of empirical parameters, and the parameter values directly determine the model's prediction accuracy. Therefore, accurate model calibration is crucial.
[0003] Traditional model parameter calibration methods primarily rely on historical monitoring data from the entire lake. However, such historical monitoring data typically suffers from limited data volume and limited operational conditions, making it difficult for calibrated models to capture the dynamic response characteristics of the system and thus failing to accurately identify response parameters under different intervention measures. Moreover, these limitations easily lead to technical dilemmas such as non-unique parameter identification and high uncertainty. That is, two or more different parameter combinations may fit the historical data, but the prediction results for future operational conditions may be diametrically opposed, severely reducing the reliability and generalization ability of the calibration results and making it difficult to adapt to the dynamic changes of actual operational conditions.
[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention discloses a method and apparatus for calibrating a lake water ecological governance model based on data from the Middle Cosmic Experiment.
[0006] The technical solutions adopted in the embodiments of the present invention are as follows: A calibration method for a lake water ecological governance model based on data from the Middle Cosmic Experiment includes the following steps: S100. Multiple experimental units are set up in the target lake, and differentiated engineering intervention measures are applied to each experimental unit; wherein, the experimental unit is an open-type mesocosmic experimental unit, and the engineering intervention measures include at least one of the following: aeration and oxygenation, bottom sediment covering, aquatic plant planting, and flocculant addition. S200: Collect water quality parameter data from each experimental unit, simultaneously collect meteorological and hydrological data, and preprocess the collected data. S300. Construct a mesocosmic-scale water ecological dynamics model for each experimental unit. Use the water quality data before the experimental intervention as the initial condition for each mesocosmic-scale water ecological dynamics model, and use the water quality data after the intervention as the objective function. Use an automatic calibration algorithm to optimize the model parameters and output the optimal values and confidence intervals of the mesocosmic-scale parameters. S400. Establish the conversion relationship between mesocosmic-scale parameters and whole-lake-scale parameters, and convert the mesocosmic-scale parameters into whole-lake-scale parameters. S500. Construct a full-lake water ecological dynamics model by inputting the full-lake scale parameters obtained in step S400 into the full-lake water ecological dynamics model. S600. Use the whole lake water ecological dynamics model determined in step S500 to simulate the whole lake water ecological parameters, compare and verify the simulation results with the whole lake historical monitoring data, calculate the model accuracy index, and iteratively correct the transformation relationship based on the verification results.
[0007] A further technical solution is that, in step S100, at least 6 experimental units are set up in the target lake, and all experimental units are divided into at least 1 control group and at least 5 experimental groups. Each experimental unit is equipped with a controllable inlet and a controllable outlet, and each outlet is equipped with an electric regulating gate, an electromagnetic flow meter, and a filter with a preset aperture.
[0008] A further technical solution is that, in step S200, water quality parameter data of each experimental unit are collected at a frequency of at least once per day. The water quality parameter data includes total nitrogen (TN), total phosphorus (TP), dissolved oxygen (DO), chlorophyll a, turbidity, and pH.
[0009] A further technical solution is that step S300 specifically includes: The water quality data of each experimental unit before and after the application of engineering intervention measures were used as the initial conditions and target values of the corresponding mesocosmic-scale aquatic ecosystem dynamics model. Based on the model framework of the mesocosmic-scale water ecosystem dynamics model, and combined with the intervention type of the mesocosmic experiment, several core parameters to be calibrated were selected. Based on empirical values and experimental data, the initial value range of each core parameter is determined; The SCE-UA global optimization algorithm is used to globally optimize the core parameters, and the optimal values and 95% confidence intervals of each core parameter are output.
[0010] A further technical solution is that step S400 specifically includes: Calculate the core scale factors between the experimental unit and the target lake, including the water depth ratio, area ratio, and hydraulic residence time ratio; Based on the physical meaning of the core parameters and the scale dependence between the core parameters and the core scale factor, the core parameters are further divided into parameters to be converted and parameters that do not need to be converted. A transformation equation is constructed, which is based on the mesocosmic-scale parameter values corresponding to the core scale factor and core parameters. The values of the parameters to be transformed are then used as the full-lake-scale parameter values. Without needing to transform the parameters, the corresponding mesocosmic-scale parameter values can be directly used as the whole-lake-scale parameter values.
[0011] The further technical solution is that, in steps S300 and S500, the mesocosmic-scale water ecological dynamics model and the whole lake water ecological dynamics model both adopt the EFDC three-dimensional surface water quality mathematical model, coupled with the hydrodynamic module, water quality module and water ecology module. The parameters to be converted include: roughness coefficient, diffusion coefficient, reoxygenation coefficient, algal growth rate, algal mortality rate, aquatic plant absorption rate, and ammonia nitrogen nitrification coefficient; the parameters that do not need to be converted include: nutrient degradation coefficient and adsorption / desorption coefficient. The transformation equation is expressed as: ,in, Indicates the water depth ratio. Indicates the area ratio. Indicates the hydraulic residence time ratio; In step S500, the accuracy of the mesocosmic scale parameter values corresponding to each mesocosmic experimental unit is evaluated, experimental group parameters whose calibration accuracy does not meet the preset standard are eliminated, and the experimental group parameters whose calibration accuracy meets the preset standard are weighted and averaged to obtain a set of comprehensive mesocosmic scale parameters. Then, based on the conversion relationship established in step S400, the comprehensive mesocosmic parameters are converted into lake-scale parameters.
[0012] A further technical solution is that step S600 specifically includes: The simulation results are compared with historical monitoring data of the entire lake obtained through at least 12 monthly monitoring sessions. The accuracy indicators of the core parameters are calculated, and it is determined whether they meet the preset requirements. The accuracy indicators include the Nash efficiency coefficient (NSE) and the coefficient of determination. At least one of the following: root mean square error (RMSE) and relative error (RE); If the accuracy index meets the preset requirements, then the model parameters of the whole lake water ecological dynamics model are qualified; If the accuracy indicators do not meet the preset requirements, the model parameters of the whole lake water ecological dynamics model are unqualified. Further, the conversion equation is adjusted according to the error characteristics of the accuracy indicators, and the adjusted conversion equation is used to return to step S500 until all accuracy indicators meet the preset requirements.
[0013] A calibration device for a lake water ecological governance model based on data from the Middle Cosmic Experiment, comprising: The data acquisition and preprocessing module is configured to acquire water quality parameter data collected from multiple experimental units, as well as synchronously collected meteorological and hydrological data, and preprocess the acquired data; wherein, the experimental unit is an open-type mesocosmic experimental unit deployed in the target lake; differentiated engineering intervention measures are applied in each experimental unit; the engineering intervention measures include at least one of the following: aeration and oxygenation, bottom sediment covering, aquatic plant planting, and flocculant addition; The mesocosmic model construction and calibration module is configured to construct mesocosmic-scale water ecological dynamics models for each experimental unit. The water quality data before the experimental intervention is used as the initial conditions for each mesocosmic-scale water ecological dynamics model, and the water quality data after the intervention is used as the objective function. An automatic calibration algorithm is used to optimize the model parameters and output the optimal values and confidence intervals of the mesocosmic-scale parameters. The conversion relationship establishment module is configured to establish a conversion relationship between mesocosmic-scale parameters and whole-lake-scale parameters, converting mesocosmic-scale parameters into whole-lake-scale parameters. The whole-lake model construction module is configured to construct a whole-lake aquatic ecosystem dynamics model, and inputs the whole-lake scale parameters obtained by the transformation relationship establishment module into the whole-lake aquatic ecosystem dynamics model. The verification and iteration module is configured to use the whole-lake aquatic ecosystem dynamics model determined by the whole-lake model construction module to simulate the whole-lake aquatic ecosystem parameters, compare and verify the simulation results with the whole-lake historical monitoring data, calculate the model accuracy index, and iteratively correct the transformation relationship based on the verification results.
[0014] An electronic device, which is a device with data processing and / or communication functions, wherein the electronic device is configured with the aforementioned lake water ecological governance model calibration device based on the data from the Middle Cosmic Experiment.
[0015] A further technical solution is that the electronic device is a computer, mobile phone, or tablet computer.
[0016] The beneficial effects of the embodiments of the present invention are as follows: (i) The lake water ecological governance model calibration method and device proposed in this invention based on mesocosmic experimental data, through the design of an open mesocosmic experiment at a scale between laboratory pilot and whole-lake experiment, simulates the real lake environment under differentiated controlled conditions, reducing the difference between experimental conditions and the real lake environment, and can obtain a large amount of multi-condition response data in a short time, increasing the amount and diversity of experimental data, breaking through the limitation of insufficient historical monitoring data of the whole lake, and calibrating mesocosmic scale model data based on multi-condition experimental data can identify parameters that are difficult to determine by traditional methods, with a narrower confidence interval and higher reliability of calibration results.
[0017] (ii) This invention further establishes the conversion relationship between mesocosmic-scale parameters and whole-lake-scale parameters, converts mesocosmic-scale parameters into whole-lake-scale parameters, and inputs the converted whole-lake-scale parameters into the whole-lake water ecological dynamics model, thereby achieving efficient and accurate calibration of the whole-lake water ecological governance model, improving the uniqueness of parameter identification and the robustness of calibration results of the whole-lake-scale model, and solving the problem of scale effect when conventional small-scale experimental data are applied to the whole-lake-scale model. Attached Figure Description
[0018] Figure 1 This is a flowchart of the lake water ecological governance model calibration method based on the data from the Middle Cosmic Experiment proposed in Embodiment 1 of the present invention.
[0019] Figure 2 This is a parameter comparison table for multiple experimental units designed using a subtropical eutrophic lake as an example in Embodiment 1 of the present invention.
[0020] Figure 3 This is a data table showing the initial value range of the core calibration parameters and intervention measures provided in Embodiment 1 of the present invention.
[0021] Figure 4 This is a comparison table for the classification, scale dependence, and transformation relationship analysis of the parameters to be transformed provided in Embodiment 1 of the present invention. Detailed Implementation
[0022] The specific embodiments of the present invention will now be described with reference to the accompanying drawings.
[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the device proposed by this invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of this invention will become clearer from the following description. It should be noted that the drawings are in a very simplified form and use non-precise proportions, only for the purpose of conveniently and clearly illustrating the embodiments of this invention. Please refer to the accompanying drawings to make the objectives, features, and advantages of this invention more apparent and understandable. It should be understood that the structures, proportions, sizes, etc., depicted in the accompanying drawings are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the implementation conditions of this invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportional relationships, or adjustments to the size, without affecting the effects and objectives achieved by this invention, should still fall within the scope of the technical content disclosed in this invention.
[0024] To clearly describe the technical solution of this invention, some terms involved in this invention are explained as follows: The Mid-Cosmic Experiment: An experimental scale between laboratory-scale experiments and full-lake experiments, capable of simulating the real lake environment under controlled conditions.
[0025] Medium Universe Experiment Unit: The carrier for conducting medium universe experiments can be divided into open and closed types. In this embodiment of the invention, an open type is used.
[0026] Open-ended mesociety experimental unit: A mesociety experimental unit with controlled exchange of matter and energy with an external lake body.
[0027] Model calibration: The process of optimizing model parameters to make the simulation results match the measured data as closely as possible.
[0028] Scale effect: When small-scale experimental data is directly applied to large-scale models, the prediction accuracy of the model decreases due to scale differences.
[0029] Unless otherwise stated, the above definitions apply to the entire contents of this specification, including but not limited to the detailed description and claims.
[0030] Example 1 Figure 1 This is a flowchart of a lake water ecological governance model calibration method based on data from the Middle Cosmic Experiment, proposed in Embodiment 1 of the present invention. Figure 1 As shown, the model calibration method in this embodiment includes the following steps.
[0031] Step S100: Design a cosmic differentiation intervention experiment.
[0032] Multiple experimental units were deployed in the target lake, including at least one control group and N-1 experimental groups. Different types or intensities of engineering interventions were then applied to each experimental group. These interventions included at least one of the following: aeration, sediment covering, planting of aquatic plants, and flocculant addition.
[0033] Step S200: High-frequency water quality data acquisition and quality control.
[0034] Water quality parameters, including total nitrogen (TN), total phosphorus (TP), dissolved oxygen (DO), chlorophyll a, turbidity, and pH, are collected from each space experiment unit at a frequency of no less than once per day. Meteorological and hydrological data are collected simultaneously. Outlier removal and missing value imputation are performed on the collected data.
[0035] Step S300: Construct a mesocosmic-scale model and optimize the model parameters.
[0036] A water ecological dynamics model was constructed for each medium universe experimental unit. The water quality data before the experimental intervention was used as the initial condition of the model, and the water quality data after the model intervention was used as the reference target to construct the objective function. Then, an automatic calibration algorithm was used to optimize the model parameters and output the maximum value and confidence interval of each model parameter.
[0037] Step S400: Establish the conversion relationship between mesocosmic-scale parameters and whole-lake-scale parameters, and convert the mesocosmic-scale parameters into whole-lake-scale parameters.
[0038] Step S500: Construct a whole-lake water ecological dynamics model by inputting the whole-lake scale parameters obtained in step S400 into the whole-lake water ecological dynamics model.
[0039] Step S600: Use the whole lake water ecological dynamics model determined in step S500 to simulate the whole lake water ecological parameters, compare and verify the simulation results with the whole lake historical monitoring data, calculate the model accuracy index, and iteratively correct the transformation relationship based on the verification results.
[0040] Specifically, in step S100, the space experiment unit is an open experiment unit. Compared with the closed experiment unit, the hydrodynamic characteristics of the open experiment unit are highly consistent with those of a real lake. Therefore, it can more accurately reflect the real impact of the treatment measures on nutrient migration and transformation. The experimental biological community is closer to that of a real lake, avoiding experimental result deviations caused by biological isolation.
[0041] This embodiment employs an open-loop space experiment unit, which achieves controlled water exchange with an external lake through a controllable inlet and outlet. The daily exchange volume is 5%-20% of the unit volume, simulating diurnal flow fluctuations. The inlet and outlet water disturbances have the following characteristics: The mixing mechanisms differ: horizontal mixing in large lakes is mainly dominated by wind-driven currents, while horizontal mixing in open cosmic units is mainly dominated by local shear flows and velocity gradients generated by inflow and outflow, with mixing intensity far exceeding that of natural large lake regions of the same scale.
[0042] Significant non-steady-state characteristics: The periodic fluctuations in inflow and outflow (simulating the diurnal hydrological changes of natural lakes) lead to obvious non-steady-state characteristics in the mixing process, which cannot be accurately described by traditional large-scale empirical formulas.
[0043] Strong spatial heterogeneity: The flow velocity and mixing intensity near the inlet and outlet are significantly higher than those in the central area of the unit, and there are local strong mixing zones.
[0044] Figure 2 This is a parameter comparison table for multiple experimental units designed using a subtropical eutrophic lake as an example in Embodiment 1 of the present invention. (See table below.) Figure 2As shown, this embodiment uses a subtropical eutrophic lake (average water depth 2.5m, TN 2.8mg / L, TP 0.12mg / L) as an example. Seven open-loop space experiment units were set up, one of which was a control group and the other six were experimental groups. The engineering intervention measures of each experimental group were different. After 30 days of intervention, the experimental data of the monitored water quality parameters showed different changes, which can reflect the impact of the engineering intervention measures.
[0045] Furthermore, the open-type experimental unit in this embodiment is equipped with two controllable inlets and two controllable outlets, each outlet being equipped with an electric regulating gate and an electromagnetic flowmeter (with control accuracy). 0.1L / s) and double-layer 100 The system includes a filter and a PLC controller. These measures ensure controlled exchange of matter and energy between the experimental unit in the central universe and the external lake water. Hydrological exchange control: The daily exchange volume (preferably 5%-20% of the unit water volume) and exchange time curve are preset by the PLC controller to simulate the hydraulic residence time and diurnal flow fluctuation of a natural lake.
[0046] Material exchange control: Double-layer 100 at the inlet and outlet The filter screen removes large planktonic organisms and debris, allowing only soluble nutrients and small phytoplankton (particle size <100 μm). This allows zooplankton to pass through while preventing experimental organisms such as aquatic plants and fish from escaping from the experimental unit.
[0047] Energy exchange control: The experimental unit receives the same meteorological conditions as the external lake, such as sunlight, temperature, and wind speed, to achieve natural exchange of solar and thermal energy and ensure that the water temperature and light intensity inside the unit are consistent with those of the external lake.
[0048] Furthermore, in step S300, the EFDC three-dimensional surface water quality mathematical model is used as the basic framework. Considering the scale characteristics of the open-ended mid-cosmic experimental unit, three core modules—hydrodynamics, water quality, and aquatic ecology—are coupled to achieve integrated simulation of hydrodynamic processes, nutrient migration and transformation, and biological communities within the experimental unit. The adaptability design of each module is as follows: Hydrodynamics module: Using measured influent flow rate, effluent flow rate, and water level data of each experimental unit as boundary conditions, the three-dimensional Reynolds-Navier-Stokes average equations are solved to calculate the velocity field, water level changes, and vertical mixing process within the experimental unit. This embodiment considers the influent and effluent disturbances of open units and modifies the empirical formula for the horizontal diffusion coefficient to adapt it to the mixing characteristics of small-scale water bodies.
[0049] Water quality module: includes total nitrogen (TN), total phosphorus (TP), and ammonia nitrogen. -N, nitrate ammonia -N, soluble orthophosphate The system uses six core state variables—P, dissolved oxygen (DO)—to simulate chemical processes such as nutrient adsorption and desorption, nitrification and denitrification, and phosphorus release.
[0050] The aquatic ecosystem module comprises three biological components: phytoplankton (cyanobacteria, green algae, diatoms), submerged plants, and zooplankton. It simulates biological processes such as algal growth / death / settling, nitrogen and phosphorus absorption / decomposition by submerged plants, and feeding by zooplankton. To address the biological community characteristics of the mesocosmic unit, the modules for large aquatic animals and benthic organisms are simplified, reducing model complexity.
[0051] Using the hydrological data obtained in step S200, the influent flow rate, effluent flow rate, and water level data of each experimental unit are used as boundary conditions to calculate the velocity field, water level changes, and water mixing process within the experimental unit. At the same time, based on the influent flow rate and influent water quality data, the external nutrient input flux is calculated, and based on the effluent flow rate, the output momentum of nutrients and algae within the experimental unit is calculated.
[0052] Using the meteorological data obtained in step S200, air temperature is used to calculate water heat exchange and water temperature stratification, wind speed is used to calculate water surface shear stress and water mixing intensity, rainfall is used to calculate the increase and dilution of water volume within the unit, and water temperature is used to correct temperature-related parameters such as nutrient degradation coefficient and reoxygenation coefficient.
[0053] Furthermore, Figure 3 This is a data table showing the initial value range of the core calibration parameters and intervention measures provided in Embodiment 1 of the present invention. For example... Figure 3 As shown, in step S300, based on the above model framework and combined with the intervention type of the medium universe experimental unit, 12 core parameters to be calibrated that have the most significant impact on the simulation results are selected, namely model parameters. These parameters are then categorized into three types: physical, chemical, and biological (consistent with the parameter classification for subsequent scale transformation), and initial value ranges based on literature values and experimental pre-experiments are given. The 12 core parameters to be calibrated, each with different physical meanings, are: bed surface roughness coefficient, horizontal diffusion coefficient, reoxygenation coefficient, settling rate, ammonia nitrogen nitrification coefficient, denitrification coefficient, sediment phosphorus release coefficient, phosphorus adsorption coefficient, maximum algal growth rate, algal mortality rate, submerged plant nitrogen absorption rate, and submerged plant phosphorus absorption rate.
[0054] In step S300 of this embodiment, the SCE-UA global optimization algorithm is used to globally optimize the above 12 core parameters. It combines the advantages of simplex method, random search, biological competitive evolution and mixed partitioning, and can effectively and quickly search for the global optimal solution of the hydrological model, and output the optimal value and 95% confidence interval of each parameter.
[0055] The specific execution flow and quantization parameters of the algorithm are as follows: 1) Construction of the objective function: Using the Nash efficiency coefficient (NSE) as the main objective function and the root mean square error (RMSE) as the auxiliary constraint objective, the expression is as follows: in, For the first The measured value at time [time]. For the first Simulated value at time, This is the average of the measured values. This represents the number of measured data points.
[0056] The closer the objective function is to 1, the higher the agreement between the simulation results and the measured data; when and When the constraints are met, the parameter calibration results are considered acceptable.
[0057] 2) Specific execution steps of the SCE-UA algorithm: Parameter space initialization: Based on the initial value range of each parameter, an initial population is generated using Latin hypercube sampling. The population size is set to 7 times the number of parameters to be calibrated to ensure that the initial population uniformly covers the entire parameter space and avoids local optima. In this embodiment, it is 12. 7 = 84 individuals.
[0058] Population sorting and complex partitioning: The 84 individuals in the initial population are sorted from best to worst according to the objective function value and divided into 6 complexes, each containing 14 individuals; the number of complexes is set to 1 / 2 of the number of parameters to balance the global search capability and convergence speed of the algorithm.
[0059] Complex evolutionary operations: Simplex operations are performed sequentially on each complex to generate new individuals. Simplex operations include: The reflection operation takes the center points of the three best individuals in the complex and reflects them in a direction away from the worst individual to generate reflection points.
[0060] The expansion operation involves further expanding along the reflection direction if the objective function value of the reflection point is better than the current best individual, thus generating an expansion point. The contraction operation involves shrinking the target function value of the reflection point towards the center point if the target function value is worse than the worst individual. The best new individual generated is used to replace the individual with the worst objective function value in the complex, thus completing one round of complex evolution.
[0061] Global competition and recombination: After all 6 complexes have completed one evolution, all 84 individuals are merged again, sorted again according to the objective function value, and 6 new complexes are re-divided, and the evolutionary process of step 3 is repeated.
[0062] Convergence Criteria and Termination Conditions: The algorithm terminates automatically when any of the following conditions are met: After 60 consecutive iterations, the change in the optimal value of the objective function is less than 10⁻⁵. The total number of iterations reached 600; Optimal value of objective function 0.85 and RMSE satisfies the constraints.
[0063] Finally, the results output and uncertainty analysis: the optimal values of the 12 core parameters to be calibrated corresponding to the optimal value of the objective function are output, and the 95% confidence interval of each core parameter to be calibrated is calculated using the Bootstrap sampling method (sampling times 1000 times) to quantify the uncertainty of the calibration results.
[0064] Furthermore, in step S400, the process of converting the mesocosmic-scale parameters into whole-lake-scale parameters is as follows: Step S401: Calculate the core scale factor between the space experiment unit and the target lake, including the water depth ratio. Area ratio Compared with hydraulic residence time They are represented as follows: = , The average water depth; = , The water surface area; = , The mean hydraulic residence time is set by a controllable hydrological exchange system for the cosmic unit.
[0065] The mesocosmic-scale parameters are converted to whole-lake-scale parameters using the following transformation equation: in, Indicates the water depth ratio. Indicates the area ratio. This indicates the hydraulic residence time ratio.
[0066] Step S402: Based on the physical meaning and scale dependence of the model parameters, the parameters to be transformed are divided into three categories, and transformation relationships are established for each category. The parameters to be transformed are further divided into parameters to be transformed and parameters that do not need to be transformed.
[0067] Figure 4 This is a comparison table showing the classification, scale dependence, and transformation relationship analysis of the parameters to be transformed provided in Embodiment 1 of the present invention. For example... Figure 4 As shown, in this embodiment, the model parameters are divided into physical parameters, chemical parameters, and biological parameters. Among the core physical parameters, the roughness coefficient, diffusion coefficient, and reoxygenation coefficient are mainly related to water depth and flow velocity, and their conversion relationship is related to a power function of the water depth ratio. Among the core chemical parameters, the nutrient degradation coefficient and adsorption-desorption coefficient are mainly related to temperature and concentration, and their conversion relationship does not change with scale; therefore, the values of the mesocosmic parameters are directly used. Among the core biological parameters, the algal growth rate, mortality rate, and aquatic plant absorption rate are mainly related to hydraulic residence time and light intensity, and their conversion relationship is related to a power function of the hydraulic residence time ratio.
[0068] Step S403: Construct a transformation equation. This transformation equation is based on the mesocosmic-scale parameter values corresponding to the core scale factor and core parameters. Calculate the values of the parameters to be transformed as the whole-lake-scale parameter values.
[0069] Construct a general transformation equation: .
[0070] Example as follows: Horizontal diffusion coefficient: = ; Reoxygenation coefficient: = ; Maximum growth rate of algae: = ; No parameter transformation is needed; the corresponding mesocosmic-scale parameter values can be directly used as the whole-lake-scale parameter values. An example is shown below: ammonia nitrogen nitrification coefficient: = .
[0071] Step S404: Verify and iteratively correct the transformation relationship. This step can be executed in conjunction with the full lake model calibration and verification step in step S600.
[0072] Furthermore, in step S500 of this embodiment, it is preferable to process the seven sets of cosmological-scale parameter values through a process of: quality screening, weighted integration, unified transformation, and iterative verification, ultimately obtaining a unique set of full-lake model parameters. The specific steps are as follows: Parameter quality screening: The accuracy of the cosmological parameters in the 7 groups was evaluated, and the experimental group parameters that did not meet the calibration accuracy (NSE<0.85 or RMSE did not meet the constraints) were removed, and the valid parameter groups were retained.
[0073] Weighted integration of multiple parameter sets: The remaining effective parameter sets are weighted and averaged to obtain a comprehensive set of mesocosmic-scale parameters. Weight allocation principle: positively correlated with the calibration accuracy (NSE value) of each experimental group and positively correlated with the lake-wide representativeness of the intervention measures (e.g., the weight of the combined measure group is higher than that of the single measure group).
[0074] Unified scale transformation: Based on the transformation relationship established in step S400 (distinguishing between parameters to be transformed and parameters that do not need to be transformed, and applying the corresponding transformation equations), the cosmological parameters in the above synthesis are transformed into a set of lake-scale parameters.
[0075] Iterative verification and optimization: Input the transformed whole-lake parameters into the whole-lake model and verify the accuracy using at least 12 months of historical monitoring data of the whole lake; if the requirements are not met (NSE < 0.7, R... 2 If the error is <0.75 and RE>20%, then the transformation equation is corrected according to the error characteristics. The transformation and verification steps are repeated until a unique set of qualified whole-lake parameters is obtained. Using this weighted integration method, the complementarity of multi-condition experimental data can be fully utilized to improve the robustness of whole-lake parameters.
[0076] It should be noted that the basic framework and core structure of the cosmological model in the seven experimental units of this embodiment are exactly the same, but the boundary conditions, initial conditions, and optimal parameter values obtained from the final calibration differ: Different boundary conditions: The influent and effluent flow rates of each experimental unit are independently controlled by a PLC controller (simulating different hydraulic exchange intensities), corresponding to different hydrodynamic boundary conditions; Different initial conditions: The measured water quality data of each experimental unit before intervention were used as the initial conditions for their respective models; Different optimal values of parameters: Due to the application of differentiated engineering interventions (aeration, sediment covering, planting of aquatic plants, etc.), the aquatic ecosystem response characteristics of each experimental unit are different. The optimal values and 95% confidence intervals of the 12 core parameters obtained by SCE-UA algorithm calibration are different (for example, the optimal value of the reoxygenation coefficient of the aeration group is significantly higher than that of the sediment covering group).
[0077] The mesocosmic model and the whole-lake model adopt the same core model framework and core module structure, but allow for adaptation modifications to suit mesocosmic scale characteristics, without needing to be completely identical. That is, they must be based on the same type of aquatic ecological dynamics model framework and have identical core modules such as hydrodynamic, water quality, and aquatic ecology modules. The physical definitions of core state variables and parameters must be consistent to achieve the derivation of scale dependence between the two based on scale factors. Simultaneously, for the small-scale characteristics of the mesocosmic unit, some modules can be locally adjusted, such as modifying the empirical formula for the horizontal diffusion coefficient and simplifying the large aquatic animal / benthic organism module.
[0078] Furthermore, the lake water ecological governance model in this embodiment is not limited to the EFDC three-dimensional surface water quality mathematical model, but can also be the CE-QUAL-W2, WASP, MIKE series and other lake water ecological models.
[0079] Furthermore, step S600 specifically includes: Step S601, Accuracy Verification Indicators and Judgment Criteria: The simulation results were compared with historical monitoring data of the entire lake, covering at least one seasonal variation, obtained through at least 12 monthly monitoring sessions, to calculate the accuracy indicators of the core parameters: NSE, RMSE and RE, all core parameters satisfy: NSE 0.7 0.75, RE 20.
[0080] Step S602: Iterative update of the transformation relationship. If the accuracy index meets the preset requirements, the model parameters of the whole lake water ecological dynamics model are qualified. If the accuracy index does not meet the preset requirements, the model parameters of the whole lake water ecological dynamics model are unqualified. Further adjust the transformation equation according to the error characteristics of the accuracy index, and use the adjusted transformation equation to enter step S500 until all accuracy indices meet the preset requirements.
[0081] If the verification result does not meet the passing standard, the conversion relationship is modified accordingly based on the error characteristics, including the following steps: If the simulated values are generally too high / too low: adjust the constant term of the transformation equation. For example, ... = Revised to: = .
[0082] If the simulation error is large in a particular season: adjust the scale factor index of the corresponding parameter. For example, during the summer when algae growth is vigorous, the index can be adjusted from... , revised to .
[0083] If a certain water quality parameter has a large error: adjust the conversion equation coefficient of that parameter alone, while keeping other parameters unchanged.
[0084] Repeat the full lake simulation and verification steps until all accuracy indicators meet the requirements.
[0085] Specifically, an example of a conversion relationship: Taking the maximum growth rate of algae as an example, the initial conversion equation is: = If the verification shows that the simulated Chl-a value for the entire lake is on average higher than the measured value, then the exponent is adjusted to -0.4, resulting in the corrected conversion equation. = 4. After resimulation, Chl's RE dropped to 18%, meeting the accuracy requirements.
[0086] Example 2 This second embodiment proposes a specific implementation method for correcting the relevant aspects in step S300 above.
[0087] 1. Correct the hydrodynamic module Specifically, the empirical formula for the horizontal diffusion coefficient in the traditional EFDC model is based on large-scale mixing dominated by wind-driven currents, and takes the form: in, This is an empirical coefficient. Because of the water depth, The velocity is the frictional velocity.
[0088] In this embodiment, to address the inlet and outlet water disturbances of the open unit, the above formula is modified as follows: 1.1) Adding a correction term for influent and effluent flow rates: A correction term positively correlated with the influent and effluent flow rates is introduced into the original formula to reflect the contribution of the influent and effluent shear flow to mixing. The corrected formula is as follows: in, This represents the total influent and effluent flow rate of the unit. For unit water surface area, The correction coefficients were obtained through calibration using preliminary experimental data. These preliminary experiments refer to small-scale preparatory experiments conducted before the formal implementation of the seven-unit medium-universe experiment. These experiments only acquire preliminary data to determine the initial range of parameters and do not participate in the final model calibration.
[0089] 1.2) Adjusting the range of empirical coefficients: Based on the measured velocity field data of the middle cosmic unit, the original empirical coefficients were adjusted... The value range has been adjusted from 0.1-0.5 applicable to large lakes to 0.3-0.8, to better suit the strong mixing characteristics of small-scale water bodies; 1.3) Unsteady-state correction: Considering the periodic fluctuations in influent and effluent flow rates, Set as a time-varying variable, linked to real-time inflow and outflow water flow.
[0090] 2. Simplify the aquatic ecosystem module Specifically, based on the biological community characteristics of the middle cosmic unit, this embodiment simplifies the large aquatic animal and benthic organism modules in the aquatic ecosystem module by "removing non-core groups + process integration", thereby reducing the model complexity without significantly affecting the simulation accuracy.
[0091] 2.1) Simplification of the large aquatic animal module: Completely remove state variables and related process equations (including feeding, excretion, death and decomposition, biological disturbance, etc.) of higher aquatic animals such as fish and large crustaceans.
[0092] The basis for this is that the inlet and outlet of the open unit are equipped with a 100μm double-layer filter, which can effectively filter large plankton and fish. Moreover, the experimental period is relatively short (30-90 days), and even if a small number of small organisms enter, their contribution to the nutrient cycle is negligible.
[0093] 2.2) Simplification of the benthic organism module: We do not set separate state variables for benthic organisms (such as benthic biomass), nor do we simulate their growth, metabolism, feeding, and other processes separately. The impact of benthic organism activity on nutrient cycling (such as phosphorus release from sediment and ammonia nitrogen excreted due to benthic disturbance) is integrated into the nutrient exchange flux at the sediment-water interface, and its impact is reflected by calibrating comprehensive parameters such as sediment phosphorus release coefficient and ammonia nitrogen release coefficient.
[0094] After simplification, the aquatic ecosystem module retains only three core biological components: phytoplankton (cyanobacteria, green algae, diatoms), submerged plants, and zooplankton. The number of parameters to be calibrated is reduced by about 40%, which significantly improves the efficiency of model calibration.
[0095] 3. The water quality module does not require structural modifications.
[0096] The nutrient migration and transformation processes simulated by the water quality module (adsorption and desorption, nitrification and denitrification, sediment phosphorus release, reoxygenation, etc.) are physicochemical processes, and their mechanisms are scale-invariant, completely consistent at both the mesocosmic and whole-lake scales.
[0097] To ensure the feasibility of parameter conversion, the core state variables of the water quality module (TN, TP, ammonia nitrogen, nitrate nitrogen, dissolved orthophosphate, DO) are exactly the same as those of the whole lake model, and no additions or subtractions are made.
[0098] The values of empirical parameters in the water quality module (such as ammonia nitrogen nitrification coefficient, denitrification coefficient, phosphorus adsorption coefficient, and sediment phosphorus release coefficient) are closely related to the specific water quality characteristics, sediment properties, and temperature conditions of the target lake, and cannot be directly adopted from general literature. Therefore, this invention uses multi-condition data from the Zhongyuzhou Experiment to accurately calibrate these parameters to adapt them to the actual conditions of the target lake.
[0099] The lake water ecological governance model calibration scheme based on the experimental data of the Middle Cosmic Space, designed according to this invention, has a narrower confidence interval and higher reliability. The calibrated model can be used for simulation and deduction of different governance scenarios, providing support for scientific decision-making.
[0100] Example 3 This embodiment proposes a calibration device for a lake water ecological governance model based on data from the Middle Cosmic Experiment, which executes some of the apple-like methods described in Embodiment 1. The device includes: The data acquisition and preprocessing module is configured to acquire water quality parameter data collected from multiple experimental units, as well as synchronously collected meteorological and hydrological data, and preprocess the acquired data; wherein, the experimental unit is an open-type mesocosmic experimental unit deployed in the target lake; differentiated engineering intervention measures are applied in each experimental unit; the engineering intervention measures include at least one of the following: aeration and oxygenation, bottom sediment covering, aquatic plant planting, and flocculant addition; The mesocosmic model construction and calibration module is configured to construct mesocosmic-scale water ecological dynamics models for each experimental unit. The water quality data before the experimental intervention is used as the initial conditions for each mesocosmic-scale water ecological dynamics model, and the water quality data after the intervention is used as the objective function. An automatic calibration algorithm is used to optimize the model parameters and output the optimal values and confidence intervals of the mesocosmic-scale parameters. The conversion relationship establishment module is configured to establish a conversion relationship between mesocosmic-scale parameters and whole-lake-scale parameters, converting mesocosmic-scale parameters into whole-lake-scale parameters. The whole-lake model construction module is configured to construct a whole-lake aquatic ecosystem dynamics model, and inputs the whole-lake scale parameters obtained by the transformation relationship establishment module into the whole-lake aquatic ecosystem dynamics model. The verification and iteration module is configured to use the whole-lake aquatic ecosystem dynamics model determined by the whole-lake model construction module to simulate the whole-lake aquatic ecosystem parameters, compare and verify the simulation results with the whole-lake historical monitoring data, calculate the model accuracy index, and iteratively correct the transformation relationship based on the verification results.
[0101] Another embodiment of the present invention proposes an electronic device, which is a device with data processing and / or communication functions. The electronic device is configured with the lake water ecological governance model calibration device based on the data of the Middle Cosmic Experiment described in Embodiment 2. The electronic device can be a computer, mobile phone or tablet computer.
[0102] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0103] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.
Claims
1. A calibration method for a lake water ecological governance model based on data from the Middle Cosmic Experiment, characterized in that, Including the following steps: S100. Multiple experimental units are set up in the target lake, and differentiated engineering intervention measures are applied to each experimental unit; wherein, the experimental unit is an open-type mesocosmic experimental unit, and the engineering intervention measures include at least one of the following: aeration and oxygenation, bottom sediment covering, aquatic plant planting, and flocculant addition. S200: Collect water quality parameter data from each experimental unit, simultaneously collect meteorological and hydrological data, and preprocess the collected data. S300. Construct a mesocosmic-scale water ecological dynamics model for each experimental unit. Use the water quality data before the experimental intervention as the initial condition for each mesocosmic-scale water ecological dynamics model, and use the water quality data after the intervention as the objective function. Use an automatic calibration algorithm to optimize the model parameters and output the optimal values and confidence intervals of the mesocosmic-scale parameters. S400. Establish the conversion relationship between mesocosmic-scale parameters and whole-lake-scale parameters, and convert the mesocosmic-scale parameters into whole-lake-scale parameters. S500. Construct a full-lake water ecological dynamics model and input the full-lake scale parameters into the full-lake water ecological dynamics model. S600. Use the whole lake water ecological dynamics model to simulate the whole lake water ecological parameters, compare and verify the simulation results with the whole lake historical monitoring data, calculate the model accuracy index, and iteratively correct the conversion relationship based on the verification results.
2. The lake ecological governance model calibration method based on the data from the Middle Cosmic Experiment as described in claim 1, characterized in that: In step S100, at least 6 experimental units are set up in the target lake, and all experimental units are divided into at least 1 control group and at least 5 experimental groups. Each experimental unit is equipped with a controllable inlet and a controllable outlet, and each outlet is equipped with an electric regulating gate, an electromagnetic flow meter, and a filter with a preset aperture.
3. The lake ecological governance model calibration method based on the data from the Middle Cosmic Experiment as described in claim 1, characterized in that: In step S200, water quality parameter data for each experimental unit are collected at least once a day. The water quality parameter data includes total nitrogen (TN), total phosphorus (TP), dissolved oxygen (DO), chlorophyll a, turbidity, and pH.
4. The lake ecological governance model calibration method based on the data from the Middle Cosmic Experiment as described in claim 1, characterized in that, Step S300 specifically includes: The water quality data of each experimental unit before and after the application of engineering intervention measures were used as the initial conditions and target values of the corresponding mesocosmic-scale aquatic ecosystem dynamics model. Based on the model framework of the mesocosmic-scale water ecosystem dynamics model, and combined with the intervention type of the mesocosmic experiment, several core parameters to be calibrated were selected. Based on empirical values and experimental data, the initial value range of each core parameter is determined; The SCE-UA global optimization algorithm is used to globally optimize the core parameters, and the optimal values and 95% confidence intervals of each core parameter are output.
5. The lake ecological governance model calibration method based on the data from the Middle Cosmic Experiment as described in claim 4, characterized in that, Step S400 specifically includes: Calculate the core scale factors between the experimental unit and the target lake, including the water depth ratio, area ratio, and hydraulic residence time ratio; Based on the physical meaning of the core parameters and the scale dependence between the core parameters and the core scale factor, the core parameters are further divided into parameters to be converted and parameters that do not need to be converted. A transformation equation is constructed, which is based on the mesocosmic-scale parameter values corresponding to the core scale factor and core parameters. The values of the parameters to be transformed are then used as the full-lake-scale parameter values. Without needing to transform the parameters, the corresponding mesocosmic-scale parameter values can be directly used as the whole-lake-scale parameter values.
6. The lake ecological governance model calibration method based on the data from the Middle Cosmic Experiment as described in claim 5, characterized in that: In steps S300 and S500, the mesocosmic-scale water ecological dynamics model and the whole lake water ecological dynamics model both adopt the EFDC three-dimensional surface water quality mathematical model, which couples the hydrodynamic module, water quality module and water ecology module. The parameters to be converted include: roughness coefficient, diffusion coefficient, reoxygenation coefficient, algal growth rate, algal mortality rate, aquatic plant absorption rate, and ammonia nitrogen nitrification coefficient; the parameters that do not need to be converted include: nutrient degradation coefficient and adsorption / desorption coefficient. The transformation equation is expressed as: ,in, Indicates the water depth ratio. Indicates the area ratio. Indicates the hydraulic residence time ratio; In step S500, the accuracy of the mesocosmic scale parameter values corresponding to each mesocosmic experimental unit is evaluated, experimental group parameters whose calibration accuracy does not meet the preset standard are eliminated, and the experimental group parameters whose calibration accuracy meets the preset standard are weighted and averaged to obtain a set of comprehensive mesocosmic scale parameters. Then, based on the conversion relationship established in step S400, the comprehensive mesocosmic parameters are converted into lake-scale parameters.
7. The lake ecological governance model calibration method based on the data from the Middle Cosmic Experiment as described in claim 5, characterized in that, Step S600 specifically includes: The simulation results are compared with historical monitoring data of the entire lake obtained through at least 12 monthly monitoring sessions. The accuracy indicators of the core parameters are calculated, and it is determined whether they meet the preset requirements. The accuracy indicators include the Nash efficiency coefficient (NSE) and the coefficient of determination. At least one of the following: root mean square error (RMSE) and relative error (RE); If the accuracy index meets the preset requirements, then the model parameters of the whole lake water ecological dynamics model are qualified; If the accuracy index does not meet the preset requirements, the model parameters of the whole lake water ecological dynamics model are unqualified. Further adjust the transformation equation according to the error characteristics of the accuracy index, and return to step S500 using the adjusted transformation equation until all accuracy indexes meet the preset requirements.
8. A calibration device for a lake water ecological governance model based on data from the Middle Cosmic Experiment, characterized in that, include: The data acquisition and preprocessing module is configured to acquire water quality parameter data collected from multiple experimental units, as well as synchronously collected meteorological and hydrological data, and preprocess the acquired data; wherein, the experimental unit is an open-type mesocosmic experimental unit deployed in the target lake; differentiated engineering intervention measures are applied in each experimental unit; the engineering intervention measures include at least one of the following: aeration and oxygenation, bottom sediment covering, aquatic plant planting, and flocculant addition; The mesocosmic model construction and calibration module is configured to construct mesocosmic-scale water ecological dynamics models for each experimental unit. The water quality data before the experimental intervention is used as the initial conditions for each mesocosmic-scale water ecological dynamics model, and the water quality data after the intervention is used as the objective function. An automatic calibration algorithm is used to optimize the model parameters and output the optimal values and confidence intervals of the mesocosmic-scale parameters. The conversion relationship establishment module is configured to establish a conversion relationship between mesocosmic-scale parameters and whole-lake-scale parameters, converting mesocosmic-scale parameters into whole-lake-scale parameters. The whole-lake model construction module is configured to construct a whole-lake aquatic ecosystem dynamics model, and inputs the whole-lake scale parameters obtained by the transformation relationship establishment module into the whole-lake aquatic ecosystem dynamics model. The verification and iteration module is configured to use the whole-lake aquatic ecosystem dynamics model determined by the whole-lake model construction module to simulate the whole-lake aquatic ecosystem parameters, compare and verify the simulation results with the whole-lake historical monitoring data, calculate the model accuracy index, and iteratively correct the transformation relationship based on the verification results.
9. An electronic device, characterized in that: The electronic device is a device with data processing and / or communication functions, and the electronic device is equipped with the lake water ecological governance model calibration device based on the data from the Middle Cosmic Experiment as described in claim 8.
10. The electronic device as claimed in claim 9, characterized in that: The electronic device is a computer, mobile phone, or tablet computer.