An ozone light pollution prevention and good decision-making method based on multi-source data dynamic research and judgment

Abstract

This invention relates to the fields of air pollution control data processing, dynamic air quality assessment, and environmental decision support, and particularly to a method for decision-making on maintaining good ozone quality during mild pollution based on dynamic assessment of multi-source data. The method receives ozone quality maintenance task data, air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, and historical quality maintenance case data. It generates basic data for quality maintenance assessment according to pollution process numbers; extracts wind direction transport, ozone and precursor temporal and historical distribution deviation characteristics based on comprehensive monitoring maps to generate ozone trend assessment data; then matches intervention measure templates to form quality maintenance intervention task data; and performs in-process tracking, task version updates, and effectiveness evaluation based on real-time monitoring updates and feedback on measure implementation, writing the results into a quality maintenance case database. This invention effectively improves the data connectivity, task matching, and feedback loop of the ozone critical transition process.

Description

Technical Field

[0001] This invention relates to the fields of atmospheric pollution control data processing, dynamic air quality assessment, and environmental decision support, and particularly to a method for maintaining good air quality in cases of mild ozone pollution based on dynamic assessment of multi-source data. Background Technology

[0002] In the field of air pollution control data processing and dynamic air quality assessment, existing solutions for ozone pollution processes typically rely on data from ambient air quality monitoring stations, ozone and its precursor observation data, meteorological data, numerical air quality forecast data, pollutant transport data, and pollution source inventory data for monitoring and early warning, source apportionment, key area identification, measure library matching, and scenario simulation assessment. While these solutions can monitor ozone concentrations, track precursor changes, identify emission areas, and record control measures under conventional ozone pollution control scenarios, they are prone to limitations during the critical transition between good and lightly polluted ozone levels. These limitations include inconsistencies in the spatiotemporal benchmarks of multi-source data, difficulties in the structured transmission of chart analysis results, and insufficient coordination between intervention tasks and the implementation of measures.

[0003] Existing solutions largely rely on numerical air quality forecasts, fixed emission source inventories, pre-set prevention and control measure databases, or manual consultation results for processing. When ozone concentration, meteorological conditions, dominant transport direction, ozone precursor concentration, and the implementation status of measures in the target area change continuously within the same pollution process, there is a lack of stable connection between the monitoring and judgment stage and the intervention task generation stage. This can easily lead to discontinuities in the correspondence between key impact periods, key impact areas, and precursor control types. It can also make it difficult to write back real-time monitoring updates and implementation feedback data into the same pollution process record during implementation.

[0004] Regarding the joint processing of ozone quality maintenance task data, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, and measure implementation feedback data, existing technologies still have common shortcomings in data correlation, critical transition determination, task version recording, and case database writing. Therefore, it is necessary to address the issue of forming ozone quality maintenance intervention tasks and writing back to the ozone quality maintenance case database based on multi-source data, spectral analysis characteristics, and measure implementation feedback, focusing on the critical transition process between good and light pollution ozone levels. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for making decisions on maintaining good air quality in cases of mild ozone pollution based on dynamic analysis of multi-source data, comprising:

[0006] S100 receives ozone protection task data, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, and historical protection case data. It unifies the time granularity, binds spatial units, and associates monitoring objects according to the pollution process number to generate basic data for protection assessment.

[0007] S200. Based on the Baoliang assessment data, generate a comprehensive monitoring map, extract wind direction transmission characteristics, ozone and precursor time-series characteristics and historical distribution deviation characteristics from the comprehensive monitoring map, and generate map assessment feature data.

[0008] S300. Based on the spectral analysis feature data and regional meteorological transmission rules, determine the critical transition between good and light pollution levels, and generate ozone trend analysis data including key impact periods, dominant transmission directions, key impact areas, and precursor control types.

[0009] S400. Based on the ozone trend analysis data and the pollution source inventory data, match the intervention measure template to generate a nutrient protection intervention task data including regional identifier, source type identifier, execution period, measure type and task version identifier;

[0010] S500: Based on the ozone pollution prevention and control intervention task data, receive real-time monitoring update data and measure implementation feedback data, generate in-process tracking data and update the task version identifier; after the ozone pollution process ends, generate light-to-good effect evaluation data, and write the spectrum analysis feature data, the ozone pollution prevention and control intervention task data, the measure implementation feedback data and the light-to-good effect evaluation data into the ozone pollution prevention and control case library according to the pollution process number.

[0011] Furthermore, in S100, the unification of time granularity includes writing the air quality monitoring data of the target area, the ozone and precursor concentration data, the meteorological data, and the pollutant transport data into a unified time period index; the binding of spatial units includes writing the target station, upwind station, pollution source control unit, and administrative grid into the spatial unit identifier; the association of monitoring objects includes writing the spatial unit identifier, the pollution source list data, and the historical pollution control case data into the pollution process number.

[0012] Furthermore, in S200, the comprehensive monitoring map includes a wind direction and speed distribution map, a time series map of ozone and precursor concentrations, a historical distribution map, a station comparison map, and a transmission direction map; the wind direction and speed distribution map is associated with the meteorological data, the ozone and precursor concentration time series map is associated with the ozone and precursor concentration data, and the historical distribution map is associated with the historical Baoliang case data.

[0013] Furthermore, in S200, the wind direction transmission characteristics include the dominant wind direction, transmission sector, and low wind speed static stability marker; the ozone and precursor time series characteristics include the ozone rise slope, peak prediction period, VOCs concentration changes, and NOx concentration changes; the historical distribution deviation characteristics include the degree of deviation of the current ozone concentration from the historical distribution range of the same meteorological pattern.

[0014] Furthermore, in S300, the critical transition determination between the good pollution level and the light pollution level includes: generating a critical transition indicator based on ozone hourly concentration, ozone 8-hour moving average trend, ozone rise slope, temperature data, humidity data, shortwave radiation data, wind speed data, wind direction data, VOCs concentration data, NOx concentration data, and historical good pollution case data; and writing the critical transition indicator into the ozone trend analysis data.

[0015] Furthermore, in S300, the key impact period is generated based on the ozone rise slope and the peak prediction period; the dominant transport direction is generated based on the dominant wind direction, the transport sector, and the pollutant transport data; the key impact area is generated based on the dominant transport direction, upwind station data, and the pollution source inventory data; and the precursor control type is generated based on the VOCs concentration change, the NOx concentration change, and the historical distribution deviation characteristics.

[0016] Further, in S400, the matching intervention measure template includes: filtering the pollution source list data according to the key impact area to generate key source category candidate data; retrieving the intervention measure template according to the key impact period, the precursor control type, and the key source category candidate data to generate measure template matching data; and associating the measure template matching data with the area identifier, the source category identifier, and the execution period to generate the nutrient protection intervention task data.

[0017] Furthermore, the Baoliang intervention task data also includes task number, trigger graph features, feedback deadline, and verification field; the verification field includes execution status, execution time, on-site record identifier, reason for non-execution, and review status; the task version identifier is associated with the task number, the measure type, and the feedback deadline.

[0018] Furthermore, in S500, the real-time tracking data includes real-time changes in ozone concentration, real-time changes in ozone precursor concentration, real-time meteorological changes, and the completeness of measure implementation. When the real-time tracking data shows that the ozone rise slope has not decreased, the peak prediction period has advanced, the dominant transmission direction has changed, or the completeness of measure implementation is abnormal, the ozone trend analysis data is regenerated based on the real-time tracking data, and the Baoliang intervention task data and the task version identifier are updated based on the regenerated ozone trend analysis data.

[0019] Furthermore, in S500, the assessment data for the effectiveness of the transition from light to good ozone includes ozone assessment level results, ozone peak changes, ozone 8-hour moving average changes, the completeness of measure implementation, the transition label, and the applicable boundaries; the case library for maintaining good ozone status includes meteorological type, dominant transmission direction, key impact period, precursor control type, intervention task data for maintaining good ozone status, feedback data on measure implementation, assessment data for the effectiveness of the transition from light to good ozone status, and weighted data for measure recommendations; in the case library for maintaining good ozone status, the meteorological type, the dominant transmission direction, the precursor control type, and the transition label are associated with the pollution process number.

[0020] The key innovations of this invention include:

[0021] (1) Using the pollution process number as an index, the ozone protection task data, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data and historical protection case data are unified in time granularity, bound in spatial units and associated with monitoring objects to form the basic data for protection assessment that can be continuously called for subsequent map generation and trend analysis.

[0022] (2) Generate a comprehensive monitoring map based on the basic data for the assessment of good air quality, and extract wind direction transmission characteristics, ozone and precursor time sequence characteristics and historical distribution deviation characteristics from the comprehensive monitoring map, and convert the monitoring map into map assessment feature data for the critical transformation between good air quality and light pollution level.

[0023] (3) Generate ozone trend analysis data based on the spectral analysis feature data and regional meteorological transmission rules, and match the ozone trend analysis data with the pollution source inventory data and intervention measure templates to form good air quality protection intervention task data containing regional identifiers, source type identifiers, execution time periods, measure types and task version identifiers; generate in-process tracking data based on real-time monitoring update data and measure implementation feedback data during the execution period, and write the light-to-good air quality effect assessment data into the good air quality protection case library according to the pollution process number.

[0024] The following are its main beneficial effects:

[0025] (1) To address the issues of inconsistent spatiotemporal benchmarks and discontinuous correlation of monitoring objects in multi-source data, the basic data for air quality monitoring and ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data and historical air quality monitoring case data are organized by using pollution process numbering, unified time period index and spatial unit identification to form a continuous data link under the same pollution process.

[0026] (2) To address the problem that existing chart analysis results are difficult to transmit in a structured manner, wind direction transmission characteristics, ozone and precursor time series characteristics and historical distribution deviation characteristics are extracted from the comprehensive monitoring map, so that the dominant transmission direction, ozone concentration change, precursor change and historical distribution deviation status can be formed into a structured input that can be called for ozone trend analysis data.

[0027] (3) To address the problem of insufficient coordination between the implementation of intervention tasks and measures, ozone trend analysis data is matched with pollution source inventory data and intervention measure templates to generate good air quality intervention task data. Based on real-time monitoring update data and measure implementation feedback data, the task version identifier is updated and the good air quality case library is written back, so that key impact periods, key impact areas, precursor control types, task execution status and light-to-good air quality effectiveness assessment results are recorded in a closed loop under the same pollution process number. Attached Figure Description

[0028] Figure 1 A flowchart illustrating a method for maintaining good air quality in cases of mild ozone pollution based on dynamic analysis of multi-source data, provided in this application embodiment;

[0029] Figure 2 This is a structural block diagram of a method for making decisions on maintaining good air quality in cases of mild ozone pollution based on dynamic analysis of multi-source data, provided in an embodiment of this application. Detailed Implementation

[0030] Example 1: Refer to Figure 1 This is a flowchart illustrating a method for making decisions on maintaining good air quality in cases of mild ozone pollution based on dynamic analysis of multi-source data, provided by an embodiment of the present invention. The process may include at least steps S100-S500:

[0031] S100 receives ozone protection task data, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, and historical protection case data. It unifies the time granularity, binds spatial units, and associates monitoring objects according to the pollution process number to generate basic data for protection assessment.

[0032] S200. Based on the Baoliang assessment data, generate a comprehensive monitoring map, extract wind direction transmission characteristics, ozone and precursor time-series characteristics and historical distribution deviation characteristics from the comprehensive monitoring map, and generate map assessment feature data.

[0033] S300. Based on the spectral analysis feature data and regional meteorological transmission rules, determine the critical transition between good and light pollution levels, and generate ozone trend analysis data including key impact periods, dominant transmission directions, key impact areas, and precursor control types.

[0034] S400. Based on the ozone trend analysis data and the pollution source inventory data, match the intervention measure template to generate a nutrient protection intervention task data including regional identifier, source type identifier, execution period, measure type and task version identifier;

[0035] S500: Based on the ozone pollution prevention and control intervention task data, receive real-time monitoring update data and measure implementation feedback data, generate in-process tracking data and update the task version identifier; after the ozone pollution process ends, generate light-to-good effect evaluation data, and write the spectrum analysis feature data, the ozone pollution prevention and control intervention task data, the measure implementation feedback data and the light-to-good effect evaluation data into the ozone pollution prevention and control case library according to the pollution process number.

[0036] S100 receives ozone protection mission data, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, and historical protection case data. It then unifies the time granularity, binds spatial units, and associates monitoring objects according to the pollution process number to generate basic data for protection assessment.

[0037] In this embodiment, S100 is executed by the data access and alignment module in the air quality dynamic assessment server. The data access and alignment module establishes data interfaces with ambient air quality standard monitoring stations, meteorological stations, volatile organic compound monitoring stations, air quality numerical forecasting systems, pollutant transport analysis systems, pollution source inventory databases, and ozone protection case databases. After receiving ozone protection task data, the data access and alignment module assigns a pollution process number to the current ozone pollution process and writes all subsequently accessed monitoring data, meteorological data, pollution source inventory data, and historical ozone protection case data under that pollution process number.

[0038] The ozone quality maintenance task data includes target area identifiers, target station identifiers, task trigger time periods, evaluation indicator types, and initial task version identifiers. The target area air quality monitoring data includes hourly ozone concentration, 8-hour moving average ozone concentration, PM2.5 concentration, and PM2.5 concentration. 10The data includes concentrations of NO2, CO, and SO2. Ozone and precursor concentration data include ozone concentration data, VOCs (volatile organic compounds) concentration data, NOx (nitrogen oxides) concentration data, and precursor monitoring time. Meteorological data includes wind speed, wind direction, temperature, humidity, shortwave radiation, air pressure, and precipitation markers. Pollutant transport data includes upwind station data, regional transport data, transport direction field, and transport time field. The pollution source inventory data includes pollution source identifier, source type identifier, emission area, emission industry, spatial location, activity level field, and manageable time period field. Historical good-quality maintenance case data includes meteorological data, dominant transport direction, key impact periods, precursor control type, good-quality maintenance intervention task data, measure implementation feedback data, mild-to-good-quality effect evaluation data, and measure recommendation weight data.

[0039] The S100 system activates upon receiving ozone quality control data. Specifically, the data access and alignment module reads the target area identifier and target site identifier, generating a target site object based on the spatial location of the target site. Simultaneously, it reads the upwind site object based on wind direction and pollutant transport data. Then, it reads the pollution source control units and administrative grids within the target site's influence area based on the pollution source inventory data. For field names of the same monitoring object in different systems, the data access and alignment module calls the field mapping table to perform field mapping, merging "Site Number," "Monitoring Point Number," and "Air Quality Site Identifier" into the target site identifier field; merging "Enterprise Number," "Source Inventory Number," and "Emission Source Number" into the pollution source identifier field; and merging "Timestamp," "Monitoring Time," and "Forecast Period" into a unified time period field.

[0040] The unified time granularity includes writing target area air quality monitoring data, ozone and precursor concentration data, meteorological data, and pollutant transmission data into a unified time period index. Specifically, the data access and alignment module takes the task triggering time period in the ozone protection task data as the starting point and establishes multiple continuous unified time period indexes according to the preset time granularity. When the target area air quality monitoring data uses hourly granularity, VOCs concentration data uses minute granularity, and meteorological data uses hourly granularity, the data access and alignment module categorizes various types of data into the corresponding unified time period index according to the collection time. For cases where multiple VOCs concentration values ​​exist within the same unified time period index, the average concentration, maximum concentration, and number of collections are recorded. For cases where precursor concentration data is not received within a certain unified time period index, a data missing marker is written, and the ozone concentration data and meteorological data for that period are retained.

[0041] The spatial unit binding includes writing the target station, upwind station, pollution source control unit, and administrative grid into the spatial unit identifier. Specifically, the data access and alignment module performs spatial mapping based on the latitude and longitude of the target station, the latitude and longitude of the upwind station, the spatial location of the pollution source, and the administrative grid boundary. When the pollution source is located within the influence range of the target station or the coverage area of ​​the dominant transmission sector, the pollution source is written into the corresponding spatial unit identifier. When the pollution source spans multiple administrative grids, the spatial unit identifier is written according to the main grid corresponding to the spatial location of the pollution source, and the identifiers of adjacent administrative grids are recorded in the extended field. The monitoring object association includes writing the spatial unit identifier, pollution source list data, and historical pollution control case data into the pollution process number. Specifically, the data access and alignment module establishes an index relationship between the target station, upwind station, pollution source control unit, administrative grid, and historical case number under the same pollution process number, forming a monitoring object association record that can be called later.

[0042] In one implementation, when the data returned by the external interface has missing fields, the data access and alignment module writes a data missing marker to the missing fields; when the ozone concentration data of the target station is missing but the ozone concentration data of the upwind station is complete, the data access and alignment module records the ozone concentration data of the upwind station as a substitute input and writes the upwind station identifier in the substitute source field; when there is a sudden change in wind direction in the meteorological data but the meteorological station status is abnormal during the same period, the data access and alignment module writes the period into a meteorological anomaly marker; when the spatial location field of the pollution source inventory data cannot match the administrative grid, the corresponding pollution source is written into the list to be reviewed and the list to be reviewed is associated with the pollution process number.

[0043] In an engineering implementation scenario, the target site within the target area receives ozone quality control task data in the morning. The data access and alignment module reads the target site's air quality monitoring data for the past 24 hours, VOCs and NOx concentration data for the same time period, wind speed and direction data from the meteorological station, and ozone concentration data from the upwind station. Simultaneously, it reads the pollution source inventory data for VOCs-related, mobile, and industrial sources within the target area from the pollution source inventory database. Furthermore, it reads historical ozone quality control case data matching the current meteorological pattern and month from the ozone quality control case database. The data access and alignment module writes the above data into the same pollution process number, forming basic data for ozone quality control analysis that includes a unified time period index, spatial unit identifier, monitoring object association records, data missing markers, and anomaly markers.

[0044] The basic data for air quality protection assessment includes pollution process numbers, unified time period indexes, spatial unit identifiers, target station objects, upwind station objects, air quality monitoring data for the target area, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, historical air quality protection case data, data missing markers, and anomaly markers. This basic data is written into the assessment database by the data access and alignment module and serves as input for S200 to generate comprehensive monitoring maps and extract map assessment feature data. Specifically, the unified time period index is used by S200 to generate time-series maps of ozone and precursor concentrations, the spatial unit identifiers are used by S200 to generate station comparison maps and transport direction maps, and the historical air quality protection case data is used by S200 to generate historical distribution maps.

[0045] S200. Based on the Baoliang assessment data, a comprehensive monitoring map is generated. Wind direction transmission characteristics, ozone and precursor time series characteristics, and historical distribution deviation characteristics are extracted from the comprehensive monitoring map to generate map assessment feature data.

[0046] S200 is executed by the integrated monitoring map generation module and the map analysis feature extraction module. The integrated monitoring map generation module receives the basic data for air quality protection analysis generated in S100, and reads from it the unified time period index, spatial unit identifier, target station object, upwind station object, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, and historical air quality protection case data; the map analysis feature extraction module receives the integrated monitoring map output by the integrated monitoring map generation module, and generates map analysis feature data according to the preset map feature extraction rules.

[0047] In this step, the comprehensive monitoring map refers to a set of multiple monitoring maps bound to the pollution process number. These include wind direction and speed distribution maps, ozone and precursor concentration time-series maps, historical distribution maps, station comparison maps, and transport direction maps. The wind direction and speed distribution maps are associated with meteorological data to record the wind speed distribution and frequency distribution within different wind direction sectors. The ozone and precursor concentration time-series maps are associated with ozone and precursor concentration data to record the changes in ozone concentration, VOCs concentration, and NOx concentration over a unified time period index. The historical distribution maps are associated with historical Baoliang case data to record historical ozone concentration distribution intervals for the same meteorological type. The station comparison maps are associated with the target station and upwind station to record the concentration difference between the target station and the upwind station under the same unified time period index. The transport direction maps are associated with pollutant transport data and spatial unit identifiers to record the mapping relationship between transport direction, transport sector, and corresponding spatial unit.

[0048] S200 is activated after the basic data for the Baoliang assessment is written into the assessment database. Specifically, the integrated monitoring map generation module first reads the basic data for the Baoliang assessment according to the pollution process number, and arranges the air quality monitoring data, ozone and precursor concentration data, and meteorological data according to a unified time period index; then it reads the target stations, upwind stations, pollution source control units, and administrative grids according to the spatial unit identifiers; subsequently, it generates wind direction and speed distribution maps, ozone and precursor concentration time series maps, historical distribution maps, station comparison maps, and transmission direction maps. When generating wind direction and speed distribution maps, the integrated monitoring map generation module divides the wind direction data into multiple wind direction sectors and writes the wind speed data corresponding to each wind direction sector into the sector wind speed sequence; when generating ozone and precursor concentration time series maps, the hourly ozone concentration, 8-hour moving average ozone concentration, VOCs concentration data and NOx concentration data are written into the same time series object according to a unified time period index; when generating historical distribution maps, historical Baoliang case data are read according to the current month, meteorological type and pollution process number, and historical distribution intervals with the same meteorological type are formed.

[0049] The map analysis and feature extraction module extracts wind direction and transmission features from the comprehensive monitoring map. Specifically, the module reads the sector wind speed sequence from the wind direction and speed distribution map and the transmission direction field from the transmission direction map. It writes the wind direction sector with the highest frequency of occurrence and consistent with the direction of pollutant transmission data into the dominant wind direction field; it writes the angular range of the dominant wind direction and its adjacent angular ranges into the transmission sector field; and when the wind speed data corresponding to the dominant wind direction is lower than the wind speed range corresponding to a preset low wind speed rule within multiple consecutive unified time period indices, it writes a low wind speed stable marker. If the wind direction and speed distribution map contains meteorological anomaly markers, the module retains the original value of the wind direction data for that time period and separates that time period from the dominant wind direction statistics.

[0050] The spectral analysis and feature extraction module extracts ozone and precursor time-series features from the ozone and precursor concentration time-series spectrum. Specifically, the module reads the change in ozone concentration data between adjacent unified time-series indices to generate an ozone rise slope field; it reads the 8-hour moving average trend of ozone and combines it with the rising segment in the time-series spectrum to generate a peak prediction time period field; it reads the change direction of VOCs concentration data and NOx concentration data under the same unified time-series index to generate VOCs concentration change fields and NOx concentration change fields, respectively. The precursor concentration phase difference can be generated in one embodiment by the time difference between the peak time period of VOCs concentration change, the peak time period of NOx concentration change, and the ozone concentration rise period, and written into the ozone and precursor time-series features.

[0051] The map analysis and feature extraction module extracts historical distribution deviation features from historical distribution maps. Specifically, the module reads ozone concentration data for the current pollution process within the target unified time period index and reads historical distribution intervals with the same meteorological pattern. It then performs interval matching between the current ozone concentration and the historical distribution intervals with the same meteorological pattern to generate the degree of deviation of the current ozone concentration relative to the historical distribution intervals with the same meteorological pattern. If multiple cases matching the current meteorological pattern exist in the historical data of pollution control, the module filters the cases according to meteorological pattern, dominant transmission direction, and month. If no historical cases with the same meteorological pattern are matched, the module uses the historical distribution map of the same month as a substitute input and writes a case substitution marker.

[0052] In an engineering implementation scenario, after receiving the baseline data for pollution control analysis generated by S100, the S200 system uses a comprehensive monitoring map generation module to generate a wind direction and speed distribution map of the current pollution process. This map shows that the prevailing wind direction in the target area is southwest during the morning period. The ozone and precursor concentration time-series map shows that ozone concentration rises continuously within multiple unified time-series indices, while VOCs concentration changes show high values ​​in the initial stage of ozone rise. The historical distribution map shows that the current ozone concentration is higher than the median range of historical distributions with similar meteorological patterns. The station comparison map shows that the concentration at upwind stations rises before that at the target station. Based on this, the map analysis feature extraction module generates the prevailing wind direction, transmission sector, ozone rise slope, peak prediction period, VOCs concentration changes, NOx concentration changes, and historical distribution deviation characteristics.

[0053] The spectral analysis feature data includes pollution process number, unified time period index, spatial unit identifier, wind direction transmission characteristics, ozone and precursor temporal characteristics, historical distribution deviation characteristics, spectral source identifier, and anomaly marker. This spectral analysis feature data is written into the analysis database and serves as input for determining the critical transition between good and light pollution levels in S300. Specifically, wind direction transmission characteristics are used to generate the dominant transmission direction in S300, ozone and precursor temporal characteristics are used to generate key impact periods and precursor control types in S300, and historical distribution deviation characteristics are used to generate critical transition markers in S300.

[0054] S300. Based on the spectral analysis feature data and regional meteorological transmission rules, determine the critical transition between good and light pollution levels, and generate ozone trend analysis data including key impact periods, dominant transmission directions, key impact areas, and precursor control types.

[0055] S300 is executed by the critical trend analysis module. This module receives the spectral analysis feature data generated by S200 and calls upon regional meteorological transmission rules, air quality standard rules, historical good-quality case data, and pollutant transmission data to form the input data required for determining the critical transition between good and light pollution levels. The regional meteorological transmission rules include the mapping relationship between the prevailing wind direction and the upwind area, the mapping relationship between transmission sectors and spatial unit identifiers, and matching rules between temperature data, humidity data, shortwave radiation data, wind speed data, and wind direction data with ozone pollution processes.

[0056] In this step, the critical transition determination between the "good" and "lightly polluted" levels refers to the critical trend analysis module generating a critical transition identifier based on hourly ozone concentration, 8-hour moving average ozone trend, ozone rise slope, temperature data, humidity data, shortwave radiation data, wind speed data, wind direction data, VOCs concentration data, NOx concentration data, and historical "good" pollution case data. This critical transition identifier is then written into the ozone trend analysis data. The critical transition identifier includes a critical state field, a critical time period field, and a critical basis field. The critical basis field records the ozone rise slope, peak prediction period, historical distribution deviation characteristics, meteorological data, and precursor concentration changes involved in the determination.

[0057] S300 starts after the spectral analysis feature data is written into the analysis database. Specifically, the critical trend analysis module first reads the ozone rise slope, peak prediction period, and historical distribution deviation characteristics from the spectral analysis feature data, and then reads the ozone hourly concentration and ozone 8-hour moving average trend from the good air quality analysis base data. Subsequently, it calls the air quality standard rules to match the evaluation level intervals corresponding to the ozone 8-hour moving average trend. When the level results corresponding to the ozone hourly concentration and the ozone 8-hour moving average trend are within the adjacent range of the good level and the light pollution level, and the ozone rise slope, shortwave radiation data, temperature data, and low wind speed static stability markers all meet the critical judgment rules, a critical transformation marker is generated. If there is a missing data marker for the ozone hourly concentration, the critical trend analysis module reads the upwind station data within the same spatial unit as a substitute input and records the substitute source in the critical basis field.

[0058] The critical trend analysis module generates key impact periods based on the temporal characteristics of ozone and its precursors. Specifically, the module reads the ozone rise slope and the peak prediction period, and writes the index of unified periods where the ozone rise slope is continuously positive into the rise segment set; then, it writes the index of unified periods before the peak prediction period that match the phase difference with the precursor concentration change into the key impact periods. If the peak prediction period changes, the critical trend analysis module updates the key impact periods, and retains both the original and updated key impact periods in the ozone trend analysis data.

[0059] The critical trend analysis module generates the dominant transmission direction based on wind direction transmission characteristics and pollutant transmission data. Specifically, the module reads the transmission direction field from the dominant wind direction, transmission sector, low-wind-speed static stability marker, and pollutant transmission data, and writes the direction that is consistent with the dominant wind direction and corresponds to the concentration change at the upwind station into the dominant transmission direction. When the transmission direction map shows that the ozone concentration at the upwind station increases before that at the target station, the spatial unit identifier corresponding to that upwind station is written into the source field of the dominant transmission direction. If the wind direction data shows a significant shift in the index of adjacent unified time periods, the critical trend analysis module triggers a recalculation of the transmission sector and writes the recalculation result into the version field of the dominant transmission direction.

[0060] The critical trend analysis module generates key impact areas based on the dominant transmission direction, upwind station data, and pollution source inventory data. Specifically, the critical trend analysis module reads pollution source control units located within the transmission sector from the pollution source inventory data according to the dominant transmission direction and performs spatial unit matching with the upwind station data. When a spatial unit simultaneously exhibits increased concentrations at upwind stations, contains controllable source types in the pollution source inventory data, and is located within the coverage area of ​​the dominant transmission direction, the spatial unit is written into the key impact area field. For pollution sources spanning multiple spatial units, the critical trend analysis module reads the adjacent administrative grid identifiers recorded in S100, writes the main spatial unit into the key impact area, and writes adjacent spatial units into the associated area field.

[0061] The critical trend analysis module generates precursor control types based on changes in VOCs concentration, NOx concentration, and historical distribution deviation characteristics. Specifically, the module reads the direction of VOCs and NOx concentration changes during key impact periods and retrieves precursor control types corresponding to the same meteorological pattern and dominant transmission direction from historical data on ozone source control. When the current VOCs concentration change matches a historical VOCs control type record, the precursor control type is written into the VOCs control type. When both VOCs and NOx concentration changes match historical collaborative control records, the precursor control type is written into the NOx collaborative control type. When historical distribution deviation characteristics show that the current ozone concentration deviation mainly occurs at upwind stations, the transmission-dominant type is written into a supplementary field of the precursor control type. In another implementation, the precursor control type can also be generated based on regional meteorological transmission rules and source class identifiers in the pollution source inventory data.

[0062] In an engineering implementation scenario, after receiving the spectral analysis feature data generated by S200, the critical trend analysis module reads the current 8-hour moving average trend of ozone and the ozone rise slope. The spectral analysis feature data records that the prevailing wind direction is southwest, the transmission sector covers multiple upwind spatial units, VOCs concentration changes increase before the ozone rise, and historical distribution deviation features show that the current ozone concentration is in the high range of the historical meteorological distribution range. Based on the above inputs, the critical trend analysis module generates a critical transformation indicator and generates key impact periods, dominant transmission directions, key impact areas, and precursor control types, forming ozone trend analysis data.

[0063] The ozone trend assessment data includes pollution process number, critical transformation indicator, key impact period, dominant transport direction, key impact area, precursor control type, critical basis field, dominant transport direction version field, and anomaly marker. This ozone trend assessment data is written into the assessment database and serves as input for S400 matching intervention measure templates and generating hygienic environment protection intervention task data. Specifically, the key impact period is used to limit the execution period, the key impact area is used to screen pollution source inventory data, the precursor control type is used to retrieve intervention measure templates, and the dominant transport direction is used to determine the matching order between spatial units and source types.

[0064] S400. Based on the ozone trend analysis data and the pollution source inventory data, match the intervention measure template to generate Baoliang intervention task data including regional identifier, source type identifier, execution period, measure type and task version identifier.

[0065] S400 is executed by the intervention task generation module. This module receives the ozone trend analysis data generated by S300 and reads the pollution source inventory data written under the pollution process number in S100, as well as the externally configured intervention measure template. The intervention measure template includes configuration fields for precursor control type, source type identification, measure type, execution period rule, feedback deadline rule, and verification. The pollution source inventory data includes fields for pollution source identification, source type identification, spatial location, emission area, emission industry, controllable period, and activity level.

[0066] S400 is activated after ozone trend assessment data is written into the assessment database. Specifically, the intervention task generation module first reads the key impact areas and dominant transmission directions from the ozone trend assessment data, and then filters the pollution source list data according to the key impact areas. During the filtering process, the intervention task generation module matches the spatial location of the pollution sources with the spatial unit identifiers corresponding to the key impact areas, and writes the pollution sources located within the key impact areas into the key source category candidate data. For pollution sources located within the coverage area of ​​the dominant transmission direction but not within the boundary of the key impact area, the intervention task generation module writes them into the candidate supplementary fields according to the adjacent spatial unit identifiers. The key source category candidate data includes the fields of pollution process number, spatial unit identifier, pollution source identifier, source category identifier, emission area, controllable period, and activity level.

[0067] The intervention task generation module retrieves intervention measure templates based on key impact periods, precursor control types, and candidate data for key source categories, and generates measure template matching data. Specifically, the intervention task generation module first reads the corresponding template group according to the precursor control type; when the precursor control type is VOCs control, it reads the intervention measure templates related to VOCs source categories; when the precursor control type is NOx synergistic control, it reads the intervention measure templates corresponding to NOx source categories and synergistic control source categories; then, it performs time period matching based on the key impact periods and the manageable periods in the pollution source inventory data, and writes the source category identifiers of overlapping periods into the measure template matching data; for cases where there is partial overlap between manageable periods and key impact periods, the intervention task generation module records the overlapping periods and writes the overlapping periods into the execution period field.

[0068] The intervention task generation module associates the measure template matching data with the regional identifier, source type identifier, and execution period to generate Baoliang intervention task data. Specifically, the intervention task generation module reads the regional identifier from the key affected areas, the source type identifier from the key source type candidate data, and the measure type and execution period from the measure template matching data; then, it generates a task number and a task version identifier based on the pollution process number. The task version identifier is associated with the task number, measure type, and feedback deadline; when the same task number is updated due to subsequent in-process tracking data, the task version identifier is used to distinguish between the original task version and the updated task version.

[0069] The Baoliang intervention task data also includes trigger spectrum features, feedback deadline time, and verification fields. The trigger spectrum features are derived from the spectrum analysis feature data generated by S200, including multiple fields from dominant wind direction, ozone rise slope, peak prediction period, VOCs concentration change, NOx concentration change, and historical distribution deviation features. The feedback deadline time is generated by the feedback deadline time rules in the execution period and intervention measure template. The verification fields include execution status, execution time, on-site record identifier, reason for non-execution, and review status. The execution status records the task execution result; the execution time records the time when the feedback data on measure implementation is returned; the on-site record identifier is used to associate on-site feedback records; the reason for non-execution records feedback anomalies; and the review status records whether the feedback data has been reviewed.

[0070] In one implementation, intervention measure templates are version-managed according to source class identifiers. When the template version is inconsistent with the task triggering time of the ozone trend assessment data, the intervention task generation module reads the currently valid template version and writes the original template version into the template version record field. In another implementation, when no source class identifier matching the precursor control type is found in the key source class candidate data, the intervention task generation module generates a template mismatch marker and writes the spatial unit into the pending review record. This pending review record is associated with the pollution process number and is read in S500 as part of the in-process tracking data.

[0071] In an engineering implementation scenario, the ozone trend assessment data records the key impact period from morning to the afternoon peak, with the dominant transport direction pointing to multiple spatial units upwind. The key impact area includes multiple administrative grids southwest of the target site, and the precursor control type is recorded as VOCs control. The intervention task generation module reads the pollution source inventory data, filters out VOCs-related pollution source control units from the key impact area, and matches controllable periods based on the key impact period. Subsequently, it retrieves intervention measure templates and generates ozone protection intervention task data including area identifier, source type identifier, execution period, measure type, task version identifier, feedback deadline, and verification fields.

[0072] The pollution control intervention task data is written into the task database and associated with the pollution control assessment baseline data formed in S100, the spectral assessment feature data formed in S200, and the ozone trend assessment data formed in S300 through the pollution process number. The pollution control intervention task data serves as an index for S500 to receive real-time monitoring update data and measure implementation feedback data; wherein, the task number is used to match the measure implementation feedback data, the task version identifier is used to record the task update process, the feedback deadline is used to generate a feedback timeout flag, and the verification field is used to determine the completeness of measure implementation.

[0073] S500: Based on the ozone pollution prevention and control intervention task data, receive real-time monitoring update data and measure implementation feedback data, generate in-process tracking data and update the task version identifier; after the ozone pollution process ends, generate light-to-good effect evaluation data, and write the spectrum analysis feature data, the ozone pollution prevention and control intervention task data, the measure implementation feedback data and the light-to-good effect evaluation data into the ozone pollution prevention and control case library according to the pollution process number.

[0074] S500 is executed by a mid-process tracking module, a task version update module, an effectiveness evaluation module, and a case library update module. The mid-process tracking module uses the ozone pollution control intervention task data generated in S400 as an index to receive real-time monitoring updates and implementation feedback data. The task version update module updates the task version identifier based on the mid-process tracking data. The effectiveness evaluation module generates a light-to-good air quality effectiveness evaluation data after the ozone pollution process ends. The case library update module writes the spectral analysis feature data, ozone pollution control intervention task data, implementation feedback data, and light-to-good air quality effectiveness evaluation data into the ozone pollution control case library according to the pollution process number.

[0075] S500 starts after the Baoliang intervention task data is written to the task database. Specifically, the in-process tracking module establishes a feedback receiving channel according to the task number, region identifier, source type identifier, and feedback deadline. When the external task feedback system returns implementation feedback data, the in-process tracking module reads the execution status, execution time, on-site record identifier, reason for non-execution, and review status, and writes them into the verification field of the Baoliang intervention task data. When the feedback deadline arrives but no implementation feedback data is received, the in-process tracking module writes a feedback timeout flag and writes the execution status to the non-feedback status. For cases where multiple implementation feedback data are received for the same task number, the in-process tracking module matches them according to the task version identifier and execution time, writes the feedback data consistent with the current task version identifier into the current version record, and writes the other feedback data into the historical feedback record.

[0076] The real-time monitoring and update data includes real-time changes in ozone concentration, real-time changes in ozone precursor concentration, and real-time meteorological changes. The in-process tracking module reads the real-time monitoring and update data according to the pollution process number and compares the fields with the spectral analysis feature data generated by S200; after reading the real-time ozone concentration change, it calculates the ozone rise slope under the current unified time period index; after reading the real-time ozone precursor concentration change, it records the changes in VOCs concentration and NOx concentration; after reading the real-time meteorological changes, it records the changes in wind speed, wind direction, temperature, humidity, and shortwave radiation data. The completeness of the measures' implementation is generated by the in-process tracking module based on the implementation status, implementation time, on-site record identifier, reason for non-implementation, and review status, and is associated with the task number and task version identifier.

[0077] When the ozone rise slope fails to decrease, the peak prediction period is advanced, the dominant transport direction changes, or the completeness of measures is abnormal in the real-time tracking data, the task version update module regenerates ozone trend analysis data based on the real-time tracking data, and updates the Baoliang intervention task data and task version identifier based on the regenerated ozone trend analysis data. Specifically, the task version update module writes the real-time ozone concentration changes, real-time meteorological changes, and completeness of measures implementation from the real-time tracking data into the update field of the spectral analysis feature data; then it calls the critical trend analysis module in S300 to regenerate the key impact period, dominant transport direction, key impact area, and precursor control type; subsequently, it calls the intervention task generation module in S400 to rematch the intervention measure template and generate updated Baoliang intervention task data. The task version update module retains the original task version identifier and the updated task version identifier, and records the version update time, version trigger field, and version source field.

[0078] In one implementation, when the dominant transmission direction changes, the task version update module reads the new dominant transmission direction and transmission sector, re-screens the pollution source list data, and generates new candidate data for key source categories. When the completeness of the measures is abnormal, the task version update module writes an execution anomaly mark into the Baoliang intervention task data and writes the corresponding task number into the review status field. When there is a missing data mark in the real-time monitoring update data, the in-process tracking module calls data from neighboring stations or upwind stations as alternative inputs and writes the alternative source field into the in-process tracking data. When the historical Baoliang case data in the case library is not similar enough to the current pollution process, the case library update module does not call the recommended weight data of the measures to participate in the current version update and retains the complete record of the current pollution process for subsequent case entries.

[0079] After the ozone pollution event concludes, the effectiveness assessment module reads the in-process tracking data and pollution event monitoring data to generate assessment data for the transition from light to good ozone levels. The pollution event monitoring data includes hourly ozone concentrations before and after the event's conclusion, changes in the 8-hour moving average of ozone, changes in ozone precursor concentrations, and changes in meteorological data. Based on the pollution event number, the effectiveness assessment module reads the ozone assessment level result, records the peak ozone level change and the 8-hour moving average of ozone; it then reads the completeness of the measures implemented, generating a transition label and applicable boundaries. The transition label is associated with the ozone assessment level result, the key impact period, and the implementation period; the applicable boundaries are associated with the meteorological type, the dominant transport direction, the precursor control type, and the completeness of the measures implemented.

[0080] The case study update module writes the spectral analysis feature data, good-quality protection intervention task data, measure implementation feedback data, and mild-to-good-quality conversion effectiveness evaluation data into the good-quality protection case study database according to the pollution process number. The good-quality protection case study database includes meteorological type, dominant transmission direction, key impact period, precursor control type, good-quality protection intervention task data, measure implementation feedback data, mild-to-good-quality conversion effectiveness evaluation data, and measure recommendation weight data. In the good-quality protection case study database, meteorological type, dominant transmission direction, precursor control type, and conversion label are associated with the pollution process number. The case study update module retains the original task version record, the updated task version record, and the measure implementation feedback record during writing. When multiple task versions exist for the same pollution process, the case study update module stores the task combinations according to the task version identifier and writes the corresponding measure execution completeness into the version record under the same pollution process number.

[0081] In an engineering implementation scenario, after the pollution control intervention task data is issued, the in-process tracking module receives feedback data on the implementation of measures from multiple pollution source control units within the key area according to the task number; simultaneously, it continuously reads the real-time ozone concentration changes at the target site, the ozone concentration changes at the upwind site, and real-time meteorological changes. If the real-time wind direction changes from southwest to westerly, the task version update module regenerates ozone trend analysis data based on the new dominant transmission direction and calls the intervention task generation module to update the key affected area and pollution control intervention task data; if a source-type task fails to return feedback data on the implementation of measures before the feedback deadline, the in-process tracking module writes a feedback timeout flag and records the completeness of measure implementation as an anomaly. After the pollution process ends, the effectiveness evaluation module reads the ozone assessment level results, ozone peak changes, ozone 8-hour moving average changes, and the completeness of measure implementation, generating light-to-good effectiveness evaluation data; the case library update module writes the spectral analysis characteristics of the pollution process, task version records, measure implementation feedback, and evaluation results into the pollution control case library.

[0082] The real-time tracking data includes pollution process number, task number, task version identifier, real-time ozone concentration change, real-time ozone precursor concentration change, real-time meteorological changes, completeness of measure implementation, feedback timeout flag, and implementation anomaly flag. The data for evaluating the effectiveness of transitioning from light to good ozone includes ozone assessment level results, ozone peak value change, 8-hour moving average ozone change, completeness of measure implementation, transition label, and applicable boundaries. Records in the ozone maintenance case library are read as historical maintenance case data when the next ozone maintenance task data enters S100; simultaneously, the measure recommendation weight data, meteorological type, dominant transmission direction, key impact period, and precursor control type in the maintenance case library can be used as input for critical trend analysis and intervention measure template matching in S300 and S400.

[0083] Example 2: Figure 2This diagram illustrates a structural block diagram of a method for maintaining good air quality in cases of mild ozone pollution based on dynamic analysis of multi-source data, according to an embodiment of the present invention. Figure 2 As shown, the structure may include:

[0084] The data access and alignment module 01 is used to receive ozone quality maintenance task data, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, and historical quality maintenance case data. It performs time granularity unification, spatial unit binding, and monitoring object association according to the pollution process number to generate basic data for quality maintenance assessment. Specifically, the data access and alignment module receives data objects from the ambient air quality monitoring terminal, meteorological data terminal, pollutant transport data terminal, pollution source inventory data terminal, and quality maintenance case database, and reads the target area identifier, target station identifier, and task triggering time period from the ozone quality maintenance task data. The data access and alignment module establishes a data index for the same pollution process based on the pollution process number, and integrates the target area air quality monitoring data, the ozone and precursor concentration data, and historical quality maintenance case data. The meteorological data and the pollutant transport data are written into a unified time period index. The target station, upwind station, pollution source control unit, and administrative grid are written into the spatial unit identifier. The spatial unit identifier, the pollution source list data, and the historical pollution protection case data are written into the pollution process number. During the data access process, if there are missing fields, the data access and alignment module writes a data missing mark. If there is a conflict in the collection time period, the original collection time period is retained and a time period conflict mark is written. If there is a spatial unit that cannot be matched, the corresponding pollution source list data is written into the record to be reviewed. The data access and alignment module outputs the generated pollution protection assessment basic data to the map assessment feature extraction module. The unified time period index, the spatial unit identifier, and the pollution process number are used as input fields for the map assessment feature extraction module to generate a comprehensive monitoring map.

[0085] The map analysis feature extraction module 02, connected to the data access and alignment module, is used to generate a comprehensive monitoring map based on the air quality protection assessment basic data. It extracts wind direction transmission characteristics, ozone and precursor time-series characteristics, and historical distribution deviation characteristics from the comprehensive monitoring map to generate map analysis feature data. Specifically, the map analysis feature extraction module receives the air quality protection assessment basic data from the data access and alignment module and reads the unified time period index, spatial unit identifier, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transmission data, and historical air quality protection case data. The map analysis feature extraction module forms a wind direction and speed distribution map based on the meteorological data, an ozone and precursor concentration time-series map based on the ozone and precursor concentration data, a historical distribution map based on the historical air quality protection case data, a station comparison map based on the target station and upwind station, and a transmission direction map based on the pollutant transmission data. The map analysis feature extraction module extracts wind direction transmission characteristics, ozone and precursor time-series characteristics, and historical distribution deviation characteristics from the comprehensive monitoring map to generate map analysis feature data. The system extracts the dominant wind direction, transmission sector, and low-wind-speed stable markers from the wind speed distribution map and the transmission direction map. It also extracts the ozone rise slope, peak prediction period, volatile organic compound (VOC) concentration changes, and nitrogen oxide (NOx) concentration changes from the ozone and precursor concentration time series map. Furthermore, it extracts the deviation of the current ozone concentration from the historical distribution map relative to the historical meteorological distribution range. When there are missing data markers in the baseline data for maintaining good air quality, the system writes the missing data source into the corresponding map source identifier. When no matching record of the historical good air quality case data is found, the system retains the historical distribution deviation field of the current pollution process as null and writes a case mismatch marker. The system outputs the generated system analysis feature data to the critical trend analysis module. The wind direction transmission characteristics, the ozone and precursor time series characteristics, and the historical distribution deviation characteristics serve as inputs for the critical trend analysis module to determine the critical transition between good and light pollution levels.

[0086] The critical trend analysis module 03, connected to the map analysis feature extraction module, is used to determine the critical transition between good and light pollution levels based on the map analysis feature data and regional meteorological transmission rules, generating ozone trend analysis data including key impact periods, dominant transmission directions, key impact areas, and precursor control types. Specifically, the critical trend analysis module receives map analysis feature data from the map analysis feature extraction module and calls regional meteorological transmission rules to read the mapping relationship between the dominant wind direction and the upwind area, the mapping relationship between the transmission sector and the spatial unit identifier, and the matching rules between meteorological data and ozone pollution processes. The critical trend analysis module generates a critical transition identifier based on ozone hourly concentration, ozone eight-hour moving average trend, ozone rise slope, temperature data, humidity data, shortwave radiation data, wind speed data, wind direction data, volatile organic compound concentration changes, nitrogen oxide concentration changes, and historical good-quality case data, and writes the critical transition identifier into the ozone trend analysis data. The ozone rise slope and the predicted peak period are used to generate key impact periods. Based on the prevailing wind direction, the transmission sector, and the pollutant transmission data, a dominant transmission direction is generated. Based on the dominant transmission direction, upwind station data, and the pollution source inventory data, a key impact area is generated. Based on the changes in volatile organic compound concentration, the changes in nitrogen oxide concentration, and the historical distribution deviation characteristics, a precursor control type is generated. When meteorological anomaly markers are present in the spectral analysis feature data, the critical trend analysis module calls meteorological data from adjacent unified time period indexes to recalculate the transmission sector and records the recalculation source in the ozone trend analysis data. When upwind station data is missing, the critical trend analysis module retains the original result of the dominant transmission direction and writes a station missing measurement marker. The critical trend analysis module outputs the generated ozone trend analysis data to the intervention task generation module. The key impact period, the dominant transmission direction, the key impact area, and the precursor control type serve as input fields for the intervention task generation module to match intervention measure templates.

[0087] Intervention task generation module 04, connected to the critical trend analysis module, is used to match intervention measure templates based on the ozone trend analysis data and the pollution source inventory data, generating hygienic control intervention task data including regional identifier, source type identifier, execution period, measure type, and task version identifier. Specifically, the intervention task generation module receives ozone trend analysis data from the critical trend analysis module and reads the pollution source inventory data and intervention measure templates. The intervention task generation module filters the pollution source inventory data according to the key impact areas, and writes the pollution source control units, administrative grids, and source type identifiers corresponding to the key impact areas into the key source type candidate data. The intervention task generation module retrieves the intervention measure templates according to the key impact period, the precursor control type, and the key source type candidate data, generates measure template matching data, and associates the measure template matching data with the regional identifier, source type identifier, and execution period. The intervention task generation module generates a task number and task version identifier based on the pollution process number, and writes the measure type, trigger map characteristics, feedback deadline, and verification field into the Baoliang intervention task data. The verification field includes execution status, execution time, on-site record identifier, reason for non-execution, and review status. When the intervention measure template does not match the precursor control type, the intervention task generation module writes a template mismatch mark and writes the corresponding spatial unit identifier into the record to be reviewed. When the same source class identifier matches multiple measure types, the intervention task generation module generates the corresponding task version identifier according to the overlap between the key impact period and the controllable period. The intervention task generation module outputs the generated Baoliang intervention task data to the feedback assessment and case library update module. The task number, the task version identifier, the feedback deadline, and the verification field serve as index fields for the feedback assessment and case library update module to receive the measure implementation feedback data.

[0088] The feedback assessment and case library update module 05 is connected to the intervention task generation module. It is used to receive real-time monitoring update data and measure implementation feedback data based on the ozone pollution prevention intervention task data, generate in-process tracking data and update the task version identifier; after the ozone pollution process ends, it generates light-to-good effect assessment data, and writes the spectrum analysis feature data, the ozone pollution prevention intervention task data, the measure implementation feedback data and the light-to-good effect assessment data into the ozone pollution prevention case library according to the pollution process number. Specifically, the feedback assessment and case library update module receives the ozone protection intervention task data from the intervention task generation module, and receives the implementation feedback data based on the task number, the task version identifier, and the feedback deadline. Simultaneously, it reads the real-time ozone concentration changes, ozone precursor concentration changes, and real-time meteorological changes from the real-time monitoring update data. The feedback assessment and case library update module writes the execution status, execution time, on-site record identifier, reason for non-execution, and review status into the verification field, and generates in-process tracking data based on the real-time monitoring update data and the implementation feedback data. When the in-process tracking data shows that the ozone rise slope has not decreased, the peak prediction period has advanced, the dominant transmission direction has changed, or the completeness of the implementation of measures is abnormal, the feedback assessment and case library update module sends the in-process tracking data back to the critical trend analysis module and the intervention task generation module to generate updated ozone trend analysis data and the number of ozone protection intervention tasks. According to the data, the task version identifier is updated; when the feedback deadline is reached and no feedback data on the implementation of measures is received, the feedback assessment and case library update module writes a feedback timeout flag and associates the corresponding task number with the task version identifier; after the ozone pollution process ends, the feedback assessment and case library update module generates light-to-good effectiveness assessment data based on the pollution process monitoring data and the in-process tracking data. The light-to-good effectiveness assessment data includes ozone evaluation level results, ozone peak change, ozone eight-hour moving average change, measure implementation completeness, good-to-good label, and applicable boundaries; the feedback assessment and case library update module writes the spectral analysis feature data, the good-to-good intervention task data, the measure implementation feedback data, and the light-to-good effectiveness assessment data into the good-to-good case library according to the pollution process number, and associates the meteorological type, dominant transmission direction, key impact period, precursor control type, and good-to-good label with the pollution process number in the good-to-good case library.

Claims

1. A method for decision-making on maintaining good air quality in cases of mild ozone pollution based on dynamic analysis of multi-source data, characterized in that, include: S100 receives ozone protection task data, target area air quality monitoring data, ozone and precursor concentration data, meteorological data, pollutant transport data, pollution source inventory data, and historical protection case data. It unifies the time granularity, binds spatial units, and associates monitoring objects according to the pollution process number to generate basic data for protection assessment. S200. Based on the Baoliang assessment data, generate a comprehensive monitoring map, extract wind direction transmission characteristics, ozone and precursor time-series characteristics and historical distribution deviation characteristics from the comprehensive monitoring map, and generate map assessment feature data. S300. Based on the spectral analysis feature data and regional meteorological transmission rules, determine the critical transition between good and light pollution levels, and generate ozone trend analysis data including key impact periods, dominant transmission directions, key impact areas, and precursor control types. S400. Based on the ozone trend analysis data and the pollution source inventory data, match the intervention measure template to generate a nutrient protection intervention task data including regional identifier, source type identifier, execution period, measure type and task version identifier; S500: Based on the ozone pollution prevention and control intervention task data, receive real-time monitoring update data and measure implementation feedback data, generate in-process tracking data and update the task version identifier; after the ozone pollution process ends, generate light-to-good effect evaluation data, and write the spectrum analysis feature data, the ozone pollution prevention and control intervention task data, the measure implementation feedback data and the light-to-good effect evaluation data into the ozone pollution prevention and control case library according to the pollution process number.

2. The method according to claim 1, characterized in that, In S100, the unification of time granularity includes writing the air quality monitoring data of the target area, the ozone and precursor concentration data, the meteorological data, and the pollutant transport data into a unified time period index; the binding of spatial units includes writing the target station, upwind station, pollution source control unit, and administrative grid into the spatial unit identifier; the association of monitoring objects includes writing the spatial unit identifier, the pollution source list data, and the historical Baoliang case data into the pollution process number.

3. The method according to claim 1, characterized in that, In S200, the comprehensive monitoring map includes a wind direction and speed distribution map, a time series map of ozone and precursor concentrations, a historical distribution map, a station comparison map, and a transmission direction map; the wind direction and speed distribution map is associated with the meteorological data, the ozone and precursor concentration time series map is associated with the ozone and precursor concentration data, and the historical distribution map is associated with the historical Baoliang case data.

4. The method according to claim 1, characterized in that, In S200, the wind direction transmission characteristics include the dominant wind direction, transmission sector, and low wind speed static stability marker; the ozone and precursor time series characteristics include the ozone rise slope, peak prediction period, VOCs concentration changes, and NOx concentration changes; the historical distribution deviation characteristics include the degree of deviation of the current ozone concentration from the historical distribution range of the same meteorological pattern.

5. The method according to claim 1, characterized in that, In S300, the critical transition determination between the good pollution level and the light pollution level includes: generating a critical transition indicator based on ozone hourly concentration, ozone 8-hour moving average trend, ozone rise slope, temperature data, humidity data, shortwave radiation data, wind speed data, wind direction data, VOCs concentration data, NOx concentration data, and historical good pollution case data; and writing the critical transition indicator into the ozone trend analysis data.

6. The method according to claim 1, characterized in that, In S300, the key impact period is generated based on the ozone rise slope and the peak prediction period; the dominant transport direction is generated based on the dominant wind direction, the transport sector, and the pollutant transport data; the key impact area is generated based on the dominant transport direction, upwind station data, and the pollution source inventory data; and the precursor control type is generated based on the VOCs concentration change, the NOx concentration change, and the historical distribution deviation characteristics.

7. The method according to claim 1, characterized in that, In S400, the matching intervention measure template includes: filtering the pollution source list data according to the key impact area to generate key source category candidate data; retrieving the intervention measure template according to the key impact period, the precursor control type, and the key source category candidate data to generate measure template matching data; and associating the measure template matching data with the area identifier, the source category identifier, and the execution period to generate the nutrient protection intervention task data.

8. The method according to claim 1, characterized in that, The Baoliang intervention task data also includes task number, trigger graph features, feedback deadline, and verification field; the verification field includes execution status, execution time, on-site record identifier, reason for non-execution, and review status; the task version identifier is associated with the task number, the measure type, and the feedback deadline.

9. The method according to claim 1, characterized in that, In S500, the real-time tracking data includes real-time ozone concentration changes, real-time ozone precursor concentration changes, real-time meteorological changes, and the completeness of measure implementation. When the real-time tracking data shows that the ozone rise slope has not decreased, the peak prediction period has advanced, the dominant transport direction has changed, or the completeness of measure implementation is abnormal, the ozone trend judgment data is regenerated based on the real-time tracking data, and the Baoliang intervention task data and the task version identifier are updated based on the regenerated ozone trend judgment data.

10. The method according to claim 1, characterized in that, In S500, the data for evaluating the effectiveness of light-to-good ozone conversion includes ozone assessment results, ozone peak changes, ozone 8-hour moving average changes, the completeness of measure implementation, the conversion label, and the applicable boundaries; the case library for maintaining good ozone status includes meteorological data, dominant transmission direction, key impact periods, precursor control types, data on intervention tasks for maintaining good ozone status, feedback data on measure implementation, data for evaluating the effectiveness of light-to-good ozone conversion, and data on the weighting of measure recommendations. In the Baoliang case database, the meteorological type, the dominant transport direction, the precursor control type, and the transformation label are associated with the pollution process number.

Images