Method, device and electronic equipment for determining water environment capacity

Abstract

The present disclosure relates to a method, device, and electronic device for determining water environment capacity, and relates to the technical field of water environment assessment, including: fusing first satellite data of a target river basin observed by a first satellite, second satellite data of the target river basin observed by a second satellite, and water environment data of the target river basin obtained by simulating a dynamic river basin model to obtain target water environment data of the target river basin; inputting the river basin flow, target water environment data, pollution source data, vegetation coverage, and meteorological data of the target river basin into the dynamic river basin model to obtain the pollutant concentration and outlet flow of the target river basin; obtaining the water environment capacity of the target river basin based on the pollutant concentration and the outlet flow of the target river basin; the water environment capacity indicates the capacity of the target river basin to accommodate pollutants. Using the method for determining water environment capacity proposed in the present disclosure, a water environment capacity with high coverage, high resolution, and accuracy can be obtained.

Description

Technical Field

[0001] The present disclosure relates to the technical field of water environment assessment, and in particular to a method, device and electronic device for determining water environment capacity. Background Art

[0002] The assessment of water environment capacity is an important part of environmental protection and water resources management. With the development of water environment assessment technology, the accuracy of water environment capacity assessed by water environment assessment technology proposed in related technologies is not high. Summary of the Invention

[0003] The purpose of the present disclosure is to provide a method, device and electronic device for determining water environment capacity to solve the above technical problems.

[0004] To achieve the above-mentioned objectives, the first aspect of the present disclosure provides a method for determining water environment capacity, comprising: fusing first satellite data of a target river basin observed by a first satellite, second satellite data of the target river basin observed by a second satellite, and water environment data of the target river basin obtained by simulating a dynamic river basin model to obtain target water environment data of the target river basin; inputting the river basin flow, target water environment data, pollution source data, vegetation coverage, and meteorological data of the target river basin into the dynamic river basin model to obtain the pollutant concentration and outlet flow of the target river basin; obtaining the water environment capacity of the target river basin based on the pollutant concentration and the outlet flow of the target river basin; the water environment capacity indicates the capacity of the target river basin to accommodate pollutants.

[0005] In a possible implementation, the fusing first satellite data of a target watershed observed by a first satellite, second satellite data of the target watershed observed by a second satellite, and water environment data of the target watershed obtained by simulation of a dynamic watershed model to obtain target water environment data of the target watershed includes: fusing the first satellite data with the water environment data to obtain first fused data; and fusing the first fused data with the second satellite data to obtain the target water environment data.

[0006] In a possible implementation, the fusing of the first satellite data with the water environment data to obtain first fused data includes: obtaining a first observation error of the first satellite based on the first satellite data and the water environment data; correcting the first observation error using a first Kalman gain to obtain a first target observation error; the first Kalman gain is used to measure the uncertainty of the water environment data; and obtaining the first fused data based on the water environment data and the first target observation error.

[0007] In a possible implementation, the fusing of the first fused data with the second satellite data to obtain the target water environment data includes: obtaining a second observation error of the second satellite based on the second satellite data and the second fused data; the second fused data is obtained by perturbing the third fused data using a disturbance term, and the third fused data is data corresponding to the first fused data; correcting the second observation error using a second Kalman gain to obtain a second target observation error; the second Kalman gain is used to balance the uncertainty of the water environment and the second observation error; and obtaining the target water environment data based on the first fused data and the second target observation error.

[0008] In a possible implementation, the watershed flow, target water environment data, pollution source data, vegetation coverage and meteorological data of the target watershed are input into the dynamic watershed model to obtain the pollutant concentration and outlet flow of the target watershed, including: screening out multiple target sub-watersheds from multiple sub-watersheds of the target watershed; inputting the watershed flow, target water environment data, vegetation coverage, future meteorological data of the target watershed and the pollution source data of the multiple target sub-watersheds into the dynamic watershed model to obtain the future pollutant concentration and outlet flow of the target watershed.

[0009] In one possible embodiment, obtaining the water environment capacity of the target basin based on the pollutant concentration of the target basin and the outlet flow includes: determining a target time period in which the pollutant concentration of the target basin is less than a target pollutant concentration; for each time point in the target time period, obtaining the pollutant flux at the time point based on the pollutant concentration and the outlet flow; and obtaining the water environment capacity based on the pollutant flux at multiple time points in the target time period.

[0010] In one possible embodiment, the method further includes: determining multiple elements that affect the water environment capacity, governance resources and ecological abnormality; the governance resources are the resources consumed to govern pollutants in the target watershed, and the ecological abnormality indicates the ecological abnormality of the target watershed; with the goal of maximizing the water environment capacity, minimizing the governance resources and minimizing the ecological abnormality, screening out multiple target elements from the multiple elements; and screening out multiple target parameters from the parameters of the multiple target elements.

[0011] In a possible embodiment, the method selects multiple target elements from the multiple elements with the goal of maximizing the water environment capacity, minimizing the governance resources, and minimizing the ecological abnormality, including: determining a penalty value when the compliance rate of the pollutant concentration does not meet the preset compliance rate; using the penalty value to respectively correct the water environment capacity, the governance resources, and the ecological abnormality to obtain a corrected water environment capacity, a corrected governance resources, and a corrected ecological abnormality; and selecting the multiple target elements from the multiple elements with the goal of maximizing the corrected water environment capacity, the corrected governance resources, and the corrected ecological abnormality.

[0012] In order to achieve the above-mentioned purpose, the second aspect of the present disclosure provides a device for determining water environment capacity, including: a fusion module, configured to fuse first satellite data of a target river basin observed by a first satellite, second satellite data of the target river basin observed by a second satellite, and water environment data of the target river basin obtained by simulation of a dynamic river basin model, to obtain target water environment data of the target river basin; a prediction module, configured to input the river basin flow, target water environment data, pollution source data, vegetation coverage and meteorological data of the target river basin into the dynamic river basin model to obtain the pollutant concentration and outlet flow of the target river basin; a water environment capacity module, configured to obtain the water environment capacity of the target river basin based on the pollutant concentration and the outlet flow of the target river basin; the water environment capacity indicates the capacity of the target river basin to accommodate pollutants.

[0013] In order to achieve the above object, the third aspect of the present disclosure provides an electronic device, comprising: a memory on which a computer program is stored; a processor for executing the computer program in the memory to implement

[0014] Through the above technical solution, after the first satellite data, the second satellite data and the water environment data are integrated, it is possible to obtain target water environment data with high coverage, high spatial resolution and high accuracy without relying on too many monitoring stations. The water environment capacity can be calculated based on the target water environment data, and the obtained water environment capacity will also be more accurate.

[0015] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following detailed description, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0017] Figure 1The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0018] Figure 2 It is a schematic diagram of dividing a target watershed into multiple target sub-watersheds according to an exemplary embodiment.

[0019] Figure 3 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0020] Figure 4 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0021] Figure 5 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0022] Figure 6 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0023] Figure 7 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0024] Figure 8 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0025] Figure 9 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0026] Figure 10 The present invention is a flowchart of a method for determining water environment capacity according to an exemplary embodiment.

[0027] Figure 11 It is a block diagram of a device for determining water environment capacity according to an exemplary embodiment.

[0028] Figure 12 is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0029] The following describes the specific embodiments of the present disclosure in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure and are not intended to limit the present disclosure.

[0030] Figure 1 A method for determining water environment capacity according to an exemplary embodiment includes the following steps:

[0031] In step S10, the first satellite data of the target river basin observed by the first satellite, the second satellite data of the target river basin observed by the second satellite, and the water environment data of the target river basin obtained by dynamic river basin model simulation are fused to obtain the target water environment data of the target river basin.

[0032] Among them, the first satellite can be a gravity (grace) satellite, and the first satellite data can be the groundwater storage (TWS) observed by the gravity satellite. The unit of groundwater storage is equivalent water height (cm). The groundwater storage is the total amount of water stored in the underground aquifer of the target basin, which can reflect the total changes in soil moisture, groundwater and surface water in the target basin.

[0033] Gravity satellites use microwave ranging technology to measure the distance between two satellites. When satellites fly over areas of varying gravity on Earth's surface, the differences in gravitational acceleration cause minute perturbations in their orbits, resulting in changes in the distance between the two satellites. By analyzing these distance changes and combining them with information such as the satellites' orbital parameters, the Earth's gravity field can be calculated. Groundwater reserves in a target basin can affect changes in the Earth's gravity field, so the observed gravity field can be used to infer the groundwater reserves in that basin.

[0034] The second satellite may be a Soil Moisture Active Passive (SMAP) satellite, and the second satellite data may be soil moisture observed by the SMAP satellite.

[0035] The soil moisture active and passive detection satellite is equipped with a passive radiometer and an active radar. Both instruments operate in the L band. The passive radiometer provides soil moisture with wider coverage and higher accuracy, while the active radar provides soil moisture with higher spatial resolution. The soil moisture provided by the passive radiometer is then fused with the soil moisture provided by the active radar to obtain soil moisture with higher coverage and accuracy.

[0036] Among them, the dynamic watershed model (LSPC) will consider parameters such as the watershed flow, water environment data, pollution source data, vegetation coverage and meteorological data of the target watershed to simulate water environment data. The water environment data is the water environment data of the target watershed, and the water environment data includes the soil moisture and groundwater level of the target watershed.

[0037] The target basin can be divided into multiple sub-basins, with monitoring stations set up in some of the target sub-basins. The monitoring stations are used to monitor the pollution source data of each target sub-basin. The initial basin flow, initial water environment data, meteorological data of the target basin, and the pollution source data of each target sub-basin monitored by the monitoring stations are then input into the dynamic basin model to obtain the water environment data of the entire target basin. The initial basin flow is the default initial basin flow. The initial water environment data includes the initial soil moisture and initial groundwater level, which are also the default initial values. The meteorological data is the parameters such as rainfall, temperature, wind speed, and solar radiation in the target area predicted by the meteorological model. This meteorological data can be predicted by the meteorological model. The pollution source data includes the pollutant emissions of each pollutant in the target basin.

[0038] Among them, since the spatial resolution of the water environment data is higher than the spatial resolution of the first satellite data and the second satellite data, and the spatial resolution of the second satellite data is higher than the spatial resolution of the first satellite data, the spatial resolutions of the water environment data, the second satellite data and the first satellite data gradually decrease, and the spatial resolution of the water environment data can compensate for the spatial resolutions of the first satellite data and the second satellite data; and the accuracy and coverage of the first satellite data and the second satellite data are higher than the accuracy of the water environment data, so the first satellite data and the second satellite data can compensate for the accuracy and coverage of the water environment data.

[0039] In step S20, the watershed flow, target water environment data, pollution source data, vegetation coverage and meteorological data of the target watershed are input into the dynamic watershed model to obtain the pollutant concentration and outlet flow of the target watershed.

[0040] Among them, when the basin flow of the target basin increases, the dilution capacity of the water body in the target basin is enhanced, and the pollutant concentration will decrease. Therefore, the basin flow will affect the pollutant concentration; the basin flow will also directly affect the outlet flow of the basin. If the basin flow in the upstream of the target basin increases, the outlet flow in the downstream will also increase. It can be seen that the basin flow will affect the outlet flow.

[0041] Among them, the target water environment data includes target soil moisture and target groundwater level. As for target soil moisture, the target soil moisture will affect the migration speed of pollutants in the soil. When the target soil moisture is high, pollutants can easily enter surface water or groundwater through soil voids, thereby affecting the pollutant concentration in the target basin. When the target soil moisture is high, the soil's infiltration capacity is stronger, and more water will flow into the target basin through underground runoff, thereby increasing the outlet flow. As for the target groundwater level, changes in the target groundwater level will affect the exchange between groundwater and surface water. If the target groundwater level rises, pollutants in the groundwater will seep into the surface water, thereby increasing the pollutant concentration. And when the groundwater level rises, the amount of groundwater replenishing surface water increases, and the outlet flow will also increase accordingly. It can be seen that target soil moisture and target groundwater level will have an impact on pollutant concentration and outlet flow.

[0042] Pollution source data includes the number of pollution sources, the amount of pollutants emitted by pollution sources, etc. If the number of pollution sources and the amount of pollutant emissions increase, it will directly affect the pollutant concentration in the target watershed. For example, an increase in pollutant emissions will lead to an increase in the pollutant concentration in the target watershed.

[0043] Among them, when the vegetation coverage of the target basin increases, the vegetation can absorb and fix some pollutants, thereby reducing the amount of pollutants entering the water body of the target basin. Therefore, in the target basin with high vegetation coverage, the pollutant concentration may be appropriately reduced; and vegetation can reduce surface runoff through transpiration and enhance soil infiltration capacity. In areas with high vegetation coverage, surface runoff is reduced and the outlet flow will be reduced accordingly.

[0044] Meteorological data includes parameters such as rainfall, temperature, wind speed, and solar radiation. For example, rainfall increases surface runoff, washing pollutants from the soil into the target watershed, thereby increasing pollutant concentrations. Rainfall is a key factor in outlet flow: increased rainfall increases both surface and groundwater runoff, leading to an increase in outlet flow.

[0045] It can be seen that the dynamic watershed model will take into account the impact of comprehensive factors such as the watershed flow of the target watershed, target water environment data, pollution source data, vegetation coverage and meteorological data on the pollutant concentration and outlet flow of the target watershed, thereby predicting relatively accurate pollutant concentration and outlet flow.

[0046] In step S30, the water environment capacity of the target watershed is obtained according to the pollutant concentration of the target watershed and the outlet flow rate.

[0047] The water environment capacity of a target basin refers to the maximum amount of water it can accommodate while maintaining a pollutant concentration below a preset concentration. The target basin is also understood to be the maximum pollutant load the water body can accommodate while still meeting water quality targets. If the water environment capacity of a target water body exceeds the standard, it indicates that the target basin contains a high level of pollutants, exceeding the standard.

[0048] Among them, pollutant concentration is the amount of pollutant emissions contained in a unit volume of water in the target basin, and its unit is mg / L or ug / L. If the pollutant concentration exceeds the self-purification capacity of the water body in the target basin, the water body will not be able to meet the water quality target. Therefore, pollutant concentration is an important factor in determining the water environment capacity.

[0049] Among them, the outlet flow is the amount of water discharged from the target basin per unit time, and its unit is m³ / s or m³ / h. The outlet flow determines the dilution and transfer capacity of the water body in the target basin. A larger outlet flow can increase the dilution effect of the water body, thereby reducing the concentration of pollutants and increasing the water environment capacity. Therefore, the outlet flow is also an important factor in determining the water environment capacity.

[0050] Based on this, we can consider using the pollutant concentration and outlet flow that affect the water environment capacity to calculate the water environment capacity of the target basin.

[0051] In the process of studying water environmental capacity, the following options can be considered to obtain the water environmental capacity of the target basin, but each option has its own advantages and disadvantages:

[0052] The first option is to use a dynamic watershed model to simulate the water environment data of the target watershed, and use the simulated water environment data to obtain the water environment capacity of the target watershed.

[0053] The advantages of dynamic watershed model simulation include: the pollution source data and other data input into the dynamic watershed model are high-resolution data monitored by monitoring stations, so the spatial resolution of the water environment data output by the dynamic watershed model is also high. For example, see Figure 2 As shown in the figure, taking the target basin as a lake as an example, the lake is divided into 10 areas, and 5 sub-basins in the 10 areas are equipped with monitoring stations ( Figure 2The black dots in the figure represent monitoring stations. The monitoring stations in the five sub-basins monitor the pollution source data in these five sub-basins. The pollution source data of these five sub-basins, the basin flow of the entire target basin, the vegetation coverage of the entire target basin, and the meteorological data of the entire target basin are input into the dynamic basin model to obtain the water environment data of each sub-basin in the target basin. It can be seen that the simulation based on the dynamic basin model can obtain water environment data of smaller areas, and the spatial resolution of the water environment data obtained is higher.

[0054] The defects of dynamic watershed model simulation include: First, the input data of the dynamic watershed model includes the pollution source data monitored by the monitoring stations, so the water environment data simulated by the dynamic watershed model depends on the monitoring of the monitoring stations; second, due to the cost of deploying monitoring stations, monitoring stations are not deployed in every area of ​​the entire target watershed, and the coverage of the monitoring stations is low, so the pollution source data monitored by the monitoring stations cannot represent the pollution source data of the entire target watershed. It represents the pollution source data of the local area in the target watershed. The water environment data obtained after inputting the pollution source data of the local area into the dynamic watershed model cannot represent the entire target watershed. Its overall representativeness and coverage are poor; third, the dynamic watershed model obtains the current water environment data based on historical pollution source data, watershed flow and other data. Since there are certain errors in the dynamic watershed model no matter how it is trained, there are still certain errors between the water environment data simulated based on the dynamic watershed model and the real water environment data, and the accuracy of the simulated water environment data is not high.

[0055] The second solution is to use the first satellite data of the target watershed to observe the first satellite to obtain the water environment capacity of the target watershed.

[0056] The advantage of the first satellite observation is that the first satellite inverts the first satellite data of the target basin (such as groundwater reserves) based on the changes in the earth's gravity field. Therefore, the first satellite data is based on actual observation data and its accuracy is higher than that of dynamic basin model simulation.

[0057] A disadvantage of the first satellite observation is that the gravity field change of the first satellite is the comprehensive result of a large area, so the first satellite data observed by the first satellite is data for a large area. For example, the first satellite data observed by the first satellite is data for an area of ​​more than 200,000 square meters. The first satellite cannot observe satellite data for an area of ​​1 km*1 km.

[0058] The third solution is to use the second satellite data of the target basin to observe the second satellite to obtain the water environment capacity of the target basin.

[0059] The advantage of second-satellite observation is that the second satellite fuses the data collected by the passive radiometer and the active radar. Since the passive radiometer provides second-satellite data with wider coverage and higher accuracy, and the active radar provides second-satellite data with higher spatial resolution, after fusing the two, the second-satellite data with higher spatial resolution than the first satellite and higher accuracy and coverage than the dynamic watershed model can be obtained.

[0060] The disadvantage of the second satellite observation is that since the second satellite is also a satellite farther away from the earth's surface, when using the second satellite observation, although the spatial resolution of the observed second satellite data is higher than that of the first satellite, the spatial resolution of the second satellite data is still lower than the spatial resolution of the water environment data of the dynamic watershed model.

[0061] Through the above technical solution, fusing the first satellite data, the second satellite data and the water environment data has the following advantages:

[0062] First, since the accuracy of the first satellite data and the second satellite data is higher than that of the water environment data, target water environment data with higher accuracy can be obtained after compensating the water environment data using the first satellite data and the second satellite data.

[0063] Secondly, since the coverage of the first satellite data and the second satellite data is higher than that of the water environment data, target water environment data with higher coverage can be obtained after compensating the water environment data with the first satellite data and the second satellite data.

[0064] Third, since the first and second satellites actually observe the water environment data of the target river basin, they do not rely on the pollution source data monitored by the monitoring stations. Therefore, even if the amount of pollution source data input into the dynamic river basin model by the monitoring stations is small and the number of monitoring stations is small, the satellite data observed by the first and second satellites can be used to compensate for the water environment data output by the dynamic river basin model, thereby obtaining accurate target water environment data and reducing dependence on monitoring stations.

[0065] Fourth, since the spatial resolution of the water environment data output by the dynamic watershed model is higher than the spatial resolution of the first satellite data and the second satellite data, after using the water environment data to compensate for the first satellite data and the second satellite data, the target water environment data with higher spatial resolution can be obtained.

[0066] In summary, after fusing the first and second satellite data with the water environment data, we can obtain target water environment data with high coverage, high spatial resolution, and high accuracy without relying on too many monitoring stations. The water environment capacity calculated based on this target water environment data will also be more accurate.

[0067] Figure 3 The present disclosure relates to an exemplary embodiment of an exemplary scheme for temporally aligning remote sensing data with watershed flow, comprising the following steps:

[0068] In step S40 , for any remote sensing data among the vegetation data and the second satellite data, time point alignment is performed between the remote sensing data and the watershed flow.

[0069] Among them, vegetation data is used to determine the vegetation coverage input into the dynamic watershed model, which is used to indicate the vegetation coverage area in the target watershed.

[0070] Among them, after the remote sensing data and the basin flow are aligned in time, there is a set of remote sensing data and basin flow at any time point.

[0071] It is understandable that the time resolution of remote sensing data is 30 minutes, which means that remote sensing data is collected every 30 minutes, and the time resolution of basin flow is 1 day, which means that basin flow is collected every 1 day. Therefore, the cubic spline interpolation method can be used to encrypt the time resolution of basin flow to the hourly level, so that the time points of basin flow can be aligned with the time points of remote sensing data as much as possible.

[0072] Through the above technical solution, the remote sensing data and the watershed flow can be aligned in time. Then, the second satellite data in the remote sensing data and the watershed flow are time-aligned, and the vegetation coverage obtained based on the vegetation data in the remote sensing data and the watershed flow are also time-aligned. Then, the watershed flow, vegetation coverage and the second satellite data input into the dynamic watershed model are time-aligned, thereby obtaining pollutant concentrations and outlet flows at different time points with higher time resolution.

[0073] Figure 4 This is an exemplary embodiment of the present disclosure, which is used to explain an exemplary solution for converting vegetation data into vegetation coverage, including the following steps:

[0074] In step S50, the vegetation data is normalized to obtain normalized vegetation data.

[0075] The vegetation data includes the collected near-infrared band reflectance and red light band reflectance. The normalized vegetation data can be obtained using the following formula:

[0076] (1)

[0077] In formula (1), is the near-infrared reflectivity, is the reflectivity in the red light band.

[0078] In step S60, the normalized vegetation data is processed to obtain vegetation coverage.

[0079] The calculation formula for vegetation coverage obtained by processing the normalized vegetation data is as follows:

[0080] FCC=0.72⋅NDVI+0.15*R 2 (2)

[0081] In formula (2), FCC is vegetation coverage, 0.72 means that for every unit increase in NDVI, FCC increases by 0.2 units on average, 0.15 is the intercept, which means that when NDVI is 0, the baseline value of FCC is 0.15, which is used to reflect the impact of non-vegetation coverage. NDVI is the normalized vegetation data, R 2 It is the coefficient of determination, which represents the extent to which NDVI can explain FCC. When R²=0.89, it means that NDVI can explain 89% of the FCC variation, and the fitting effect is good.

[0082] In the related art, the vegetation coverage is measured manually, which is cumbersome and inefficient.

[0083] Through the above technical solution, vegetation coverage can be calculated in real time through vegetation data in remote sensing data, thereby solving the problems of tediousness and inefficiency brought about by manual measurement of vegetation coverage, and eliminating the nonlinear deviation between the normalized vegetation data and vegetation coverage through normalization, thereby ensuring the credibility of the obtained vegetation coverage.

[0084] Figure 5 This is an exemplary embodiment involved in the above step S10, which is used to explain an exemplary solution for fusing the first satellite data, the water environment data, and the second satellite data, including the following steps:

[0085] In step S11, the first satellite data and the water environment data are fused to obtain first fused data.

[0086] In one possible implementation, fusing the first satellite data with the water environment data to obtain first fused data includes the following sub-steps:

[0087] Sub-step A1: obtaining a first observation error of the first satellite based on the first satellite data and the water environment data.

[0088] Among them, the water environment data includes first water environment data and second water environment data. The first water environment data is the water environment data in the first state, and the first state indicates that the first predicted noise is not 0; the second water environment data is the water environment data in the second state, and the second state indicates that the first predicted noise is 0, that is, the first predicted noise does not exist.

[0089] For the first water environment data, the first water environment data at the current moment can be obtained based on the first water environment data predicted by the dynamic watershed model at the previous moment, the state transfer matrix and the first prediction noise.

[0090] The calculation formula for obtaining the first water environment data at the current moment based on the first water environment data predicted by the dynamic watershed model at the previous moment, the state transfer matrix, and the first prediction noise is as follows:

[0091] (3)

[0092] In formula (3), x k is the first water environment data at the current moment, A is the state transfer matrix, is the first water environment data of the previous moment, w k is the first predicted noise.

[0093] The first water environment data at the current moment includes the soil moisture and groundwater level at the current moment. The first water environment data at the current moment is data obtained by simulating the dynamic watershed model, which may have errors due to structural errors of the dynamic watershed model or parameter uncertainties of the dynamic watershed model.

[0094] The state transfer matrix is ​​in the form of a 2X2 matrix, which corresponds to the state transfer of soil moisture and groundwater level respectively. The state transfer matrix can quantitatively describe the physical connection between soil moisture and groundwater level, and drive the forward movement of the first water environment data.

[0095] The first prediction noise is used to describe the prediction uncertainty of the dynamic watershed model, which is represented by the covariance matrix Q. The covariance matrix Q is used to describe the prediction uncertainty of the dynamic watershed model. The covariance matrix Q can be determined by the error statistics of the dynamic watershed model verification period. The error indicates the error (RMSE) between the first water environment data simulated by the dynamic watershed model and the real water environment data. The covariance matrix Q = error RMSE 2 .

[0096] As can be seen from formula (3), it describes the dynamic process of water environment data such as soil moisture and groundwater level over time. It has the following functions: First, temporal continuity. It can predict the first water environment data at the current moment based on the first water environment data at the previous moment, reflecting the dynamic evolution law of the target watershed system. Second, physical constraints. The state transition matrix is ​​embedded in the water environment data to ensure that the changes in water environment data conform to natural physical laws. Third, it provides a basis for generating subsequent first observation errors.

[0097] For the second water environment data at the current moment, the second water environment data is the first predicted noise w in the first water environment data at the current moment. k The calculation formula for the second water environment data at the current moment is as follows:

[0098] (4)

[0099] In formula (4), It is the second water environment data at the current moment. is the state transition matrix, It is the second water environment data of the previous moment.

[0100] Obtaining a first observation error of the first satellite based on the first satellite data and the water environment data includes: mapping the second water environment data at the current moment using a first mapping matrix to obtain mapped second water environment data; and then subtracting the mapped second water environment data from the first satellite data to obtain the first observation error, which is calculated using the following formula:

[0101] vk GRACE = (5)

[0102] In formula (5), vk GRACE is the first observation error, is the first satellite data observed by the first satellite, is the first mapping matrix, It is the second water environment data at the current moment.

[0103] The first observation error indicates the measurement error of the first satellite itself (for example, the inversion error of the grace gravity field) and the atmospheric correction error of the first satellite (for example, the interference caused by changes in air pressure on the first satellite's observation). The first observation error has the characteristics of Gaussian white noise and is used to quantify the reliability of the first satellite data. The larger the gap between the first satellite data and the mapped second water environment data, the less reliable the first satellite data.

[0104] The first satellite data is satellite data observed by the first satellite, which may be terrestrial water storage.

[0105] The first mapping matrix is ​​used to map the two-dimensional secondary water environment data (soil moisture and groundwater level) simulated by the dynamic watershed model into a one-dimensional form, thereby aligning it with the one-dimensional terrestrial water storage. The first mapping matrix converts the weighted sum of the soil moisture and groundwater level simulated by the dynamic watershed model into observational data in the same dimension as the terrestrial water storage. This allows for the coupling of the high-spatial-resolution soil moisture and groundwater level with the low-spatial-resolution terrestrial water storage. Essentially, it calculates the contribution of the soil moisture and groundwater level simulated by the dynamic watershed model to the terrestrial water storage observed by the first satellite.

[0106] It can be seen from the above formula (5) that formula (5) maps the first satellite data obtained by the first satellite inversion and the second water environment data obtained by the dynamic watershed model simulation to the same dimension, achieving the following effects: First, spatial constraint, the two second water environment data of soil moisture and groundwater level that cannot be observed are associated and assimilated with the first satellite data of terrestrial water storage that can be observed; second, error transfer, through the first mapping matrix, the error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data is converted into the error in the satellite observation space; third, different first mapping matrices can be designed to adapt to different types of first satellite data.

[0107] Sub-step A2: correcting the first observation error using a first Kalman gain to obtain a first target observation error.

[0108] The first Kalman gain is used to measure the uncertainty of the second water environment data. The first Kalman gain matrix can be obtained based on the error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data, the first mapping matrix, and the first observation noise. The calculation formula is as follows:

[0109] (6)

[0110] In formula (6), is the first Kalman gain, is the error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data, is the first mapping matrix, is the first observation noise.

[0111] The first Kalman gain is used to correct the first observation error and measure the importance between the first observation error and the second water environment data obtained by simulating the dynamic watershed model.

[0112] The error between the second water environment data predicted by the dynamic watershed model and the actual water environment data reflects the error of the dynamic watershed model itself, such as inaccurate parameters of the dynamic watershed model itself or simulation errors caused by simplification of the simulation process.

[0113] The first observation noise is used to indicate the uncertainty of the first satellite data obtained by inverting the first satellite. The uncertainty comes from the measurement error of the first satellite, the atmospheric correction error, etc.

[0114] Sub-step A3: obtaining the first fused data based on the water environment data and the first target observation error.

[0115] Obtaining the first fused data according to the water environment data and the first target observation error includes: obtaining the second fused data according to the second water environment data at the current moment and the first target observation error.

[0116] The product of the first Kalman gain and the first observation error can be used as the first target observation error, and the first target observation error can be superimposed on the second water environment data to obtain the first fusion data. The calculation formula is as follows:

[0117] (7)

[0118] In formula (7), is the first fusion data, It is the second water environment data at the current moment. is the first Kalman gain, is the first observation error.

[0119] The first fused data represents coarse-scale or low-resolution fused data. It is the fusion of low-resolution but highly accurate first satellite data with high-resolution but less accurate second water environment data, resulting in high-accuracy, low-resolution first fused data. It is understood that the low resolution of the first fused data refers to the fact that the resolution of the first fused data is lower than that of the target water environment data obtained by the final fusion, but objectively the resolution of the first fused data is not low.

[0120] The first Kalman gain is used to measure the weight between the second water environment data simulated by the dynamic watershed model and the first satellite data simulated by the first satellite. Combining formula (6) and formula (7), it can be seen that when the first observation noise of the first satellite data When it becomes larger, the first Kalman gain becomes smaller, thereby reducing the first observation error of the first satellite data The impact on the first fusion data increases the second water environment data obtained by dynamic watershed model simulation. Impact on the first fusion data, the second water environment data The weight is greater than the first observation error The weight of , indicating that the prediction results of the dynamic watershed model are more trusted at this time. When the error between the second water environment data simulated by the dynamic watershed model and the real water environment data When it is large, the first Kalman gain becomes larger, thereby increasing the first observation error of the first satellite data The impact of the first fusion data is reduced by the second water environment data obtained by the dynamic watershed model simulation. Impact on the first fusion data, the first observation error The weight is greater than the second water environment data The weight of indicates that the observation results of the first satellite are more trusted at this time.

[0121] It can be seen from the above formula (7) that the first Kalman gain can be used to measure the degree of influence of the first satellite data and the second water environment data obtained by simulating the dynamic watershed model on the first fused data. If the accuracy of the first satellite data is higher than that of the second water environment data, the first Kalman gain will make the weight of the first satellite data higher than that of the second water environment data, so that the first fused data obtained is mainly affected by the first satellite data; conversely, if the accuracy of the first satellite data is lower than that of the second water environment data, the first Kalman gain will make the weight of the first satellite data lower than that of the second water environment data, so that the first fused data obtained is mainly affected by the second water environment data, and ultimately ensure the accuracy of the first fused data obtained.

[0122] In step S12, the first fused data and the second satellite data are fused to obtain the target water environment data.

[0123] In a possible implementation, fusing the first fused data with the second satellite data to obtain target water environment data includes the following sub-steps B1 to B3:

[0124] Sub-step B1: obtaining a second observation error of the second satellite based on the second satellite data and the second fusion data.

[0125] Among them, the second fused data is the data obtained after the third fused data is disturbed by the disturbance term, and the third fused data is the data corresponding to the first fused data. For example, the third fused data is the data obtained based on the first water environment data when the first observation noise is not 0, and the first fused data is the data obtained based on the second water environment data when the first observation noise is 0.

[0126] The second fused data is obtained through the following sub-steps C1 and C2.

[0127] Sub-step C1, obtaining third fusion data according to the first satellite data, the first water environment data and the first Kalman gain.

[0128] For example, the second observation error can be obtained based on the first satellite data and the first water environment data, and then the second observation error is corrected using the first Kalman gain to obtain the second target observation error; and the third fusion data is obtained based on the first water environment data and the second target observation error. The calculation formula is as follows:

[0129] (8)

[0130] In formula (8), is the third fusion data, It is the first water environment data at the current moment. is the first Kalman gain, is the first observation error, which is the same as the first observation error in the above formula (5) The essence is the same, both represent the observation error of the first satellite, It is the first satellite data, is the first mapping matrix.

[0131] From the above formula (8), formula (5) and formula (7), it can be seen that after substituting formula (5) into formula (7), the first fusion data obtained is essentially the same as the third fusion data obtained based on formula (8), but the first fusion data obtained by formula (7) is based on w k The third fusion data in formula (8) is based on the second water environment data when w is 0. k The first water environment data when it is not 0 is obtained.

[0132] Sub-step C2, obtaining second fused data based on the disturbance term and the third fused data.

[0133] For example, the third fused data at the current moment may be disturbed by a disturbance term to obtain the second fused data, which is calculated as follows:

[0134] (9)

[0135] In formula (9), is the second fused data, is the third fusion data, is the disturbance term.

[0136] It can be seen from formula (9) that the disturbance term can be superimposed on the third fused data to obtain the second fused data.

[0137] The third fusion data is obtained based on the first water environment data when the first observation noise wk is not 0. It is coarse-scale, low-resolution data obtained by combining the first water environment data obtained by simulating the dynamic watershed model with the first observation error of the first satellite.

[0138] The disturbance term is a local fine-scale, high-resolution disturbance term, which represents the high-resolution spatial details that are not captured in the coarse-scale, low-resolution background field. For example, it can be represented by high-resolution soil texture differences such as the water holding capacity of sand and clay, local vegetation cover changes caused by differences in crop evapotranspiration, and micro-topography effects such as slope runoff convergence.

[0139] The second fused data is fine-scale and higher-resolution data obtained by taking fine-scale disturbance terms into account on the basis of the second fused data.

[0140] The above formulas (8) and (9) have the following functions: the first water environment data obtained by simulating the dynamic watershed model is fused with the first observation error of the first satellite data, and the coarse-scale, low-resolution third fused data is used as the initial background of the disturbance term. The disturbance term is then used to capture the impact of terrain and vegetation coverage differences on soil moisture that cannot be distinguished in a coarse-scale environment. The individual differences of each region can be taken into account, so that the second fused data obtained has higher resolution and higher accuracy.

[0141] Obtaining the second observation error based on the second fused data and the second satellite data includes: mapping the second fused data using a second mapping matrix to obtain mapped second fused data; and then subtracting the mapped second fused data from the second satellite data to obtain the second observation error, which is calculated as follows:

[0142] (10)

[0143] In formula (10), is the second observation error, is the second satellite data, is the second mapping matrix, is the second fused data.

[0144] The second observation error indicates an observation error of the second satellite itself.

[0145] The second satellite data is satellite data observed by a second satellite, which may be soil moisture with high spatial resolution.

[0146] The second mapping matrix is ​​a local observation operator, which is used to map the fine-scale, high-resolution second fused data to the observation space of the second satellite, so that the mapped second fused data can have the same physical dimension and spatial resolution as the second satellite data obtained by inversion from the second satellite.

[0147] It can be seen from formula (10) that formula (10) maps the second satellite data and the second fused data observed by the second satellite to the same dimension, achieving the following effects: first, the second satellite data is the data in the observation space of the second satellite data, and the second fused data is the data in the prediction space of the dynamic watershed model, which solves the scale and unit differences between the observation space of the second satellite and the prediction space simulated by the dynamic watershed model; second, the high-resolution observation of the second satellite is used to correct the local deviations in the second fused data at fine scale and high resolution.

[0148] Sub-step B2: correcting the second observation error using the second Kalman gain to obtain a second target observation error.

[0149] The second Kalman gain is used to measure the uncertainty of the first fusion data and the second satellite data. The second Kalman gain can be obtained based on the error between the first water environment data predicted by the dynamic watershed model and the actual water environment data, the second mapping matrix, and the second observation noise. The calculation formula is as follows:

[0150] (11)

[0151] In formula (11), is the second Kalman gain, is the error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data, is the second mapping matrix, is the second observation noise.

[0152] The second Kalman gain is a weight matrix used to balance the uncertainty between the error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data and the second observation noise. The error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data can also be the error between the first water environment data predicted by the dynamic watershed model and the actual first water environment data.

[0153] The error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data reflects the error of the dynamic watershed model itself, such as the simulation error caused by inaccurate parameters of the dynamic watershed model itself or the simplification of the simulation process. This error is the uncertainty of the second water environment data / first water environment data output by the dynamic watershed model. This uncertainty is due to changes in water environment data such as soil moisture and groundwater level caused by terrain influences and human activities, which leads to uncertainty in the second water environment data / first water environment data output by the dynamic watershed model.

[0154] The second observation noise is used to indicate uncertainty of the second satellite data obtained by inversion of the second satellite.

[0155] The error between the second water environment data predicted by the dynamic watershed model and the actual second water environment data can be calculated using the following formula:

[0156] (12)

[0157] In formula (12), is the error between the second water environment data and the true second water environment data, is the i-th second water environment data, It is the average value of multiple second water environment data.

[0158] Sub-step B3, obtaining target water environment data based on the first fusion data and the second target observation error.

[0159] The product of the second Kalman gain and the second observation error can be used as the second target observation error, and the second target observation error can be superimposed on the first fusion data to obtain the target water environment data. The calculation formula is as follows:

[0160] (13)

[0161] In formula (13), is the target water environment data, is the first fusion data, is the second Kalman gain, is the second observation error.

[0162] The target water environment data represents the fused water environment data at fine scale and high resolution. It is the target water environment data with higher accuracy and resolution obtained by secondary fusion of the first fused data with high accuracy and high resolution and the second satellite data with high accuracy and high resolution.

[0163] The first fused data represents the fused data at a coarse scale or low resolution (coarse scale and low resolution are relative to the target water environment data). It is the fusion of the low-resolution but high-accuracy first satellite data with the high-resolution but low-accuracy second water environment data, thereby obtaining the first fused data with high accuracy and low resolution.

[0164] The second Kalman gain is used to measure the weight between the first fused data after the second water environment data simulated by the dynamic watershed model is fused with the first satellite data observed by the first satellite, and the second satellite data. Combining formula (11) and formula (13), it can be seen that when the second observation noise of the second satellite data When it increases, the second Kalman gain becomes smaller, thereby reducing the second observation error of the second satellite data. The influence of the first fusion data on the target water environment data is increased, and the weight of the first fusion data is greater than the weight of the second observation error, indicating that the dynamic watershed model and the first fusion data of the first satellite are more trusted at this time. When the error between the second water environment data simulated by the dynamic watershed model and the real second water environment data is When it increases, the second Kalman gain becomes larger, thereby increasing the second observation error of the second satellite data Regarding the impact on the target water environment data, the weight of the second observation error of the second satellite data is greater than that of the first fusion data, which means that the observation results of the second satellite are more trusted at this time.

[0165] As can be seen from formula (13), the second Kalman gain can be used to measure the degree of influence of the second satellite data and the first fusion data on the target water environment data. If the accuracy of the second satellite data is higher than that of the first fusion data, the second Kalman gain will make the weight of the second satellite data higher than that of the first fusion data, so that the target water environment data obtained is mainly affected by the second satellite data; conversely, if the accuracy of the second satellite data is lower than that of the first fusion data, the second Kalman gain will make the weight of the second satellite data lower than that of the first fusion data, so that the target water environment data obtained is mainly affected by the first fusion data.

[0166] Through the above technical solution, the first satellite data with high accuracy observed by the first satellite is first fused with the water environment data with low accuracy obtained by dynamic watershed model simulation, so as to obtain first fused data with low spatial resolution and high accuracy; then, the first fused data is fused again using the second satellite data with high resolution and high accuracy observed by the second satellite, so as to further improve the resolution and accuracy of the first fused data.

[0167] Figure 6 The exemplary embodiment of the present disclosure is used to explain an exemplary solution for training a dynamic watershed model, including the following steps:

[0168] In step S70, the initial model is trained using historical training samples and labels corresponding to the historical training samples. During multiple training processes, the network parameters of the initial model are updated until the error between the predicted water environment data output by the trained initial model and the actual water environment data is less than a preset value, thereby obtaining a trained dynamic watershed model.

[0169] Among them, historical training samples include historical watershed flow, water environment data, pollution source data, vegetation coverage and meteorological data, and the labels corresponding to historical training samples include historical pollutant concentrations and export flows.

[0170] Among them, the network parameters of the initial model include roughness and permeability coefficient. Roughness refers to the roughness of the river channel in the target basin. The higher the roughness of the river channel in the target basin, the lower the river channel fluidity, which affects the outlet flow; the permeability coefficient includes the permeability coefficient of the soil in the target basin and the permeability coefficient of the soil around the target basin. The higher the permeability coefficient of the soil in the target basin, the more water will be lost, and the outlet flow will decrease. The higher the permeability coefficient of the soil around the target basin, the more water will infiltrate into the target basin, and the outlet flow will increase.

[0171] In the process of training the initial model, an online particle optimization algorithm can be used to obtain the target network parameters of the initial model.

[0172] For example, network parameters such as roughness and permeability can be treated as particles. Each particle has its own position and velocity. The position represents the particle's position in the search space, and the velocity represents the particle's flight speed and direction in the search space. The optimal solution found by each particle is called the individual optimal solution, and each particle knows the optimal solution in the entire search space, which is called the global optimal solution. Each particle adjusts its own flight speed and direction. If a particle discovers that another particle has found a better individual optimal solution, it adjusts its own flight speed and direction to move closer to the other individual optimal solution. Repeat the above steps until all particles in the search space are close to the global optimal solution, thus obtaining the network parameters such as roughness and permeability at the global optimal solution.

[0173] Through the above technical solution, a dynamic watershed model can be trained and used to simulate high-resolution water environment data. The particle optimization algorithm can also help the dynamic watershed model quickly find its own network parameters, thereby realizing rapid training of the dynamic watershed model.

[0174] Figure 7 This is an exemplary embodiment involved in the above step S20, which is used to explain an exemplary solution for obtaining the pollutant concentration and outlet flow of the target watershed based on the watershed flow of the target watershed, target water environment data, pollution source data, vegetation coverage and meteorological data, including the following steps:

[0175] In step S21, a plurality of target sub-basins are screened out from the plurality of sub-basins of the target basin.

[0176] The multiple target sub-basins are sub-basins in the target basin where monitoring stations are set.

[0177] For example, see Figure 2 As shown in FIG, the target basin is divided into 10 sub-basins, 5 of which are equipped with monitoring stations. The sub-basins where these 5 monitoring stations are located can be used as target sub-basins.

[0178] In step S22, the watershed flow of the target watershed, target water environment data, vegetation coverage, future meteorological data and pollution source data of the multiple target sub-watersheds are input into the dynamic watershed model to obtain the future pollutant concentration and outlet flow of the target watershed.

[0179] For example, the basin flow of the entire target basin, target water environment data, vegetation coverage, future meteorological data, and pollution source data of the sub-basins where the five monitoring stations are located can be input into the dynamic basin model to simulate the future pollutant concentration and export flow of each sub-basin in the target basin. Of course, the future pollutant concentration and export flow of the entire target basin can also be simulated.

[0180] Through the above technical solution, the pollution source data measured by a small number of monitoring stations and the target water environment data obtained by the above fusion can be used to obtain the pollutant concentration and outlet flow of each sub-basin under the entire target basin. When the number of monitoring stations deployed in the target basin is small, the data monitored by the monitoring stations cannot reflect the entire target basin. Therefore, the data monitored by the monitoring stations can be fused and compensated by the data observed by the second satellite data and the first satellite data. Then, the data input into the dynamic basin model is comprehensive and accurate input data for the target basin. Based on this data, the dynamic basin model can predict more comprehensive and accurate pollutant concentrations and outlet flows, thereby reducing dependence on monitoring stations.

[0181] Figure 8 This is an exemplary embodiment involved in the above step S30, which is used to explain an exemplary solution for obtaining the water environment capacity of the target basin based on the pollutant concentration and outlet flow of the target basin, including the following steps:

[0182] In step S31 , a target time period during which the pollutant concentration in the target watershed is less than a target pollutant concentration is determined.

[0183] Among them, the dynamic watershed model outputs the pollutant concentration and outlet flow of the target watershed in the future. Therefore, the target time period in which the pollutant concentration is lower than the target pollutant concentration can be screened out from the pollutant concentration in the target watershed in the future, and the target time period in which the water quality of the target watershed meets the standard can also be screened out.

[0184] For example, the pollutant concentration in the target watershed is C sim(t) , the target pollutant concentration is Ctarget For example, we can select C from the target basin in the future. sim(t) Less than or equal to C target target time period.

[0185] In step S32 , for each time point in the target time period, the pollutant flux at the time point is obtained according to the pollutant concentration and the outlet flow rate.

[0186] The product of the pollutant concentration and the outlet flow rate can be used as the pollutant flux. The pollutant flux is the amount of pollutant discharged from the outlet of the target basin per unit time. The calculation formula is as follows:

[0187] F(t)=C sim(t) ×Q(t)(14)

[0188] In formula (14), F(t) is the pollutant flux at the current time point, C sim(t) is the pollutant concentration at the current time point, and Q(t) is the outlet flow rate at the current time point.

[0189] In step S33, the water environment capacity is obtained according to the pollutant fluxes at multiple time points in the target time period.

[0190] The pollutant fluxes at multiple time points in the target time period can be integrated to obtain the water environment capacity, which is calculated as follows:

[0191] (15)

[0192] In formula (15), W is the water environment capacity within the target time period for achieving water quality standards, and F(t) is the pollutant flux at the current time point.

[0193] It is understandable that after obtaining the water environment capacity, the water capacity threshold corresponding to the current pollutant concentration can be compared with the water environment capacity. If the water environment capacity is greater than the water capacity threshold, it means that the water environment capacity exceeds the standard and there are more pollutants in the target watershed. Of course, if the water environment capacity is less than or equal to the water capacity threshold and the water environment capacity does not exceed the standard, the greater the water environment capacity, the more water body can accommodate pollutants.

[0194] Through the above technical solution, the water environment capacity of the target river basin during the period when the pollutant concentration is lower than the target pollutant concentration can be determined. If the water environment capacity is greater than the water environment capacity threshold, it means that the water environment capacity of the target river basin during the period when the water quality meets the standard is still exceeded, and further means that the water environment capacity of the target river basin during the period when the water quality does not meet the standard will also exceed the standard. There is no need to measure the water environment capacity of the target river basin during the remaining periods when the water quality does not meet the standard, thereby reducing the additional cost brought about by measuring the water environment capacity of the target river basin.

[0195] Figure 9 This is an exemplary embodiment of the present disclosure, which is used to explain that after obtaining the water environment capacity, the target parameters affecting the water environment capacity, the management resources and the ecological flow are screened out, including the following steps:

[0196] In step S80, multiple elements that affect the water environment capacity, management resources and ecological abnormality are determined.

[0197] Water environmental capacity refers to the maximum amount of water a target watershed can accommodate, provided the pollutant concentration in the target watershed is below a preset pollutant concentration. Factors influencing water environmental capacity include chemical oxygen demand (COD), ammonia nitrogen emission thresholds, sewage treatment equipment costs, and ecological flow thresholds.

[0198] Chemical oxygen demand is the main indicator for measuring the content of organic matter in water bodies. It represents the amount of oxygen consumed when strong oxidants are used to oxidize organic matter in water bodies under certain conditions. The higher the chemical oxygen demand, the more serious the organic pollution in the water body.

[0199] The ammonia nitrogen emission threshold is the maximum amount of pollutants allowed to be discharged into water bodies.

[0200] The cost of sewage treatment equipment indicates the cost of hardware equipment required to treat pollutants in water bodies.

[0201] The ecological flow threshold is the minimum downstream flow to ensure the ecological health of the river. If the outlet flow of the target basin is less than or equal to the ecological flow threshold, it means that the river ecology of the target basin is unhealthy.

[0202] Among them, governance resources refer to the resource costs required to control pollutants. Elements affecting governance resources include equipment investment costs and equipment operation and maintenance costs.

[0203] The equipment investment cost is determined by the initial investment cost, equipment scale and scale effect. The expression of equipment investment cost is as follows:

[0204] (16)

[0205] In formula (16), is the equipment investment cost, and b are parameters related to the device type. is the proportional coefficient, representing the initial investment cost under the unit equipment scale, b is the index, representing the scale effect, Represents the device scale, such as the number of devices.

[0206] It can be seen from the above formula (16) that if b is less than 1, it means that there is a scale effect. As the equipment scale increases, the equipment input cost of a single device will decrease, and the total equipment input cost will decrease; if b is equal to 1, it means that there is a linear relationship between the equipment input cost and the equipment scale; if b is greater than 1, it means that there is diseconomy of scale. As the equipment scale increases, the equipment input cost of a single device will increase, and the total equipment input cost will increase.

[0207] Equipment operation and maintenance costs are time-dependent and are typically influenced by factors such as equipment load, energy consumption, equipment scale, and equipment aging. For example, equipment aging can lead to higher equipment operation and maintenance costs over time. Similarly, larger equipment scale leads to higher equipment operation and maintenance costs.

[0208] Ecological anomaly refers to the degree of ecological anomaly in the target basin, which can also be expressed as ecological risk. Factors influencing ecological instability include the target duration of time within a specified period during which the outlet flow of the target basin is less than or equal to the outlet flow threshold. A longer target duration indicates a longer period of time during which the outlet flow is less than or equal to the outlet flow threshold, and a longer period during which the outlet flow does not meet the standard. This indicates a more abnormal ecological environment in the target basin.

[0209] In step S90, a plurality of target elements are screened out from the plurality of elements with the goal of maximizing the water environment capacity, minimizing the management resources, and minimizing the ecological abnormality.

[0210] Among them, maximizing the water environment capacity means maximizing the amount of water that carries pollutants, thereby reducing the concentration of pollutants.

[0211] For example, the expression for maximizing water environment capacity is as follows:

[0212] (17)

[0213] In formula (17), is the objective function of maximizing the water environment capacity, is the water environment capacity of the i-th pollutant calculated in step S30 above, It is the true and accurate water environment capacity of the i-th pollutant.

[0214] It can be seen from formula (17) that the simulated water environment capacity of the i-th pollutant can be divided by the actual water environment of the i-th pollutant to obtain the water capacity ratio; then the water capacity ratios of various pollutants are added together to obtain the total water capacity. With the goal of maximizing the total water capacity, the target element that can maximize the water environment capacity can be screened out from the four elements that affect the water environment capacity, namely, chemical oxygen demand, ammonia nitrogen emission threshold, sewage treatment equipment cost, and ecological flow threshold.

[0215] Among them, minimizing governance resources means minimizing the cost of controlling pollutants, thereby reducing costs.

[0216] For example, the expression for minimizing governance resources is as follows:

[0217] (18)

[0218] In formula (18), is the objective function of minimizing governance resources, is the equipment investment cost, It is the equipment operation and maintenance cost.

[0219] It can be seen from formula (18) that the sum of equipment investment cost and equipment operation and maintenance cost can be used as governance resources, and the target element that can minimize governance resources can be screened out from these two elements.

[0220] Among them, minimizing ecological abnormality means minimizing the abnormality of the ecological environment, thereby improving the health of the ecological environment.

[0221] For example, the expression for minimizing ecological abnormality is as follows:

[0222] (19)

[0223] In formula (19), is the objective function of minimizing ecological abnormality, is the outlet flow of the target basin calculated in real time, is the egress flow threshold, Indicates the target duration for determining that the egress flow rate is less than or equal to the egress flow rate threshold; T is the total duration of the observation.

[0224] From the above formula (19), it can be seen that the target duration of the export flow not meeting the standard as a percentage of the total duration can be calculated, and this duration ratio can be used as the ecological abnormality. The larger the duration ratio, the higher the ecological abnormality. Then, the duration ratio of the export flow not meeting the standard or the target duration can be used as the target element that can minimize the ecological abnormality.

[0225] In some scenarios, three target elements can be selected from the elements that affect water environment capacity, governance resources and ecological abnormality for optimization. For example, chemical oxygen demand can be selected from the elements that affect water environment capacity, equipment investment cost can be selected from the elements that affect governance resources, and the target duration when the outlet flow is less than the outlet flow threshold can be selected from the elements that affect ecological abnormality. Chemical oxygen demand, equipment investment cost and target duration can be used as target elements.

[0226] In step S100 , a plurality of target parameters are respectively screened out from the parameters of the plurality of target elements.

[0227] Among them, each target element has multiple parameters. For example, multiple chemical oxygen demands can be used to measure the target watershed, there can be multiple sizes of equipment investment costs to invest in the treatment of pollutants in the target watershed, and there can be multiple target durations for the outlet flow of the target watershed to not meet the standards. Therefore, multiple target parameters can be screened out from multiple chemical oxygen demands, multiple equipment investment costs and multiple target durations, and then the target watershed can be measured and treated with multiple target parameters, so that the water environment capacity of the target watershed can be maximized, the treatment resources can be minimized and the ecological abnormality can be minimized.

[0228] For example, the chemical oxygen demand is 5gm / L, 10mg / L, and 15mg / L, the equipment investment cost is 100,000 yuan, 200,000 yuan, and 300,000 yuan, and the target time for the outlet flow to fail to meet the standard is 1h, 2h, and 3h. From these three target elements, the chemical oxygen demand of 5gm / L, the equipment investment cost of 100,000 yuan, and the target time of 1h can be selected as the target parameters for operation. Ultimately, the water environment capacity of the target basin can be larger, the management resources can be less, and the ecological abnormality can be smaller, so that the ecological environment of the target basin can be better with less management cost.

[0229] In one possible implementation, for multiple objective functions such as maximizing water environment capacity, minimizing governance resources, and minimizing ecological abnormality, a non-dominated sorting method can be used to prioritize them. The higher the priority of the objective function, the higher the weight it occupies in the optimization process, and the objective function with a higher priority will be given priority in optimization.

[0230] For non-dominant relationships, in the process of multi-objective optimization with the three objectives of maximizing water environment capacity, minimizing governance resources, and minimizing ecological abnormality, the solution of one objective function (such as an element) may not be better than another solution in all objective functions, but may be better than another solution in some objective functions, and not inferior to another solution in other objective functions. This relationship is called a non-dominant relationship. The three objective functions can be divided into levels such as F1, F2, and F3. The levels of F1, F2, and F3 gradually decrease, and the level of F1 is the highest non-dominated level. All influencing elements of the objective function under the F1 level are not dominated by other elements.

[0231] A parallel bubble sort can be used to sort the levels of multiple target functions, with 3241 as the level. Bubble sort compares two adjacent numbers from left to right. If the number on the left is larger than the number on the right, they swap positions. For example, if 3 is larger than 2, it swaps to 23. Then compare 41. If 4 is larger than 1, it swaps to 14, and so on to 2314. Then repeat the sorting with 2314 until all the data becomes 1234, completing the hierarchical order. The parallel bubble sort is to divide 21 and 41 into two groups in each round of comparison, and sort and compare them in parallel, so as to quickly arrange the order to 1234.

[0232] In one possible implementation, the crowding degree of the target element can be approximated to retain the target element and remove the non-target elements. For multiple target elements under the same objective function, the sum of the distances between the element and other elements in the objective function can be calculated, and the distance can be used to decide whether to retain the target element.

[0233] For example, crowding is used to measure the distribution density of elements in the target function population. For an element, the sum of the distances between the element and its adjacent elements in the population can be calculated. The farther the distance, the fewer elements around the element, and the better the diversity of the individual.

[0234] Finally, the three objective functions can be regarded as three populations, and three target elements can be selected from the three populations respectively. For example, chemical oxygen demand can be selected from the population corresponding to the water environment capacity, equipment investment cost can be selected from the governance resources, and target duration can be selected from the ecological abnormality. This will enable the target element with the highest non-dominated hierarchy and the largest crowding degree in the current population. The highest non-dominated hierarchy indicates that the element is not dominated by other elements in the population, and the element performs relatively small in each objective function. The maximum crowding degree indicates that the diversity of the target element is good and the diversity of the population is good.

[0235] Figure 10This is an exemplary embodiment involved in the above step S90, which is used to explain an exemplary solution for screening multiple target elements when a penalty function is introduced, including the following steps:

[0236] In step S91 , a penalty value is determined when the compliance rate of the pollutant concentration does not meet a preset compliance rate.

[0237] The pollutant concentration compliance rate refers to the ratio of the duration of time when the pollutant concentration is lower than the preset concentration to the preset duration. For example, if the pollutant concentration is lower than the preset concentration for 8 days out of a 10-day period, the compliance rate is 80%.

[0238] The penalty value can be obtained based on the difference between the target rate and the preset target rate and the number of iterations. The calculation formula is as follows:

[0239] (20)

[0240] In formula (20), is the penalty value; It is the penalty coefficient caused by the number of iterations, and also the penalty coefficient caused by the number of times the compliance rate does not meet the preset compliance rate. The more times the compliance rate does not meet the preset compliance rate, the higher the penalty coefficient. It is the difference between the preset compliance rate and the compliance rate; p is the penalty intensity coefficient, for example, p is greater than or equal to 2.

[0241] From the above formula (20), it can be seen that the greater the difference between the preset target rate and the target rate, the greater the penalty value. The greater the difference between the preset target rate and the target rate, the greater the gap between the current target rate and the expected target rate, and therefore the higher the penalty value. The more iterations the target rate does not meet the preset target rate, the greater the penalty coefficient and the corresponding penalty value.

[0242] The penalty coefficient is obtained by the following formula:

[0243] (twenty one)

[0244] In formula (21), is the penalty coefficient, is the initial penalty factor, is the increment rate, which determines the speed of constraint tightening, and t is the number of iterations.

[0245] From the above formula (21), it can be seen that the larger the number of iterations, the larger the penalty coefficient.

[0246] In step S92, the penalty value is used to correct the water environment capacity, the governance resources and the ecological flow respectively to obtain a corrected water environment capacity, a corrected governance resources and a corrected ecological flow.

[0247] Among them, the penalty value can be subtracted from the water environment capacity to obtain the corrected water environment capacity. The calculation formula is as follows:

[0248] (twenty two)

[0249] In formula (22), is the corrected water environment capacity, is the water environment capacity before correction, is the weight of the penalty value, is the penalty value.

[0250] From formula (22), it can be seen that the penalty value is inversely proportional to the water environment capacity. When the penalty value is higher, the revised water environment capacity is smaller. The reason is that when the gap between the target basin's pollutant concentration compliance rate and the preset compliance rate is larger, the more serious the violation of the target basin's pollutant concentration constraint is. Therefore, the penalty value can be controlled to increase. When the penalty value increases, the revised water environment capacity becomes smaller. Then, in the process of selecting the target parameters with the goal of maximizing the revised water environment capacity, the water environment capacity before correction can be maximized while minimizing the penalty value to obtain the maximized revised water environment capacity. While maximizing the water environment capacity, the penalty value can also be reduced, so that the target basin's pollutant concentration compliance rate reaches the preset compliance rate.

[0251] Among them, the penalty value can be added to the governance resource to obtain the revised governance resource. The calculation formula is as follows:

[0252] (twenty three)

[0253] In formula (23), is the revised governance resource, is the governance resource before the amendment, is the weight of the penalty value, is the penalty value.

[0254] From formula (23), it can be seen that the penalty value is proportional to the governance resources. The higher the penalty value, the larger the revised governance resources. The reason is that when the gap between the target basin's pollutant concentration compliance rate and the preset compliance rate is large, more governance resources are needed to govern the target basin in order to achieve the preset compliance rate for the target basin's pollutant concentration. Therefore, the penalty value can be controlled to increase. When the penalty value increases, the revised governance resources become larger. Then, in the process of selecting target parameters with the goal of minimizing governance resources, the governance resources before correction can be minimized while minimizing the penalty value to obtain the minimized revised governance resources. While minimizing governance resources, the penalty value can also be reduced, thereby reducing the pollutant concentration in the target basin and increasing the compliance rate of the pollutant concentration in the target basin.

[0255] Among them, a penalty value can be added to the ecological abnormality to obtain the corrected ecological abnormality. The calculation formula is as follows:

[0256] (twenty four)

[0257] In formula (24), is the corrected ecological abnormality, is the ecological abnormality before correction, is the weight of the penalty value, is the penalty value.

[0258] From formula (24), it can be seen that the penalty value is proportional to the degree of ecological anomaly. When the penalty value is higher, the corrected degree of ecological anomaly increases. The reason is that when the gap between the target basin's pollutant concentration compliance rate and the preset compliance rate is large, the target basin's ecology becomes more abnormal, and thus the corrected degree of ecological anomaly increases. Therefore, the penalty value can be controlled to increase. When the penalty value increases, the corrected degree of ecological anomaly increases. In the process of selecting target parameters with the goal of minimizing the degree of ecological anomaly, the pre-corrected degree of ecological anomaly can be minimized while minimizing the penalty value to obtain the minimized corrected degree of ecological anomaly. When minimizing the corrected degree of ecological anomaly, the penalty value will also be reduced, thereby reducing the pollutant concentration in the target basin and increasing the compliance rate of the pollutant concentration in the target basin.

[0259] In step S93, the multiple target elements are selected from the multiple elements with the goal of maximizing the corrected water environment capacity, the corrected governance resources and the corrected ecological flow.

[0260] Through the above technical solution, the penalty value can be reduced while maximizing the corrected water environment capacity, corrected governance resources and corrected ecological flow, thereby increasing the compliance rate of pollutant concentration in the target basin and reducing the severity of pollution in the target basin.

[0261] Figure 11 This is a water environment capacity determination device proposed in the present disclosure. The water environment capacity determination device 1100 includes a fusion module 1110 , a prediction module 1120 and a water environment capacity module 1130 .

[0262] a fusion module 1110 configured to fuse first satellite data of a target watershed observed by a first satellite, second satellite data of the target watershed observed by a second satellite, and water environment data of the target watershed obtained by simulation of a dynamic watershed model to obtain target water environment data of the target watershed;

[0263] The prediction module 1120 is configured to input the watershed flow, target water environment data, pollution source data, vegetation coverage and meteorological data of the target watershed into the dynamic watershed model to obtain the pollutant concentration and outlet flow of the target watershed;

[0264] The water environment capacity module 1130 is configured to obtain the water environment capacity of the target basin based on the pollutant concentration of the target basin and the outlet flow; the water environment capacity indicates the capacity of the target basin to accommodate pollutants.

[0265] In a possible implementation, the fusion module 1110 is further configured to fuse the first satellite data with the water environment data to obtain first fused data; and fuse the first fused data with the second satellite data to obtain the target water environment data.

[0266] In a possible embodiment, the fusion module 1110 is further configured to obtain a first observation error of the first satellite based on the first satellite data and the water environment data; correct the first observation error using a first Kalman gain to obtain a first target observation error; the first Kalman gain is used to measure the uncertainty of the water environment data; and obtain the first fusion data based on the water environment data and the first target observation error.

[0267] In a possible embodiment, the fusion module 1110 is further configured to obtain a second observation error of the second satellite based on the second satellite data and the second fused data; the second fused data is obtained by perturbing the third fused data using a disturbance term, and the third fused data is data corresponding to the first fused data; the second observation error is corrected using a second Kalman gain to obtain a second target observation error; the second Kalman gain is used to balance the uncertainty of the water environment and the second observation error; and the target water environment data is obtained based on the first fused data and the second target observation error.

[0268] In a possible embodiment, the prediction module 1120 is also configured to screen out multiple target sub-basins from the multiple sub-basins of the target basin; input the basin flow, target water environment data, vegetation coverage, future meteorological data of the target basin and the pollution source data of the multiple target sub-basins into the dynamic basin model to obtain the future pollutant concentration and outlet flow of the target basin.

[0269] In one possible embodiment, the water environment capacity module 1130 is also configured to determine a target time period in which the pollutant concentration in the target watershed is less than a target pollutant concentration; for each time point in the target time period, the pollutant flux at the time point is obtained based on the pollutant concentration and the outlet flow; and the water environment capacity is obtained based on the pollutant flux at multiple time points in the target time period.

[0270] In a possible implementation, the water environment capacity determination device 1100 further includes:

[0271] an element determination module configured to determine a plurality of elements affecting the water environment capacity, treatment resources, and ecological abnormality; the treatment resources are resources consumed to treat pollutants in the target watershed, and the ecological abnormality indicates ecological abnormality of the target watershed;

[0272] an element screening module configured to screen out a plurality of target elements from the plurality of elements with the goal of maximizing the water environment capacity, minimizing the treatment resources, and minimizing the ecological abnormality;

[0273] The parameter screening module is configured to screen out a plurality of target parameters from the parameters of the plurality of target elements.

[0274] In one possible embodiment, the element screening module is also configured to determine a penalty value when the compliance rate of the pollutant concentration does not meet the preset compliance rate; use the penalty value to correct the water environment capacity, the governance resources and the ecological abnormality respectively to obtain the corrected water environment capacity, the corrected governance resources and the corrected ecological abnormality; with the goal of maximizing the corrected water environment capacity, the corrected governance resources and the corrected ecological abnormality, screen out the multiple target elements from the multiple elements.

[0275] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0276] Figure 12 FIG. 1 is a block diagram of an electronic device 1200 according to an exemplary embodiment. Figure 12 As shown, the electronic device 1200 may include: a processor 1201 , a memory 1202 , and may further include one or more of a multimedia component 1203 , an input / output (I / O) interface 1204 , and a communication component 1205 .

[0277] The processor 1201 is used to control the overall operation of the electronic device 1200 to complete all or part of the steps in the above-mentioned method for determining the water environment capacity. The memory 1202 is used to store various types of data to support the operation of the electronic device 1200. This data may include, for example, instructions for any application or method operating on the electronic device 1200, as well as application-related data, such as contact information, sent and received messages, images, audio, video, etc. The memory 1202 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 1203 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 1202 or transmitted via the communication component 1205. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 1204 provides an interface between the processor 1201 and other interface modules. The aforementioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 1205 is used for wired or wireless communication between the electronic device 1200 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more thereof, and therefore the corresponding communication component 1205 may include: a Wi-Fi module, a Bluetooth module, an NFC module.

[0278] In an exemplary embodiment, the electronic device 1200 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the above-mentioned method for determining the water environment capacity.

[0279] In another exemplary embodiment, a computer-readable storage medium including program instructions is further provided. When executed by a processor, the program instructions implement the steps of the above-described method for determining water environment capacity. For example, the computer-readable storage medium may be the aforementioned memory 1202 including the program instructions. The program instructions may be executed by the processor 1201 of the electronic device 1200 to implement the above-described method for determining water environment capacity.

[0280] In another exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program that can be executed by a processor. When the computer program is executed by the processor, the steps of the above-mentioned method for determining the water environment capacity are implemented.

[0281] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings. However, the present disclosure is not limited to the specific details of the above embodiments. Within the technical concept of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all fall within the scope of protection of the present disclosure.

[0282] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any appropriate manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0283] In addition, the various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A method for determining water environment capacity, characterized in that: include: fusing first satellite data of a target watershed observed by a first satellite, second satellite data of the target watershed observed by a second satellite, and water environment data of the target watershed obtained by simulating a dynamic watershed model to obtain target water environment data of the target watershed, including: obtaining a first observation error of the first satellite based on the first satellite data and the water environment data; correcting the first observation error using a first Kalman gain to obtain a first target observation error; the first Kalman gain is used to measure the uncertainty of the water environment data; obtaining first fused data based on the water environment data and the first target observation error; obtaining a second observation error of the second satellite based on the second satellite data and the second fused data; the second fused data is obtained by perturbing third fused data using a disturbance term, the third fused data being data corresponding to the first fused data; correcting the second observation error using a second Kalman gain to obtain a second target observation error; the second Kalman gain is used to balance the uncertainty between the water environment data and the second observation error; and obtaining the target water environment data based on the first fused data and the second target observation error. Inputting the watershed flow, target water environment data, pollution source data, vegetation coverage and meteorological data of the target watershed into the dynamic watershed model to obtain the pollutant concentration and outlet flow of the target watershed; The water environment capacity of the target basin is obtained based on the pollutant concentration of the target basin and the outlet flow; the water environment capacity indicates the capacity of the target basin to accommodate pollutants.

2. The method according to claim 1, characterized in that The method of inputting the watershed flow, target water environment data, pollution source data, vegetation coverage and meteorological data of the target watershed into the dynamic watershed model to obtain the pollutant concentration and outlet flow of the target watershed includes: Selecting a plurality of target sub-basins from the plurality of sub-basins of the target basin; The target basin flow, target water environment data, vegetation coverage, future meteorological data and pollution source data of the multiple target sub-basins are input into the dynamic basin model to obtain the future pollutant concentration and outlet flow of the target basin.

3. The method according to claim 1, characterized in that Obtaining the water environment capacity of the target watershed according to the pollutant concentration of the target watershed and the outlet flow rate includes: determining a target time period during which the pollutant concentration in the target watershed is less than a target pollutant concentration; For each time point in the target time period, obtaining the pollutant flux at the time point according to the pollutant concentration and the outlet flow rate; The water environment capacity is obtained according to the pollutant fluxes at multiple time points in the target time period.

4. The method according to claim 1, wherein The method further comprises: Determining multiple factors that affect the water environment capacity, treatment resources, and ecological abnormality; the treatment resources are the resources consumed to treat pollutants in the target watershed, and the ecological abnormality indicates the ecological abnormality of the target watershed; With the goal of maximizing the water environment capacity, minimizing the management resources, and minimizing the ecological abnormality, a plurality of target elements are screened out from the plurality of elements; A plurality of target parameters are respectively screened out from the parameters of the plurality of target elements.

5. The method according to claim 4, characterized in that The goal of maximizing the water environment capacity, minimizing the management resources, and minimizing the ecological abnormality is to screen out multiple target elements from the multiple elements, including: Determine a penalty value when the compliance rate of the pollutant concentration does not meet a preset compliance rate; Using the penalty value to respectively correct the water environment capacity, the governance resources, and the ecological abnormality, to obtain a corrected water environment capacity, a corrected governance resources, and a corrected ecological abnormality; With the goal of maximizing the corrected water environment capacity, the corrected governance resources and the corrected ecological abnormality, the multiple target elements are screened out from the multiple elements.

6. A device for determining water environment capacity, characterized in that: include: a fusion module configured to fuse first satellite data of a target watershed observed by a first satellite, second satellite data of the target watershed observed by a second satellite, and water environment data of the target watershed obtained by simulating a dynamic watershed model to obtain target water environment data of the target watershed, including: obtaining a first observation error of the first satellite based on the first satellite data and the water environment data; correcting the first observation error using a first Kalman gain to obtain a first target observation error; the first Kalman gain is used to measure the uncertainty of the water environment data; obtaining first fused data based on the water environment data and the first target observation error; obtaining a second observation error of the second satellite based on the second satellite data and the second fused data; the second fused data is obtained by perturbing third fused data using a disturbance term, the third fused data being data corresponding to the first fused data; correcting the second observation error using a second Kalman gain to obtain a second target observation error; the second Kalman gain is used to balance the uncertainty between the water environment data and the second observation error; and obtaining the target water environment data based on the first fused data and the second target observation error; A prediction module is configured to input the watershed flow, target water environment data, pollution source data, vegetation coverage and meteorological data of the target watershed into the dynamic watershed model to obtain the pollutant concentration and outlet flow of the target watershed; The water environment capacity module is configured to obtain the water environment capacity of the target basin based on the pollutant concentration of the target basin and the outlet flow; the water environment capacity indicates the capacity of the target basin to accommodate pollutants.

7. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 5.

Images