Intelligent city air state monitoring method and system based on multi-source data fusion

Abstract

The application relates to the technical field of environmental monitoring, in particular to a smart city air state monitoring method and system based on multi-source data fusion. The method comprises the following steps: acquiring multi-source pollutant emission data and weather forecast information; based on the multi-source pollutant emission data, the process of the coupling of ammonia and sulfur dioxide to generate secondary particulate matters under different meteorological scenarios is analyzed in combination with the weather forecast information, and a multi-pollutant coupling information set is obtained; based on the multi-pollutant coupling information set, the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the concentration of secondary particulate matters in a pollution area are simulated and analyzed, and source contribution information is obtained; based on the source contribution information, a dynamic differentiated collaborative control instruction set is generated, the emission intensity limit time sequence of pollution sources of different contribution levels and risk levels in the corresponding period is dynamically planned in a differentiated manner, and a smart city air state monitoring log is output. The method promotes the modernization of ecological environment governance and provides high-quality air environment for a smart city.

Description

Technical Field

[0001] This application relates to the field of environmental monitoring technology, and in particular to a smart city air condition monitoring method and system based on multi-source data fusion. Background Technology

[0002] Currently, the monitoring and control of urban atmospheric compound pollution (such as secondary particulate matter) mainly relies on independent emission monitoring of various pollution sources and static management based on fixed thresholds or simple diffusion models. Common methods include: separately monitoring the concentration and setting emission limits for primary pollutants such as ammonia and sulfur dioxide; predicting pollution trends based on historical data and empirical formulas; and taking indiscriminate overall emission restrictions or shutdowns for all relevant industrial sources in the region when air pollution events occur. These methods treat pollutants as isolated factors, ignoring the transport, convergence, and complex nonlinear chemical reaction processes of different pollutants (such as ammonia and sulfur dioxide) under dynamic meteorological conditions. Their control models are crude and have a lagging response.

[0003] Existing methods cannot dynamically simulate the synergistic effects and transformation processes of different pollutants under complex meteorological conditions, thus failing to accurately trace the main causes of pollution. This leads to "one-size-fits-all" control measures, which are both inefficient and cause unnecessary economic and social costs. Summary of the Invention

[0004] This application provides a smart city air condition monitoring method and system based on multi-source data fusion to solve the above problems.

[0005] Firstly, this application provides a smart city air quality monitoring method based on multi-source data fusion, the method comprising:

[0006] The system acquires multi-source pollutant emission data and weather forecast information. Based on the multi-source pollutant emission data and the weather forecast information, it analyzes the process of ammonia and sulfur dioxide coupling to generate secondary particulate matter under different meteorological scenarios, obtaining a multi-pollutant coupling information set. Based on the multi-pollutant coupling information set, it simulates and analyzes the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the concentration of secondary particulate matter in the polluted area, obtaining source contribution information. Based on the source contribution information, it generates a dynamic differentiated collaborative control instruction set, dynamically plans the emission intensity limit time sequence of pollution sources with different contribution levels and risk levels in the corresponding time period, and outputs the smart city air status monitoring log.

[0007] Through the above technical solutions, air monitoring is upgraded from simple concentration measurement to in-depth analysis of the formation mechanism of secondary pollution, which significantly improves the scientific understanding and early warning capabilities of complex pollution processes. Based on detailed source apportionment, key pollution sources under different meteorological conditions can be accurately identified, providing a basis for targeted governance. The generated dynamic, differentiated, and collaborative control instructions can flexibly adjust the control intensity according to changes in contribution and risk, greatly improving management efficiency and precision. The panoramic monitoring log provides a complete chain of evidence for decision-making, enhancing the traceability and continuous optimization capabilities of the governance process.

[0008] Optionally, the multi-source pollutant emission data includes ammonia emission source location information, ammonia emission intensity information, sulfur dioxide emission source location information, and sulfur dioxide emission intensity information; the weather forecast information includes wind speed and direction information, and atmospheric temperature and humidity information for the forecast period; based on the ammonia emission source location information and the sulfur dioxide emission source location information, combined with the wind speed and direction information, the transmission paths and spatial diffusion characteristics of pollutants emitted from different emission sources are analyzed during the forecast period through a combination of trajectory simulation and diffusion analysis, to obtain pollutant dynamic distribution information; based on the pollutant dynamic distribution information, combined with the atmospheric temperature and humidity information, the promotion or inhibition information of meteorological conditions at different spatial locations on the chemical reaction process of ammonia and sulfur dioxide during the forecast period is analyzed, to obtain meteorological impact information; based on the meteorological impact information, combined with the ammonia emission intensity information and the sulfur dioxide emission intensity information, the spatial distribution and intensity time series of secondary particulate matter generated by the coupling of ammonia and sulfur dioxide under the differentiated reaction environment characterized by the meteorological impact information are analyzed, to obtain a multi-pollutant coupling information set.

[0009] Optionally, based on the dynamic distribution information of the pollutants and combined with the atmospheric temperature and humidity information, the degree of compatibility of the suitable conditions required for the reaction of ammonia and sulfur dioxide is analyzed to obtain regional reaction suitability information; based on the regional reaction suitability information, through time-series correlation analysis, the influence of the pollutant transport time sequence characteristics and temperature and humidity changes at different times on the reaction rate abrupt change nodes of the coupling of ammonia and sulfur dioxide to generate secondary particulate matter is analyzed to obtain time-series reaction abrupt change information; based on the time-series reaction abrupt change information, through dynamic effect assessment, the influence of temperature and humidity fluctuation amplitude on the initiated reaction of the coupling of ammonia and sulfur dioxide to generate secondary particulate matter is analyzed to obtain meteorological impact information.

[0010] Optionally, based on the dynamic distribution information of the pollutants, spatiotemporal matching analysis is performed on the effective reaction pairings. This analysis examines the time-varying effective reaction pairing relationships between different ammonia emission sources and different sulfur dioxide emission sources during the predicted period, formed by the combined effects of spatial relative positions and meteorological conditions, to obtain dynamic source synergy information. The effective reaction pairings are those formed between ammonia emission sources and sulfur dioxide emission sources. Based on the dynamic source synergy information, the matching degree and effectiveness between the emission intensity ratio of each effective reaction pairing and the reaction conditions characterized by the time-series reaction mutation information are quantified by calculating the dynamic synergy contribution intensity, thus obtaining dynamic synergy contribution information. Based on the dynamic synergy contribution information, combined with the wind speed and direction information, a spatiotemporal mapping table is constructed, using the effective reaction pairing identifier and the corresponding wind speed and direction information as indexes, and the concentration distribution of secondary particulate matter on the transmission grid as values. This table characterizes the trajectory interweaving and spatial superposition effects of secondary particulate matter generated by different effective reaction pairings in subsequent transmission, resulting in the multi-pollutant coupling information set.

[0011] Optionally, based on the aforementioned spatiotemporal mapping table, a panoramic pollution source map covering all ammonia and sulfur dioxide emission sources is constructed to clarify the spatial attributes, emission intensity time series, and effective response pair correlations of each emission source, resulting in a panoramic pollution source map. Based on the panoramic pollution source map, the concentration accumulation process of secondary particulate matter at each receptor point in the pollution area is simulated under the normal operating conditions of all emission sources during the predicted period, obtaining baseline concentration field information. Based on the baseline concentration field information, the single variation scenarios of each ammonia emission source and each sulfur dioxide emission source under shutdown or doubling emission intensity are simulated sequentially, and the change characteristics of secondary particulate matter concentration at each receptor point compared to the baseline concentration field are analyzed to obtain single-source contribution intensity information. Based on the single-source contribution intensity information, through contribution ranking and regional mapping analysis, the contribution ratio of each emission source to the overall pollution area and key receptor point concentration changes is analyzed to obtain sub-source contribution information.

[0012] Optionally, based on the dynamic source synergy information, dynamic synergy impact analysis is used to analyze the temporal sequence of the influence intensity of each synergistic contribution of the effective response pair combination on the secondary particulate matter concentration at the receptor point within the corresponding time period, thereby obtaining single-source time-effect contribution information; based on the single-source time-effect contribution information, spatiotemporal overlay analysis is used to integrate the influence of all emission sources on each receptor point within the predicted time period, thereby obtaining the temporal sequence change information of receptor point concentration; based on the temporal sequence change information of receptor point concentration, the polluted area is spatially gridded, and the temporal sequence change of concentration at each receptor point is mapped to the corresponding spatial grid to form the spatial distribution temporal sequence of secondary particulate matter concentration within the polluted area, thereby obtaining the baseline concentration field information.

[0013] Optionally, based on the panoramic pollution source map, a set of single variation scenarios is constructed for each ammonia emission source and each sulfur dioxide emission source; the set of single variation scenarios includes a first scenario where the corresponding emission source is completely shut down and a second scenario where the emission intensity of the corresponding emission source is doubled, resulting in a set of source-specific variation scenarios; based on the set of source-specific variation scenarios and combined with the baseline concentration field information, simulation of each variation scenario is executed sequentially, and air quality numerical technology is used to analyze the spatiotemporal changes of secondary particulate matter concentration at each receptor point compared with the baseline concentration during the prediction period when the variation scenario is executed, to obtain the single-source contribution intensity information.

[0014] Optionally, based on the single-source contribution intensity information, the impact ratio of each emission source on the total cumulative concentration of the pollution area during the predicted period is analyzed through dynamic contribution ranking to obtain initial contribution ratio information; based on the temporal change information of the receptor point concentration, receptor points with excessive concentrations are identified as key receptor points through regional correlation mapping, and the distribution characteristics of the contribution intensity of each emission source during the key period leading to the exceedance are analyzed to obtain key point source correlation information; based on the initial contribution ratio information and combined with the key point source correlation information, the overall impact of each emission source and its local impact on key points are weighted and fused to quantify the final contribution ratio, thus obtaining the sub-source contribution information.

[0015] Optionally, based on the single-source contribution intensity information and combined with the weather forecast information, dynamic limit planning is used to analyze the time-varying emission intensity upper limit sequence that different emission sources should follow to ensure air quality targets in the predicted future period, thus obtaining emission intensity limit time series information. Based on the emission intensity limit time series information and combined with the panoramic pollution source map, instruction entries are generated for each emission source through instruction generation and mapping, thus obtaining the dynamic differentiated collaborative control instruction set. Based on the dynamic differentiated collaborative control instruction set, through monitoring log integration, pollution simulation results, control instructions, and expected performance evaluations within the predicted period are recorded in chronological order, thus obtaining and outputting the smart city air status monitoring log.

[0016] Secondly, this application provides a smart city air quality monitoring system based on multi-source data fusion, the system comprising:

[0017] The coupling analysis module is used to acquire multi-source pollutant emission data and weather forecast information. Based on the multi-source pollutant emission data and the weather forecast information, it analyzes the process of ammonia and sulfur dioxide coupling to generate secondary particulate matter under different meteorological scenarios, and obtains a multi-pollutant coupling information set. The source contribution module is used to simulate and analyze the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the concentration of secondary particulate matter in the polluted area based on the multi-pollutant coupling information set, and obtain source contribution information. The dynamic planning module is used to generate a dynamic differentiated collaborative control instruction set based on the source contribution information, and dynamically plan the emission intensity limit time sequence of pollution sources with different contribution levels and risk levels in the corresponding time period, and output the smart city air status monitoring log.

[0018] Through the above technical solution, this application achieves the following beneficial effects:

[0019] Upgrading air monitoring from simple concentration measurement to in-depth analysis of the formation mechanism of secondary pollution significantly improves the scientific understanding and early warning capabilities of complex pollution processes. Based on detailed source apportionment, it can accurately identify key pollution sources under different meteorological conditions, providing a basis for targeted governance. The generated dynamic, differentiated, and collaborative control instructions can flexibly adjust the control intensity according to changes in contribution and risk, greatly improving management efficiency and precision. The panoramic monitoring log provides a complete chain of evidence for decision-making, enhancing the traceability and continuous optimization capabilities of the governance process.

[0020] It enables precise monitoring and efficient control of air conditions. Multi-source data fusion makes the secondary particulate matter generation mechanism clearer, source contribution analysis accurately locates key pollution sources, and dynamic differentiated control instructions are highly targeted. It not only reduces the cost of ineffective control, but also responds quickly to changes in pollution, helps environmental protection departments make scientific decisions, and promotes the continuous improvement of air quality in smart cities.

[0021] Accurately capture the multi-dimensional impact of meteorological conditions on the coupled reaction of ammonia and sulfur dioxide, clarify the regional reaction adaptability, reaction rate mutation nodes, and the dynamic effects of temperature and humidity fluctuations, make up for existing monitoring shortcomings, provide accurate meteorological support for the construction of multi-pollutant coupled information sets, ensure data accuracy and timeliness, help accurately identify emission source contribution characteristics, make monitoring fit the actual scenario, improve the reliability of results, provide a scientific basis for the generation of control instructions, and promote the improvement of air monitoring and control level in smart cities;

[0022] It accurately identifies the core impacts of factory emissions, clearly presents the synergistic contribution of different effective response pair combinations of factories and the superposition effect of secondary particulate matter transport, and solves the problem of insufficient targeting of factory emission control;

[0023] It quantifies the contribution ratio of each factory's emission sources, allowing control to focus on high-polluting factories and avoid resource waste and blind spots in control. It can both set targeted emission limits for factories to reduce secondary particulate matter concentrations (such as a decrease in concentrations in core areas) and ensure reasonable production needs of factories, balancing environmental protection and economic development.

[0024] It provides factories with accurate pollution concentration benchmarks, clearly showing the scope and extent of the impact of factory emissions on the surrounding area. This ensures that factories meet environmental compliance requirements, avoids production losses caused by excessive emission reduction, and provides data support for the coordinated governance of factories and other pollution sources at the regional level, promoting the harmonious development of factories and the surrounding environment.

[0025] To achieve precise source tracing and differentiated management of pollution sources in the factory area, focus on targeted treatment of high-contribution sources, significantly improve emission reduction efficiency, and dynamically adapt to changes in weather and production conditions to reduce management lag, thereby ensuring the orderly operation of production in the factory area and effectively reducing secondary particulate matter emissions.

[0026] By balancing the overall regional emission reduction needs with the local compliance requirements, the source contribution information is made more relevant to the actual management and control scenarios, avoiding the inefficiency and waste of a "one-size-fits-all" approach. Its quantitative results provide precise targeting for subsequent management and control instructions, enabling key management of high-contribution and high-risk sources and reasonable regulation of low-impact sources. This optimizes the allocation of management and control resources, reduces emission reduction costs, and significantly improves the scientific and efficient nature of smart city air governance.

[0027] To achieve precise and differentiated management and control of pollution sources, avoid resource waste and management gaps; dynamically adapt to meteorological changes to ensure real-time and effective management; monitoring logs support governance optimization and accountability, improve the systematicness and accuracy of air governance, optimize resource allocation, help achieve air quality standards, promote the modernization of ecological and environmental governance, and provide a high-quality air environment for smart cities. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a schematic diagram of an application scenario provided in an embodiment of this application;

[0030] Figure 2 A flowchart illustrating a smart city air condition monitoring method based on multi-source data fusion, provided as an embodiment of this application;

[0031] Figure 3This is a schematic diagram of the structure of a smart city air condition monitoring system based on multi-source data fusion, provided as an embodiment of this application. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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 some, not all, of the embodiments of this application. 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.

[0033] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0034] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0035] In the process of air quality monitoring and management in smart cities, existing technologies are unable to dynamically simulate the pollutant generation mechanism and accurately quantify the contribution of emission sources, resulting in inaccurate early warning of complex pollution and extensive management. There is an urgent need to develop a smart solution that can accurately simulate the pollution process, trace key sources, and support differentiated dynamic management.

[0036] Based on this, this application provides a smart city air condition monitoring method and system based on multi-source data fusion. Air monitoring is upgraded from concentration measurement to secondary pollution mechanism analysis, which deepens the scientific understanding and early warning capabilities of pollution processes. Based on accurate source apportionment, key pollution sources under different meteorological conditions can be identified to achieve targeted governance. Through dynamic and differentiated collaborative control instructions, the control intensity can be flexibly adjusted according to pollution contribution and risk, improving management efficiency and precision. The panoramic monitoring log provides a complete chain of evidence for decision-making, enhancing the traceability and continuous optimization capabilities of the governance process.

[0037] Figure 1 This application provides an illustration of an application scenario. In the process of air quality monitoring and control in smart cities, the method provided in this application has upgraded air monitoring from static concentration measurement to dynamic analysis of the causes of secondary pollution, realizing accurate identification and intelligent dynamic control of pollution sources. This transformation not only strengthens the ability of pollution early warning and targeted governance, but also promotes the continuous optimization of management efficiency through full-process traceability.

[0038] Specifically, the method provided in this application can be applied to any server. The server interacts with sensor networks and meteorological data service platforms to obtain multi-source pollutant emission data provided by the sensor networks and weather forecast information provided by the meteorological data service platforms. This enables precise and differentiated management of pollution sources, avoiding resource waste and management gaps, and outputs smart city air condition monitoring logs to environmental monitoring and inspection personnel, improving the systematicness and accuracy of air pollution control. Specific implementation methods can be found in the following embodiments.

[0039] Figure 2 This is a flowchart illustrating a smart city air quality monitoring method based on multi-source data fusion, provided in one embodiment of this application. The method of this embodiment can be applied to servers in the above scenarios. Figure 2 As shown, the method includes:

[0040] S201. Obtain multi-source pollutant emission data and weather forecast information. Based on the multi-source pollutant emission data and combined with weather forecast information, analyze the process of ammonia and sulfur dioxide coupling to generate secondary particulate matter under different meteorological scenarios, and obtain a multi-pollutant coupling information set.

[0041] Multi-source pollutant emission data can be a collection of concentration or emission flux data for multiple air pollutants (such as sulfur dioxide and ammonia), with the sensor network deployed by environmental monitoring departments as the data source. Weather forecast information can be meteorological element prediction data for a future period issued by meteorological departments, with the meteorological data service platform as the data source. Multi-pollutant coupling information sets can be a collection of information used to characterize the interaction of multiple precursor pollutants through atmospheric physicochemical processes under different meteorological conditions, generating secondary particulate matter.

[0042] Specifically, air quality monitoring and control in smart cities often face challenges. On the one hand, existing monitoring focuses on single-point pollutant concentration measurements, lacking in-depth analysis of complex pollution causes, especially the key process of secondary pollutant generation, resulting in "seeing the concentration but not the source and mechanism." On the other hand, meteorological conditions have a decisive influence on the pollution process, but existing methods often simply overlay meteorological data with emission data, failing to simulate the nonlinear pollution generation process driven by the coupling of the two in a refined manner, resulting in insufficient targeting of early warning and control.

[0043] S202. Based on the multi-pollutant coupled information set, the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the secondary particulate matter concentration in the polluted area are simulated and analyzed to obtain the source contribution information.

[0044] Source contribution information can be the impact of different ammonia emission sources and different sulfur dioxide emission sources on the concentration of secondary particulate matter in a polluted area. Ammonia emission sources can be any type of pollution source that emits ammonia into the atmosphere. Sulfur dioxide emission sources can be any type of pollution source that emits sulfur dioxide into the atmosphere. The polluted area can be a specific area requiring air quality monitoring and pollution control.

[0045] Specifically, one of the core reasons for the inefficiency of existing control measures is the inability to clearly define the specific contribution of each emission source to secondary particulate matter pollution, resulting in a lack of targeted control. By using the existing multi-pollutant coupled information set, pollution source tracking analysis methods and contribution assessment models, and by simulating the correlation between the emission intensity, emission time period and changes in secondary particulate matter concentration in the polluted area of ​​different emission sources, we can accurately quantify the contribution ratio, scope of influence and key period of influence of each ammonia emission source and each sulfur dioxide emission source to the secondary particulate matter concentration. This will generate detailed source contribution information and provide a scientific basis for the subsequent formulation of differentiated control strategies.

[0046] S203. Based on the source contribution information, generate a dynamic and differentiated collaborative management and control instruction set, dynamically plan the emission intensity limit time sequence of pollution sources with different contribution levels and risk levels in the corresponding time period, and output the smart city air status monitoring log.

[0047] A dynamic, differentiated, and collaborative control instruction set can be a dynamically formulated set of differentiated control instructions for different pollution sources. Contribution level can be a classification based on the degree of impact of the emission source on secondary particulate matter concentration. Risk level can be an environmental risk level classified based on factors such as the emission intensity of the emission source, the environmental sensitivity of the area, and the hazardousness of the pollutants. Emission intensity limit time series can be dynamically changing emission intensity upper limits set for different levels of pollution sources at different time periods. A smart city air condition monitoring log can be a complete file recording data throughout the entire monitoring process.

[0048] Specifically, the existing "one-size-fits-all" static control model is unable to effectively control high-pollution sources, may impose unnecessary restrictions on low-contribution pollution sources, and cannot adapt to dynamic changes in meteorological conditions and pollution status. Based on the contribution information of each source, combined with the contribution level classification standards and risk level assessment system, the emission intensity limit sequence for each time period should be dynamically planned in a differentiated manner to ensure that the control measures are accurate and efficient.

[0049] The method provided in this embodiment upgrades air monitoring from simple concentration measurement to in-depth analysis of the formation mechanism of secondary pollution, significantly improving the scientific understanding and early warning capabilities of complex pollution processes. Based on detailed source apportionment, it can accurately identify key pollution sources under different meteorological conditions, providing a basis for targeted governance. The generated dynamic, differentiated, and collaborative control instructions can flexibly adjust the control intensity according to changes in contribution and risk, greatly improving management efficiency and precision. The panoramic monitoring log provides a complete chain of evidence for decision-making, enhancing the traceability and continuous optimization capabilities of the governance process.

[0050] In some embodiments, multi-source pollutant emission data includes ammonia emission source location information, ammonia emission intensity information, sulfur dioxide emission source location information, and sulfur dioxide emission intensity information; weather forecast information includes wind speed and direction information, and atmospheric temperature and humidity information for the forecast period; based on the ammonia emission source location information and sulfur dioxide emission source location information, combined with wind speed and direction information, the transmission paths and spatial diffusion characteristics of pollutants emitted from different emission sources are analyzed during the forecast period through a combination of trajectory simulation and diffusion analysis, to obtain pollutant dynamic distribution information; based on the pollutant dynamic distribution information, combined with atmospheric temperature and humidity information, the promotion or inhibition information of meteorological conditions at different spatial locations on the chemical reaction process of ammonia and sulfur dioxide during the forecast period is analyzed, to obtain meteorological impact information; based on the meteorological impact information, combined with ammonia emission intensity information and sulfur dioxide emission intensity information, the spatial distribution and intensity time series of secondary particulate matter generated by the coupling of ammonia and sulfur dioxide under the differentiated reaction environment characterized by the meteorological impact information are analyzed, to obtain a multi-pollutant coupling information set.

[0051] Ammonia emission source location information can be data identifying the geographic coordinates of ammonia release sources within a city and surrounding areas. Ammonia emission intensity information can be data characterizing the mass of ammonia released by each ammonia emission source per unit time. Sulfur dioxide emission source location information can be data identifying the geographic coordinates of sulfur dioxide release sources. Sulfur dioxide emission intensity information can be data characterizing the mass of sulfur dioxide released by each sulfur dioxide emission source per unit time. Wind speed and direction information can be a data sequence from meteorological forecasts regarding the speed and direction of horizontal atmospheric movement near the ground or at different altitudes within a specific future time period. Atmospheric temperature and humidity information can be a data sequence from meteorological forecasts regarding atmospheric temperature and relative humidity within a specific future time period. Trajectory simulation can be a numerical method that uses known or predicted wind field data to trace the movement paths of air masses or pollutant particles through backward or forward calculations. Diffusion analysis can simulate the process of pollutants gradually diluting and decreasing in concentration in space due to turbulent mixing during transport. Pollutant dynamic distribution information can be a quantitative description of the temporal changes in the concentrations of ammonia and sulfur dioxide in the three-dimensional space of the monitoring area within the predicted time period, obtained by combining trajectory simulation and diffusion analysis results.

[0052] Specifically, in urban air quality monitoring and management, ammonia and sulfur dioxide emissions from factories are key pollution sources that couple to generate secondary particulate matter. Existing methods struggle to accurately capture the coupling reaction process between these two substances in complex environments, failing to clearly define the specific impact of factory emissions on secondary pollution. This easily leads to a lack of targeted control measures and a state of blind action. To address these issues: Collect and integrate ammonia emission source location information (e.g., the geographical coordinates of a farm) and ammonia emission intensity information (e.g., emission rates in kilograms per hour calculated based on activity levels and emission factors) covering sources such as agriculture and industry; and sulfur dioxide emission source location information and sulfur dioxide emission intensity information covering sources such as power plants and industrial boilers; simultaneously, integrate numerical weather prediction products to obtain wind speed and direction information (e.g., hourly, multi-layer wind field data for the next 48 hours) and atmospheric temperature and humidity information (e.g., gridded temperature and relative humidity data for the corresponding time period) for the forecast period. Next, dynamic transport and diffusion simulations are performed: starting from the aforementioned emission sources, wind speed and direction information is used to drive trajectory simulation (e.g., using the HYSPLIT Lagrange particle model) to calculate the movement paths of each pollutant cloud during the prediction period; simultaneously, diffusion analysis is combined (e.g., using a Gaussian plume / puff model or a more complex CFD model) to quantify the dilution and spatial distribution of pollutants during transport due to atmospheric turbulence, thereby generating high spatiotemporal resolution dynamic distribution information of pollutants, accurately depicting the concentration field evolution of ammonia and sulfur dioxide in three-dimensional space. Subsequently, Assessing the chemical reaction environment: Based on the dynamic distribution information of pollutants and atmospheric temperature and humidity information, the meteorological conditions of each grid point at each moment are analyzed. Using the principles of chemical reaction kinetics (e.g., applying the Arrhenius formula to describe the effect of temperature on the reaction rate, and combining the parameterization scheme of humidity on the heterogeneous reaction rate), the promotion or inhibition information of the environment on the formation of secondary particulate matter such as ammonium sulfate by the reaction of ammonia and sulfur dioxide is analyzed and quantified. For example, it is determined that under high humidity (e.g., relative humidity exceeding a certain critical value) and suitable temperature conditions in a certain region and at a certain moment, the reaction rate will be significantly increased. Finally, the coupling process is integrated: The above meteorological impact information is used as a dynamic environmental factor field and coupled with the ammonia emission intensity information and sulfur dioxide emission intensity information of each emission source within the framework of the chemical transport model. The spatiotemporal distribution of secondary particulate matter generation rate and cumulative concentration is simulated and calculated throughout the region and the prediction period. The output is a multi-pollutant coupled information set that can comprehensively reflect the results of the "emission-meteorology-chemistry" coupling process. This information set can be specifically represented as a grid distribution map of secondary particulate matter concentration and its change sequence over time.

[0053] The method provided in this embodiment enables precise monitoring and efficient control of air conditions. Multi-source data fusion makes the secondary particulate matter generation mechanism clearer, source contribution analysis accurately locates key pollution sources, and dynamic differentiated control instructions are highly targeted. This not only reduces the cost of ineffective control but also enables rapid response to changes in pollution, assisting environmental protection departments in making scientific decisions and promoting the continuous improvement of air quality in smart cities.

[0054] In some embodiments, based on the dynamic distribution information of pollutants and combined with atmospheric temperature and humidity information, the degree of compatibility of the suitable conditions required for the reaction of ammonia and sulfur dioxide is analyzed to obtain regional reaction suitability information; based on the regional reaction suitability information, through time-series correlation analysis, the influence of pollutant transport time-series characteristics and temperature and humidity changes at different times on the reaction rate abrupt change nodes of ammonia and sulfur dioxide coupling to generate secondary particulate matter is analyzed to obtain time-series reaction abrupt change information; based on the time-series reaction abrupt change information, through dynamic effect assessment, the influence of temperature and humidity fluctuation amplitude on the initiated reaction of ammonia and sulfur dioxide coupling to generate secondary particulate matter is analyzed to obtain meteorological impact information.

[0055] The suitable conditions for the reaction of ammonia and sulfur dioxide can be defined as the range of suitable meteorological parameters such as temperature and humidity when ammonia and sulfur dioxide react chemically to generate secondary particulate matter. Regional reaction suitability information can be information on the degree of compatibility between the meteorological conditions of a specific region and the suitable conditions for the ammonia-sulfur dioxide reaction. Temporal correlation analysis can be used to explore the impact of pollutant transport timing and temperature and humidity changes on the reaction process. Pollutant transport timing characteristics can be the characteristics of the transport distance and coverage of pollutants at different time points after they originate from the emission source, changing over time. Temporal temperature and humidity changes can be used to predict the fluctuations in atmospheric temperature and humidity at different time points within a given time period. Secondary particulate matter can be new types of particulate matter generated by the chemical reaction of ammonia and sulfur dioxide. The reaction rate abrupt change node can be the time point at which the reaction rate of ammonia and sulfur dioxide suddenly changes significantly. Temporal reaction abrupt change information can be information reflecting the relationship between the pollutant transport timing, temporal temperature and humidity changes, and the reaction rate abrupt change node. Dynamic effect assessment can be an assessment method used to analyze the dynamic impact of environmental parameter fluctuations on initiated reactions. Temperature and humidity fluctuation amplitude can be the range of atmospheric temperature and humidity changes at a specific time point within a given time period. The initiated reaction could be a chemical reaction process in which ammonia and sulfur dioxide have begun to couple and generate secondary particulate matter.

[0056] Specifically, in smart city air quality monitoring, the coupling of ammonia and sulfur dioxide to generate secondary particulate matter is a key chemical reaction affecting air quality. This reaction is highly dependent on dynamic changes in meteorological conditions. Meteorological factors not only directly determine whether the reaction occurs but also significantly influence its rate, intensity, and spatiotemporal distribution. To address these issues: First, regional similarity analysis is employed. Based on the dynamic distribution information of pollutants, the spatial location of pollutants is locked and matched with a gridded meteorological field provided by atmospheric temperature and humidity information. For each spatial grid point, the proximity index between its real-time temperature and humidity and the pre-set optimal temperature range (e.g., 20 to 25 degrees Celsius) and humidity range (e.g., 60% to 80% relative humidity) for the ammonia-sulfur reaction is calculated. For example, if a grid point currently has a temperature of 23 degrees Celsius and humidity of 70%, the percentage of overlap with the optimal conditions is calculated using a pre-set function model (e.g., piecewise linear or Gaussian decay function), thereby generating quantified regional reaction fit information. A higher value indicates that the current meteorological conditions are more favorable for the reaction to start. Next, time-series correlation analysis is used to analyze the continuous time of the above fit information. The sequence and pollutant transport model outputs trajectory time-series data reflecting the movement path of the same pollutant mass are overlaid and analyzed to identify the "resonance points" between the two in space and time. For example, when tracking a pollutant mass transported downwind from an emission source, it is found that when it moves to a grid about 10 kilometers downwind, the grid's fit index jumps sharply from 50% to 90% between 2 pm and 3 pm. This specific spatiotemporal node is identified as a temporal reaction abrupt change, meaning that a sudden increase in the reaction rate may occur at this time. Finally, the dynamic effect assessment method is applied to analyze the actual fluctuation range of temperature and humidity in the area in the following hours for the identified abrupt change node (such as the reaction acceleration point at 2 pm mentioned above). For example, if at 4 pm after the reaction acceleration starts, the temperature in the area drops sharply by 5 degrees and the humidity drops by 20% due to weather changes, the preset reaction process decay model is used to assess whether this drastic fluctuation causes the already generated ammonium sulfate precursor or particulate matter growth process to stop the reaction, change the product morphology, or only cause the rate to decrease gradually. Thus, a comprehensive quantification of meteorological impact information that reflects both the reaction initiation potential and the process disturbance effect is obtained.

[0057] The method provided in this embodiment accurately captures the multi-dimensional impact of meteorological conditions on the coupled reaction of ammonia and sulfur dioxide, clarifies the regional reaction adaptability, reaction rate mutation nodes, and the dynamic effects of temperature and humidity fluctuations, makes up for the shortcomings of existing monitoring, provides accurate meteorological support for the construction of multi-pollutant coupled information sets, ensures data accuracy and timeliness, helps to accurately identify the contribution characteristics of emission sources, makes monitoring more relevant to actual scenarios, improves the reliability of results, provides a scientific basis for the generation of control instructions, and promotes the improvement of air monitoring and control in smart cities.

[0058] In some embodiments, based on pollutant dynamic distribution information, spatiotemporal matching analysis is performed on effective reaction pairings. This analysis examines the time-varying effective reaction pairing relationships between different ammonia emission sources and different sulfur dioxide emission sources within the predicted time period, formed by the combined effects of spatial relative location and meteorological conditions, yielding dynamic source synergy information. An effective reaction pairing is a combination of effective reaction pairs formed between ammonia emission sources and sulfur dioxide emission sources. Based on the dynamic source synergy information, the dynamic synergy contribution intensity is calculated to quantify the matching degree and effectiveness between the emission intensity ratio of each effective reaction pairing and the reaction conditions characterized by the time-series reaction mutation information within the corresponding time period, obtaining dynamic synergy contribution information. Based on the dynamic synergy contribution information, combined with wind speed and direction information, a spatiotemporal mapping table is constructed, indexed by the effective reaction pairing identifier and corresponding wind speed and direction information, and taking the concentration distribution of secondary particulate matter on the transport grid as the value. This table characterizes the trajectory interweaving and spatial superposition effects of secondary particulate matter generated by different effective reaction pairings in subsequent transport, resulting in a multi-pollutant coupling information set.

[0059] Dynamic source synergy information can be the information on effective reaction pairing relationships between different ammonia and sulfur dioxide emission sources within a predicted time period, formed by the combined effects of spatial relative location and meteorological conditions, and changing over time. Dynamic synergy contribution intensity can be an indicator quantifying the matching degree and effectiveness between the emission intensity ratio of each effective reaction pairing combination and the reaction conditions characterized by time-series reaction abrupt change information within a corresponding time period. Time-series reaction abrupt change information can be information reflecting the temporal characteristics of pollutant transport and the impact of temperature and humidity changes at different times on the reaction rate abrupt change nodes of ammonia and sulfur dioxide coupling to generate secondary particulate matter. Trajectory interweaving and spatial superposition effects can be the phenomenon of interweaving paths and spatial superposition of concentrations of secondary particulate matter generated by different effective reaction pairing combinations during transport.

[0060] Specifically, in the collaborative management and control of air pollution in smart cities, ammonia and sulfur dioxide emissions from factories are the main sources of secondary particulate matter. Differences in emission intensity, spatial location, and meteorological conditions among different factories lead to varying synergistic effects of effective reaction pair combinations. Without precise analysis, it is easy to misjudge high-contribution factories (such as factories in a chemical industrial park) or overlook dynamically formed highly synergistic factory combinations, resulting in a lack of targeted control instructions. Only by clearly defining the synergistic relationships and contributions of effective reaction pair combinations from factories can a basis for differentiated management be provided, effectively reducing secondary particulate matter pollution. To address these issues: First, based on the acquired dynamic distribution information of pollutants (e.g., showing that an ammonia mass in an industrial area is spreading downstream with the northwest wind), the spatiotemporal matching analysis of effective reaction pair combinations is used to systematically scan and analyze the spatiotemporal coupling possibility between all ammonia emission sources and sulfur dioxide emission sources within the predicted period. The algorithm comprehensively considers the real-time distance, relative orientation, and pollutant transport trajectory caused by wind speed and direction between each pair of sources, and combines key nodes in meteorological impact information such as "relative humidity rising above 70% in the afternoon" to dynamically determine which effective reaction pair combinations are in specific situations. At a given time, pollutants emitted from these sources can converge along the transport path and be in a suitable meteorological environment for chemical reactions, thus forming effective reaction pairs (i.e., "effective reaction pair combinations"). This process outputs structured dynamic source synergy information. Then, for each identified effective reaction pair combination, its actual impact is quantified by calculating the dynamic synergistic contribution intensity. This method integrates the real-time emission intensity ratio of ammonia to sulfur dioxide in the effective reaction pair combination (e.g., the ammonia emission rate in a certain source pair is twice that of sulfur dioxide) and evaluates this ratio in relation to the optimal reaction conditions revealed by the current time-series reaction abrupt change information. The matching degree of conditions (such as "high temperature and high humidity conditions are more conducive to ammonium salt formation") is used to calculate the quantitative contribution value of the effective reaction pairing combination to the formation of secondary particulate matter in the current time period, i.e., dynamic synergistic contribution information. Finally, in order to present a panoramic view of the spatial distribution and superposition effect of the products of all effective reaction pairing combinations, a spatiotemporal mapping table is constructed using gridded spatial analysis and data indexing technology: using the unique identifier of each effective reaction pairing combination and the dominant wind speed and direction of its location (such as "effective reaction pairing combination A, wind speed 3m / s, wind direction southeast") as the composite index key, and using atmospheric diffusion models (such as CA) LPUFF or AERMOD simulations measure the concentration distribution of secondary particulate matter generated by the effective reaction pairings within a predefined three-dimensional spatial transport grid (e.g., dividing the urban area into 1km×1km grids) over its lifetime. The results are then entered into a table. This table clearly shows, for example, how secondary particulate matter generated by several key effective reaction pairings from the northern industrial area is transported and superimposed on the sensitive area of ​​the city center under a specific wind field. This ultimately forms a comprehensive and detailed set of multi-pollutant coupling information, providing a targeted map for precise control.

[0061] The method provided in this embodiment accurately identifies the core impacts of factory emissions, clearly presents the synergistic contribution of different effective response pair combinations of factories and the superposition effect of secondary particulate matter transport, and solves the problem of insufficient targeting of factory emission control.

[0062] In some embodiments, a panoramic pollution source map covering all ammonia and sulfur dioxide emission sources is constructed based on a spatiotemporal mapping table. This map clarifies the spatial attributes, emission intensity time series, and effective response pair correlations of each emission source, resulting in a panoramic pollution source map. Based on this panoramic pollution source map, the concentration accumulation process of secondary particulate matter at each receptor point within the pollution area is simulated under normal operating conditions of all emission sources during the predicted period, yielding baseline concentration field information. Based on the baseline concentration field information, the changes in secondary particulate matter concentration at each receptor point compared to the baseline concentration field are analyzed under single-variable scenarios of closure or doubling of emission intensity for each ammonia and sulfur dioxide emission source, obtaining single-source contribution intensity information. Based on the single-source contribution intensity information, the contribution ratio of each emission source to the overall pollution area and key receptor point concentration changes is analyzed through contribution ranking and regional mapping analysis, yielding sub-source contribution information.

[0063] A panoramic pollution source map can be a structured network map integrating the spatial locations, emission intensity time series, and dynamic effective response pairings of all ammonia and sulfur dioxide emission sources. The baseline concentration field information can be the spatiotemporal distribution data field of secondary particulate matter concentration within the polluted area, obtained through numerical simulation under the assumption that all pollution sources operate according to their normal emission sequence during the prediction period. Single-source contribution intensity information can be obtained through scenario simulation, quantifying the change characteristics of secondary particulate matter concentration at each receiver point relative to the baseline concentration when a single change occurs in the emission status (e.g., shutdown or doubling) of a particular ammonia or sulfur dioxide emission source. Receiving points can be selected spatial locations within the polluted area used to monitor or assess pollutant concentrations; these can be actual monitoring stations or virtual grid points. Scenario simulation is a research method that involves changing model input conditions and running an air quality model to predict pollutant concentration distributions under different assumed scenarios.

[0064] Specifically, in the practical scenarios of smart city air quality monitoring and control, factories (such as chemical and metallurgical plants) are major sources of ammonia and sulfur dioxide emissions. Their emission intensity and timing vary significantly, and the contributions of pollutants emitted by different factories to the formation of secondary particulate matter differ. Simply knowing the multi-pollutant coupling effect is insufficient to accurately identify high-contribution factories, easily leading to control imbalances—either over-controlling low-contribution factories, increasing operating costs, or under-controlling high-polluting factories, resulting in pollution exceeding standards. To address these issues, the approach begins with a deep analysis of the "spatiotemporal mapping table" of multi-pollutant coupling information. Using graph theory and Geographic Information System (GIS) spatial analysis techniques, discrete effective reaction pairs are combined. By reconstructing the pollutant grid data, a "panoramic pollution source map" is constructed, with each independent ammonia and sulfur dioxide source as nodes, dynamic effective reaction pairing relationships as edges, and the precise geographical coordinates and emission intensity time series curves of each node integrated. Based on this structured map, a third-generation air quality numerical model (such as CMAQ) with localized parameter verification is used to perform high spatiotemporal resolution simulation, driven by the normal emission time series of all sources within the prediction period, to generate the spatiotemporal field of the baseline concentration of secondary particulate matter at each receptor point in the pollution area (e.g., the national monitoring station A located in the city center). (For example, the simulation shows that the peak baseline concentration at this station will be approximately 0.12 mg / L tomorrow evening.) Subsequently, a systematic single-source contribution analysis loop was initiated: using scenario simulation technology, the emission status of each node (source) in the spectrum was sequentially set to "off" (emissions of 0) or "doubled" (emissions multiplied by 2), while ensuring that the emission status of all other nodes strictly remained unchanged from the baseline scenario. The numerical model was then rerun. After each simulation, the difference between the concentration at each receptor point and the baseline concentration field under the new scenario was calculated using a differential analysis algorithm (for example, when a thermal power plant B in the eastern part of the city is shut down, the peak concentration at monitoring station A drops to 0.09 mg / L, which means it contributes 0.03 mg / L). The data is analyzed to accurately quantify the "contribution intensity" of a single source under all other conditions remaining constant. Finally, the contribution intensity data of all sources are collected, and the contribution ranking algorithm (such as ranking by absolute amount or proportion of contribution) and spatial mapping analysis (e.g., using Kriging interpolation to visualize the spatial contribution values ​​of discrete receptor points) are used to calculate the contribution rate of the overall average concentration of each effective response pair combination area (e.g., thermal power plant B contributes 8% of the average concentration of the entire area) and the influence weight of the concentration of key grid points such as schools and hospitals. Finally, a structured "source contribution information" report and spatial distribution map are formed.

[0065] The method provided in this embodiment quantifies the contribution ratio of each factory's emission sources, allowing control to focus on high-polluting factories and avoiding resource waste and blind spots in control. It can both formulate targeted emission limit sequences for factories to reduce secondary particulate matter concentrations (such as a decrease in concentrations in core areas) and ensure reasonable production needs of factories, balancing environmental protection and economic development.

[0066] In some embodiments, based on dynamic source synergy information, dynamic synergy impact analysis is used to analyze the temporal sequence of the influence intensity of each synergistic contribution effective response pair combination on the secondary particulate matter concentration at the receptor point within the corresponding time period, thereby obtaining single-source time-effect contribution information; based on single-source time-effect contribution information, spatiotemporal overlay analysis is used to integrate the influence of all emission sources on each receptor point within the predicted time period, thereby obtaining the temporal sequence change information of receptor point concentration; based on the temporal sequence change information of receptor point concentration, the polluted area is spatially gridded, and the temporal sequence change of concentration at each receptor point is mapped to the corresponding spatial grid to form the spatial distribution temporal sequence of secondary particulate matter concentration within the polluted area, thereby obtaining the baseline concentration field information.

[0067] Single-source temporal contribution information can be the information on the intensity of the influence of each synergistic contributing effective response pair on the secondary particulate matter concentration at the receptor site over time within the corresponding time period. Receptor site concentration temporal variation information can be the information on the variation of secondary particulate matter concentration at each receptor site within the polluted area over the prediction period. Spatial grid division can be a processing method that regularly divides the polluted area according to a certain scale.

[0068] Specifically, subsequent source contribution analysis requires quantifying the contribution ratio of a single emission source by comparing its concentration changes before and after the change. The baseline concentration field information, serving as the concentration benchmark for all emission sources under normal operating conditions, is the prerequisite and foundation for this comparison. Without a unified and accurate baseline concentration field, it will be impossible to distinguish the cumulative effects of concentration differences caused by single-source changes on other emission sources, leading to deviations in the calculation of source contribution ratios. To address this issue: First, dynamic synergistic impact analysis is employed to perform refined simulations of each effective "effective reaction pairing combination." Using a Lagrange particle diffusion model or an Eulerian chemical transport model coupled with chemical mechanisms, the ammonia and sulfur dioxide emitted by this effective reaction pairing combination are simulated under wind speed and direction (e.g., a southerly wind of 2.5 m / s). The transmission trajectory and spatial diffusion are driven by real-time meteorological fields such as temperature and humidity (e.g., temperature 25℃, relative humidity 70%). The rate and yield of secondary particulate matter (such as ammonium sulfate) generated by chemical reactions along the transmission path are calculated in real time. Finally, the time-series curves of the secondary particulate matter concentration impact of the effective reaction pair combination at various preset receptor points downwind (e.g., 50 air quality monitoring stations distributed in the urban area) are quantified, thus obtaining structured single-source time-effect contribution information. Next, a blank concentration field time-series covering all receptor points and all time steps is created using spatiotemporal overlay analysis. Then, the concentration contribution time-series data of all effective reaction pair combinations (potentially thousands of pairs) to each receptor point, calculated in the previous steps, are linearly overlaid and integrated according to their corresponding spatial location and time label. For example, the contribution values ​​of effective reaction pair combination S1 to receptor point R1 at time t1, the contribution values ​​of effective reaction pair combination S2 to R1 at time t1, and so on, are all accumulated to generate a complete concentration time series for each receptor point throughout the entire prediction period, formed by the combined effects of all emission sources—that is, the time-series concentration change information of the receptor point. Finally, based on this information, a global concentration field is constructed using spatial interpolation and grid mapping techniques. The entire polluted area of ​​interest (e.g., the entire city area of ​​approximately 5000 square kilometers) is divided into regular spatial grids (e.g., 1 km × 1 km square grids). Using algorithms such as Kriging interpolation or inverse distance weighted interpolation, the discrete receptor point concentration time-series data are calculated and allocated to the center point of each grid unit based on their geographic coordinates and spatial correlation. This process not only fills the data gaps between monitoring points but also fully considers the potential impact of topography, land use, and other factors on pollutant diffusion. Ultimately, a three-dimensional data volume (spatial dimension X*Y, time dimension T) is output, with grids as spatial units and time steps (e.g., 1 hour) as time units. The value of each data unit represents the secondary particulate matter simulated concentration of that grid at that moment. This is the high spatiotemporal resolution baseline concentration field information, providing a panoramic and mechanistic dynamic background concentration reference for subsequent contribution assessment.

[0069] The method provided in this embodiment offers factories a precise reference for pollution concentration, clearly showing the scope and extent of the impact of factory emissions on the surrounding area. This ensures that factories meet environmental compliance requirements while avoiding production losses caused by excessive emission reductions. At the same time, it provides data support for the coordinated management of factories and other pollution sources at the regional level, promoting the harmonious development of factories and the surrounding environment.

[0070] In some embodiments, based on a panoramic pollution source map, a set of single variation scenarios is constructed for each ammonia emission source and each sulfur dioxide emission source. The set of single variation scenarios includes a first scenario where the corresponding emission source is completely shut down and a second scenario where the emission intensity of the corresponding emission source is doubled, resulting in a set of source-specific variation scenarios. Based on the set of source-specific variation scenarios and combined with baseline concentration field information, simulations of each variation scenario are executed sequentially. Using air quality numerical techniques, the spatiotemporal changes in the secondary particulate matter concentration at each receptor point compared to the baseline concentration are analyzed during the prediction period when the variation scenario is executed, to obtain information on the contribution intensity of each source.

[0071] A single variation scenario set can be a combination of scenarios designed for a single emission source to simulate the impact of changes in its emission status. The first scenario can be a complete shutdown scenario of the emission source within the single variation scenario set. The second scenario can be a doubling scenario of the emission intensity of an emission source within the single variation scenario set, where the emission intensity of a certain emission source doubles from its original level. A source-specific variation scenario set can be a complete scenario library formed by summarizing the single variation scenario sets of all ammonia and sulfur dioxide emission sources. Air quality numerical techniques are professional analytical techniques based on atmospheric physicochemical principles that simulate pollutant transport, diffusion, and concentration changes. The spatiotemporal variation of secondary particulate matter concentration can be the difference in secondary particulate matter concentration at each receiver point over time and spatial location under varying scenarios.

[0072] Specifically, existing pollution source contribution assessments largely rely on emission inventory proportion allocation or simple sensitivity tests. Their core flaw lies in simplifying complex, nonlinear atmospheric chemical-transport processes into linear assumptions, failing to isolate the true impact of individual sources within a dynamically coupled network. For example, the contribution of an upwind ammonia source to secondary particulate matter in a downwind residential area is not accurately reflected by its emission proportion, but rather depends on the spatiotemporal convergence probability of the ammonia transport path with multiple sulfur dioxide sources, and the real-time modulation of reaction rates by meteorological conditions at the convergence point. This macroscopic estimation, ignoring process mechanisms, leads to distorted source tracing, causing control measures to "miss the main contradictions." To address these issues: utilizing the full data generated from preceding stages... The pollution source map records the identity IDs, geographical coordinates, baseline emission intensity time-series curves, and collaborative relationships with other sources for all ammonia and sulfur dioxide emission sources in the region in structured data form. Based on this, an automated script program is written to traverse each independent emission source ID in the map (e.g., the ID "ammonia emission outlet of Unit #1 of a thermal power plant") and programmatically generate two corresponding scenario configuration files, thereby constructing a set of source-specific change scenarios covering all assessment objects: the first configuration file overwrites the emission intensity time-series of the source with 0 for the entire prediction period (e.g., the next 48 hours), i.e., the first scenario (completely shut down); the second configuration file sets the baseline emission intensity of the source to 0. The intensity time series data is multiplied by a set coefficient (e.g., 2), resulting in the second scenario (emission intensity doubling). This scenario set, along with baseline concentration field information reflecting the background field of all sources' normal operating conditions (a dataset containing spatiotemporally gridded concentration values), is then imported into an air quality numerical model system (e.g., using the WRF-CMAQ framework) for simulation calculations. This system, based on atmospheric dynamics and chemical reaction mechanisms (such as simulating the process of ammonia reacting with sulfur dioxide to form ammonium sulfate and other particulate matter), sequentially reads each scenario configuration file. While keeping all other source emissions at the baseline state unchanged, it only changes the state of the target source and reruns the simulation. For the "ammonia emission outlet of Unit #1 of a certain thermal power plant," the simulation is then performed with the "closed" setting applied. In the "closed" scenario simulation, the model calculates a new concentration field. Through spatiotemporal data comparison and analysis technology, this new concentration field is subtracted from the baseline concentration field hourly and grid-by-grid to obtain the concentration difference field. The value at each position in this difference field quantifies the contribution intensity of the ammonia source closure to local pollution. Similarly, by simulating and comparing the "doubling" scenario, the concentration increment caused by the increase in emissions can be obtained. By systematically performing simulations and comparative analyses of all sources and all scenarios, a structured single-source contribution intensity information dataset is finally generated, which clearly records the specific impact values ​​and spatiotemporal distribution characteristics of each source on the secondary particulate matter concentration at various locations in the region under two key intervention states.

[0073] The method provided in this embodiment enables precise source tracing and differentiated management of pollution sources in the plant area, focuses on targeted treatment of high-contribution sources, significantly improves emission reduction efficiency, and dynamically adapts to changes in weather and production conditions to reduce management lag. This ensures the orderly operation of plant production while effectively reducing secondary particulate matter emissions.

[0074] In some embodiments, based on single-source contribution intensity information, the impact ratio of each emission source on the total cumulative concentration of the pollution area during the prediction period is analyzed through dynamic contribution ranking to obtain initial contribution ratio information; based on the temporal change information of receptor point concentration, receptor points with excessive concentrations are identified as key receptor points through regional correlation mapping, and the distribution characteristics of the contribution intensity of each emission source during the key period leading to the exceedance are analyzed to obtain key point source correlation information; based on the initial contribution ratio information, combined with the key point source correlation information, the overall impact of each emission source and its local impact on key points are weighted and fused to quantify the final contribution ratio and obtain sub-source contribution information.

[0075] Contribution ranking and regional mapping analysis can be a composite data processing approach that integrates overall ranking and spatial correlation analysis. The overall pollution area can refer to the entire monitored or simulated custom-defined gridded study area. Key receptor points can refer to specific geographical locations within the pollution area that are environmentally sensitive or prone to concentration exceedances. The contribution ratio can be a quantitative indicator characterizing the percentage change in secondary particulate matter concentration caused by a change in emissions from a single emission source, relative to the total concentration change caused by the combined effect of all emission sources. Initial contribution ratio information can be data considering only the proportion of the emission source's impact on the overall cumulative concentration of the pollution area. Temporal variation information of receptor point concentrations can be the hourly secondary particulate matter concentration data sequence for each receptor point within the pollution area during the prediction period. Key point source correlation information can be the distribution of contribution intensity of each emission source corresponding to the key receptor point during the critical period of concentration exceedance.

[0076] Specifically, in the practical scenario of air quality monitoring and control in smart cities, relying solely on single-source contribution intensity information cannot distinguish the differentiated impact of emission sources on the overall regional and local exceedances, easily leading to control imbalances—insufficient control of high-contribution sources or excessive control of low-contribution sources. The specific implementation is as follows: First, based on single-source contribution intensity information, numerical simulation post-processing technology is used to perform "dynamic contribution ranking" on each ammonia and sulfur dioxide emission source. This involves calculating the reduction in the cumulative secondary particulate matter concentration across all grids within the entire pollution area during the prediction period under varying scenarios (e.g., shutdown) for each source. This reduction is then divided by the sum of reductions from all similar sources, thus quantifying the "initial contribution ratio information" of each emission source to the overall regional area. For example, the contribution ratio of a power plant's sulfur dioxide source might be calculated as 18.5%, representing its macroscopic impact weight. Simultaneously, based on "receptor point concentration time-series change information," geographic information spatial analysis and time series mining techniques are used for "regional association mapping": a concentration exceedance threshold is set (e.g., 0.15 mg / m³). 3 The simulation results automatically identify all grid points where the concentration exceeds this threshold and define them as "critical receptor points." Then, for each critical receptor point, the concentration composition during the period of exceedance (e.g., 20:00-24:00 on a certain day) is traced back, and the contribution intensity sequence of each emission source to this point during that period is extracted to analyze its distribution characteristics (e.g., the peak contribution intensity of ammonia from a chemical plant during this period reaches 0.025 mg / m³ per hour). 3 This forms "key point source association information." Finally, a weighted fusion algorithm is designed to integrate the "initial contribution ratio" with the "key point source association information." For example, a weight coefficient is set (e.g., 0.7 for overall contribution and 0.3 for local key influence), and the overall proportion of each source and its average contribution intensity during all key point exceedance periods are normalized and weighted to generate a comprehensive "final contribution ratio," thus completing the construction of "source-specific contribution information."

[0077] The method provided in this embodiment takes into account both the overall emission reduction needs of the region and the local compliance needs, making the source contribution information more in line with the actual management and control scenarios, avoiding the inefficiency and waste of "one-size-fits-all" approaches. Its quantitative results provide precise targeting for subsequent management and control instructions, enabling key management of high-contribution and high-risk sources and reasonable regulation of low-impact sources, optimizing the allocation of management and control resources, reducing emission reduction costs, and significantly improving the scientific and efficient nature of smart city air governance.

[0078] In some embodiments, based on single-source contribution intensity information and combined with weather forecast information, dynamic limit planning is used to analyze the time-varying emission intensity upper limit sequence that different emission sources should follow in order to ensure air quality targets during the predicted future period, thus obtaining emission intensity limit time series information. Based on the emission intensity limit time series information and combined with a panoramic pollution source map, instruction entries are generated for each emission source through instruction generation and mapping, resulting in a dynamic differentiated collaborative control instruction set. Based on the dynamic differentiated collaborative control instruction set, through monitoring log integration, pollution simulation results, control instructions, and expected performance evaluations during the predicted period are recorded in chronological order, resulting in and outputting the smart city air status monitoring log.

[0079] Dynamic limit planning can be a method that combines multi-source data to dynamically analyze the reasonable upper limit of emission intensity for emission sources. Emission intensity limit time-series information can be a sequence of emission intensity upper limits that different emission sources should follow at different times. Instruction generation and mapping can be the process of accurately matching emission intensity limit time-series information with each emission source to generate specific control instructions. Monitoring log integration can be the process of summarizing pollution simulation results, control instructions, and expected implementation effects in chronological order.

[0080] Specifically, in the process of air quality monitoring and governance in smart cities, the aforementioned analysis identified key pollution sources and their contribution characteristics. However, if this scientific understanding is not transformed into specific, executable, and condition-dependent control actions, the entire monitoring system will remain at the "diagnosis" level and will not achieve the "treatment" objective. The existing "one-size-fits-all" or static emission restriction model cannot cope with the spatiotemporal heterogeneity of pollution contributions under dynamic changes in meteorological conditions, which may lead to insufficient or excessive control, resulting in increased economic costs or environmental risks. To address the above issues: First, based on the calculated "single-source contribution intensity information" (for example, it has been quantified that the contribution rate of coal-fired power plant A to PM2.5 in the urban area is 15%, while that of chemical plant B is only 2%) and the input refined "weather forecast information" (such as hourly wind direction, wind speed, and humidity forecasts for the next 72 hours), the embedded "dynamic limit planning" optimization algorithm is invoked. The algorithm takes achieving regional air quality targets (such as the daily average PM2.5 concentration not exceeding 75 micrograms per cubic meter in the next 24 hours) as a constraint, treats each pollution source as an adjustable variable, and, with the support of the fast response surface of an air quality numerical model (such as CMAQ or WRF-Chem), solves inversely the optimal emission intensity limit sequence that each source should follow in order to achieve the overall target during the future forecast period, thereby generating "emission intensity limit time series information". For example, the algorithm may output instructions: Power plant A's emission limit from 8 pm tonight to 8 am tomorrow (when the expected static and high humidity and poor diffusion conditions) needs to be tightened to 60% of the normal value (i.e., about 30 tons / day), while chemical plant B only needs to reduce it to 90% during the same period. Subsequently, the system uses the "Instruction Generation and Mapping" module to automatically match and bind this limit time series information with the "Panoramic Pollution Source Map" (which is associated with control attributes such as source ID, geographical coordinates, affiliated enterprise, and contact information). For each controlled source in the map (such as an ammonia emission source with ID 'NH3_Plant_001'), a machine-readable and human-executable instruction entry containing specific limit requirements and execution time periods is generated. All entries are aggregated to form the "Dynamic Differentiated Collaborative Control Instruction Set". Finally, through the "Monitoring Log Integration" service, the key data of the entire process in this cycle (input meteorological and emission data), model simulation results (multi-pollutant coupled spatial distribution, source contribution map), generated control instruction set, and model-based instruction execution expected effect assessment (such as "after executing this instruction set, the peak concentration is expected to be reduced by 20%) are automatically integrated and formatted according to timestamps to generate a complete, traceable, and auditable "Smart City Air Status Monitoring Log" document, which is then output to the management platform for decision-makers to sign and archive.

[0081] The method provided in this embodiment enables precise and differentiated management and control of pollution sources, avoiding resource waste and management gaps; dynamically adapts to meteorological changes to ensure real-time and effective management; monitoring logs support governance optimization and accountability, improve the systematicness and accuracy of air governance, optimize resource allocation, help achieve air quality standards, promote the modernization of ecological and environmental governance, and provide a high-quality air environment for smart cities.

[0082] Figure 3 This is a schematic diagram of the structure of a smart city air condition monitoring system based on multi-source data fusion, provided in an embodiment of this application. Figure 3 As shown, the smart city air condition monitoring system 300 based on multi-source data fusion in this embodiment includes: a coupling analysis module 301, a source contribution module 302, and a dynamic programming module 303.

[0083] The coupling analysis module 301 is used to acquire multi-source pollutant emission data and weather forecast information. Based on the multi-source pollutant emission data and the weather forecast information, it analyzes the process of ammonia and sulfur dioxide coupling to generate secondary particulate matter under different meteorological scenarios, and obtains a multi-pollutant coupling information set. The source contribution module 302 is used to simulate and analyze the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the concentration of secondary particulate matter in the polluted area based on the multi-pollutant coupling information set, and obtain source contribution information. The dynamic planning module 303 is used to generate a dynamic differentiated collaborative control instruction set based on the source contribution information, and dynamically plan the emission intensity limit time sequence of pollution sources with different contribution levels and risk levels in the corresponding time period, and output the smart city air status monitoring log.

[0084] Optionally, when the coupling analysis module 301 analyzes the process of ammonia and sulfur dioxide coupling to generate secondary particulate matter under different meteorological scenarios based on the multi-source pollutant emission data and combined with the weather forecast information to obtain a multi-pollutant coupling information set, it is specifically used for: the multi-source pollutant emission data including ammonia emission source location information, ammonia emission intensity information, sulfur dioxide emission source location information, and sulfur dioxide emission intensity information; the weather forecast information including wind speed and direction information, and atmospheric temperature and humidity information during the forecast period; based on the ammonia emission source location information and the sulfur dioxide emission source location information, combined with the wind speed and direction information, and through trajectory simulation and diffusion analysis... The study analyzes the transport paths and spatial diffusion characteristics of pollutants emitted from different emission sources during the predicted period to obtain dynamic distribution information of pollutants. Based on the dynamic distribution information of pollutants, combined with the atmospheric temperature and humidity information, the study analyzes the promotion or inhibition of the chemical reaction process between ammonia and sulfur dioxide by meteorological conditions at different spatial locations during the predicted period to obtain meteorological impact information. Based on the meteorological impact information, combined with the ammonia emission intensity information and the sulfur dioxide emission intensity information, the study analyzes the spatial distribution and intensity time series of secondary particulate matter generated by the coupling of ammonia and sulfur dioxide under the differentiated reaction environment characterized by the meteorological impact information to obtain a multi-pollutant coupling information set.

[0085] Optionally, the coupling analysis module 301, during the construction of the meteorological impact information, is specifically used for: analyzing the degree of compatibility of the suitable conditions required for the reaction of ammonia and sulfur dioxide based on the dynamic distribution information of pollutants and the atmospheric temperature and humidity information, to obtain regional reaction suitability information; based on the regional reaction suitability information, analyzing the influence of pollutant transport time sequence characteristics and temperature and humidity changes at different times on the reaction rate abrupt change nodes of ammonia and sulfur dioxide coupling to generate secondary particulate matter through time-series correlation analysis, to obtain time-series reaction abrupt change information; based on the time-series reaction abrupt change information, analyzing the influence of temperature and humidity fluctuation amplitude on the initiated reaction of ammonia and sulfur dioxide coupling to generate secondary particulate matter through dynamic effect assessment, to obtain meteorological impact information.

[0086] Optionally, when the coupling analysis module 301 analyzes the spatial distribution and intensity time series of secondary particulate matter generated by the coupling of ammonia and sulfur dioxide under the differentiated reaction environment characterized by the meteorological impact information to obtain a multi-pollutant coupling information set, it is specifically used to: perform spatiotemporal matching analysis on effective reaction pairings based on the pollutant dynamic distribution information; analyze the effective reaction pairing relationships that change over time between different ammonia emission sources and different sulfur dioxide emission sources during the predicted period due to the combined effects of spatial relative positions and meteorological conditions; and obtain dynamic source synergy information; the effective reaction pairing combination is the effective reaction pairing formed between ammonia emission sources and sulfur dioxide emission sources. For the combination; based on the dynamic source synergy relationship information, by calculating the dynamic synergy contribution intensity, the matching degree and effectiveness between the emission intensity ratio of each effective reaction pair combination and the reaction conditions characterized by the time-series reaction mutation information in the corresponding time period are quantified to obtain dynamic synergy contribution information; based on the dynamic synergy contribution information, combined with the wind speed and direction information, a spatiotemporal mapping table is constructed with the effective reaction pair combination identifier and the corresponding wind speed and direction information as the index and the concentration distribution of secondary particulate matter on the transport grid as the value, to characterize the trajectory interweaving and spatial superposition effect of secondary particulate matter generated by different effective reaction pair combinations in subsequent transport, and to obtain the multi-pollutant coupling information set.

[0087] Optionally, when the source contribution module 302 simulates and analyzes the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the secondary particulate matter concentration in the polluted area based on the multi-pollutant coupling information set to obtain source contribution information, it is specifically used for: constructing a panoramic pollution source map covering all ammonia emission sources and sulfur dioxide emission sources based on the spatiotemporal mapping table, clarifying the spatial attributes, emission intensity time sequence, and effective response pair correlation of each emission source, and obtaining a panoramic pollution source map; simulating the concentration accumulation process of secondary particulate matter at each receptor point in the polluted area under the normal operating conditions of all emission sources during the prediction period, based on the panoramic pollution source map, and obtaining baseline concentration field information; simulating the change characteristics of secondary particulate matter concentration at each receptor point relative to the baseline concentration field under the single change scenario of each ammonia emission source and each sulfur dioxide emission source being closed or having its emission intensity doubled, based on the baseline concentration field information, and obtaining single-source contribution intensity information; and analyzing the contribution ratio of each emission source to the overall concentration change of the polluted area and key receptor points through contribution ranking and regional mapping analysis, based on the single-source contribution intensity information, and obtaining source contribution information.

[0088] Optionally, the source contribution module 302, during the construction of the baseline concentration field information, is specifically used for: based on the dynamic source synergy relationship information, analyzing the temporal sequence of the influence intensity of the effective response pair combination of each synergy contribution on the secondary particulate matter concentration of the receptor point within the corresponding time period through dynamic synergy influence analysis, to obtain single-source time-effect contribution information; based on the single-source time-effect contribution information, comprehensively analyzing the influence of all emission sources on each receptor point within the predicted time period through spatiotemporal overlay analysis, to obtain receptor point concentration temporal change information; based on the receptor point concentration temporal change information, dividing the pollution area into spatial grids, and mapping the concentration temporal changes of each receptor point to the corresponding spatial grids to form the spatial distribution temporal sequence of secondary particulate matter concentration within the pollution area, to obtain the baseline concentration field information.

[0089] Optionally, the source contribution module 302, during the construction of the single-source contribution intensity information, is specifically used to: construct a set of single variation scenarios for each ammonia emission source and each sulfur dioxide emission source based on the panoramic pollution source map; the set of single variation scenarios includes a first scenario where the corresponding emission source is completely shut down and a second scenario where the emission intensity of the corresponding emission source is doubled, thus obtaining a set of source variation scenarios; based on the set of source variation scenarios and combined with the baseline concentration field information, sequentially execute the simulation of each variation scenario, and use air quality numerical technology to analyze the spatiotemporal changes of secondary particulate matter concentration at each receptor point compared with the baseline concentration during the prediction period when the variation scenario is executed, thereby obtaining the single-source contribution intensity information.

[0090] Optionally, when the source contribution module 302 analyzes the contribution ratio of each emission source to the overall pollution area and key receptor point concentration changes through contribution ranking and regional mapping analysis to obtain source contribution information, it is specifically used for: based on the single-source contribution intensity information, analyzing the influence ratio of each emission source on the overall pollution area's cumulative concentration within the predicted period through dynamic contribution ranking to obtain initial contribution ratio information; based on the receptor point concentration time-series change information, identifying receptor points with excessive concentrations as key receptor points through regional correlation mapping, and analyzing the distribution characteristics of each emission source's contribution intensity during the key period leading to the excessive concentration to obtain key point source correlation information; based on the initial contribution ratio information, combined with the key point source correlation information, weightedly fusing the overall impact of each emission source and its local impact on key points to quantify the final contribution ratio and obtain the source contribution information.

[0091] Optionally, when the dynamic planning module 303 generates a dynamic differentiated collaborative control instruction set based on the source contribution information, dynamically plans the emission intensity limit time sequence of pollution sources with different contribution levels and risk levels in the corresponding time period, and outputs the smart city air status monitoring log, it is specifically used for: based on the single-source contribution intensity information and combined with the weather forecast information, analyzing the time-varying emission intensity upper limit sequence that different emission sources should follow in order to ensure air quality targets in the predicted future period through dynamic limit planning, and obtaining emission intensity limit time sequence information; based on the emission intensity limit time sequence information and combined with the panoramic pollution source map, generating instruction entries for each emission source through instruction generation and mapping, and obtaining the dynamic differentiated collaborative control instruction set; based on the dynamic differentiated collaborative control instruction set, recording the pollution simulation results, control instructions and expected execution effect evaluation in the predicted period in chronological order through monitoring log integration, and obtaining and outputting the smart city air status monitoring log.

[0092] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A smart city air quality monitoring method based on multi-source data fusion, characterized in that, include: Acquire multi-source pollutant emission data and weather forecast information. Based on the multi-source pollutant emission data and the weather forecast information, analyze the process of ammonia and sulfur dioxide coupling to generate secondary particulate matter under different meteorological scenarios to obtain a multi-pollutant coupling information set, including: The multi-source pollutant emission data includes ammonia emission source location information, ammonia emission intensity information, sulfur dioxide emission source location information, and sulfur dioxide emission intensity information; The weather forecast information includes wind speed and direction information, and atmospheric temperature and humidity information for the forecast period. Based on the ammonia emission source location information and the sulfur dioxide emission source location information, combined with the wind speed and direction information, the transmission path and spatial diffusion characteristics of pollutants emitted from different emission sources during the predicted period are analyzed by combining trajectory simulation and diffusion analysis, so as to obtain the dynamic distribution information of pollutants. Based on the dynamic distribution information of the pollutants and the atmospheric temperature and humidity information, the meteorological conditions at different spatial locations during the predicted period are analyzed to determine whether they promote or inhibit the chemical reaction between ammonia and sulfur dioxide, thus obtaining meteorological impact information. The process of constructing the meteorological impact information includes: Based on the dynamic distribution information of the pollutants and the atmospheric temperature and humidity information, the degree of compatibility of the suitable conditions required for the reaction of ammonia and sulfur dioxide is analyzed to obtain the regional reaction suitability information. Based on the regional reaction adaptation information, through time-series correlation analysis, the influence of pollutant transport time-series characteristics and temperature and humidity changes at different times on the reaction rate mutation nodes of ammonia and sulfur dioxide coupling to generate secondary particulate matter is analyzed, and time-series reaction mutation information is obtained. Based on the aforementioned time-series reaction mutation information, the impact of temperature and humidity fluctuations on the initiated reaction of ammonia and sulfur dioxide coupling to generate secondary particulate matter is analyzed through dynamic effect assessment, and meteorological impact information is obtained. Based on the meteorological impact information, combined with the ammonia emission intensity information and the sulfur dioxide emission intensity information, the spatial distribution and intensity temporal sequence of secondary particulate matter generated by the coupling of ammonia and sulfur dioxide under the differentiated reaction environment characterized by the meteorological impact information are analyzed, resulting in a multi-pollutant coupling information set, including: Based on the dynamic distribution information of the pollutants, spatiotemporal matching analysis is performed on the effective reaction pairing combinations. The analysis is conducted on the effective reaction pairing relationships between different ammonia emission sources and different sulfur dioxide emission sources during the predicted period, which are formed by the combined effects of spatial relative location and meteorological conditions, and the dynamic source synergy relationship information is obtained. The effective reaction pairing combination is the effective reaction pairing combination formed between ammonia emission sources and sulfur dioxide emission sources; Based on the dynamic source synergy relationship information, by calculating the dynamic synergy contribution intensity, the matching degree and effectiveness of the emission intensity ratio of each effective reaction pair combination in the corresponding time period and the reaction conditions characterized by the time-series reaction mutation information are quantified, and dynamic synergy contribution information is obtained. Based on the dynamic collaborative contribution information and combined with the wind speed and direction information, a spatiotemporal mapping table is constructed with the identifier of the effective reaction pairing combination and the corresponding wind speed and direction information as the index and the concentration distribution of secondary particulate matter on the transmission grid as the value. This table characterizes the trajectory interweaving and spatial superposition effect of secondary particulate matter generated by different effective reaction pairing combinations in subsequent transmission, and obtains the multi-pollutant coupling information set. Based on the multi-pollutant coupling information set, the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the secondary particulate matter concentration in the polluted area are simulated and analyzed to obtain source contribution information. Based on the source contribution information, a dynamic differentiated collaborative management and control instruction set is generated. The emission intensity limit time sequence of pollution sources with different contribution levels and risk levels is dynamically planned in a differentiated manner for the corresponding time period, and the smart city air status monitoring log is output.

2. The method according to claim 1, characterized in that, Based on the multi-pollutant coupling information set, the simulation analysis of the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the secondary particulate matter concentration in the polluted area yields source contribution information, including: Based on the aforementioned spatiotemporal mapping table, a panoramic pollution source map covering all ammonia and sulfur dioxide emission sources is constructed, clarifying the spatial attributes, emission intensity time sequence, and effective reaction pairing correlation of each emission source, thus obtaining a panoramic pollution source map. Based on the panoramic pollution source map, the concentration accumulation process of secondary particulate matter at each receptor point in the pollution area is simulated under the normal operating conditions of all emission sources during the predicted period to obtain baseline concentration field information. Based on the baseline concentration field information, each ammonia emission source and each sulfur dioxide emission source are simulated sequentially under a single variation scenario of closure or doubling of emission intensity. The variation characteristics of secondary particulate matter concentration at each receptor point relative to the baseline concentration field are analyzed to obtain single-source contribution intensity information. Based on the single-source contribution intensity information, the contribution ratio of each emission source to the overall concentration change of the polluted area and key receptor points is analyzed through contribution ranking and regional mapping analysis, thus obtaining source contribution information.

3. The method according to claim 2, characterized in that, The process of constructing the baseline concentration field information includes: Based on the dynamic source synergy relationship information, through dynamic synergy impact analysis, the time series of the influence intensity of the effective response pairing combination of each synergy contribution on the secondary particulate matter concentration at the receptor point is analyzed in the corresponding time period to obtain single-source time-effect contribution information. Based on the single-source time-effect contribution information, the temporal and spatial superposition analysis is used to comprehensively analyze the impact of all emission sources on each receptor point during the prediction period, and obtain the temporal change information of receptor point concentration. Based on the temporal variation information of the receptor point concentration, the polluted area is divided into spatial grids, and the temporal variation of the concentration of each receptor point is mapped to the corresponding spatial grid to form the spatial distribution temporal sequence of secondary particulate matter concentration in the polluted area, thus obtaining the reference concentration field information.

4. The method according to claim 3, characterized in that, The process of constructing the single-source contribution intensity information includes: Based on the panoramic pollution source map, a set of single variation scenarios is constructed for each ammonia emission source and each sulfur dioxide emission source. The single variation scenario set includes a first scenario where the corresponding emission source is completely shut down and a second scenario where the emission intensity of the corresponding emission source is doubled, thus obtaining a source variation scenario set; Based on the set of source variation scenarios and the baseline concentration field information, simulations of each variation scenario are executed sequentially. Using air quality numerical techniques, the spatiotemporal changes in secondary particulate matter concentrations at each receptor point compared to the baseline concentration are analyzed during the prediction period when the variation scenarios are executed, thereby obtaining the single-source contribution intensity information.

5. The method according to claim 4, characterized in that, The contribution ranking and regional mapping analysis are used to analyze the proportion of each emission source's contribution to the overall concentration changes of the pollution area and key receptor points, obtaining source contribution information, including: Based on the single-source contribution intensity information, the impact ratio of each emission source on the total cumulative concentration of the pollution area during the prediction period is analyzed by dynamic contribution ranking to obtain the initial contribution ratio information. Based on the temporal variation information of the receptor point concentration, the receptor points with excessive concentrations are identified as key receptor points through regional correlation mapping, and the distribution characteristics of the contribution intensity of each emission source during the key period leading to the excessive concentration are analyzed to obtain key point source correlation information. Based on the initial contribution ratio information and combined with the key point source association information, the overall impact of each emission source and its local impact on the key points are weighted and fused to quantify the final contribution ratio, thus obtaining the source contribution information.

6. The method according to claim 5, characterized in that, Based on the source contribution information, a dynamic differentiated collaborative management and control instruction set is generated. This set dynamically plans the emission intensity limit time series for pollution sources with different contribution levels and risk levels in corresponding time periods, and outputs a smart city air condition monitoring log, including: Based on the single-source contribution intensity information and combined with the weather forecast information, through dynamic limit planning, the emission intensity upper limit sequence that different emission sources should follow in order to ensure air quality targets in the predicted future period is analyzed as a time-varying sequence, and emission intensity limit time series information is obtained. Based on the emission intensity limit time series information and combined with the panoramic pollution source map, instruction entries are generated for each emission source through instruction generation and mapping to obtain the dynamic differentiated collaborative control instruction set. Based on the aforementioned dynamic differentiated collaborative control instruction set, the monitoring logs are integrated to record pollution simulation results, control instructions, and expected performance evaluations within the predicted time period in chronological order, thereby obtaining and outputting the smart city air condition monitoring logs.

7. A smart city air quality monitoring system based on multi-source data fusion, characterized in that, The method applied to any one of claims 1-6 includes: The coupling analysis module is used to acquire multi-source pollutant emission data and weather forecast information. Based on the multi-source pollutant emission data and the weather forecast information, it analyzes the process of ammonia and sulfur dioxide coupling to generate secondary particulate matter under different meteorological scenarios, and obtains a multi-pollutant coupling information set. The source contribution module is used to simulate and analyze the contribution characteristics of each ammonia emission source and each sulfur dioxide emission source to the secondary particulate matter concentration in the polluted area based on the multi-pollutant coupling information set, and obtain source contribution information. The dynamic programming module is used to generate a dynamic differentiated collaborative management and control instruction set based on the source contribution information, and to dynamically plan the emission intensity limit time sequence of pollution sources with different contribution levels and risk levels in the corresponding time period, and output the smart city air status monitoring log.

Images