Constructed wetland space optimization arrangement method and system

Abstract

The invention discloses an artificial wetland space optimization arrangement method and system, a non-point source pollution model is constructed based on a hydrological response unit, target meteorological data, target pollution data and an SWAT model, and the simulation precision of the non-point source pollution process can be improved. Climate scene data is constructed through future climate data, a preset path rule and a simple quantile algorithm, and the influence of climate change on pollution load can be reflected. Furthermore, a wetland optimization model is constructed based on a non-point source pollution model and a target function, the scientificity of an optimization strategy is enhanced, candidate wetland areas are identified by a spatial analysis algorithm based on a complex terrain index CTI, and the maximum potential area of each candidate wetland area is counted, so that efficient identification of potential arrangement areas can be realized. The constructed wetland optimization regulation and control scheme is generated based on the preset genetic algorithm, and the constructed wetland space optimization arrangement is performed according to the scheme, so that the optimization and practicability of the constructed wetland layout can be realized.

Description

Technical Field

[0001] This invention relates to the field of wetland optimization technology, and in particular to a method and system for optimizing the spatial layout of artificial wetlands. Background Technology

[0002] Climate change has exacerbated the frequency and intensity of extreme precipitation events, significantly altering watershed hydrological processes and thus intensifying the export of non-point source pollution, accelerating eutrophication and ecosystem degradation, affecting approximately 4 billion people annually, and leading to severe water shortages. Constructed wetlands, as a widely used engineering best practice, not only possess continuous water purification capabilities on a multi-year scale but also offer multiple synergistic benefits such as ecological restoration and water regulation. Therefore, they are considered a crucial measure for controlling watershed-scale non-point source pollution under the influence of climate change.

[0003] When deploying constructed wetlands at the watershed scale, it is necessary to comprehensively weigh multiple factors, including pollutant reduction efficiency, land availability, and long-term operation and maintenance costs. In the context of climate change, the variability of hydrological processes has increased, the spatial distribution of non-point source pollution loads exhibits dynamic changes, and future climate scenarios themselves are highly uncertain. These factors further increase the complexity of constructed wetland design and site selection. Furthermore, constructed wetlands typically have a long design life. As the spatial distribution pattern of non-point source pollution changes in the future, the spatial priority and cost of constructed wetlands will also change. This means that the optimal configuration developed under current conditions may lose its original advantages in the future and be unable to cope with the challenges posed by future agricultural non-point source pollution. At the same time, climate change may also directly affect the governance function of constructed wetlands. For example, an increase in drought events will shorten the effective water retention time of wetlands, reducing their water purification capacity; extreme rainfall may generate high-intensity runoff in a short period of time, exceeding the wetland's treatment capacity, leading to pollutant bypassing.

[0004] Existing methods for optimizing the spatial layout of constructed wetlands, while employing simulation-optimization (SOM) frameworks and heuristic algorithms such as genetic algorithms and NSGA-II to seek a balance between environmental benefits and economic costs, still have shortcomings: First, the boundaries of decision variables lack scientific basis, often failing to fully consider the topography, soil, and water accumulation conditions required for wetland site selection, leading to physical or environmental mismatch issues in actual deployment. Second, although GIS-based composite topographic indices (CTI) can identify potential wetland candidate areas, most current studies only use them for location constraints, without incorporating the evolution of precipitation, evapotranspiration, and hydrological processes under future climate change scenarios into the optimization process. Third, existing studies mostly focus on assessing the impact of climate change on non-point source pollution and the effectiveness of wetland interception, but rarely couple future climate scenarios such as CMIP6-SSP with wetland layout optimization, neglecting the differences in the long-term performance and adaptability of constructed wetlands under different shared socio-economic pathways. Summary of the Invention

[0005] This invention provides a method and system for optimizing the spatial layout of constructed wetlands, thereby improving the scientific nature, feasibility, and sustainable adaptability of constructed wetland configuration.

[0006] To address the aforementioned technical problems, this invention provides a method for optimizing the spatial layout of constructed wetlands, comprising: Obtain the region to be optimized and its basic information, and divide the region to be optimized into several hydrological response units based on the basic information; Acquire meteorological data and pollution data, and preprocess the meteorological data, point source pollution data and non-point source pollution data respectively to obtain target meteorological data and target pollution data; A non-point source pollution model is constructed based on the aforementioned hydrological response unit, target meteorological data, target pollution data, and STAW model; Acquire future climate data, and construct climate scenario data based on the future climate data, preset path rules, and simple quantile algorithm; Construct an objective function, and then construct a wetland optimization model based on the non-point source pollution model and the objective function. Candidate wetland areas were identified based on GIS spatial analysis algorithms, land use data, soil infiltration rate, and composite topographic index, and the maximum potential area of ​​each candidate wetland area was calculated. An artificial wetland optimization and control scheme is generated based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data. The artificial wetland is then spatially optimized and arranged based on the artificial wetland optimization and control scheme.

[0007] This invention acquires the region to be optimized and its basic information, and divides the region into several hydrological response units based on this information. This enables regional division at a geographic scale, providing a clear unit basis for subsequent simulations. Furthermore, by acquiring meteorological and pollution data, and preprocessing the meteorological, point source pollution, and non-point source pollution data to obtain target meteorological and target pollution data, it helps improve the completeness and consistency of the input data. Constructing a non-point source pollution model based on the hydrological response units, target meteorological data, target pollution data, and the STAW model improves the simulation accuracy of non-point source pollution processes. Acquiring future climate data and constructing climate scenario data based on this future climate data, preset path rules, and a simple quantile algorithm helps reflect the impact of climate change on pollution loads. Moreover, by constructing a wetland optimization model based on the non-point source pollution model and the objective function, the simulation results can be coupled with the governance objectives, enhancing the scientific nature of the optimization strategy. By identifying candidate wetland areas based on GIS spatial analysis algorithms, land use data, soil infiltration rate, and composite topographic indices, and calculating the maximum potential area of ​​each candidate wetland area, efficient identification of potential deployment areas can be achieved. Finally, based on the preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, an optimized control scheme for artificial wetlands is generated, and the spatial layout of artificial wetlands is optimized accordingly. This is conducive to achieving the optimization and practicality of artificial wetland layout, thereby improving the scientificity, feasibility, and sustainable adaptability of artificial wetland configuration.

[0008] Furthermore, the basic information includes a digital elevation model, land use data, and soil type data; the process of acquiring the area to be optimized and its basic information, and dividing the area to be optimized into several hydrological response units based on the basic information, includes: Acquire the area to be optimized, its digital elevation model, land use data, and soil type data; Based on the digital elevation model, land use data, and soil type data, the area to be optimized is divided into several hydrological response units.

[0009] This invention clarifies the source of geographic information upon which regional division depends by specifically defining the basic information as digital elevation models, land use data, and soil type data. Furthermore, by acquiring the region to be optimized and its digital elevation model, land use data, and soil type data, and dividing the region to be optimized into several hydrological response units based on the above three types of data, the objectivity and consistency of the division process can be improved, making the division results closer to the actual terrain and land attributes, and facilitating subsequent pollution simulation and layout assessment.

[0010] Furthermore, the acquisition of meteorological data and pollution data involves preprocessing the meteorological data, point source pollution data, and non-point source pollution data to obtain target meteorological data and target pollution data, respectively, including: Acquire meteorological data, fill in missing values ​​and format the meteorological data to generate target meteorological data; Acquire pollution data, including point source pollution data and non-point source pollution data; The point source pollution emissions are calculated based on the point source pollution data, and the point source pollution emissions are converted according to a preset format to obtain the target point source pollution data. Based on preset farmland management measures and the non-point source pollution data, fertilization event sequences and irrigation time sequences are obtained, and target non-point source pollution data are generated based on the fertilization event sequences and irrigation time sequences. Target pollution data is generated based on the target point source pollution data and the target non-point source pollution data.

[0011] This invention acquires meteorological data, performs missing value imputation and formatting on the meteorological data to generate target meteorological data, effectively solving the problems of incomplete and inconsistent structure of the original meteorological data. Simultaneously, it acquires pollution data, including point source pollution data and non-point source pollution data, ensuring the comprehensiveness of pollution sources. Based on the point source pollution data, it calculates the point source pollution emissions and converts the emissions according to a preset format to obtain target point source pollution data, which helps improve the model adaptability of point source pollution data. By acquiring fertilization event sequences and irrigation time sequences based on preset farmland management measures and the non-point source pollution data, it generates target non-point source pollution data, making the non-point source pollution information more consistent with actual agricultural activities. Finally, based on the target point source pollution data and target non-point source pollution data, it generates target pollution data, providing unified, formatted, and complete input data for the non-point source pollution model.

[0012] Furthermore, the construction of the non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and STAW model includes: An initial non-point source pollution model is constructed based on the aforementioned hydrological response unit, target meteorological data, target pollution data, and STAW model. Historical observed flow rates and historical pollution concentrations are collected, and the initial non-point source pollution model is calibrated based on the historical observed flow rates and historical pollution concentrations to obtain the non-point source pollution model.

[0013] This invention constructs an initial non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and STAW model, enabling the establishment of a simulation mechanism based on multi-source data. Furthermore, it collects historical observed flow rates and historical pollution concentrations, and calibrates the initial non-point source pollution model based on these data to obtain a true non-point source pollution model. This makes the simulation results more closely resemble actual hydrological and pollution processes, improving the reliability and interpretability of the model output.

[0014] Furthermore, the future climate data includes raw daily precipitation data and temperature grid data corresponding to several GCMs, as well as locally observed climate sequences; the acquisition of future climate data, and the construction of climate scenario data based on the future climate data and preset path rules, includes: Acquire raw daily precipitation data and temperature grid data corresponding to several GCMs, as well as local meteorological observation sequences; Climate data for each GCM is generated based on the raw daily precipitation data and temperature grid data corresponding to each GCM. Climate-driven data is generated based on preset path rules and the climate data. The climate data is downscaled based on the Delta variability method and the local observed climate sequence to obtain future climate sequences. Pollution load time series are generated based on the simple quantile algorithm and the future climate sequence corresponding to each GCM. Climate scenario data are constructed based on the climate-driven data and the pollution load time series.

[0015] This invention ensures the breadth and local adaptability of future climate data by acquiring raw daily precipitation and temperature grid data corresponding to several Global Climate Change Models (GCMs) and local meteorological sequences. Climate data corresponding to each GCM is generated based on the raw daily precipitation and temperature grid data, and climate driving data is generated based on preset path rules and the climate data, reflecting climate responses under different scenarios. Furthermore, the climate data is downscaled using the Delta variability method and the local observed climate sequences to obtain future climate sequences, improving the spatial resolution of the climate data. Pollution load time series are generated based on the simple quantile algorithm and the future climate sequences corresponding to each GCM, and climate scenario data is constructed based on the climate driving data and the pollution load time series, providing diverse and reliable future scenario inputs for the model.

[0016] Furthermore, the identification of candidate wetland areas based on GIS spatial analysis algorithms, land use, soil infiltration rate, and composite topographic indices, and the calculation of the maximum potential area of ​​each candidate wetland area, includes: Obtain land use data and soil infiltration rate of the area to be optimized; Based on the land use data, farmland data was selected as the first candidate area; Based on the soil infiltration rate, areas in the region to be optimized that are lower than the preset infiltration threshold are eliminated to obtain a second candidate region; Calculate the composite terrain index of the area to be optimized, and select areas whose composite terrain index is greater than a preset index threshold as the third candidate area; Based on GIS spatial analysis algorithms, the first candidate region, the second candidate region, and the third candidate region are overlaid to determine the candidate wetland regions, and the maximum potential area of ​​each sub-basin is calculated.

[0017] This invention acquires land use data and soil infiltration rate of the area to be optimized, and selects farmland data as the first candidate area based on the land use data, thus prioritizing areas with concentrated agricultural non-point source pollution. Based on the soil infiltration rate, areas below a preset infiltration threshold are eliminated to obtain the second candidate area, which helps ensure the actual infiltration function of the constructed wetland. By calculating the composite topographic index of the area to be optimized and selecting areas above a preset index threshold as the third candidate area, the runoff potential of the area can be further improved. Finally, based on GIS spatial analysis algorithms, the above three types of areas are overlaid to determine the candidate wetland areas, and the maximum potential area of ​​each sub-basin is calculated, which effectively defines the layout boundary and provides area constraints for the optimization model.

[0018] Furthermore, the construction of the wetland optimization model based on the non-point source pollution model and the preset objective function includes: A target function is constructed based on cost parameters and the output parameters of the non-point source pollution model. A wetland optimization model is constructed based on the objective function and the non-point source pollution model.

[0019] This invention constructs an objective function based on cost parameters and the output parameters of the non-point source pollution model, enabling the optimization model to strike a balance between pollution reduction targets and costs. Furthermore, by constructing a wetland optimization model based on the objective function and the non-point source pollution model, pollution simulation and resource allocation processes can be closely integrated, thereby improving the practicality and controllability of the optimization process and providing clear quantitative basis for subsequent optimization layout.

[0020] Furthermore, the step of generating an artificial wetland optimization and control scheme based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, and then optimizing the spatial layout of the artificial wetland based on the artificial wetland optimization and control scheme, includes: The wetland optimization model is used as the fitness function of the genetic algorithm, the maximum potential area of ​​each candidate wetland region is used as the decision variable of the genetic algorithm, and the candidate wetland information and the climate scenario data are used as the input data of the genetic algorithm. Based on the genetic algorithm, an artificial wetland optimization and control scheme is generated. Based on the aforementioned constructed wetland optimization and control scheme, the spatial layout of constructed wetlands is optimized.

[0021] This invention constructs a multi-variable, multi-constraint optimization search framework by using the wetland optimization model as the fitness function of the genetic algorithm, the maximum potential area of ​​each candidate wetland region as the decision variable of the genetic algorithm, and the candidate wetland information and climate scenario data as the input data of the genetic algorithm. The generation of artificial wetland optimization and control schemes based on the genetic algorithm helps to quickly converge to near-optimal solutions. Furthermore, the spatial optimization and layout of artificial wetlands based on the artificial wetland optimization and control scheme ensures that the final layout is operable and scientific, adaptable to multiple scenario requirements.

[0022] In a second aspect, the present invention provides an artificial wetland spatial optimization layout system, comprising: a unit division module, a data processing module, a non-point source construction module, a climate construction module, an optimization model construction module, a wetland identification module, and an optimization layout module; The unit division module is used to obtain the region to be optimized and its basic information, and to divide the region to be optimized into several hydrological response units based on the basic information. The data processing module is used to acquire meteorological data and pollution data, and to preprocess the meteorological data, point source pollution data and non-point source pollution data to obtain target meteorological data and target pollution data respectively. The non-point source construction module is used to construct a non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and SWAT model. The climate construction module is used to acquire future climate data and construct climate scenario data based on the future climate data, preset path rules, and a simple quantile algorithm. The optimization model construction module is used to construct an objective function and to construct a wetland optimization model based on the non-point source pollution model and the objective function. The wetland identification module is used to identify candidate wetland areas based on composite topographic index, land use data and soil infiltration rate, and to calculate the maximum potential area of ​​each candidate wetland area. The optimization layout module is used to generate an artificial wetland optimization and control scheme based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, and to optimize the spatial layout of artificial wetlands based on the artificial wetland optimization and control scheme.

[0023] Furthermore, the basic information includes a digital elevation model, land use data, and soil type data; the unit division module is used to acquire the area to be optimized and its basic information, and to divide the area to be optimized into several hydrological response units based on the basic information, including: Acquire the area to be optimized, its digital elevation model, land use data, and soil type data; Based on the digital elevation model, land use data, and soil type data, the area to be optimized is divided into several hydrological response units. Attached Figure Description

[0024] Figure 1 A flowchart illustrating a method for optimizing the spatial layout of artificial wetlands, as provided in an embodiment of the present invention; Figure 2 This is a flowchart of a solution for an adaptive constructed wetland configuration based on GA, provided for an embodiment of the present invention. Detailed Implementation

[0025] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0026] The terms "first" and "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0027] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0028] Example 1 See Figure 1 , Figure 1 This is a flowchart illustrating a method for optimizing the spatial layout of constructed wetlands according to an embodiment of the present invention. The embodiment of the present invention provides a method for optimizing the spatial layout of constructed wetlands, including steps 101 to 107, as detailed below: Step 101: Obtain the region to be optimized and its basic information, and divide the region to be optimized into several hydrological response units based on the basic information; In this embodiment, the basic information includes a digital elevation model, land use data, and soil type data; the step of obtaining the area to be optimized and its basic information, and dividing the area to be optimized into several hydrological response units based on the basic information, includes: Acquire the area to be optimized, its digital elevation model, land use data, and soil type data; Based on the digital elevation model, land use data, and soil type data, the area to be optimized is divided into several hydrological response units.

[0029] In this embodiment, for the area to be optimized, its digital elevation model (DEM), land use data, and soil type data are first acquired as the basic information basis for dividing hydrological response units. Specifically, topographic relief information is constructed using 30 m resolution DEM data, and spatial preprocessing operations are performed on the study area using the ArcSWAT tool. Based on this, the DEM data is used to divide the sub-basin range, and further, combined with land use maps and soil type maps, various land cover types and their corresponding soil physical properties are defined in the SWAT model. Subsequently, based on the spatial overlay analysis of slope classification, land use type, and soil type, the area to be optimized is divided into several hydrological response units (HRUs). The above division results can well reflect the comprehensive characteristics of topography, hydrology, and land cover, providing unified and physically based spatial unit support for subsequent hydrological process simulation and non-point source pollution load assessment.

[0030] In this embodiment, by specifically defining the basic information as digital elevation model, land use data, and soil type data, the source of geographic information on which the regional division depends is clarified. Furthermore, by acquiring the region to be optimized and its digital elevation model, land use data, and soil type data, and dividing the region to be optimized into several hydrological response units based on the above three types of data, the objectivity and consistency of the division process can be improved, making the division results closer to the actual terrain and land attributes, which facilitates subsequent pollution simulation and layout assessment.

[0031] Step 102: Obtain meteorological data and pollution data, and preprocess the meteorological data, point source pollution data and non-point source pollution data to obtain target meteorological data and target pollution data respectively; In this embodiment, the acquisition of meteorological data and pollution data, including preprocessing the meteorological data, point source pollution data, and non-point source pollution data to obtain target meteorological data and target pollution data, includes: Acquire meteorological data, fill in missing values ​​and format the meteorological data to generate target meteorological data; Acquire pollution data, including point source pollution data and non-point source pollution data; The point source pollution emissions are calculated based on the point source pollution data, and the point source pollution emissions are converted according to a preset format to obtain the target point source pollution data. Based on preset farmland management measures and the non-point source pollution data, fertilization event sequences and irrigation time sequences are obtained, and target non-point source pollution data are generated based on the fertilization event sequences and irrigation time sequences. Target pollution data is generated based on the target point source pollution data and the target non-point source pollution data.

[0032] In this embodiment, daily precipitation, maximum temperature, and minimum temperature data are first obtained from meteorological observation stations or the National Meteorological Data Center within the study area. For missing values ​​in the raw data due to equipment failure, transmission interruption, or human error, the missing records are supplemented using time-series interpolation (such as linear interpolation or polynomial interpolation based on preceding and following observations) or spatial proximity station substitution (selecting data from the same day from the observation point closest to the missing station with similar topography and climate). After supplementation, all observed variables are organized in year-month-day order and converted to the standard format file required by the SWAT model. For example, precipitation data is output as a ".pcp" file, and temperature data as a ".cli" file, with spatial information such as station number, latitude, longitude, and altitude added to the file header. The resulting continuous, complete meteorological input file conforming to the SWAT format is the target meteorological data, used to drive the subsequent SWAT model simulations of daily runoff, evapotranspiration, surface runoff, and nutrient transport.

[0033] In this embodiment, there are various methods for calculating point source pollution. Taking the empirical parameter method as an example, and in combination with the requirements of the SWAT model for the input format of point sources, the pollution load of urban domestic sewage and industrial wastewater discharges were estimated and the data format was converted, and the point source pollution data construction work that conforms to the input specifications of the SWAT model was completed.

[0034] First, point source pollution data was acquired and integrated. Pollutant production data for 10 districts and counties (Wuhua County, Xingning City, Dapu County, Meixian District, Jiaoling County, Pingyuan County, Yongding County, Shanghang County, Wuping County, and Changting County) within the study area from 1980 to 2020 were collected and calculated, including annual production of urban and industrial wastewater, COD, ammonia nitrogen, total nitrogen, and total phosphorus. A relatively simple empirical parameter method was adopted. This method can estimate both urban wastewater production and the production of various specific pollutants; the only difference lies in the pollution generation coefficient. For the SWAT point source input file, four pollutants—Chemical Oxygen Demand (COD), ammonia nitrogen, total nitrogen, and total phosphorus—were selected for statistical analysis. The empirical parameter method is as follows: (1) in, This represents the amount of domestic / industrial products, which can be wastewater (unit: L / a) or pollutants (unit: t / a). The pollution generation coefficient represents the per capita pollution generation. For wastewater, the unit is L / (person*day), and for COD, ammonia nitrogen, total nitrogen, and total phosphorus pollutants, the unit is g / (person*day). It is divided into two categories: domestic and industrial pollution generation coefficients, and is obtained from the "Pollution Generation and Discharge Coefficient Handbook" and the "National Pollution Source Census Bulletin of Longyan City / Meizhou City". The number of days in a year; The unit conversion factor represents the unit of measurement.

[0035] In this embodiment, it is assumed that industrial and urban wastewater are mixed before entering the wastewater treatment plant, and data on the treatment capacity of wastewater treatment plants in each district and county are collected. Next, the discharge path is divided, with pollution emissions from each district and county divided into treated emissions and untreated emissions (exceeding the treatment capacity of the wastewater treatment plants). The pollutant concentrations of treated wastewater are calculated according to the wastewater treatment plant's discharge standards, while the concentrations of untreated wastewater are calculated based on their original values. The output of each pollutant type is then calculated by combining the two. Next, the time scale is converted. SWAT point source input requires monthly daily average emissions (m³ / day; kg / day) format. The calculated annual wastewater and pollutant outputs need to be allocated to each month and then converted to daily average format. Simultaneously, the input amounts of other uncalculated pollutants (such as nitrates) are estimated based on empirical proportional relationships. Finally, the ten districts and counties within the study area are simplified as their center points, serving as the locations for SWAT point source input. The processed data is then input into the corresponding point source file for point source simulation.

[0036] In this embodiment, based on the collected data on urban domestic sewage volume, industrial wastewater volume, and annual production of pollutants such as COD, ammonia nitrogen, total nitrogen, and total phosphorus in ten districts and counties (Wuhua County, Xingning City, Dapu County, Meixian District, Jiaoling County, Pingyuan County, Yongding County, Shanghang County, Wuping County, and Changting County) within the study area from 1980 to 2020, the annual average point source emissions for each district and county were calculated using an empirical parameter method. Subsequently, based on the treatment capacity of the sewage treatment facilities in each district and county, the aforementioned annual average emissions were divided into treated emissions and untreated emissions. The former was converted according to the emission standard concentration, while the latter retained its original concentration. The total emissions of each pollutant were then combined to obtain the total emissions of each pollutant. The total annual emissions were then allocated to each month according to a predetermined ratio and converted into daily average emissions (m³ / day or kg / day). At the same time, the daily average dosage of other unlisted pollutants (such as nitrates) was estimated with reference to empirical coefficients. Finally, each district / county is simplified to its center point coordinates and a ".pnd" file is generated according to the SWAT point source input specification format, which includes fields such as site identifier, latitude and longitude, and daily average emissions of each pollutant, thereby obtaining target point source pollution data that meets the requirements for model calling.

[0037] In this embodiment, pre-determined field management measures for double-cropping rice include tillage, transplanting, application rates of basal and topdressing fertilizers, and irrigation frequency and volume. Specifically, based on the time nodes and operational content shown in Table 1, the application times and application rates of basal and topdressing fertilizers are sequentially converted into fertilization event sequences (Fertilizer ApplicationEvents) recognizable by the SWAT model, and irrigation, automatic irrigation, and shallow wet irrigation are converted into irrigation time sequences (IrrigationSchedule), which are recorded in the farmland HRU management file (.mgt). Subsequently, combined with the original non-point source pollution data (including soil infiltration rate, crop absorption coefficient, and nutrient transfer parameters after fertilization), the daily nitrogen and phosphorus losses caused by fertilization and irrigation are calculated, and the corresponding daily non-point source nutrient loading is output. The target non-point source pollution data generated through the above process not only includes nutrient input fluctuations during fertilization and irrigation but also reflects the dynamic pollution contribution of each HRU during the crop growth period, providing accurate spatiotemporal data support for subsequent watershed pollution simulation and optimization configuration. Table 1 shows the irrigation and fertilization schedule for double-cropping rice in the study area.

[0038] Table 1 In this embodiment, by acquiring meteorological data, performing missing value imputation and formatting processing on the meteorological data to generate target meteorological data, the problems of incompleteness and inconsistent structure of the original meteorological data are effectively solved. Simultaneously, pollution data is acquired, including point source pollution data and non-point source pollution data, ensuring the comprehensiveness of pollution sources. The point source pollution emissions are calculated based on the point source pollution data, and the point source pollution emissions are converted according to a preset format to obtain target point source pollution data, which helps improve the model adaptability of the point source pollution data. By acquiring fertilization event sequences and irrigation time sequences based on preset farmland management measures and the non-point source pollution data, target non-point source pollution data is generated, making the non-point source pollution information more consistent with actual agricultural activities. Finally, target pollution data is generated based on the target point source pollution data and target non-point source pollution data, providing unified, formatted, and complete input data for the non-point source pollution model.

[0039] Step 103: Construct a non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and STAW model; In this embodiment, the construction of a non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and STAW model includes: An initial non-point source pollution model is constructed based on the aforementioned hydrological response unit, target meteorological data, target pollution data, and STAW model. Historical observed flow rates and historical pollution concentrations are collected, and the initial non-point source pollution model is calibrated based on the historical observed flow rates and historical pollution concentrations to obtain the non-point source pollution model.

[0040] In this embodiment, an initial non-point source pollution simulation system for the watershed is first established in the SWAT model based on the pre-divided hydrological response units, pre-processed target meteorological data, and target pollution data. This system uses key parameters such as evapotranspiration, surface runoff, channel hydrodynamics, soil water processes, groundwater cycle, surface and channel sediment transport, and nitrogen cycle to drive the water and nutrient transport processes of each HRU. Subsequently, representative hydrological stations and water quality monitoring points within the watershed are selected, and daily runoff and daily pollutant concentration data from multiple years of actual observation are collected. The SWAT-CUP platform then utilizes a combination of SUFI-2 automatic optimization and manual adjustment to progressively calibrate sensitive parameters in the initial model, such as CANMX, CN2, SURLAG, SOL_AWC, RCHRG_DP, USLE_K, CDN, and SOL_NO3, ensuring that the simulation results simultaneously meet preset requirements in evaluation indicators such as the Nash-Sutcliffe efficiency coefficient (NSE), coefficient of determination (R²), and bias ratio (PBIAS). After calibration and verification, a non-point source pollution model that can truly reflect the hydrology and non-point source pollution processes in the study area is obtained, providing reliable dynamic simulation support for subsequent wetland optimization.

[0041] In this embodiment, the key parameters selected for the evapotranspiration process module in the non-point source pollution model include: the maximum canopy index (CANMX) at the HRU level, the plant transpiration compensation factor (EPCO), and the soil evaporation compensation factor (ESCO); GW_REVAP (.gw) is the groundwater re-evaporation coefficient, and REVAPMN (.gw) represents the shallow groundwater level threshold depth (mm). The surface runoff process module uses the effective hydraulic conductivity CH_K2 (.rte) in the main channel, the runoff curve number CN2 (.mgt) under SCS soil wetting conditions II, and the surface runoff lag time SURLAG (.bsn) to describe the water flow characteristics. The channel water module relies on the baseflow α factor ALPHA_BNK (.rte) of the groundwater storage area on both banks of the river, the Manning roughness coefficient CH_N1 (.sub) of the tributary channel, and the Manning roughness coefficient CH_N2 (.rte) of the main channel and the surface runoff Manning roughness coefficient OV_N (.hru) to simulate the channel flow. The soil water process module includes effective water capacity SOL_AWC (.sol) and soil hydraulic conductivity SOL_K (.sol); the groundwater module uses baseflow α factor ALPHA_BF (.gw), shallow groundwater level threshold depth GWQMN (.gw), groundwater lag time GW_DELAY (.gw), and deep groundwater permeability fraction RCHRG_DP (.gw) to describe baseflow and recharge processes. The surface sediment module simulates soil erosion and scouring using the sub-basin sediment transport peak rate adjustment factor ADJ_PKR (.bsn), average slope length SLSUBBSN (.hru), USLE soil erodibility factor USLE_K (.sol), and support measure factor USLE_P (.mgt). The channel sediment module describes the channel deposition process using the channel erodibility factor CH_COV1 (.rte), channel cover factor CH_COV2 (.rte), sediment transport method CH_EQN (.rte), and the maximum resuspended sediment linear parameter SPCON (.bsn) and resuspension index parameter SPEXP (.bsn). The nitrogen cycle module uses the following parameters to comprehensively reflect the characteristics of nitrogen cycling in the watershed: denitrification rate coefficient CDN (.bsn), organic nitrogen mineralization rate coefficient CMN (.bsn), organic nitrogen enrichment ratio ERORGN (.hru), nitrogen permeability coefficient NPERCO (.bsn), nitrogen absorption distribution parameter N_UPDIS (.bsn), precipitation nitrogen concentration RCN (.bsn), residue decomposition coefficient RSDCO (.bsn), denitrification threshold moisture content SDNCO (.bsn), soil nitrate concentration SOL_NO3 (.chm), and soil organic nitrogen concentration SOL_ORGN (.chm).

[0042] In this embodiment, an initial non-point source pollution model is constructed based on the hydrological response unit, target meteorological data, target pollution data, and STAW model, enabling the establishment of a simulation mechanism based on multi-source data. Furthermore, historical observed flow rates and historical pollution concentrations are collected, and the initial non-point source pollution model is calibrated based on these historical data to obtain the actual non-point source pollution model. This makes the simulation results more closely resemble actual hydrological and pollution processes, improving the reliability and interpretability of the model output.

[0043] Step 104: Obtain future climate data, and construct climate scenario data based on the future climate data, preset path rules, and simple quantile algorithm; In this embodiment, the future climate data includes raw daily precipitation data and temperature grid data corresponding to several GCMs, as well as locally observed climate sequences; the acquisition of future climate data, and the construction of climate scenario data based on the future climate data and preset path rules, includes: Acquire raw daily precipitation data and temperature grid data corresponding to several GCMs, as well as local meteorological observation sequences; Climate data for each GCM is generated based on the raw daily precipitation data and temperature grid data corresponding to each GCM. Climate-driven data is generated based on preset path rules and the climate data. The climate data is downscaled based on the Delta variability method and the local observed climate sequence to obtain future climate sequences. Pollution load time series are generated based on the simple quantile algorithm and the future climate sequence corresponding to each GCM. Climate scenario data are constructed based on the climate-driven data and the pollution load time series.

[0044] In this embodiment, raw daily precipitation and temperature grid data corresponding to several GCMs, as well as local meteorological observation sequences, are acquired. First, the raw daily precipitation and temperature grid output files for each GCM model under four paths (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5) are downloaded from the CMIP6 database. Simultaneously, the daily precipitation and maximum / minimum temperature observation sequences of local stations for the corresponding period are obtained from the regional meteorological observation network. The grid data and station data are organized into a dedicated database according to time index, and the spatial resolution, time span, and quality control information of each data source are recorded to provide a complete input set for subsequent processing.

[0045] In this embodiment, climate data corresponding to each GCM is generated based on the original daily precipitation and temperature grid data. For each GCM model, the original gridded precipitation and temperature data are cropped to the study area using a batch script, and the NetCDF format is converted into a standardized raster TIFF or CSV file to form directly callable "GCM climate data". During this process, the data units are standardized (e.g., precipitation units are standardized to mm / day, and temperature units are standardized to °C) and metadata is labeled to ensure data consistency.

[0046] In this embodiment, when generating climate-driven data based on preset path rules and the climate data, the climate data of the corresponding GCM is mapped to path-driven input according to the time nodes and greenhouse gas emission scenarios of each SSP path, generating climate-driven datasets under different scenarios. For example, for the SSP2-4.5 path, the cropped climate data of all GCM models under this scenario are selected and archived in chronological order into a folder called "SSP2-4.5 Climate-Driven," providing scenario differentiation for subsequent downscaling and simulation calls.

[0047] In this embodiment, additive (temperature) and relative (precipitation) variation factors are calculated for precipitation and temperature data of each GCM in the historical period (e.g., 1995–2014) and the observation period, respectively. These variation factors are then superimposed or multiplied / divided into the observation sequence to generate future temperature and precipitation sequences with local observation characteristics. The "future climate sequence" output by this process has a diurnal temporal resolution and can be directly used for refined hydrological and non-point source pollution simulations.

[0048] In this embodiment, the future climate series downscaled from each GCM is used as driving data to simulate the daily pollution load in the future period in a pre-established non-point source pollution model. The quantile algorithm is applied to the output results of multiple models to extract the Q25, Q50 and Q75 values ​​in sequence, forming three pollution load time series curves representing different risk levels (optimistic, neutral and conservative), providing dynamic pollution input for optimization decision-making.

[0049] In this embodiment, climate scenario data is constructed based on the climate-driven data and the pollution load time series. Specifically, the climate-driven data under each SSP path is integrated with the corresponding three quantile pollution load time series to generate a complete "climate scenario data" set. This dataset is classified by path and risk level, and each subset contains three indicators: daily precipitation, temperature, and pollution load, which facilitates the simultaneous consideration of both meteorological and pollution-driven factors in the wetland optimization model.

[0050] In this embodiment, raw daily precipitation and temperature grid data corresponding to various global climate models (GCMs) are first obtained from the CMIP6 project, while measured daily climate sequences from local meteorological observation stations are also collected. For each GCM model, its historical and future daily precipitation and temperature grid outputs are read, and corresponding climate data are generated. Then, according to the preset shared socioeconomic path rules (such as SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5), the climate data of each model are mapped to climate driving data under the corresponding path. Next, using the Delta variability method, based on the locally observed climate sequences, additive (temperature) or relative (precipitation) corrections are applied to the future climate data of each GCM, completing the downscaling process from grid scale to station scale, and obtaining high-resolution future climate sequences. Subsequently, for the future climate series of each GCM, a simple quantile algorithm was applied to extract multiple quantile values ​​(Q25, Q50, Q75) of the pollution load time series to reflect the non-point source pollution load predictions for different risk preferences. Finally, the above climate driving data were combined with the pollution load time series to construct climate scenario data for each SSP pathway and risk level, providing multi-model and multi-scenario input support for wetland optimization models.

[0051] In this embodiment, by acquiring raw daily precipitation and temperature grid data corresponding to several Global Climate Change Models (GCMs) and local observation meteorological sequences, the breadth and local adaptability of future climate data are ensured. Climate data corresponding to each GCM is generated based on the raw daily precipitation and temperature grid data, and climate driving data is generated based on preset path rules and the climate data, reflecting climate responses under different scenarios. Furthermore, the climate data is downscaled using the Delta variability method and the local observation climate sequences to obtain future climate sequences, improving the spatial resolution of the climate data. Pollution load time series are generated based on the simple quantile algorithm and the future climate sequences corresponding to each GCM, and climate scenario data is constructed based on the climate driving data and the pollution load time series, providing diverse and reliable future scenario inputs for the model.

[0052] Step 105: Identify candidate wetland areas based on GIS spatial analysis algorithms, land use data, soil infiltration rate and composite topographic index, and calculate the maximum potential area of ​​each candidate wetland area; In this embodiment, the process of identifying candidate wetland areas based on GIS spatial analysis algorithms, land use, soil infiltration rate, and composite topographic indices, and calculating the maximum potential area of ​​each candidate wetland area, includes: Obtain land use data and soil infiltration rate of the area to be optimized; Based on the land use data, farmland data was selected as the first candidate area; Based on the soil infiltration rate, areas in the region to be optimized that are lower than the preset infiltration threshold are eliminated to obtain a second candidate region; Calculate the composite terrain index of the area to be optimized, and select areas whose composite terrain index is greater than a preset index threshold as the third candidate area; Based on GIS spatial analysis algorithms, the first candidate region, the second candidate region, and the third candidate region are overlaid to determine the candidate wetland regions, and the maximum potential area of ​​each sub-basin is calculated.

[0053] In this embodiment, firstly, land cover type information of all grid cells in the study area is extracted from the regional land use database, and the soil infiltration rate of each grid cell is estimated from soil attribute data or measured sample point data using equation (2). : (2) Among them, This represents the estimated soil infiltration rate. , , and These correspond to the physical properties of the soil, namely, bulk density, sand content, silt content, and clay content. The land use raster and soil infiltration rate raster mentioned above are loaded into the ArcMap platform as input data for subsequent spatial analysis.

[0054] In this embodiment, the "Extract by Attributes" tool is called in ArcMap, and all pixels marked as farmland are extracted from the land use raster of the entire area with "Land use type = farmland" as the filter condition. This generates a binary raster layer containing only farmland, which serves as the first candidate area and provides a preliminary spatial range for subsequent wetland site selection.

[0055] In this embodiment, the "Extract by Attributes" tool is applied to the soil infiltration rate raster. A preset infiltration rate threshold, such as 10 mm / h, is used as the screening condition to retain the areas where the infiltration rate is lower than the threshold, generating a second candidate raster layer to indicate potential plots with poor drainage and easy runoff retention.

[0056] In this embodiment, the Composite Topographic Index (CTI) is calculated using the Spatial Analyst toolbox based on a 30m DEM. The CTI is a widely used index in hydrological analysis, combining upstream catchment area and surface slope to characterize topographic wetness. A CTI raster layer is generated, showcasing the water accumulation potential of different areas, typically used to identify potential wetland regions. These high CTI value areas generally have high water accumulation capacity and are suitable locations for wetland construction, thus providing a scientific basis for the optimal allocation of wetlands. (3) in, This is an estimate of the CTI value. The upstream catchment area per unit width (m²) The slope angle is (°). This reflects the influence of terrain slope on water flow velocity.

[0057] Then, using a preset threshold, such as 11.5, the "Extract by Attributes" function is called to extract high CTI areas, obtaining a third candidate raster layer to identify low-lying locations where the terrain is prone to water accumulation.

[0058] In this embodiment, the first, second, and third candidate regions are overlaid using GIS spatial analysis algorithms to determine candidate wetland regions. The maximum potential area of ​​each sub-basin is then calculated. In ArcMap, the "Raster Calculator" or "Boolean AND" tool is used to logically overlay the first, second, and third candidate rasters to generate the final candidate wetland region raster. Subsequently, the "Raster to Polygon" tool is used to convert the raster into polygon features, and the "Zonal Statistics" function is used to calculate the total area of ​​all candidate wetland polygons within each sub-basin. This total area is taken as the maximum potential area of ​​each sub-basin, and boundary constraint data for optimizing the model is exported.

[0059] In this embodiment, by acquiring land use data and soil infiltration rate of the area to be optimized, and selecting farmland data as the first candidate area based on the land use data, areas with concentrated agricultural non-point source pollution can be prioritized. Based on the soil infiltration rate, areas below a preset infiltration threshold are eliminated to obtain the second candidate area, which helps ensure the actual infiltration function of the constructed wetland. By calculating the composite topographic index of the area to be optimized and selecting areas greater than a preset index threshold as the third candidate area, the runoff potential of the deployment area can be further improved. Finally, based on GIS spatial analysis algorithms, the above three types of areas are overlaid to determine candidate wetland areas, and the maximum potential area of ​​each sub-basin is calculated, which effectively defines the deployment boundary and provides area constraints for the optimization model.

[0060] Step 106: Construct an objective function and build a wetland optimization model based on the non-point source pollution model and the objective function; In this embodiment, the construction of a wetland optimization model based on the non-point source pollution model and the preset objective function includes: A target function is constructed based on cost parameters and the output parameters of the non-point source pollution model. A wetland optimization model is constructed based on the objective function and the non-point source pollution model.

[0061] In this embodiment, an artificial wetland configuration optimization toolbox is first built on the MATLAB platform. This toolbox includes the wetland deployment area boundary, control parameters required by the SWAT model (such as WET_FR, WET_NSA, WET_MXVOL, NSETLW1, PSETLW1, WET_SED, WET_NO3, PNDEVCOEFF, etc.), economic parameters such as unit volume construction cost and design life NN, and identifies the maximum potential area of ​​each sub-basin based on the topographic relief, land use status, and soil type of the study area, thereby determining the value range of the decision variables. During the optimization iteration process, the toolbox first dynamically modifies the wetland parameters in the SWAT project file (.pnd) according to the candidate schemes, such as adjusting WET_FR to represent the runoff area ratio, setting the surface area and volume corresponding to WET_NSA and WET_MXVOL, configuring NSETLW1 and PSETLW1 to simulate nitrogen and phosphorus deposition efficiency, and updating the initial water quality indicators WET_SED, WET_NO3, etc. Subsequently, the SWAT model is invoked to run a simulation, extracting the non-point source pollution load output under this configuration scheme. Combining the objective function and pollution reduction constraints, a genetic algorithm performs selection, crossover, and mutation operations within the decision variable space, continuously screening and generating new wetland configuration schemes until convergence to the optimal combination that minimizes cost and satisfies the pollution reduction objective. Through this process, high-resolution spatial optimization of constructed wetlands can be achieved for different cost, lifespan, and environmental benefit requirements.

[0062] In this embodiment, cost parameters such as unit construction cost, maintenance rate, interest rate, and design life (30 years) are first obtained from the constructed wetland configuration optimization toolbox. Simultaneously, a calibrated non-point source pollution model is invoked to extract the annual average NH3–N load output under different wetland area configurations in each sub-basin. Based on the aforementioned cost parameters and model output data, a total cost objective function is constructed. Subsequently, this objective function is combined with the non-point source pollution model, and constraints are set according to the preset reduction rate and the maximum buildable area of ​​the sub-basin to form a nonlinear wetland optimization model. This model, while ensuring that the NH3–N load meets the standard, aims to minimize construction and maintenance costs, providing a clear mathematical basis and boundary constraints for subsequent genetic algorithm solutions.

[0063] In this embodiment, in the wetland optimization model, the decision variables ( ) represents the first The area within a sub-basin used for constructing artificial wetlands is defined by the objective function as the total cost. That is, to minimize the construction cost of artificial wetlands ( ) and maintenance costs ( The sum of these variables is used, while constraints are applied to ensure that non-point source pollution loads (NH3-N pollution in this example) are effectively controlled. Furthermore, decision variables... Boundary conditions must be met, namely, the area of ​​constructed wetlands within each sub-basin must not exceed its maximum constructible area. Based on this, the following nonlinear simulation optimization model can be established: min (3) subject to: (4) (5) (6) in, The construction cost per unit volume of retained artificial wetlands; Maintenance rate (as a percentage of construction costs); Discount factor; The interest rate is 3% in this project. N The design life of the constructed wetland (30 years for this project); The baseline year represents the NH3-N load at the watershed outlet where no constructed wetlands were implemented. The expected NH3-N load reduction rate after the implementation of constructed wetlands (10% in this project); Sub-basins identified by ArcMap tools i The maximum potential area of ​​constructed wetlands; Representing meteorological variables The SWAT model is driven to simulate NH3-N load under constructed wetland conditions by changing the parameter set. To characterize the implementation status of constructed wetlands. When meteorological variables... When considering future climate change scenarios, This represents the NH3-N load with Q50, Q70, and Q90 estimated based on multiple GCMs sets, solved for... This represents the near-optimal constructed wetland layout scheme corresponding to the GCM set estimates for Q50, Q70, and Q90. yes The function can be described as follows: (7) (8) (9) (10) (11) (12) (13) (14) (15) in, This is the normal water level for the wetland. This is the maximum water level in the wetland. This represents the proportion of the sub-basin's watershed area flowing into the wetland. The surface area of ​​the wetland at its normal water level. The volume of the wetland at its normal water level. The surface area of ​​the wetland below the maximum water level. WET_MXVOL represents the volume of the wetland at its maximum water level. The normal sediment concentration in wetland water bodies, WET_NSED. NSETLW1 represents the nitrogen deposition rate of constructed wetlands IPND1 to IPND2. NSETLW2 represents wetland nitrogen deposition rate for months other than IPND1 to IPND2 (m / year).

[0064] In this embodiment, an objective function is constructed based on cost parameters and the output parameters of the non-point source pollution model, enabling the optimization model to strike a balance between pollution reduction targets and costs. Furthermore, by constructing a wetland optimization model based on the objective function and the non-point source pollution model, pollution simulation and resource allocation processes can be closely integrated, thereby improving the practicality and controllability of the optimization process and providing clear quantitative basis for subsequent optimization layout.

[0065] Step 107: Generate an artificial wetland optimization and control scheme based on the preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, and optimize the spatial layout of artificial wetlands based on the artificial wetland optimization and control scheme.

[0066] In this embodiment, the step of generating an artificial wetland optimization and control scheme based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, and then optimizing the spatial layout of the artificial wetland based on the artificial wetland optimization and control scheme, includes: The wetland optimization model is used as the fitness function of the genetic algorithm, the maximum potential area of ​​each candidate wetland region is used as the decision variable of the genetic algorithm, and the candidate wetland information and the climate scenario data are used as the input data of the genetic algorithm. Based on the genetic algorithm, an artificial wetland optimization and control scheme is generated. Based on the aforementioned constructed wetland optimization and control scheme, the spatial layout of constructed wetlands is optimized.

[0067] Please refer to Figure 2 , Figure 2 This is a flowchart of a solution for an adaptive constructed wetland configuration based on GA, provided for an embodiment of the present invention.

[0068] In this embodiment, the constructed nonlinear wetland optimization model is used as the fitness function of the MATLAB Genetic Algorithm (GA). First, the maximum potential area and spatial boundary information of candidate wetland regions in each sub-basin are read as upper and lower limits for decision variables in the genetic algorithm. Simultaneously, climate scenario data (historical pollution load sequences at Q25, Q50, and Q75 quantiles under each SSP path) generated by the multi-model CMIP6 GCM and candidate wetland polygon attributes (area, location, land use type, etc.) are imported as input data for the algorithm. During GA population initialization, individuals are composed of chromosomes randomly generated within the variable range, with each chromosome encoding a wetland configuration area vector for each sub-basin. Subsequently, for each chromosome, the artificial wetland parameter toolbox is used to automatically update the wetland parameters in the SWAT.pnd file, and the SWAT model is run under historical and future climate scenarios to obtain the annual average NH3–N load value at the basin outlet and the cost output required by the objective function. The cost and reduction constraints are combined into a fitness value using the Lagrange multiplier method. The Genetic Algorithm (GA) performs selection, crossover, and mutation operations, iteratively optimizing until convergence. Finally, the GA outputs a set of optimal or near-optimal constructed wetland management schemes, clearly defining the optimal wetland area for each sub-basin. Based on this optimization scheme, constructed wetlands can be spatially deployed on a GIS platform according to sub-basin divisions and potential boundaries, achieving adaptive wetland configuration for multiple historical and future scenarios.

[0069] In this embodiment, by using the wetland optimization model as the fitness function of the genetic algorithm, the maximum potential area of ​​each candidate wetland region as the decision variable of the genetic algorithm, and the candidate wetland information and the climate scenario data as the input data of the genetic algorithm, a multi-variable, multi-constraint optimization search framework can be constructed. Generating an artificial wetland optimization and control scheme based on the genetic algorithm helps to quickly converge to a near-optimal solution. Furthermore, the spatial optimization and layout of artificial wetlands based on the artificial wetland optimization and control scheme makes the final layout result operable and scientific, adaptable to multiple scenario requirements.

[0070] In this embodiment, by acquiring the region to be optimized and its basic information, the region is divided into several hydrological response units based on the basic information. This enables regional division at a geographic scale, providing a clear unit basis for subsequent simulations. Furthermore, by acquiring meteorological and pollution data, and preprocessing the meteorological data, point source pollution data, and non-point source pollution data to obtain target meteorological and target pollution data, the completeness and consistency of the input data are improved. Constructing a non-point source pollution model based on the hydrological response units, target meteorological data, target pollution data, and the STAW model improves the simulation accuracy of non-point source pollution processes. Acquiring future climate data and constructing climate scenario data based on the future climate data, preset path rules, and a simple quantile algorithm helps to reflect the impact of climate change on pollution loads. Furthermore, by constructing a wetland optimization model based on the non-point source pollution model and the objective function, the simulation results can be coupled with the governance objectives, enhancing the scientific nature of the optimization strategy. Candidate wetland areas can be identified using GIS spatial analysis algorithms, land use data, soil infiltration rates, and composite topographic indices, and the maximum potential area of ​​each candidate wetland area can be calculated, enabling efficient identification of potential deployment areas. Finally, based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, an optimized control scheme for constructed wetlands is generated, and the spatial optimization of constructed wetlands is carried out accordingly. This facilitates the optimization and practical application of constructed wetland layout, improving the scientific nature, feasibility, and sustainable adaptability of constructed wetland configuration.

[0071] This invention provides a system for optimizing the spatial layout of artificial wetlands, comprising: a unit division module, a data processing module, a non-point source construction module, a climate construction module, an optimization model construction module, a wetland identification module, and an optimization layout module; The unit division module is used to obtain the region to be optimized and its basic information, and to divide the region to be optimized into several hydrological response units based on the basic information. The data processing module is used to acquire meteorological data and pollution data, and to preprocess the meteorological data, point source pollution data and non-point source pollution data to obtain target meteorological data and target pollution data respectively. The non-point source construction module is used to construct a non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and STAW model. The climate construction module is used to acquire future climate data and construct climate scenario data based on the future climate data, preset path rules, and a simple quantile algorithm. The optimization model construction module is used to construct an objective function and to construct a wetland optimization model based on the non-point source pollution model and the objective function. The wetland identification module is used to identify candidate wetland areas based on GIS spatial analysis algorithms, land use data, soil infiltration rate and composite topographic index, and to calculate the maximum potential area of ​​each candidate wetland area. The optimization layout module is used to generate an artificial wetland optimization and control scheme based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, and to optimize the spatial layout of artificial wetlands based on the artificial wetland optimization and control scheme.

[0072] In this embodiment, the basic information includes a digital elevation model, land use data, and soil type data; the unit division module is used to acquire the area to be optimized and its basic information, and to divide the area to be optimized into several hydrological response units based on the basic information, including: Acquire the area to be optimized, its digital elevation model, land use data, and soil type data; Based on the digital elevation model, land use data, and soil type data, the area to be optimized is divided into several hydrological response units.

[0073] In this embodiment of the invention, a terminal device is also provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the above-described artificial wetland spatial optimization layout method.

[0074] In this embodiment of the invention, a computer-readable storage medium is also provided, which includes a stored computer program, wherein the computer program controls the device where the computer-readable storage medium is located to execute the above-described artificial wetland spatial optimization arrangement method when it is running.

[0075] For example, a computer program can be divided into one or more modules, one or more of which are stored in memory and executed by a processor to perform the present invention. The one or more modules can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a terminal device.

[0076] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor, memory, and display. Those skilled in the art will understand that the above components are merely examples of terminal devices and do not constitute a limitation on the terminal device. It may include more or fewer components, or combinations of certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0077] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device through various interfaces and lines.

[0078] Memory can be used to store computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory, and by accessing data stored in the memory. Memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, application programs required for at least one function (such as sound playback, text conversion, etc.), etc.; the data storage area can store data created based on the use of the mobile phone (such as audio data, text message data, etc.). In addition, memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0079] In this invention, modules based on the optimized spatial arrangement of constructed wetlands, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. Those skilled in the art can understand and implement this invention without any inventive effort.

[0080] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for optimizing the spatial layout of constructed wetlands, characterized in that, include: Obtain the region to be optimized and its basic information, and divide the region to be optimized into several hydrological response units based on the basic information; Acquire meteorological data and pollution data, and preprocess the meteorological data, point source pollution data and non-point source pollution data respectively to obtain target meteorological data and target pollution data; A non-point source pollution model is constructed based on the aforementioned hydrological response unit, target meteorological data, target pollution data, and STAW model; Acquire future climate data, and construct climate scenario data based on the future climate data, preset path rules, and simple quantile algorithm; Candidate wetland areas were identified based on composite topographic index, land use data, and soil infiltration rate, and the maximum potential area of ​​each candidate wetland area was calculated. Construct an objective function, and then construct a wetland optimization model based on the non-point source pollution model and the objective function; An artificial wetland optimization and control scheme is generated based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data. The artificial wetland is then spatially optimized and arranged based on the artificial wetland optimization and control scheme.

2. The method for optimizing the spatial layout of an artificial wetland as described in claim 1, characterized in that, The basic information includes a digital elevation model, land use data, and soil type data; the process of acquiring the area to be optimized and its basic information, and dividing the area to be optimized into several hydrological response units based on the basic information, includes: Acquire the area to be optimized, its digital elevation model, land use data, and soil type data; Based on the digital elevation model, land use data, and soil type data, the area to be optimized is divided into several hydrological response units.

3. The method for optimizing the spatial layout of an artificial wetland as described in claim 2, characterized in that, The acquisition of meteorological and pollution data involves preprocessing the meteorological data, point source pollution data, and non-point source pollution data to obtain target meteorological and pollution data, including: Acquire meteorological data, fill in missing values ​​and format the meteorological data to generate target meteorological data; Acquire pollution data, including point source pollution data and non-point source pollution data; The point source pollution emissions are calculated based on the point source pollution data, and the point source pollution emissions are converted according to a preset format to obtain the target point source pollution data. Based on preset farmland management measures and the non-point source pollution data, fertilization event sequences and irrigation time sequences are obtained, and target non-point source pollution data are generated based on the fertilization event sequences and irrigation time sequences. Target pollution data is generated based on the target point source pollution data and the target non-point source pollution data.

4. The method for optimizing the spatial layout of an artificial wetland as described in claim 3, characterized in that, The construction of a non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and STAW model includes: An initial non-point source pollution model is constructed based on the aforementioned hydrological response unit, target meteorological data, target pollution data, and STAW model. Historical observed flow rates and historical pollution concentrations are collected, and the initial non-point source pollution model is calibrated based on the historical observed flow rates and historical pollution concentrations to obtain the non-point source pollution model.

5. The method for optimizing the spatial layout of an artificial wetland as described in claim 4, characterized in that, The future climate data includes raw daily precipitation data and temperature grid data corresponding to several GCMs, as well as locally observed climate sequences; the acquisition of future climate data, and the construction of climate scenario data based on the future climate data and preset path rules, includes: Acquire raw daily precipitation data and temperature grid data corresponding to several GCMs, as well as local meteorological observation sequences; Climate data for each GCM is generated based on the raw daily precipitation data and temperature grid data corresponding to each GCM. Climate-driven data is generated based on preset path rules and the climate data. The climate data is downscaled based on the Delta variability method and the local observed climate sequence to obtain future climate sequences. Pollution load time series are generated based on the simple quantile algorithm and the future climate sequence corresponding to each GCM. Climate scenario data are constructed based on the climate-driven data and the pollution load time series.

6. The method for optimizing the spatial layout of an artificial wetland as described in claim 5, characterized in that, The candidate wetland areas are based on composite topographic indices, land use, and soil infiltration rates, and the maximum potential area of ​​each candidate wetland area is calculated, including: Obtain land use data and soil infiltration rate of the area to be optimized; Based on the land use data, farmland data was selected as the first candidate area; Based on the soil infiltration rate, areas in the region to be optimized that are lower than the preset infiltration threshold are eliminated to obtain a second candidate region; Calculate the composite terrain index of the area to be optimized, and select areas whose composite terrain index is greater than a preset index threshold as the third candidate area; Based on GIS spatial analysis algorithms, the first candidate region, the second candidate region, and the third candidate region are overlaid to determine the candidate wetland regions, and the maximum potential area of ​​each sub-basin is calculated.

7. The method for optimizing the spatial layout of an artificial wetland as described in claim 6, characterized in that, The construction of the wetland optimization model based on the non-point source pollution model and the preset objective function includes: A target function is constructed based on cost parameters and the output parameters of the non-point source pollution model. A wetland optimization model is constructed based on the objective function and the non-point source pollution model.

8. The method for optimizing the spatial layout of an artificial wetland as described in claim 7, characterized in that, The process of generating an artificial wetland optimization and control scheme based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, and then optimizing the spatial layout of the artificial wetland based on the artificial wetland optimization and control scheme, includes: The wetland optimization model is used as the fitness function of the genetic algorithm, the maximum potential area of ​​each candidate wetland region is used as the decision variable of the genetic algorithm, and the candidate wetland information and the climate scenario data are used as the input data of the genetic algorithm. Based on the genetic algorithm, an artificial wetland optimization and control scheme is generated. Based on the aforementioned constructed wetland optimization and control scheme, the spatial layout of constructed wetlands is optimized.

9. A system for optimizing the spatial layout of constructed wetlands, characterized in that, include: The module includes a unit partitioning module, a data processing module, a non-point source construction module, a climate construction module, an optimization model construction module, a wetland identification module, and an optimization layout module. The unit division module is used to obtain the region to be optimized and its basic information, and to divide the region to be optimized into several hydrological response units based on the basic information. The data processing module is used to acquire meteorological data and pollution data, and to preprocess the meteorological data, point source pollution data and non-point source pollution data to obtain target meteorological data and target pollution data respectively. The non-point source construction module is used to construct a non-point source pollution model based on the hydrological response unit, target meteorological data, target pollution data, and STAW model. The climate construction module is used to acquire future climate data and construct climate scenario data based on the future climate data, preset path rules, and a simple quantile algorithm. The wetland identification module is used to identify candidate wetland areas based on composite topographic index, land use data and soil infiltration rate, and to calculate the maximum potential area of ​​each candidate wetland area. The optimization model construction module is used to construct an objective function and to construct a wetland optimization model based on the non-point source pollution model and the objective function. The optimization layout module is used to generate an artificial wetland optimization and control scheme based on a preset genetic algorithm, the candidate wetland information, the wetland optimization model, and the climate scenario data, and to optimize the spatial layout of artificial wetlands based on the artificial wetland optimization and control scheme.

10. The artificial wetland spatial optimization layout system as described in claim 9, characterized in that, The basic information includes a digital elevation model, land use data, and soil type data; the unit division module is used to acquire the area to be optimized and its basic information, and divide the area to be optimized into several hydrological response units based on the basic information, including: Acquire the area to be optimized, its digital elevation model, land use data, and soil type data; Based on the digital elevation model, land use data, and soil type data, the area to be optimized is divided into several hydrological response units.

Images