Catchment-based stormwater runoff non-point source pollutant load forecasting method and apparatus
By using satellite remote sensing data and deep learning models, the non-point source pollution load of urban stormwater runoff can be accurately predicted, solving the problem of unpredictable non-point source pollution, realizing an effective strategy for reducing non-point source pollution, and ensuring stable river water quality.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- POWERCHINA WATER ENVIRONMENT GOVERANCE
- Filing Date
- 2024-12-04
- Publication Date
- 2026-05-21
AI Technical Summary
Existing technologies struggle to accurately predict urban non-point source pollution due to the numerous influencing factors.
By acquiring satellite remote sensing image data and rainfall prediction information of the target watershed section, functional areas are divided, and a deep learning model is used to predict the rainwater runoff non-source pollution load curve of each functional area. The curves are then fused to determine the predicted results of the rainwater runoff non-source pollution load of the target watershed section.
It enables accurate prediction of non-point source pollution load, helps formulate non-point source pollution reduction strategies, weakens the adverse effects of the initial rainfall on the water quality of the assessment section, and ensures the long-term cleanliness of river water.
Smart Images
Figure CN2024136585_21052026_PF_FP_ABST
Abstract
Description
Method and apparatus for predicting non-point source pollution loads from rainwater runoff based on watershed.
[0001] This patent application claims priority to Chinese Patent Application No. CN 202411618904.0, filed on November 13, 2024. The disclosure of the earlier application is incorporated herein by reference in its entirety. Technical Field
[0002] This application belongs to the field of pollution control technology, and in particular relates to a method and device for predicting non-point source pollution loads of rainwater runoff based on watershed. Background Technology
[0003] Urban non-point source pollution refers to the pollution of the urban environment caused by various pollutants on the urban surface entering urban rainwater receiving water bodies in a large-scale form through runoff or surface flow under the washing and leaching action of rainwater. In the process of rapid urbanization, urban rainwater runoff carries a large amount of pollutants into water bodies, which has become an important cause of urban water quality deterioration.
[0004] Rainwater runoff pollution control is a key focus of current and future water management and quality improvement efforts. To further mitigate the adverse impact of initial rainfall on water quality standards at monitoring sections and ensure long-term cleanliness of river channels, it is necessary to predict urban non-point source pollution emissions and implement corresponding treatment and control measures. However, numerous factors influence urban non-point source pollution, and its causes are complex, making accurate prediction difficult. Technical issues
[0005] In view of this, embodiments of this application provide a method and apparatus for predicting non-point source pollution loads of rainwater runoff based on watershed, so as to accurately predict non-point source pollution loads of rainwater runoff. Technical solutions
[0006] The first aspect of this application provides a watershed-based method for predicting non-point source pollution loads from stormwater runoff, including:
[0007] Acquire satellite remote sensing image data of each catchment area of the target watershed section, as well as rainfall forecast information for each catchment area during future rainfall periods;
[0008] Each catchment area is divided into multiple functional zones;
[0009] Based on the satellite remote sensing image data, determine the underlying surface distribution information for each functional area;
[0010] Based on the functional type, rainfall forecast information and underlying surface distribution information of each functional area, predict the non-point source pollution load curve of stormwater runoff for each functional area during the future rainfall period;
[0011] Based on the non-point source pollution load curve of rainwater runoff in each functional area during the future rainfall period, the predicted results of the non-point source pollution load of rainwater runoff at the target watershed section during the future rainfall period are determined.
[0012] In one possible implementation of the first aspect, the rainfall forecast information includes rainfall level, and the underlying surface distribution information includes the area percentage of each type of underlying surface;
[0013] The method of predicting the stormwater runoff non-point source pollution load curve for each functional area during future rainfall periods, based on the functional type, rainfall forecast information, and underlying surface distribution information of each functional area, includes:
[0014] The functional type, rainfall level, and area ratio of each underlying surface of each functional area are input into a pre-trained non-point source pollution prediction model to obtain the non-point source pollution load curve of rainwater runoff for each functional area during future rainfall periods; wherein, the non-point source pollution prediction model is a deep learning model.
[0015] In one possible implementation of the first aspect, the method further includes: pre-training the non-point source pollution prediction model;
[0016] The pre-trained non-point source pollution prediction model includes:
[0017] Obtain time series of pollutant concentrations and runoff discharge volumes monitored in multiple functional areas under different rainfall levels;
[0018] Based on the pollutant concentration time series and the runoff drainage series, the non-point source pollution load curves of stormwater runoff for each functional area under different rainfall levels are determined.
[0019] The initial model is obtained by using the functional type, rainfall level, and area ratio of each underlying surface of each functional area as inputs and the corresponding stormwater runoff non-point source pollution load curve as outputs. The connection weights of the initial model are then trained to obtain the non-point source pollution prediction model.
[0020] In one possible implementation of the first aspect, determining the predicted stormwater runoff non-source pollution load of the target watershed section during the future rainfall period based on the stormwater runoff non-source pollution load curve of each functional area during the future rainfall period includes:
[0021] The non-point source pollution load curves of rainwater runoff in each functional area of each catchment zone are merged to obtain the non-point source pollution load curve of rainwater runoff in each catchment zone.
[0022] The rainwater runoff non-source pollution load curves of each catchment area are fused to obtain the rainwater runoff non-source pollution load curve of the target watershed section.
[0023] Based on the stormwater runoff non-source pollution load curve of the target watershed section, the predicted results of stormwater runoff non-source pollution load of the target watershed section during future rainfall periods are determined.
[0024] In one possible implementation of the first aspect, prior to the fusion of the stormwater runoff non-source pollution load curves for each functional area of each catchment zone, the method further includes:
[0025] Acquire air pollution data and / or pipeline pollutant deposition data for each functional area;
[0026] Based on the air pollution data and / or pipeline pollutant deposition data of each functional area, determine the pollution load adjustment curve for each functional area during the future rainfall period;
[0027] Based on the pollution load adjustment curve of each functional area, the non-point source pollution load curve of rainwater runoff in each functional area is adjusted.
[0028] In one possible implementation of the first aspect, after determining the predicted stormwater runoff non-source pollution load of the target watershed section during the future rainfall period, the method further includes:
[0029] Based on the stormwater runoff non-source pollution load curve of the target watershed section, determine the maximum pollution load of the target watershed section;
[0030] Obtain the pollution absorption rate and water environmental capacity of the target watershed section;
[0031] The recovery time of the target watershed section is determined based on the maximum pollution load, the water environment capacity, and the pollution absorption rate.
[0032] If the recovery time is greater than a preset time threshold, then it is determined that a preset non-point source pollution reduction strategy needs to be implemented during the future rainfall period.
[0033] In one possible implementation of the first aspect, after determining that a preset non-point source pollution reduction strategy needs to be implemented during the future rainfall period, the method further includes:
[0034] Based on the rainwater runoff non-source pollution load curve of the target watershed section, determine the target time period with a slope greater than a preset threshold during the future rainfall period;
[0035] The target time period is defined as the period during which the non-point source pollution reduction strategy is implemented.
[0036] A second aspect of this application provides a watershed-based stormwater runoff non-point source pollution load prediction device, comprising:
[0037] The acquisition module is used to acquire satellite remote sensing image data of each catchment area of the target watershed section, as well as rainfall forecast information for each catchment area during future rainfall periods.
[0038] The partitioning module is used to divide each catchment area into multiple functional zones;
[0039] The processing module is used to determine the underlying surface distribution information of each functional area based on the satellite remote sensing image data;
[0040] The prediction module is used to predict the stormwater runoff non-point source pollution load curve of each functional area during the future rainfall period, based on the functional type of each functional area, rainfall prediction information and underlying surface distribution information.
[0041] The calculation module is used to determine the predicted results of the rainwater runoff non-source pollution load of the target watershed section during the future rainfall period based on the rainwater runoff non-source pollution load curve of each functional area during the future rainfall period.
[0042] In one possible implementation of the second aspect, the rainfall forecast information includes rainfall levels, and the underlying surface distribution information includes the area percentage of each type of underlying surface.
[0043] The prediction module is used for:
[0044] The functional type, rainfall level, and area ratio of each underlying surface of each functional area are input into a pre-trained non-point source pollution prediction model to obtain the non-point source pollution load curve of rainwater runoff for each functional area during future rainfall periods; wherein, the non-point source pollution prediction model is a deep learning model.
[0045] In one possible implementation of the second aspect, the prediction module is further configured to: pre-train the non-point source pollution prediction model;
[0046] The pre-trained non-point source pollution prediction model includes:
[0047] Obtain time series of pollutant concentrations and runoff discharge volumes monitored in multiple functional areas under different rainfall levels;
[0048] Based on the pollutant concentration time series and the runoff drainage series, the non-point source pollution load curves of stormwater runoff for each functional area under different rainfall levels are determined.
[0049] The initial model is obtained by using the functional type, rainfall level, and area ratio of each underlying surface of each functional area as inputs and the corresponding stormwater runoff non-point source pollution load curve as outputs. The connection weights of the initial model are then trained to obtain the non-point source pollution prediction model.
[0050] In one possible implementation of the second aspect, the computing module is used for:
[0051] The non-point source pollution load curves of rainwater runoff in each functional area of each catchment zone are merged to obtain the non-point source pollution load curve of rainwater runoff in each catchment zone.
[0052] The rainwater runoff non-source pollution load curves of each catchment area are fused to obtain the rainwater runoff non-source pollution load curve of the target watershed section.
[0053] Based on the stormwater runoff non-source pollution load curve of the target watershed section, the predicted results of stormwater runoff non-source pollution load of the target watershed section during future rainfall periods are determined.
[0054] In one possible implementation of the second aspect, before fusing the stormwater runoff non-source pollution load curves for each functional area of each catchment zone, the calculation module is further configured to:
[0055] Acquire air pollution data and / or pipeline pollutant deposition data for each functional area;
[0056] Based on the air pollution data and / or pipeline pollutant deposition data of each functional area, determine the pollution load adjustment curve for each functional area during the future rainfall period;
[0057] Based on the pollution load adjustment curve of each functional area, the non-point source pollution load curve of rainwater runoff in each functional area is adjusted.
[0058] In one possible implementation of the second aspect, after determining the predicted stormwater runoff non-source pollution load of the target watershed section during the future rainfall period, the calculation module is further configured to:
[0059] Based on the stormwater runoff non-source pollution load curve of the target watershed section, determine the maximum pollution load of the target watershed section;
[0060] Obtain the pollution absorption rate and water environmental capacity of the target watershed section;
[0061] The recovery time of the target watershed section is determined based on the maximum pollution load, the water environment capacity, and the pollution absorption rate.
[0062] If the recovery time is greater than a preset time threshold, then it is determined that a preset non-point source pollution reduction strategy needs to be implemented during the future rainfall period.
[0063] In one possible implementation of the second aspect, after determining that a preset non-point source pollution reduction strategy needs to be implemented during the future rainfall period, the calculation module is further configured to:
[0064] Based on the rainwater runoff non-source pollution load curve of the target watershed section, determine the target time period with a slope greater than a preset threshold during the future rainfall period;
[0065] The target time period is defined as the period during which the non-point source pollution reduction strategy is implemented.
[0066] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect or any implementation thereof.
[0067] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any implementation thereof. Beneficial effects
[0068] The advantages of this application compared to the prior art are:
[0069] This application considers that urban non-point source pollution is mainly affected by the coupling effect of regional functional type, rainfall forecast information, and underlying surface distribution information. Therefore, by dividing the target watershed section into functional areas based on catchment area division, and determining the underlying surface distribution information of the functional areas based on satellite remote sensing image data, the application then predicts the rainwater runoff non-point source pollution load curve of each functional area during the future rainfall period based on the functional type, rainfall forecast information, and underlying surface distribution information of each functional area. This curve can reflect the change of non-point source pollution load over time. Finally, based on the rainwater runoff non-point source pollution load curve of each functional area, the prediction result of rainwater runoff non-point source pollution load of the target watershed section is obtained. This can help relevant personnel predict non-point source pollution load, formulate corresponding non-point source pollution reduction strategies in advance, and help mitigate the adverse effects of initial rainfall on the water quality of the assessment section, ensuring the long-term cleanliness of river water. Attached Figure Description
[0070] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0071] Figure 1 is a schematic diagram of the application scenario of the watershed-based rainwater runoff non-point source pollution load prediction method provided in the embodiments of this application;
[0072] Figure 2 is a schematic diagram of the implementation process of the watershed-based stormwater runoff non-point source pollution load prediction method provided in the embodiments of this application;
[0073] Figure 3 is a schematic diagram of the structure of the watershed-based rainwater runoff non-source pollution load prediction device provided in the embodiment of this application;
[0074] Figure 4 is a schematic diagram of the structure of the electronic device provided in an embodiment of this application. Embodiments of the present invention
[0075] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0076] The technical solution of this application will be described below through specific embodiments.
[0077] Figure 1 is a schematic diagram of the application scenario of the watershed-based rainwater runoff non-point source pollution load prediction method provided in the embodiments of this application.
[0078] Referring to Figure 1, urban non-point source pollution generally forms and discharging into the natural environment along with the formation and transfer of urban runoff. The urban surface is composed of various underlying surfaces. Urban runoff first forms on different underlying surfaces and forms urban non-point source pollution through scouring. Within a certain catchment area, the runoff from different underlying surfaces eventually converges and enters the river basin.
[0079] Urban non-point source pollution is affected by many factors such as rainfall, underlying surface, and land use. Rainfall is highly random, and the spatial heterogeneity of urban underlying surface and land use is high, resulting in a complex formation mechanism of urban non-point source pollution, which is highly random, uncertain, and difficult to predict and control.
[0080] To this end, this application proposes a watershed-based method for predicting non-point source pollution loads from stormwater runoff. As shown in Figure 2, in one embodiment, the prediction method provided by this application includes steps S201 to S205, which are described below.
[0081] Step S201: Obtain satellite remote sensing image data of each catchment area of the target watershed section, as well as rainfall forecast information for each catchment area during future rainfall periods.
[0082] Urban catchment area delineation can be based on regional topographic data and drainage network information. For example, firstly, the catchment boundaries of the main streams within the watershed are delineated based on the watershed boundaries; then, based on the main streams, the boundaries of the main tributaries are further refined; finally, based on the catchment units of the main tributaries, the delineation is further refined in conjunction with the characteristics of the drainage network in each area, thereby determining the final catchment areas.
[0083] Referring to Figure 1, for a watershed, the target watershed section is set by controlling the cross-section. Based on the flow patterns of the pre-divided catchment areas, the catchment areas flowing into the target watershed section are determined, which serve as the catchment areas corresponding to the target watershed section.
[0084] Satellite remote sensing images should be able to reflect the current state of the underlying surface in each water catchment area as much as possible. First, the spatial resolution of the remote sensing data should be as high as possible, and second, the remote sensing data should be as cloudless as possible.
[0085] Rainfall forecast information includes duration and rainfall level. For example, rainfall levels can be categorized as light rain, moderate rain, heavy rain, torrential rain, extremely heavy rain, and exceptionally heavy rain, generally measured by daily rainfall. Specifically, light rain refers to daily rainfall below 10 mm; moderate rain to 10–24.9 mm; heavy rain to 25–49.9 mm; torrential rain to 50–99.9 mm; extremely heavy rain to 100–250 mm; and exceptionally heavy rain to over 250 mm.
[0086] Step S202: Divide each catchment area into multiple functional zones.
[0087] With the rapid development of urbanization, resources and factors are constantly converging on cities. Urbanization and human socio-economic activities have significantly changed the composition, form, and structural pattern of urban landscapes, forming different types of functional areas (such as commercial areas, residential areas, and industrial areas).
[0088] A large catchment area is often composed of several different functional zones, and the pollution load characteristics of these zones vary significantly. In step S202, a land planning map of the city can be obtained, and the functional zone composition of each catchment area can be determined based on the land planning map.
[0089] Step S203: Determine the underlying surface distribution information for each functional area based on satellite remote sensing image data.
[0090] The underlying surface is the interface between the atmosphere and the solid ground or liquid water surface below it. By preprocessing satellite remote sensing images, including geometric correction, atmospheric correction, and image fusion, and combining professional remote sensing interpretation software and manual interpretation, a refined classification of land use can be achieved. According to the composition of the underlying surface, land cover types can be divided into water bodies (reservoirs, canals, rivers, lakes), roofs, roads, and green spaces (woodlands, grasslands, etc.).
[0091] Step S204: Based on the functional type, rainfall forecast information and underlying surface distribution information of each functional area, predict the non-point source pollution load curve of rainwater runoff for each functional area during the future rainfall period.
[0092] For example, in step S204, a deep learning non-point source pollution prediction model can be pre-trained. By inputting the functional type, rainfall level and area ratio of each underlying surface of each functional area into the non-point source pollution prediction model, the non-point source pollution load curve of rainwater runoff during future rainfall periods can be obtained.
[0093] The pollution load varies significantly across different underlying surfaces; therefore, information on the distribution of underlying surfaces can be the area percentage of different underlying surfaces.
[0094] Rainfall forecast information can be in the form of rainfall levels. For example, during light rain, although the initial runoff is small and the scouring capacity is limited, the pollutant concentration is very high during the light rain event. During moderate rain, the pollutant concentration decreases due to the dilution effect of water. The larger the rain type, the greater the runoff and the greater the scouring capacity of surface pollutants. The scouring effect on the road surface is greater than the dilution effect of water, so the concentration level is the highest during the entire heavy rain process.
[0095] Meanwhile, 20% to 30% of the runoff typically carries more than 50% to 80% of the pollution load, resulting in a significant initial scouring effect.
[0096] By using non-point source pollution prediction models, the coupling effect of functional type, rainfall prediction information and underlying surface distribution information on pollution load can be considered, thereby enabling the prediction of changes in pollution load.
[0097] Here, the non-point source pollution prediction model can be pre-trained using the following steps:
[0098] Obtain time series of pollutant concentrations and runoff discharge volumes monitored in multiple functional areas under different rainfall levels;
[0099] Based on the pollutant concentration time series and runoff drainage volume series, the pollutant concentration at the same moment multiplied by the runoff drainage volume is the pollution load generated. This allows us to determine the stormwater runoff non-point source pollution load curve for each functional area under different rainfall levels.
[0100] The initial model is obtained by using the functional type, rainfall level, and area ratio of each underlying surface of each functional area as inputs and the corresponding stormwater runoff non-point source pollution load curve as outputs. The connection weights of the initial model are trained to obtain the non-point source pollution prediction model.
[0101] Step S205: Based on the rainwater runoff non-source pollution load curve of each functional area during the future rainfall period, determine the predicted results of rainwater runoff non-source pollution load of the target watershed section during the future rainfall period.
[0102] Here, the predicted results of stormwater runoff non-point source pollution load at the target watershed section during the future rainfall period can be determined through the following steps:
[0103] The non-point source pollution load curves of rainwater runoff in each functional area of each catchment zone are merged to obtain the non-point source pollution load curve of rainwater runoff in each catchment zone.
[0104] The non-point source pollution load curves of rainwater runoff in each catchment area are fused to obtain the non-point source pollution load curve of rainwater runoff in the target watershed section.
[0105] Based on the stormwater runoff non-point source pollution load curves of the target watershed section, the predicted stormwater runoff non-point source pollution load of the target watershed section during future rainfall periods is determined. Curve fusion refers to adding the load values of multiple curves at the same time.
[0106] As one possible embodiment, before merging the stormwater runoff non-source pollution load curves of each functional area in each catchment zone, the following may also be included:
[0107] Acquire air pollution data and / or pipeline pollutant deposition data for each functional area;
[0108] Based on the air pollution data and / or pipeline pollutant deposition data of each functional area, determine the pollution load adjustment curve for each functional area during the future rainfall period;
[0109] Based on the pollution load adjustment curve of each functional area, the non-point source pollution load curve of rainwater runoff in each functional area is adjusted.
[0110] In this embodiment, air pollution and pipeline pollutant deposition are considered. Air pollution causes rainwater to carry pollutants, and pollutants deposited in pipelines can also flow into rivers under the washout of rainwater. Both have a significant impact on the non-point source pollution load of rainwater runoff. Therefore, air pollution data for future periods can be obtained through weather forecast data. Pipeline pollutant deposition data for future periods are obtained by sampling sediments from some typical sampling pipe sections. Based on these two types of data, an intelligent algorithm model is used to determine the pollution load adjustment curve for each functional area during the future rainfall period. Finally, the rainwater runoff non-point source pollution load curve is adjusted according to the pollution load adjustment curve to improve accuracy.
[0111] The final stormwater runoff non-point source pollution load curve of the target watershed section can not only analyze the stormwater runoff non-point source pollution load of the target watershed section after rainfall, but also analyze the changes in stormwater runoff non-point source pollution load during rainfall, helping managers to formulate appropriate load reduction strategies.
[0112] Compared with existing technologies, the beneficial effects of the prediction method provided in this application are as follows: Considering that urban non-point source pollution is mainly affected by the coupling effect of regional functional type, rainfall forecast information and underlying surface distribution information, the method divides the target watershed section into functional areas by water catchment zones, determines the underlying surface distribution information of the functional areas based on satellite remote sensing image data, and then predicts the rainwater runoff non-point source pollution load curve of each functional area during the future rainfall period based on the functional type, rainfall forecast information and underlying surface distribution information of each functional area. This curve can reflect the change of non-point source pollution load over time. Finally, based on the rainwater runoff non-point source pollution load curve of each functional area, the prediction result of rainwater runoff non-point source pollution load of the target watershed section is obtained. This can help relevant personnel predict non-point source pollution load, formulate corresponding non-point source pollution reduction strategies in advance, and help weaken the adverse effects of initial rainfall on the water quality of the assessment section, ensuring the long-term cleanliness of river water.
[0113] The above embodiments illustrate how to predict the non-point source pollution load of stormwater runoff at a target watershed section during future rainfall periods. In one embodiment below, based on the predicted non-point source pollution load of stormwater runoff, i.e., the stormwater runoff non-point source pollution load curve of the target watershed section, a non-point source pollution reduction strategy for the target watershed section is analyzed, which will be detailed below.
[0114] As one possible embodiment, after determining the predicted non-point source pollution load of stormwater runoff at the target watershed section during future rainfall periods, the method may further include:
[0115] Based on the rainwater runoff non-source pollution load curve of the target watershed section, determine the maximum pollution load of the target watershed section;
[0116] Obtain the pollution absorption rate and water environmental capacity of the target watershed section;
[0117] The recovery time of the target watershed section is determined based on the maximum pollution load, water environment capacity, and pollution absorption rate.
[0118] If the recovery time exceeds the preset time threshold, then it is determined that a preset non-point source pollution reduction strategy needs to be implemented during future rainfall periods.
[0119] Water environmental capacity refers to the maximum load of pollutants that a water body can accommodate while meeting water quality requirements; therefore, it is also called water body load or pollution carrying capacity. Here, the maximum value of the rainwater runoff non-point source pollution load curve is the maximum pollution load of the target watershed section. Subtracting the water environmental capacity from this maximum pollution load and then dividing by the pollution absorption rate gives the recovery time of the target watershed section.
[0120] For example, if pollutants at a target watershed section can recover to levels within the water environment capacity within 48 hours, then no non-point source pollution reduction strategy is needed, and the river can absorb them on its own. Otherwise, a non-point source pollution reduction strategy is required.
[0121] As one possible implementation, the target period with a slope greater than a preset threshold can be determined based on the stormwater runoff non-point source pollution load curve of the target watershed section. This target period is then designated as the period for implementing non-point source pollution reduction strategies. These strategies may include intercepting and transporting rainwater runoff to wastewater treatment plants, temporarily storing it in stormwater regulation facilities for later input into the wastewater system on sunny days, or subjecting it to other targeted treatments. By implementing non-point source pollution reduction strategies during the target period when the pollution load rises rapidly, the pollution load can be effectively reduced, and wastewater system overload can be avoided.
[0122] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and the sequence number of each step should not constitute any limitation on the implementation process of the embodiments of this application.
[0123] Figure 3 is a schematic diagram of the structure of the watershed-based stormwater runoff non-source pollution load prediction device provided in an embodiment of this application, including:
[0124] The acquisition module 31 is used to acquire satellite remote sensing image data of each catchment area of the target watershed section, as well as rainfall forecast information of each catchment area during future rainfall periods.
[0125] The partitioning module 32 is used to divide each catchment area into multiple functional areas;
[0126] Processing module 33 is used to determine the underlying surface distribution information of each functional area based on satellite remote sensing image data;
[0127] Prediction module 34 is used to predict the stormwater runoff non-point source pollution load curve of each functional area during the future rainfall period based on the functional type of each functional area, rainfall prediction information and underlying surface distribution information.
[0128] The calculation module 35 is used to determine the predicted results of the rainwater runoff non-source pollution load of the target watershed section during the future rainfall period based on the rainwater runoff non-source pollution load curve of each functional area during the future rainfall period.
[0129] As one possible embodiment, the rainfall forecast information includes rainfall level, and the underlying surface distribution information includes the area percentage of each type of underlying surface;
[0130] In this embodiment, the prediction module 34 is used for:
[0131] The functional type, rainfall level, and area ratio of each underlying surface of each functional area are input into a pre-trained non-point source pollution prediction model to obtain the non-point source pollution load curve of rainwater runoff for each functional area during future rainfall periods; the non-point source pollution prediction model is a deep learning model.
[0132] As one possible embodiment, the prediction module 34 is further configured to: pre-train a non-point source pollution prediction model; the pre-trained non-point source pollution prediction model includes:
[0133] Obtain time series of pollutant concentrations and runoff discharge volumes monitored in multiple functional areas under different rainfall levels;
[0134] Based on the time series of pollutant concentrations and the runoff drainage series, the non-point source pollution load curves of stormwater runoff for each functional area under different rainfall levels were determined.
[0135] The initial model is obtained by using the functional type, rainfall level, and area ratio of each underlying surface of each functional area as inputs and the corresponding stormwater runoff non-point source pollution load curve as outputs. The connection weights of the initial model are trained to obtain the non-point source pollution prediction model.
[0136] As one possible embodiment, the computing module 35 is used for:
[0137] The non-point source pollution load curves of rainwater runoff in each functional area of each catchment zone are merged to obtain the non-point source pollution load curve of rainwater runoff in each catchment zone.
[0138] The non-point source pollution load curves of rainwater runoff in each catchment area are fused to obtain the non-point source pollution load curve of rainwater runoff in the target watershed section.
[0139] Based on the non-point source pollution load curve of rainwater runoff at the target watershed section, the predicted results of the non-point source pollution load of rainwater runoff at the target watershed section during future rainfall periods are determined.
[0140] As one possible implementation, before fusing the stormwater runoff non-point source pollution load curves of each functional area in each catchment zone, the calculation module 35 is also used to:
[0141] Acquire air pollution data and / or pipeline pollutant deposition data for each functional area;
[0142] Based on the air pollution data and / or pipeline pollutant deposition data of each functional area, determine the pollution load adjustment curve for each functional area during the future rainfall period;
[0143] Based on the pollution load adjustment curve of each functional area, the non-point source pollution load curve of rainwater runoff in each functional area is adjusted.
[0144] As one possible embodiment, after determining the predicted stormwater runoff non-point source pollution load of the target watershed section during future rainfall periods, the calculation module 35 is further used for:
[0145] Based on the rainwater runoff non-source pollution load curve of the target watershed section, determine the maximum pollution load of the target watershed section;
[0146] Obtain the pollution absorption rate and water environmental capacity of the target watershed section;
[0147] The recovery time of the target watershed section is determined based on the maximum pollution load, water environment capacity, and pollution absorption rate.
[0148] If the recovery time exceeds the preset time threshold, then it is determined that a preset non-point source pollution reduction strategy needs to be implemented during future rainfall periods.
[0149] As one possible embodiment, after determining that a preset non-point source pollution reduction strategy needs to be implemented during future rainfall periods, the calculation module 35 is further configured to:
[0150] Based on the stormwater runoff non-point source pollution load curve of the target watershed section, determine the target time period during which the slope is greater than the preset threshold during future rainfall periods;
[0151] The target period is defined as the period during which the non-point source pollution reduction strategy is implemented.
[0152] The prediction device provided in this application can implement the prediction method provided in this application, and therefore can achieve the beneficial effects of the prediction method described above, which will not be repeated here.
[0153] This application also provides an electronic device. As shown in FIG4, as a possible embodiment, the electronic device 40 includes: a processor 41, a memory 42, and a computer program 43 stored in the memory 42 and executable on the processor 41, such as a watershed-based stormwater runoff non-point source pollution load prediction program. When the processor 41 executes the computer program 43, it implements the steps in the above-described embodiments of the watershed-based stormwater runoff non-point source pollution load prediction methods, such as steps S201 to S205 shown in FIG2. Alternatively, when the processor 41 executes the computer program 43, it implements the functions of each module in the above-described device embodiments, such as the functions of modules 31 to 35 shown in FIG3.
[0154] For example, the computer program 43 may be divided into one or more modules / units, which are stored in the memory 42 and executed by the processor 41 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program 43 in the electronic device 40.
[0155] The electronic device 40 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The electronic device 40 may include, but is not limited to, a processor 41 and a memory 42. Those skilled in the art will understand that Figure 4 is merely an example of an electronic device and does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.
[0156] The processor 41 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any other conventional processor.
[0157] The memory 42 can be an internal storage unit of the electronic device 40, such as a hard disk or memory of the electronic device 40. The memory 42 can also be an external storage device of the electronic device 40, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 40. Furthermore, the memory 42 can include both internal and external storage units of the electronic device 40. The memory 42 is used to store the computer program and other programs and data required by the electronic device 40. The memory 42 can also be used to temporarily store data that has been output or will be output.
[0158] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above-described prediction device and electronic device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0159] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.
[0160] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can implement the described functions using different methods for each specific application, but such implementation should not be considered beyond the scope of protection of this application.
[0161] In the embodiments provided in this application, it should be understood that the disclosed devices / electronic devices can be implemented in other ways. For example, it should be understood that the embodiments of the devices / electronic devices described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation; for example, multiple units or components can be combined or integrated into another system; and for example, some features can be ignored or not executed. Furthermore, the mutual couplings shown or discussed (e.g., direct coupling or communication connections) can be implemented through some interfaces, which can be electrical, mechanical, or other forms.
[0162] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this application's solution, depending on actual needs.
[0163] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0164] If an integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above method embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can be any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.
[0165] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A method for predicting non-point source pollution load of stormwater runoff based on a catchment area, characterized in that, include: Acquire satellite remote sensing image data of each catchment area of the target watershed section, as well as rainfall forecast information for each catchment area during future rainfall periods; Each catchment area is divided into multiple functional zones; Based on the satellite remote sensing image data, determine the underlying surface distribution information for each functional area; Based on the functional type, rainfall forecast information and underlying surface distribution information of each functional area, predict the non-point source pollution load curve of stormwater runoff for each functional area during the future rainfall period; Based on the non-point source pollution load curve of rainwater runoff in each functional area during the future rainfall period, the predicted results of the non-point source pollution load of rainwater runoff at the target watershed section during the future rainfall period are determined.
2. The watershed-based prediction method of stormwater runoff nonpoint source pollution load according to claim 1, wherein, The rainfall forecast information includes rainfall levels, and the underlying surface distribution information includes the area percentage of each type of underlying surface. The method of predicting the stormwater runoff non-point source pollution load curve for each functional area during future rainfall periods, based on the functional type, rainfall forecast information, and underlying surface distribution information of each functional area, includes: The functional type, rainfall level, and area ratio of each underlying surface of each functional area are input into a pre-trained non-point source pollution prediction model to obtain the non-point source pollution load curve of rainwater runoff for each functional area during future rainfall periods; wherein, the non-point source pollution prediction model is a deep learning model.
3. The watershed-based prediction method of rainwater runoff non-point source pollution load according to claim 2, wherein, The method further includes: pre-training the non-point source pollution prediction model; The pre-trained non-point source pollution prediction model includes: Obtain time series of pollutant concentrations and runoff discharge volumes monitored in multiple functional areas under different rainfall levels; Based on the pollutant concentration time series and the runoff drainage series, the non-point source pollution load curves of stormwater runoff for each functional area under different rainfall levels are determined. The initial model is obtained by using the functional type, rainfall level, and area ratio of each underlying surface of each functional area as inputs and the corresponding stormwater runoff non-point source pollution load curve as outputs. The connection weights of the initial model are then trained to obtain the non-point source pollution prediction model.
4. The catchment-based prediction method of stormwater runoff non-point source pollution load according to any one of claims 1 to 3, characterized in that, The step of determining the predicted stormwater runoff non-source pollution load of the target watershed section during the future rainfall period based on the stormwater runoff non-source pollution load curve of each functional area during the future rainfall period includes: The non-point source pollution load curves of rainwater runoff in each functional area of each catchment zone are merged to obtain the non-point source pollution load curve of rainwater runoff in each catchment zone. The rainwater runoff non-source pollution load curves of each catchment area are fused to obtain the rainwater runoff non-source pollution load curve of the target watershed section. Based on the stormwater runoff non-source pollution load curve of the target watershed section, the predicted results of stormwater runoff non-source pollution load of the target watershed section during future rainfall periods are determined.
5. The watershed-based prediction method of stormwater runoff nonpoint source pollution load according to claim 4, wherein, Before fusing the stormwater runoff non-source pollution load curves of each functional area in each catchment zone, the following steps are also included: Acquire air pollution data and / or pipeline pollutant deposition data for each functional area; Based on the air pollution data and / or pipeline pollutant deposition data of each functional area, determine the pollution load adjustment curve for each functional area during the future rainfall period; Based on the pollution load adjustment curve of each functional area, the non-point source pollution load curve of rainwater runoff in each functional area is adjusted.
6. The watershed-based rainwater runoff nonpoint source pollution load prediction method of claim 4, wherein, After determining the predicted non-point source pollution load of stormwater runoff at the target watershed section during the future rainfall period, the method further includes: Based on the stormwater runoff non-source pollution load curve of the target watershed section, determine the maximum pollution load of the target watershed section; Obtain the pollution absorption rate and water environmental capacity of the target watershed section; The recovery time of the target watershed section is determined based on the maximum pollution load, the water environment capacity, and the pollution absorption rate. If the recovery time is greater than a preset time threshold, then it is determined that a preset non-point source pollution reduction strategy needs to be implemented during the future rainfall period.
7. The watershed-based rainwater runoff nonpoint source pollution load prediction method of claim 6, wherein, After determining that a preset non-point source pollution reduction strategy needs to be implemented during the future rainfall period, the method further includes: Based on the rainwater runoff non-source pollution load curve of the target watershed section, determine the target time period with a slope greater than a preset threshold during the future rainfall period; The target time period is defined as the period during which the non-point source pollution reduction strategy is implemented.
8. A watershed-based rainwater runoff non-point source pollution load prediction device, characterized in that, include: The acquisition module is used to acquire satellite remote sensing image data of each catchment area of the target watershed section, as well as rainfall forecast information for each catchment area during future rainfall periods. The partitioning module is used to divide each catchment area into multiple functional zones; The processing module is used to determine the underlying surface distribution information of each functional area based on the satellite remote sensing image data; The prediction module is used to predict the stormwater runoff non-point source pollution load curve of each functional area during the future rainfall period, based on the functional type of each functional area, rainfall prediction information and underlying surface distribution information. The calculation module is used to determine the predicted results of the rainwater runoff non-source pollution load of the target watershed section during the future rainfall period based on the rainwater runoff non-source pollution load curve of each functional area during the future rainfall period.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.