Methods, systems, and media for dynamic load scheduling of pollution from multiple tributaries in a watershed.

Abstract

This invention discloses a method, system, and medium for dynamic scheduling of pollution loads in multiple tributaries of a watershed, relating to the field of water environment data management technology. The method includes: deploying an online sensor network in the watershed to collect pollutant data in real time and upload it to a control center to establish a pollution load database; performing time-series source tracing analysis by combining a surface pollutant inventory and the pollution load database to establish source identification; collecting dynamic indicators such as hydrology, water quality, and rainfall to output time-varying pollutant carrying capacity thresholds for different zones; and using the pollution load database, source identification, and carrying capacity thresholds to perform dynamic load scheduling optimization and management. This invention solves the technical problem of inaccurate pollution load scheduling caused by the single monitoring indicator and lack of comprehensive big data assessment and dynamic control capabilities in existing watershed pollution control technologies. It achieves the technical effect of improving the accuracy of pollution discharge load scheduling within the watershed through multi-dimensional water environment big data assessment and dynamic control management.

Description

Technical Field

[0001] This invention relates to the field of water environment data management technology, specifically to a method, system, and medium for dynamic scheduling of pollution loads in multiple tributaries of a watershed. Background Technology

[0002] Traditional governance methods are mostly single-point treatment or passive monitoring, lacking the analysis and management of big data on the overall watershed water environment. Furthermore, existing technologies rely heavily on local monitoring data, lacking real-time tracking and precise control of the overall pollution situation and pollution sources in the watershed, and failing to fully consider the dynamic changes of multi-dimensional data such as hydrological, meteorological, and ecological factors. Summary of the Invention

[0003] This application provides a method, system, and medium for dynamic scheduling of pollution loads in multiple tributaries of a watershed, which addresses the technical problem of inaccurate pollution load scheduling caused by the lack of single monitoring indicators, comprehensive big data assessment, and dynamic control capabilities in existing watershed pollution control technologies.

[0004] The first aspect of this application provides a method for dynamic load scheduling of pollution in multiple tributaries of a watershed. The method includes: deploying an online monitoring sensor network within the watershed to collect pollutant data in real time and uploading the time-series pollutant collection results to a control center to establish a pollution load database; obtaining a surface pollutant inventory, performing time-series source tracing analysis based on the surface pollutant inventory and the pollution load database, and establishing source identification; collecting a dynamic indicator set, including watershed hydrological, water quality, rainfall, water temperature, and biological indicators, constructing a time-series evolution model of ecological capacity based on the dynamic indicator set, considering pollution background, self-purification capacity, and ecological response, and outputting time-varying pollutant carrying capacity thresholds for different zones; jointly constructing a pollution discharge load-ecological capacity adaptation matrix using the pollution load database, the source identification, and the time-varying pollutant carrying capacity thresholds for different zones, using the pollution discharge load-ecological capacity adaptation matrix as a constraint to perform dynamic load scheduling optimization and establish a scheduling strategy; and performing dynamic load scheduling management based on the scheduling strategy.

[0005] A second aspect of this application provides a dynamic load scheduling system for pollution in multiple tributaries of a watershed. The system includes: a pollutant data acquisition module, used to deploy an online monitoring sensor network within the watershed to collect pollutant data in real time and upload the time-series pollutant collection results to a control center to establish a pollution load database; a time-series source tracing analysis module, used to obtain a surface pollutant inventory, perform time-series source tracing analysis based on the surface pollutant inventory and the pollution load database, and establish source identification; and a time-varying carrying capacity threshold generation module, used to collect a dynamic indicator set, which includes watershed hydrological data. The system includes a set of dynamic indicators for water quality, rainfall, water temperature, and biological indicators. Based on these indicators, a time-series evolution model of ecological capacity, considering pollution background, self-purification capacity, and ecological response, is constructed, outputting time-varying carrying capacity thresholds for pollutants in different zones. A load dynamic scheduling optimization module is used to jointly construct a pollution discharge load-ecological capacity adaptation matrix using the pollution load database, source identification, and time-varying carrying capacity thresholds for pollutants in different zones. This matrix serves as a constraint for executing dynamic load scheduling optimization and establishing a scheduling strategy. A load dynamic scheduling management module is used to manage dynamic load scheduling according to the scheduling strategy.

[0006] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of the first aspect.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The method, system, and medium for dynamic scheduling of pollution loads in multiple tributaries of a watershed provided in this application relate to the field of water environment data management technology. By real-time online monitoring of pollutant data, combined with a pollution load database and source tracing analysis, a pollution discharge load and ecological capacity matching matrix is ​​constructed. Based on this matrix, dynamic scheduling optimization is performed using an ecological capacity time-series evolution model, thereby achieving scientific regulation of watershed pollution discharge loads. This solves the technical problem of existing watershed pollution control technologies having single monitoring indicators and lacking comprehensive big data assessment and dynamic regulation capabilities, leading to inaccurate pollution load scheduling. It achieves the technical effect of improving the accuracy of pollution discharge load scheduling within the watershed through multi-dimensional water environment big data assessment and dynamic regulation management. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic flowchart of a method for dynamic load scheduling of pollution in multiple tributaries of a watershed, provided in an embodiment of this application;

[0011] Figure 2 This is a schematic diagram of the structure of a dynamic load scheduling system for pollution in multiple tributaries of a watershed, provided in an embodiment of this application.

[0012] Figure labeling: 11 Pollutant data acquisition module, 12 Time-series source analysis module, 13 Time-varying load capacity threshold generation module, 14 Load dynamic scheduling optimization module, 15 Load dynamic scheduling management module. Detailed Implementation

[0013] This application provides a method, system, and medium for dynamic scheduling of pollution loads in multiple tributaries of a watershed, which addresses the technical problem of inaccurate pollution load scheduling caused by the lack of single monitoring indicators, comprehensive big data assessment, and dynamic control capabilities in existing watershed pollution control technologies.

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0015] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0016] Example 1

[0017] like Figure 1As shown, this application provides a method for dynamic load scheduling of pollution in multiple tributaries of a watershed, the method comprising:

[0018] P10: Deploy an online monitoring sensor network within the watershed to collect pollutant data in real time, and upload the time-series pollutant collection results to the control center to establish a pollution load database.

[0019] Specifically, the primary task of deploying an online monitoring sensor network within the watershed is to ensure the real-time collection of data on various pollutants and to upload this data to the control center for subsequent analysis and scheduling. First, multiple sensors need to be deployed at the inlets of key tributaries and water sections within the watershed. These sensors should employ appropriate detection methods based on the characteristics of the pollutants. For example, conventional pollutants such as ammonia nitrogen, total phosphorus, and chemical oxygen demand (COD) can be monitored using electrochemical sensors and photometric sensors. Simultaneously, appropriate sensors should also be configured to detect emerging pollutants such as antibiotics.

[0020] These sensors should upload real-time data on pollutant concentrations and flow rates to the regional control center via wireless communication modules (such as LoRa and NB-IoT). The uploaded data includes pollutant concentrations at each monitoring point, corresponding water flow rates, and other hydrological and meteorological data (such as water temperature and precipitation). This data will be recorded and organized chronologically to form time-series pollutant data. The real-time pollutant data will provide accurate real-time data support for subsequent pollution source analysis, load forecasting, and scheduling optimization.

[0021] After pollutant data is uploaded to the control center, the system uses this data to calculate the pollution load. The formula for calculating the pollution load is: Pollution Load = Flow Rate × Pollutant Concentration. This calculation allows for real-time acquisition of the pollution load at each monitoring point, such as tributaries and cross-sections. The control center stores this data uniformly in a pollution load database, and establishes a database based on time-series data to ensure historical data traceability, facilitating subsequent analysis and decision-making. By dynamically updating the historical data in the database, the control center can monitor the pollution status of each tributary and cross-section within the watershed at any time. Furthermore, as the amount of data accumulates, the system can generate pollution load trends for each tributary, providing a scientific basis for subsequent scheduling decisions.

[0022] Based on this, the pollution load database provides precise data support for subsequent steps such as pollutant source tracing, pollutant scheduling, and ecological capacity assessment. Through real-time monitoring and data aggregation, the control center can comprehensively grasp the water pollution status within the basin, providing strong data support for subsequent pollution source management, implementation of control measures, and ecological restoration assessment.

[0023] P20: Obtain a ground pollutant inventory, perform time-series source tracing analysis based on the ground pollutant inventory and the pollution load database, and establish source identification.

[0024] Optionally, a surface pollutant inventory can be obtained and analyzed, and combined with a pollution load database to conduct time-series source tracing analysis of pollutants, thereby establishing source identification. First, it is necessary to collect an inventory of potential surface pollutants within the watershed. These pollutants typically originate from multiple sectors such as agriculture, industry, and urban emissions. Therefore, the surface pollutant inventory includes all substances that may pollute water bodies, such as pesticides, fertilizers, industrial wastewater discharges, toxic heavy metals, microplastics, antibiotics, and other emerging pollutants. This inventory should not only include common pollutants (such as COD, ammonia nitrogen, and total phosphorus) but also cover substances such as drug residues, heavy metals, and other chemical pollutants.

[0025] Subsequently, based on real-time monitoring data from the surface pollutant inventory and pollution load database, a time-series source tracing analysis was conducted. The pollution load database stores pollution load data for each tributary and cross-section within the watershed at different time periods, including pollutant concentration and flow data at each monitoring point. This analysis process utilizes data processing and analysis techniques, combined with the physicochemical properties of pollutants and hydrological and hydrodynamic conditions, to conduct an in-depth investigation into the spatiotemporal distribution characteristics of pollutants within the watershed. Specifically, by comparing pollutant concentration and flow data at different time points and different monitoring points, and combining this with pollution source information recorded in the surface pollutant inventory, the transport path of pollutants from their source to the monitoring point can be traced, and their migration and transformation patterns within the watershed can be analyzed. For example, if a monitoring point detects a significant increase in the concentration of a specific pollutant within a specific time period, by combining the hydrological conditions during that period and the emission records of the corresponding pollution source in the surface pollutant inventory, the possible source and transport path of the pollutant can be inferred, thereby achieving precise source tracing of pollutants.

[0026] Finally, based on the results of time-series source tracing analysis, source identification tags are established for each pollution source. These tags serve as markers to identify the source, type, geographical location, and historical trajectory of pollutants. Through these tags, the control center can track pollutant changes in real time and accurately identify pollution hotspots within the watershed. These tags not only provide clear direction for watershed pollution control but also serve as crucial decision-making support in subsequent dynamic pollution load scheduling, helping to formulate more scientific and rational control strategies. For example, after determining that the main pollutant in a high-pollution-load tributary originates from a specific upstream industrial emission source, this tributary can be specially marked in the source identification tags, and subsequent scheduling strategies can prioritize pollution control measures for this tributary, such as adjusting emission times and optimizing treatment processes, to reduce its pollution contribution to the main stream.

[0027] P30: Collect a dynamic indicator set, which includes watershed hydrology, water quality, rainfall, water temperature, and biological indicators. Based on the dynamic indicator set, construct a time-series evolution model of ecological capacity for pollution background, self-purification capacity, and ecological response, and output the time-varying carrying capacity threshold of pollutants in the zone.

[0028] Furthermore, step P30 in this embodiment of the application also includes:

[0029] P31: After standardizing the dynamic indicator set, pollution mutations in the standardized results are removed using discharge records. Trend modeling and distribution modeling are performed for each pollutant based on the pollution mutation removal results. A pollution background fitting layer is established based on the time-series modeling results. P32: A self-purification time-series feature set is extracted from the standardized results. Self-purification capacity is assessed based on the self-purification time-series feature set. Time-series data on the self-purification capacity of each pollutant is established. A self-purification capacity layer is established based on the time-series data on the self-purification capacity. P33: Biological indicator data and pollutant concentration indicator data are extracted from the standardized results. The biological indicator data includes biological response indicators. P34: Critical response identification is performed based on the biological indicator data and pollutant concentration indicator data. An ecological response layer is constructed based on the critical response identification results. P35: An ecological capacity time-series evolution model is constructed based on the pollution background fitting layer, the self-purification capacity layer, and the ecological response layer.

[0030] It should be understood that after completing the source analysis of pollutants within the watershed and establishing source identification, the next focus is on the dynamic assessment of the watershed's ecological capacity. This involves collecting a set of dynamic indicators for the watershed and constructing a time-series evolution model of ecological capacity, ultimately outputting the time-varying carrying capacity thresholds for pollutants in different zones.

[0031] First, a dynamic indicator set was collected for the watershed, covering multiple aspects including hydrology, water quality, rainfall, water temperature, and biological indicators. The collection of these indicators aims to provide comprehensive and accurate data support for subsequent model building. The dynamic indicator set includes watershed hydrological indicators (such as flow rate, velocity, and water level), water quality indicators (including detection data for emerging pollutants in addition to conventional pollutant concentrations), rainfall indicators (recording rainfall amount, intensity, and temporal distribution), water temperature indicators (monitoring changes in water temperature), and a biological indicator set (covering biological response indicators such as biodiversity indices and specific species population sizes). The collection of these indicators provides the foundational data for subsequent ecological capacity assessment.

[0032] Next, the collected dynamic index set is standardized. The purpose of standardization is to eliminate dimensional differences between different indicators, enabling them to be compared and analyzed on the same scale. Common techniques such as Z-score standardization and Min-Max standardization can be used. After standardization, pollution abrupt changes in the results are removed using discharge records. Pollution abrupt changes are usually caused by sudden discharge events, which, while significantly affecting pollutant concentrations in the short term, do not reflect the normal pollution status of the watershed. Removing these abrupt changes allows for a more accurate assessment of the watershed's pollution background. Subsequently, trend modeling and distribution modeling are performed for each pollution factor, and a pollution background fitting layer is established based on the time-series modeling results. This fitting layer provides fundamental pollution background information for subsequent ecological capacity assessment.

[0033] Next, a self-purification time-series feature set was extracted from the standardized data. These features reflect the self-purification capacity of the water body. Self-purification capacity refers to the ability of a water body to eliminate pollutants through natural processes (such as sedimentation and degradation) without external intervention. By analyzing the self-purification time-series feature set, self-purification capacity was assessed, and time-series data on the self-purification capacity of each pollutant were established. Based on this data, a self-purification capacity layer was further constructed. This layer can dynamically reflect changes in the self-purification capacity of the water body, providing support for subsequent pollutant discharge control.

[0034] Next, biometric data and pollutant concentration data are extracted from the standardized processing results. Biometric data includes not only conventional indicators such as biodiversity indices and the population size of specific organisms, but also, and more specifically, bioresponsiveness indicators. Bioresponsiveness indicators refer to the sensitive responses of organisms to changes in pollutant concentrations, such as physiological changes and behavioral alterations. These indicators can more directly reflect the degree of impact of pollutants on the ecosystem. Simultaneously, corresponding pollutant concentration data are extracted to provide a data foundation for subsequent critical response identification. Based on the extracted biometric and pollutant concentration data, critical response identification is performed. A critical response refers to a significant change in biometrics when the pollutant concentration reaches a certain threshold. This response usually indicates that the ecosystem's tolerance to pollutants has reached its limit, and further pollution may lead to irreversible damage to the ecosystem. By identifying critical responses, the impact threshold of pollutants on the ecosystem can be determined, thereby constructing an ecological response layer. The ecological response layer reflects the sensitivity and responsiveness of the watershed ecosystem to changes in pollutant concentrations and is an indispensable part of ecological capacity assessment.

[0035] Finally, the pollution background fitting layer, self-purification capacity layer, and ecological response layer were combined to construct a time-series evolution model of ecological capacity. This model comprehensively considers changes in pollutant concentration, the improvement of self-purification capacity, and changes in ecosystem response, enabling it to accurately predict the pollutant carrying capacity and ecological capacity of various regions within the watershed. Through this model, the control center can monitor and adjust pollution emissions in real time, outputting the time-varying pollutant carrying capacity threshold for each region. The time-varying pollutant carrying capacity threshold for a region refers to the maximum pollutant load that the watershed can withstand under specific time and spatial conditions; exceeding this threshold may lead to irreversible damage to the ecosystem. The time-series evolution model of ecological capacity allows for real-time monitoring of changes in the watershed's ecological capacity, providing precise threshold guidance for the dynamic scheduling of pollution loads.

[0036] P40: Using the pollution load database, the source identification, and the time-varying carrying capacity threshold of pollutants in the zone, a pollution emission load-ecological capacity adaptation matrix is ​​jointly constructed. The pollution emission load-ecological capacity adaptation matrix is ​​used as a constraint to perform dynamic load scheduling optimization and establish a scheduling strategy.

[0037] Furthermore, step P40 in this embodiment of the application also includes:

[0038] P41: Read the watershed branch data and flow velocity data; P42: Perform upstream and downstream spatial dependency modeling based on the watershed branch data and flow velocity data, and generate spatial dependency modeling results; P43: Configure an objective function based on the spatial dependency modeling results. The evaluation features of the objective function include carrying capacity adaptation terms, pollution load regulation cost terms, load timing peak shifting incentive terms, and upstream and downstream coordination constraint terms; P44: Perform load dynamic scheduling optimization under the constraints of the pollution emission load-ecological capacity adaptation matrix according to the objective function, and establish a scheduling strategy.

[0039] Optionally, by jointly using pollution load databases, source identification, and time-varying pollutant carrying capacity thresholds for different zones, a pollution emission load-ecological capacity matching matrix can be constructed. This matrix can then be used as a constraint to perform dynamic load scheduling optimization, ultimately establishing a scientific and reasonable scheduling strategy.

[0040] Before constructing the pollution discharge load-ecological capacity adaptation matrix, it is first necessary to obtain watershed branch data and flow velocity data. Watershed branch data records detailed information such as the distribution and confluence points of each tributary within the watershed, while flow velocity data reflects the speed of water flow within each tributary. This data is crucial for understanding the transport paths and time delays of pollutants within the watershed. By analyzing watershed branch data and flow velocity data, the upstream and downstream relationships between tributaries can be clarified, providing a foundation for subsequent spatial dependency modeling.

[0041] Based on watershed branch data and flow velocity data, spatial dependency modeling of upstream and downstream areas is conducted. This modeling process aims to quantify the interdependencies among tributaries in terms of pollutant transport and ecological impact. Specifically, by analyzing the transport paths, transport times, and transport volumes of pollutants among different tributaries, a spatial dependency model between upstream and downstream areas is established. This model can reflect the potential impact of pollutant emissions from a particular tributary on downstream tributaries and the main stream, as well as the response patterns of pollutant changes in upstream tributaries to downstream areas. The spatial dependency modeling results provide important spatial dimension information for subsequent scheduling strategy formulation.

[0042] Next, based on the results of spatial dependency modeling, an objective function is configured. The evaluation characteristics of the objective function include a carrying capacity adaptation term, a pollution load regulation cost term, a load temporal staggering incentive term, and an upstream-downstream coordination constraint term. Among these, the carrying capacity adaptation term ensures that the dynamic scheduling of pollutant emissions conforms to the ecological capacity of each region within the watershed; the pollution load regulation cost term considers the costs of regulation measures, such as flow regulation and facility activation; the load temporal staggering incentive term guides the time-sharing scheduling of pollution loads, avoiding concentrated pollutant emissions during peak periods; and the upstream-downstream coordination constraint term ensures coordinated pollution emissions between upstream and downstream areas, preventing excessive upstream emissions from affecting downstream water quality.

[0043] Finally, under the constraint of the pollution emission load-ecological capacity matching matrix, dynamic load scheduling optimization is performed according to the above objective function. For example, optimization algorithms such as linear programming and dynamic programming can be used to find the optimal pollution load scheduling scheme while satisfying ecological capacity constraints. During the optimization process, factors such as carrying capacity matching, regulation costs, temporal peak shifting, and upstream-downstream coordination are comprehensively considered to strive for effective pollution load regulation and rational resource utilization while ensuring the ecological security of the watershed. Ultimately, a scheduling strategy is established based on the optimization results. This strategy clarifies the pollution load emission arrangements for each tributary under different temporal and spatial conditions, providing scientific decision support for watershed pollution control.

[0044] This process not only takes into account the spatial dependence between the tributaries within the basin, but also comprehensively weighs various regulatory needs, providing strong technical support for the dynamic management of the basin's pollution load and the sustainable development of the ecosystem.

[0045] Furthermore, step P44 in this embodiment of the application also includes:

[0046] P44-1: Randomly generate a population of scheduling schemes that meet the constraints, and complete the population initialization; P44-2: Calculate the fitness value of each scheduling individual using the objective function, and generate the calculation result; P44-3: Perform population individual selection based on the calculation result, and perform crossover and mutation operations using the population individual selection result to iteratively update the population; P44-4: When the iterative update result meets the termination condition, the load dynamic scheduling optimization is completed.

[0047] In one possible embodiment of this application, a genetic algorithm (or other optimization algorithm) may be introduced to perform dynamic load scheduling optimization in order to obtain the optimal pollution load scheduling strategy.

[0048] When performing dynamic load scheduling optimization, a population of scheduling schemes satisfying the constraints is first randomly generated to complete the population initialization. This population of scheduling schemes is randomly generated based on the constraints of the pollution emission load-ecological capacity adaptation matrix, and each scheduling scheme represents a possible pollution load allocation method. By randomly generating these schemes, multiple possibilities can be covered in the search space, ensuring that potential optimal solutions can be found in subsequent optimization processes.

[0049] Subsequently, the fitness value of each scheduling individual is calculated using the objective function, generating the calculation results. The objective function comprehensively considers multiple evaluation characteristics such as carrying capacity adaptation, regulation cost, temporal peak shifting, and upstream and downstream coordination. The fitness value reflects the degree of optimization of each scheduling scheme to the objective function while meeting ecological capacity constraints. By calculating the fitness value of each individual, the merits of each scheduling scheme can be evaluated.

[0050] Next, population selection is performed based on the calculation results. The selection process can be based on fitness values ​​to determine whether an individual is selected for the next generation. Individuals with higher fitness have a higher probability of being selected, thus preserving the best scheduling scheme. After the selection operation, crossover and mutation operations are performed using the population selection results to iteratively update the population. Crossover generates new individuals by combining some features of two or more individuals; mutation introduces new genetic variations by randomly changing some features of individuals. These operations help explore a wider solution space and avoid the algorithm getting trapped in local optima.

[0051] When the iterative update result meets the termination condition, the load dynamic scheduling optimization is completed. The termination condition may be reaching a preset number of iterations, the fitness value stabilizing, or meeting specific optimization accuracy requirements, etc. Once the termination condition is met, the algorithm stops iterating and outputs the current optimal scheduling scheme as the final load dynamic scheduling strategy.

[0052] The population optimization algorithm described above can effectively find the optimal pollution load scheduling scheme under complex constraints. This process not only considers multiple evaluation characteristics but also gradually improves the quality of the scheduling scheme through iterative optimization, providing scientific and reasonable decision support for the dynamic management of watershed pollution load and the sustainable development of the ecosystem.

[0053] Furthermore, the step P44-3 of this application embodiment, which involves performing population individual selection based on the calculation results, further includes:

[0054] P44-31: Divide the population proportionally according to the calculation results to establish a first divided population and a second divided population; P44-32: Set an enhancing random factor in the first divided population and a weakening random factor in the second divided population; P44-33: Randomly select individuals in the population according to the enhancing random factor and the calculation results of the first divided population to establish a first selection result; P44-34: Randomly select individuals in the population according to the weakening random factor and the calculation results of the second divided population to establish a second selection result; P44-35: Combine the first selection result and the second selection result to complete the selection of individuals in the population.

[0055] Optionally, the process of individual selection in the population can be further refined by proportionally dividing the population and introducing enhancing and weakening random factors to improve the flexibility and diversity of the selection process, thereby improving the global search capability and convergence speed of the optimization algorithm.

[0056] When performing population selection, the population is first divided proportionally based on the calculation results. Specifically, the current population is divided into two parts according to fitness values ​​or other evaluation criteria: a first segment and a second segment. This division aims to differentiate individuals with different fitness levels to better balance the capabilities of global and local searches. For example, individuals with higher fitness values ​​can be assigned to the first segment, while those with lower fitness values ​​can be assigned to the second segment, or the division can be based on other specific evaluation criteria.

[0057] An enhanced random factor is introduced into the first segment population. This factor increases the randomness of the selection process for these individuals, thereby improving the algorithm's global search capability. The enhanced random factor can be implemented by adjusting the selection probability distribution or introducing random perturbations, resulting in greater randomness in the selection process of individuals in the first segment population and preventing the algorithm from prematurely falling into local optima.

[0058] A reduced randomness factor is introduced in the second-segment population. This factor reduces the randomness of the selection process for these individuals, thereby improving the algorithm's local search capability. The reduced randomness factor can be achieved by narrowing the range of the selection probability distribution and reducing random perturbations, making individuals in the second-segment population more inclined to make deterministic selections based on fitness values, thus accelerating the algorithm's convergence.

[0059] Based on the calculation results of the enhanced random factor and the first split population, individuals in the population are randomly selected to establish the first selection result. In this process, the enhanced random factor makes the selection of individuals in the first split population more random, which helps to explore a wider solution space and discover potential high-quality solutions.

[0060] Based on the calculation results of the reduced random factor and the second-segment population, random selection of individuals in the population is performed to establish a second selection result. In this process, the reduced random factor makes individuals in the second-segment population more inclined to make deterministic selections based on fitness values, which helps to further optimize based on existing high-quality solutions and accelerates the convergence of the algorithm.

[0061] Finally, the first and second selection results are merged to complete the selection of individuals in the population. By fusing selection results with two different levels of randomness, a balance can be achieved between global and local search, avoiding the algorithm from getting trapped in local optima too early and enabling rapid convergence after finding a high-quality solution. This fusion method can be a simple merging or a weighted merging based on specific weights, depending on the algorithm design and optimization objectives.

[0062] This process provides a more effective population selection mechanism for dynamic load scheduling optimization, which helps to find the optimal pollution load scheduling scheme under complex constraints.

[0063] Furthermore, the process of performing crossover and mutation operations using the selection results of individuals in the population, as described in step P44-3 of this application embodiment, further includes:

[0064] P44-36: Configure gene correction strategies after crossover and mutation operations; P44-37: Perform individual correction discrimination based on the gene correction strategy, and update individuals using the individual correction discrimination results.

[0065] Specifically, the processing steps after performing crossover and mutation operations based on the selection results of individuals in the population can be further refined to ensure that the generated individuals have better adaptability and characteristics that meet the actual constraints.

[0066] First, configure gene correction strategies. Crossover and mutation operations typically generate new individuals, whose genes (i.e., scheduling schemes) may contain parts that do not conform to problem constraints or actual needs, leading to decreased fitness. Configuring gene correction strategies aims to correct the crossover and mutation-induced individuals, ensuring that the generated individuals still achieve improved fitness while meeting constraints. These correction strategies may include adjusting parts that do not conform to scheduling rules, optimizing unreasonable time series, or ensuring that adjustments still meet ecological capacity constraints.

[0067] Next, based on the configured gene correction strategy, the individuals generated after crossover and mutation operations are corrected and judged. Specifically, each newly generated individual is first checked for constraints to verify whether it meets the constraints of the pollution emission load-ecological capacity fit matrix, such as whether it exceeds the ecological capacity threshold or whether the pollution load allocation is reasonable. If the individual meets the constraints, it is retained; if it does not meet the constraints, it is adjusted according to the preset correction rules. The correction rules may include adjusting certain gene values ​​of the individual to make it meet the constraints again, or optimizing the individual through a specific repair algorithm.

[0068] After constraint checks and corrections are completed, the individuals in the population are updated based on the correction results. Corrected individuals replace the original individuals that did not meet the constraints, thus ensuring that every individual in the population meets the requirements of the optimization algorithm. Simultaneously, during the correction process, specific information about the correction is recorded, including the individual states before and after the correction, and the application of the correction rules. This information can be used for subsequent analysis and optimization, helping to further refine the gene correction strategy.

[0069] By configuring gene correction strategies and performing correction judgments and updates on individuals, the stability and effectiveness of the optimization algorithm can be effectively ensured. This process not only avoids individuals that do not meet the constraints due to crossover and mutation operations from entering subsequent iterations, but also improves the convergence speed and solution quality of the algorithm, providing a more reliable mechanism for dynamic load scheduling optimization and helping to find the optimal pollution load scheduling scheme under complex constraints.

[0070] P50: Perform dynamic load scheduling management according to the aforementioned scheduling strategy.

[0071] Furthermore, step P50 in this embodiment of the application also includes:

[0072] P51: Invoke the deployed online monitoring sensor network to perform watershed pollution monitoring and establish a feedback time series dataset; P52: Verify the response consistency of the scheduling strategy based on the feedback time series dataset and generate verification results; P53: Update the scheduling strategy based on the verification results.

[0073] It should be understood that load dynamic scheduling management is carried out according to the previously determined scheduling strategy. The core objective of this process is to ensure that, through the optimized scheduling strategy, pollution loads can be managed efficiently in actual operation, and that the watershed water quality is always kept within a safe range.

[0074] First, the deployed online monitoring sensor network is invoked to perform real-time monitoring of pollutants in the watershed. Through the sensor network, data such as pollutant concentrations and flow rates at various monitoring points within the watershed are acquired and uploaded to the control center in real time. This data forms a feedback time-series dataset, representing the actual trend of pollutant changes and the real-time dynamics of the pollution load within the watershed. Using this feedback data, the system can instantly monitor the effectiveness of the scheduling strategy, ensuring that pollutant emissions remain within predetermined limits.

[0075] Next, based on the acquired feedback time-series dataset, the executed scheduling strategy is validated for response consistency. The purpose of response consistency validation is to check whether the scheduling strategy meets the expected objectives in actual operation and can effectively respond to changes in pollutant concentrations. For example, the system will compare the actual pollution load in the feedback dataset with the target value set in the scheduling strategy, analyzing the execution effect of the scheduling strategy in different times and regions. If there is a deviation between the actual response and the expected target, further analysis of the reasons for the deviation is needed, such as whether it is due to errors in monitoring data, unreasonable model assumptions, or changes in the external environment.

[0076] Finally, the scheduling strategy is updated based on the verification results. If the verification results show good consistency in the response of the scheduling strategy, it indicates that the current strategy can effectively achieve dynamic scheduling of pollution load and can continue to be implemented. However, if the verification results show significant deviations, the scheduling strategy needs to be adjusted based on the information in the feedback time-series dataset. Adjustments may include reconfiguring pollution load allocation, optimizing the scheduling time window, and adjusting the tributary discharge sequence to ensure that the scheduling strategy can better adapt to the actual conditions of the watershed and achieve effective control of pollution load and rational utilization of ecological capacity.

[0077] This process not only enables real-time monitoring of the pollution status of the basin, but also allows for dynamic adjustment of scheduling strategies based on actual conditions, ensuring that the basin always meets the goals and requirements of pollution control. This achieves precise pollutant control, improves management effectiveness, and ultimately safeguards the stability of water quality and ecological security in the basin.

[0078] Furthermore, step P51 in the embodiments of this application also includes:

[0079] P51-1: Perform ecological species distribution analysis in the watershed and configure key monitoring points; P51-2: Set up ecological monitoring points at the key monitoring points and establish ecological change indicators; P51-3: Add the ecological change indicators to the feedback time series dataset.

[0080] Specifically, in addition to conducting real-time monitoring of pollutants in the watershed, it also includes further analysis of the distribution of ecological species and the configuration of key monitoring points, aiming to ensure that the pollutant dispatch strategy can not only effectively manage water quality, but also monitor the health status of the watershed ecosystem in real time, so as to achieve comprehensive environmental protection.

[0081] First, before conducting watershed pollution monitoring, an analysis of the watershed's ecological species distribution is performed. This analysis aims to comprehensively understand the distribution of different ecological species within the watershed, including fish, benthic animals, and plankton. Through this analysis, key areas and species sensitive to pollutant changes can be identified, and the health of these areas and species reflects the overall health of the watershed ecosystem. Based on the results of the ecological species distribution analysis, key monitoring points are established. These points should cover important locations such as ecologically sensitive areas and the confluence of major tributaries within the watershed to ensure that the monitoring data comprehensively reflects the watershed ecosystem's response.

[0082] Next, ecological monitoring stations will be established in the key ecological areas or habitats identified in the above analysis. These monitoring stations will be used to collect ecological change indicators of ecological species in real time. These indicators may include biodiversity indices, population changes of specific species, biological health indicators (such as fish growth rate and reproduction rate), and other indicators reflecting the health of the ecosystem. Through these monitoring stations, the system can detect changes in ecological species in a timely manner, especially their response to changes in pollutant concentrations. For example, when pollutant concentrations exceed safe thresholds, sensitive ecological species may exhibit signs such as reduced numbers and decreased population diversity. Monitoring these changes can provide timely feedback for adjusting pollution control measures.

[0083] Finally, the collected ecological change indicators will be added to the feedback time-series dataset. These indicators are long-term feedback data, reflecting the gradual changes in the ecosystem during pollution load scheduling. This dataset, together with the time-series dataset of pollutants, will form a comprehensive environmental change database. Through this database, the system can not only monitor changes in pollutants but also track the response and recovery of the ecosystem. Over time, the system can observe the gradual recovery or degradation of the ecosystem, thereby continuously optimizing the scheduling strategy and ensuring that pollutant emissions within the watershed do not cause irreversible damage to the ecosystem.

[0084] This process not only enables real-time monitoring of pollution levels in the watershed, but also allows for assessment of the long-term health of the ecosystem, ensuring that pollutant control not only focuses on water quality itself, but also takes into account the health of the ecosystem, ultimately achieving the dual goals of water quality protection and ecological restoration.

[0085] In summary, the embodiments of this application have at least the following technical effects:

[0086] This application utilizes an online sensor network to collect pollutant data in real time and establish a pollution load database, enabling dynamic monitoring and real-time assessment of pollutants within the watershed. By combining the surface pollutant inventory and the pollution load database, time-series source tracing analysis is performed to accurately identify pollution sources and their spatiotemporal distribution, providing a scientific basis for pollution control. By constructing a matching matrix between pollution discharge load and ecological capacity, pollution load control strategies are optimized to ensure effective matching between pollution discharge and ecological capacity, guaranteeing water quality remains within ecologically safe limits. Multi-dimensional data is used to provide real-time feedback and optimization of control strategies, improving pollution treatment efficiency. Furthermore, by combining various dynamic indicators such as hydrology, water quality, and ecology, the application promotes the intelligent and scientific development of watershed pollution control.

[0087] The technology has achieved the goal of improving the accuracy of pollution discharge load scheduling within the basin through multi-dimensional water environment big data assessment and dynamic regulation and management.

[0088] Example 2

[0089] Based on the same inventive concept as the dynamic load scheduling method for pollution in multiple tributaries of a watershed in the foregoing embodiments, such as Figure 2 As shown, this application provides a dynamic load scheduling system for pollution in multiple tributaries of a watershed. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0090] The pollutant data acquisition module 11 is used to deploy an online monitoring sensor network within the watershed, collect pollutant data in real time, and upload the time-series pollutant collection results to the control center to establish a pollution load database.

[0091] The time-series source tracing analysis module 12 is used to obtain a ground pollutant inventory, perform time-series source tracing analysis based on the ground pollutant inventory and the pollution load database, and establish source identification.

[0092] The time-varying carrying capacity threshold generation module 13 is used to collect a dynamic indicator set, which includes watershed hydrology, water quality, rainfall, water temperature, and biological indicators. Based on the dynamic indicator set, an ecological capacity time-series evolution model of pollution background, self-purification capacity, and ecological response is constructed, and the time-varying carrying capacity threshold of pollutants in the zone is output.

[0093] The load dynamic scheduling optimization module 14 is used to jointly construct a pollution emission load-ecological capacity adaptation matrix by utilizing the pollution load database, the source identification, and the time-varying carrying capacity threshold of pollutants in the zone, and to perform load dynamic scheduling optimization by using the pollution emission load-ecological capacity adaptation matrix as a constraint to establish a scheduling strategy.

[0094] The load dynamic scheduling management module 15 is used to perform load dynamic scheduling management according to the scheduling strategy.

[0095] Furthermore, the time-varying load threshold generation module 13 is also used to perform the following steps:

[0096] After standardizing the dynamic indicator set, pollution mutations in the standardized results are removed using discharge records. Trend modeling and distribution modeling are then performed for each pollutant based on the pollution mutation removal results. A pollution background fitting layer is established based on the time-series modeling results. A self-purification time-series feature set is extracted from the standardized results. Self-purification capacity is assessed based on this feature set, and time-series data on the self-purification capacity of each pollutant is established. A self-purification capacity layer is then established based on this time-series data. Biological indicator data and pollutant concentration indicator data, including biological response indicators, are extracted from the standardized results. Critical response identification is performed based on the biological indicator data and pollutant concentration indicator data. An ecological response layer is constructed based on the critical response identification results. Finally, an ecological capacity time-series evolution model is constructed based on the pollution background fitting layer, the self-purification capacity layer, and the ecological response layer.

[0097] Furthermore, the load dynamic scheduling optimization module 14 is also used to perform the following steps:

[0098] Read the watershed branch data and flow velocity data; perform upstream and downstream spatial dependency modeling based on the watershed branch data and flow velocity data to generate spatial dependency modeling results; configure an objective function based on the spatial dependency modeling results, the evaluation features of the objective function including carrying capacity adaptation term, pollution load regulation cost term, load time-series peak-shifting incentive term, and upstream and downstream coordination constraint term; perform dynamic load scheduling optimization under the constraints of the pollution emission load-ecological capacity adaptation matrix according to the objective function, and establish a scheduling strategy.

[0099] Furthermore, the load dynamic scheduling optimization module 14 is also used to perform the following steps:

[0100] Randomly generate a population of scheduling schemes that meet the constraints to complete population initialization; calculate the fitness value of each scheduling individual using the objective function to generate calculation results; perform population individual selection based on the calculation results, and perform crossover and mutation operations using the population individual selection results to iteratively update the population; when the iterative update results meet the termination condition, the load dynamic scheduling optimization is completed.

[0101] Furthermore, the load dynamic scheduling optimization module 14 is also used to perform the following steps:

[0102] Based on the calculation results, the population is proportionally divided to establish a first segmented population and a second segmented population. An enhancing random factor is set in the first segmented population, and a weakening random factor is set in the second segmented population. Individuals in the population are randomly selected based on the enhancing random factor and the calculation results of the first segmented population to establish a first selection result. Individuals in the population are randomly selected based on the weakening random factor and the calculation results of the second segmented population to establish a second selection result. The first selection result and the second selection result are then merged to complete the selection of individuals in the population.

[0103] Furthermore, the load dynamic scheduling optimization module 14 is also used to perform the following steps:

[0104] After crossover and mutation operations, a gene correction strategy is configured; individual correction is performed based on the gene correction strategy, and the individual is updated using the individual correction results.

[0105] Furthermore, the load dynamic scheduling management module 15 is also used to perform the following steps:

[0106] The deployed online monitoring sensor network is invoked to perform watershed pollution monitoring and establish a feedback time-series dataset; the response consistency of the scheduling strategy is verified based on the feedback time-series dataset, and a verification result is generated; the scheduling strategy is updated based on the verification result.

[0107] Furthermore, the load dynamic scheduling management module 15 is also used to perform the following steps:

[0108] Perform an analysis of the ecological species distribution in the watershed and configure key monitoring points; set up ecological monitoring points at the key monitoring points and establish ecological change indicators; add the ecological change indicators to the feedback time series dataset.

[0109] Example 3

[0110] Based on the same inventive concept as the dynamic load scheduling method for pollution in multiple tributaries of a watershed in the foregoing embodiments, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as described in Embodiment 1.

[0111] Through the foregoing detailed description of the dynamic load scheduling method for pollution in multiple tributaries of a watershed, those skilled in the art can clearly understand the dynamic load scheduling method, system, and medium for pollution in multiple tributaries of a watershed in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here. As for the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and relevant parts can be referred to in the method section.

[0112] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0113] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0114] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0115] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for dynamic load scheduling of pollution from multiple tributaries in a watershed, characterized in that, The method includes: Deploy an online monitoring sensor network within the watershed to collect pollutant data in real time, and upload the time-series pollutant collection results to the control center to establish a pollution load database; Obtain a ground pollutant inventory, perform time-series source tracing analysis based on the ground pollutant inventory and the pollution load database, and establish source tracing identifiers; A dynamic indicator set is collected, which includes watershed hydrology, water quality, rainfall, water temperature, and biological indicators. Based on the dynamic indicator set, an ecological capacity time-series evolution model of pollution background, self-purification capacity, and ecological response is constructed, and the time-varying carrying capacity threshold of pollutants in each zone is output. A pollution emission load-ecological capacity adaptation matrix is ​​jointly constructed using the pollution load database, the source identification, and the time-varying carrying capacity threshold of pollutants in the region. The pollution emission load-ecological capacity adaptation matrix is ​​used as a constraint to perform dynamic load scheduling optimization and establish a scheduling strategy. Dynamic load scheduling management is performed according to the aforementioned scheduling strategy; The step of using the pollution emission load-ecological capacity adaptation matrix as a constraint to perform dynamic load scheduling optimization and establish a scheduling strategy includes: Read the watershed branch data and flow velocity data; Based on the watershed branch data and flow velocity data, spatial dependency modeling of upstream and downstream areas is performed to generate spatial dependency modeling results. The objective function is configured based on the spatial dependency modeling results. The evaluation features of the objective function include load adaptation term, pollution load regulation cost term, load time-series peak-shifting incentive term, and upstream and downstream coordination constraint term. Based on the objective function and constrained by the pollution emission load-ecological capacity adaptation matrix, dynamic load scheduling optimization is performed to establish a scheduling strategy; The dynamic load scheduling management according to the scheduling strategy includes: The deployed online monitoring sensor network is invoked to perform watershed pollution monitoring and establish a feedback time-series dataset; The consistency of the scheduling strategy's response is verified based on the feedback time series dataset, and verification results are generated. Update the scheduling strategy based on the verification results; The process of calling upon the deployed online monitoring sensor network to perform watershed pollution monitoring and establish a feedback time-series dataset includes: Perform ecological species distribution analysis in the watershed and configure key monitoring points; Ecological monitoring points were set up at the key monitoring points, and ecological change indicators were established. The ecological change indicators are added to the feedback time series dataset.

2. The method for dynamic load scheduling of pollution in multiple tributaries of a watershed as described in claim 1, characterized in that, The step of constructing a time-series evolution model of ecological capacity based on the dynamic indicator set, considering pollution background, self-purification capacity, and ecological response, and outputting time-varying carrying capacity thresholds for pollutants in different zones, includes: After standardizing the dynamic index set, pollution mutations in the standardization results are removed using the discharge records. Based on the pollution mutation removal results, trend modeling and distribution modeling are performed on each pollution factor. A pollution background fitting layer is established based on the time series modeling results. Extract the self-cleaning time series feature set from the standardized processing results, evaluate the self-cleaning capacity based on the self-cleaning time series feature set, establish the self-cleaning capacity time series data for each pollutant, and establish a self-cleaning capacity layer based on the self-cleaning capacity time series data. Extract biological indicator data and pollutant concentration indicator data from the standardized processing results, wherein the biological indicator data includes biological response indicators; Critical responses are identified based on the bioindicator data and pollutant concentration data, and an ecological response layer is constructed based on the critical response identification results. An ecological capacity time-series evolution model is constructed based on the pollution background fitting layer, self-purification capacity layer, and ecological response layer.

3. The method for dynamic load scheduling of pollution in multiple tributaries of a watershed as described in claim 1, characterized in that, The step of performing dynamic load scheduling optimization based on the objective function under the constraints of the pollution emission load-ecological capacity adaptation matrix includes: Randomly generate a population of scheduling schemes that satisfy the constraints, and complete the population initialization; The fitness value of each scheduled individual is calculated using the objective function, and the calculation results are generated. Based on the calculation results, population individual selection is performed, and crossover and mutation operations are performed using the population individual selection results to iteratively update the population; When the iterative update result meets the termination condition, the load dynamic scheduling optimization is completed.

4. The method for dynamic load scheduling of pollution in multiple tributaries of a watershed as described in claim 3, characterized in that, The step of performing population individual selection based on the calculation results includes: Based on the calculation results, the population is divided proportionally to establish a first divided population and a second divided population; An enhanced random factor is set in the first split population, and a weakened random factor is set in the second split population. Based on the calculation results of the enhanced random factor and the first segmented population, individuals in the population are randomly selected to establish the first selection result; Based on the calculation results of the reduced random factor and the second segmentation of the population, random selection of individuals in the population is performed to establish the second selection result; The selection of individuals in the population is completed by combining the first selection result and the second selection result.

5. The method for dynamic load scheduling of pollution in multiple tributaries of a watershed as described in claim 3, characterized in that, The process of performing crossover and mutation operations using the selection results of individuals in the population includes: Configure gene correction strategies after crossover and mutation operations; Individual correction is performed based on the gene correction strategy, and the individual is updated using the results of the individual correction.

6. A dynamic load scheduling system for pollution in multiple tributaries of a river basin, characterized in that, The system is used to execute the dynamic load scheduling method for multi-tributary pollution in a watershed according to any one of claims 1 to 5, the system comprising: The pollutant data acquisition module is used to deploy an online monitoring sensor network within the watershed, collect pollutant data in real time, and upload the time-series pollutant collection results to the control center to establish a pollution load database. The time-series source tracing analysis module is used to obtain a ground pollutant inventory, perform time-series source tracing analysis based on the ground pollutant inventory and the pollution load database, and establish source identification. A time-varying carrying capacity threshold generation module is used to collect a dynamic indicator set, which includes watershed hydrology, water quality, rainfall, water temperature, and biological indicators. Based on the dynamic indicator set, an ecological capacity time-series evolution model of pollution background, self-purification capacity, and ecological response is constructed, and the time-varying carrying capacity threshold of pollutants in the zone is output. The load dynamic scheduling optimization module is used to jointly construct a pollution emission load-ecological capacity adaptation matrix by utilizing the pollution load database, the source identification, and the time-varying carrying capacity threshold of pollutants in the zone, and to perform load dynamic scheduling optimization by using the pollution emission load-ecological capacity adaptation matrix as a constraint to establish a scheduling strategy. A load dynamic scheduling management module is used to perform load dynamic scheduling management according to the scheduling strategy.

7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method steps of any one of claims 1 to 5.

Images