Urban air quality real-time monitoring system and method based on multi-source data fusion

Abstract

The invention belongs to the technical field of air quality monitoring, and particularly relates to an urban air quality real-time monitoring system and method based on multi-source data fusion. According to the invention, by establishing the comprehensive space grid set combining the macro-scale and the micro-scale, the overall air condition of a city and the block-level fine-grained pollution distribution can be reflected at the same time, the monitoring precision is improved, the grid division is automatically adjusted according to the condition that the pollutant concentration is too high or too low, the matching between the space division and the actual pollutant distribution is realized, and the accuracy of the monitoring is improved. By introducing a spatial distance coefficient and fusing historical pollution data, an obtained pollutant concentration fusion value can better reflect the real pollution state of a region, the influence caused by accidental data fluctuation is reduced, and the change amplitude, the rate and the direction of an air quality feature vector are utilized for analysis, so that the monitoring result is more accurate and reliable. The future air quality trend is deduced, and an advance is provided for environmental governance and early warning.

Description

Technical Field

[0001] This invention belongs to the field of air quality monitoring technology, specifically relating to a real-time urban air quality monitoring system and method based on multi-source data fusion. Background Technology

[0002] With the acceleration of urbanization and industrial development, urban air pollution has become increasingly serious, severely impacting residents' health, urban environmental quality, and sustainable development. Effective monitoring and accurate forecasting of air quality have become crucial for environmental management and decision support. Traditional air quality monitoring methods primarily rely on fixed monitoring stations or single data sources.

[0003] In recent years, with the development of the Internet of Things, big data analytics, remote sensing technology, and artificial intelligence, multi-source data fusion has become an effective means to improve the accuracy and timeliness of air quality monitoring. By integrating data from fixed monitoring stations, mobile sensing devices, and satellite remote sensing, and combining it with meteorological information, macroscopic and microscopic spatial grids can be constructed to achieve refined modeling of air quality.

[0004] However, existing methods still have shortcomings. They are difficult to dynamically adjust the grid structure in real time to adapt to changes in pollutant concentrations, difficult to combine historical data and spatial relationships for pollutant concentration fusion analysis, lack pollution trend prediction and high-concentration area identification mechanisms based on air quality feature vectors, and cannot optimize the scheduling of monitoring equipment and patrol commands in key areas in real time, affecting monitoring efficiency and early warning accuracy. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time urban air quality monitoring system and method based on multi-source data fusion, which can effectively solve the problems of insufficient monitoring coverage, limitations of single data sources, and lack of dynamic control in the existing technology, and provide a reliable technical means for urban air quality management and pollution prevention and control.

[0006] The specific technical solution adopted by this invention is as follows: Real-time urban air quality monitoring methods based on multi-source data fusion include: Acquire air quality monitoring data and geographic meteorological data within the city, and establish a comprehensive spatial grid set based on the air quality monitoring data and geographic meteorological data at both the macro-scale of the city and the micro-scale of the street blocks; Based on pollutant concentration data, the integrated spatial grid set is dynamically adjusted to obtain a multi-scale spatial grid; The pollutant concentration fusion set of the multi-scale spatial grid is obtained based on the spatial location of the multi-scale spatial grid and historical pollution data; Based on the pollutant concentration fusion set and geographic meteorological data, air quality feature vectors of multi-scale spatial grids are generated. Based on the changes in air quality feature vectors, future air quality change trends are predicted, and high-concentration spatial grids are determined. Based on air quality feature vectors, wind direction information, and wind speed information, the upstream spatial region of the high-concentration spatial grid is identified as a key monitoring area; Based on key monitoring areas, generate alarm information and increase the sampling frequency and data reporting frequency of all monitoring devices within the key monitoring areas; Based on the actual monitoring results of key monitored areas, the spatial grid is dynamically adjusted, and alarm information and patrol instructions are updated in real time.

[0007] In a preferred embodiment, air quality monitoring data and geographic meteorological data within the urban area are acquired. Based on the air quality monitoring data and geographic meteorological data, a comprehensive spatial grid set is established at both the macro-scale of the city and the micro-scale of the street blocks; including: Acquire air quality monitoring data from multiple air quality monitoring stations within the city, and acquire geographic meteorological data covering the city area, including wind speed, wind direction, temperature, humidity, topography, and road network information; Based on the collection coordinates of each air quality monitoring point, coordinate transformation is performed on the spatial location corresponding to the air quality monitoring data and geographic meteorological data to obtain multiple collection spatial coordinates, which are then summarized into a collection spatial coordinate reference set. Based on the urban area boundary, a grid construction range is established. Combined with the collected spatial coordinate reference set, a macro-scale spatial grid is constructed for the urban area, and a micro-scale spatial grid is constructed for the block area. The grids are then combined into a comprehensive spatial grid set.

[0008] In a preferred embodiment, the integrated spatial grid set is dynamically adjusted based on pollutant concentration data to obtain a multi-scale spatial grid, including: Obtain pollutant concentration data for each spatial grid in the integrated spatial grid set, and extract the pollutant concentration values; Obtain the range of pollutant concentration values, determine whether the pollutant concentration values ​​are within the range, and make adjustments accordingly; If the pollutant concentration value in each spatial grid is within the pollutant concentration value range, then the pollutant concentration of the spatial grid is determined to meet the requirements. If the pollutant concentration value in a spatial grid is outside the pollutant concentration value range and exceeds the upper limit of the pollutant concentration value range, the pollutant concentration in the spatial grid is determined to be too high. Based on the pollutant concentration exceeding the value of the spatial grid, a corresponding preset spatial grid adjustment strategy is matched to dynamically adjust the comprehensive spatial grid set. If the pollutant concentration value in a spatial grid is not within the pollutant concentration value range and is less than the lower limit of the pollutant concentration value range, the pollutant concentration in the spatial grid is determined to be too low. Based on the lower limit of the pollutant concentration in the spatial grid, a corresponding preset spatial grid adjustment strategy is matched to dynamically adjust the comprehensive spatial grid set. The adjusted spatial grids are then aggregated to obtain a multi-scale spatial grid.

[0009] In a preferred embodiment, a pollutant concentration fusion set of a multi-scale spatial grid is obtained based on the spatial location of the multi-scale spatial grid and historical pollution data, including: Acquire the spatial location, current pollutant concentration data, and historical pollution data of each spatial grid in a multi-scale spatial grid; Obtain the spatial distance between each spatial grid and the corresponding air quality monitoring point; Obtain the baseline spatial distance; The corresponding spatial distance coefficient is obtained based on the spatial distance of each spatial grid and the reference spatial distance; Extract the current pollutant concentration value based on the current pollutant concentration data, and extract multiple corresponding historical pollutant concentration values ​​based on historical pollution data; The pollutant concentration fusion value of each spatial grid in the multi-scale spatial grid is obtained based on the spatial distance coefficient, the current pollutant concentration value, and multiple historical pollutant concentration values; The pollutant concentration fusion set is obtained by summing the pollutant concentration fusion values ​​from multiple spatial grids.

[0010] In a preferred embodiment, air quality feature vectors of a multi-scale spatial grid are generated based on a fused set of pollutant concentrations and geographic meteorological data. Future air quality trends are predicted based on changes in these feature vectors, and high-concentration spatial grids are identified, including: Based on geographic meteorological data, obtain corresponding wind speed, wind direction, temperature, humidity, topography, and road network information; The pollutant concentration fusion set of the multi-scale spatial grid is combined with the corresponding wind speed, wind direction, temperature, humidity, topography and road network information in a preset order to generate the air quality feature vector of each spatial grid in the multi-scale spatial grid. Multiple benchmark monitoring periods are obtained, and the feature vector change information of the air quality feature vector is obtained based on the multiple benchmark monitoring periods. The feature vector change information includes the change amplitude, change rate and change direction. Predict future air quality trends based on feature vector changes, where future air quality trends may include rising, falling, or remaining stable. High-concentration spatial grids are obtained based on future air quality change trends and the fusion value of pollutant concentrations for each spatial grid.

[0011] In a preferred embodiment, based on air quality feature vectors, wind direction information, and wind speed information, the upstream spatial region of the high-concentration spatial grid is identified as a key monitoring area, including: Based on the air quality feature vector of each spatial grid, the dominant wind direction and wind speed parameters of the high-concentration spatial grid in the current monitoring period are obtained; Obtain the prevailing wind direction relationship in the spatial grid, and obtain multiple upstream spatial grid cells in the upstream direction of the high-concentration spatial grid based on the wind direction relationship; Based on the fused pollutant concentration values, air quality feature vectors, and wind speed parameters of multiple upstream spatial grid cells, pollutant propagation paths are identified. Based on the pollutant propagation path, upstream spatial areas that affect high-concentration spatial grids are identified, and these upstream spatial areas are designated as key monitoring areas.

[0012] In a preferred embodiment, alarm information is generated based on key monitored areas, and the sampling frequency and data reporting frequency of all monitoring devices within the key monitored areas are increased, including: Obtain the pollutant concentration fusion value of key monitoring areas, obtain the corresponding pollution risk level based on the pollutant concentration fusion value, and generate the pollutant concentration fusion value; Acquire information on mobile monitoring devices surrounding key monitoring areas and issue patrol instructions to these devices to guide them to concentrate in key monitoring areas. Based on the pollution risk level, obtain the corresponding sampling strategy, and adjust the sampling frequency and data reporting frequency of all monitoring equipment in the key regulatory area according to the sampling strategy.

[0013] In a preferred embodiment, the spatial grid is dynamically adjusted based on actual monitoring results of key monitored areas, the identified pollution source areas and air quality trends are corrected, and alarm information and patrol instructions are updated in real time, including: Obtain the actual monitoring results of key regulatory areas over multiple consecutive monitoring periods, and based on the actual monitoring results, obtain the corresponding spatial grid adjustment strategy and adjust the spatial grid accordingly; Obtain the differences in monitoring values ​​between fixed and mobile monitoring equipment within key regulatory areas; Based on the differences in the monitored values, the corresponding patrol command correction strategy is obtained, and the patrol command of the mobile monitoring device is corrected. The patrol command correction strategy includes adjusting the travel direction, stopping point position and patrol frequency of the patrol path. Based on the revised patrol instructions, new alarm information is generated in real time, and updated patrol instructions are issued to relevant mobile monitoring devices.

[0014] The present invention also provides a real-time urban air quality monitoring system based on multi-source data fusion, used in the above-mentioned real-time urban air quality monitoring method based on multi-source data fusion, comprising: The spatial grid module is used to acquire air quality monitoring data and geographic meteorological data within the city. Based on the air quality monitoring data and geographic meteorological data, a comprehensive spatial grid set is established at both the macro-scale of the city and the micro-scale of the street blocks. The spatial grid adjustment module is used to dynamically adjust the integrated spatial grid set based on pollutant concentration data to obtain a multi-scale spatial grid. The pollutant concentration fusion module is used to obtain a fused set of pollutant concentrations from a multi-scale spatial grid based on the spatial location of the multi-scale spatial grid and historical pollution data. The high-concentration determination module is used to generate air quality feature vectors of multi-scale spatial grids based on pollutant concentration fusion sets and geographic meteorological data. Based on the changes in air quality feature vectors, it predicts future air quality change trends and determines high-concentration spatial grids. The key monitoring module identifies upstream spatial areas of high-concentration spatial grids as key monitoring areas based on air quality feature vectors, wind direction information, and wind speed information. The alarm module is used to generate alarm information based on key monitoring areas and increase the sampling frequency and data reporting frequency of all monitoring devices in key monitoring areas; The correction module is used to dynamically adjust the spatial grid based on the actual monitoring results of key monitored areas, and to update alarm information and patrol instructions in real time.

[0015] And, a real-time urban air quality monitoring terminal based on multi-source data fusion, including: One or more processors; A storage device on which one or more programs are stored; When one or more programs are executed by one or more processors, the one or more processors implement a method for real-time monitoring of urban air quality based on multi-source data fusion.

[0016] The technical effects achieved by this invention are as follows: This invention establishes a comprehensive spatial grid set combining macro and micro scales, simultaneously reflecting the overall air quality of a city and the fine-grained pollution distribution at the street level, thus improving monitoring accuracy. By automatically adjusting the grid division based on excessively high or low pollutant concentrations, it achieves a match between spatial division and actual pollutant distribution, making monitoring results more accurate and reliable. Introducing a spatial distance coefficient and fusing historical pollution data ensures that the obtained pollutant concentration fusion value better reflects the true pollution state of the region, reducing the impact of random data fluctuations. Utilizing the analysis of the amplitude, rate, and direction of changes in air quality feature vectors, it infers future air quality trends, providing advance warning for environmental governance and early warning. Combining multi-scale grids and wind direction and speed characteristics to identify upstream pollutant propagation paths, it can locate upstream pollution source areas affecting high-concentration regions, improving pollution tracking and control efficiency. Based on key monitoring areas, it automatically generates patrol instructions, increasing the sampling frequency in those areas and enabling collaboration between mobile and fixed monitoring equipment, thereby improving monitoring intensity and spatial coverage in key areas. By continuously adjusting the grid division, fusion strategy, and patrol route based on actual monitoring results, this invention forms a real-time adaptive feedback mechanism, improving overall monitoring accuracy and response speed. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method provided by the present invention; Figure 2 This is a system module diagram provided by the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0019] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0020] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0021] Furthermore, the present invention will be described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention in detail, the schematic diagrams are merely examples for ease of explanation and should not limit the scope of protection of the present invention.

[0022] Please see the appendix Figure 1 As shown, a real-time urban air quality monitoring method based on multi-source data fusion is provided, including: S1. Obtain air quality monitoring data and geographic meteorological data within the city. Based on the air quality monitoring data and geographic meteorological data, establish a comprehensive spatial grid set at the macro scale of the city and the micro scale of the street blocks. S2. Based on pollutant concentration data, the integrated spatial grid set is dynamically adjusted to obtain a multi-scale spatial grid; S3. Obtain the pollutant concentration fusion set of the multi-scale spatial grid based on its spatial location and historical pollution data; S4. Based on the pollutant concentration fusion set and geographic meteorological data, generate air quality feature vectors for multi-scale spatial grids. Based on the changes in air quality feature vectors, predict future air quality change trends and determine high-concentration spatial grids. S5. Based on air quality feature vectors, wind direction information, and wind speed information, identify the upstream spatial areas of high-concentration spatial grids as key monitoring areas; S6. Generate alarm information based on key monitoring areas, and increase the sampling frequency and data reporting frequency of all monitoring devices in key monitoring areas; S7. Based on the actual monitoring results of key regulatory areas, dynamically adjust the spatial grid and update alarm information and patrol instructions in real time.

[0023] As described in steps S1 to S7 above, air quality monitoring data from fixed monitoring stations, mobile monitoring equipment, and remote sensing devices within the city are acquired. Simultaneously, geographic and meteorological data covering the city area, such as wind speed, wind direction, temperature, humidity, topography, and road network, are also acquired. By uniformly transforming the collected spatial coordinates, the data is mapped to a unified spatial reference system. A macroscopic spatial grid is constructed at the city-wide scale, and a microscopic spatial grid is constructed at the street block scale, forming a comprehensive spatial grid set. Based on the pollutant concentration data of each spatial grid, areas of abnormal concentration are identified. The size, density, or boundaries of the grid units are dynamically adjusted according to a preset grid adjustment strategy to ensure consistency between the grid division and the spatial distribution of pollutants, thereby obtaining a more suitable result for the real-time pollution status. A multi-scale spatial grid is used to acquire the spatial location, current pollutant concentration, and corresponding historical pollution data of each grid. Combined with the distance relationship between each grid and monitoring points, a spatial distance coefficient is obtained. Based on this coefficient, the current pollution concentration value and multiple historical pollution values ​​are weighted and fused to obtain a pollutant concentration fusion set for the multi-scale spatial grid. This fusion set is then combined with wind speed, wind direction, temperature, humidity, topography, and road network information according to preset rules to generate air quality feature vectors for each grid. Based on the amplitude, rate, and direction of change of these feature vectors over multiple consecutive monitoring periods, future air quality trends (including increases, decreases, or stabilization) are predicted, thereby identifying areas of high pollution concentration in the future. By using air quality feature vectors, prevailing wind direction, and wind speed parameters, the upstream direction of high-concentration spatial grids within urban areas is determined. Combined with the fused pollution concentration values ​​and wind speed and direction characteristics of the upstream areas, potential pollutant propagation paths and source directions affecting high-concentration areas are identified, thus defining relevant upstream spatial areas as key monitoring areas. Based on the pollutant concentrations in these key monitoring areas, pollution risk levels are determined, and alarm information is generated accordingly. Simultaneously, patrol instructions are issued to mobile monitoring devices surrounding the key monitoring areas, guiding them to concentrate in these areas. The sampling frequency and data reporting frequency of all monitoring devices within the area are adjusted according to the pollution risk level, significantly improving the spatial and temporal monitoring density of key areas. Based on actual monitoring results from multiple consecutive monitoring cycles, the current spatial grid division method, pollutant concentration fusion weights, and mobile monitoring equipment patrol paths are dynamically adjusted to correct the judgment results of pollution source areas and air quality trends. New alarm information and patrol instructions are generated in real time, enabling the monitoring system to form a closed-loop feedback mechanism and continuously adaptively optimize. By establishing a comprehensive spatial grid set combining macro and micro scales, it can simultaneously reflect the overall air quality of the city and the fine-grained pollution distribution at the street level, improving monitoring accuracy. By automatically adjusting the grid division according to excessively high or low pollutant concentrations, the spatial division is matched with the actual distribution of pollutants, making the monitoring results more accurate and reliable. Furthermore, by introducing a spatial distance coefficient and fusing historical pollution data…This invention aims to make the obtained pollutant concentration fusion values ​​more reflective of the true pollution status of a region, reducing the impact of random data fluctuations. By analyzing the amplitude, rate, and direction of changes in air quality feature vectors, it can predict future air quality trends, providing advance warning for environmental governance and early warning. Combining multi-scale grids and wind direction and speed characteristics to identify upstream pollutant propagation paths, it can locate upstream pollution source areas affecting high-concentration regions, improving the efficiency of pollution tracking and control. Based on key monitoring areas, it automatically generates patrol instructions, increasing the sampling frequency in those areas and enabling collaboration between mobile and fixed monitoring equipment to improve monitoring intensity and spatial coverage in key areas. By continuously adjusting grid division, fusion strategies, and patrol routes based on actual monitoring results, this invention forms a real-time adaptive feedback mechanism, improving overall monitoring accuracy and response speed.

[0024] In a preferred embodiment, air quality monitoring data and geographic meteorological data within the urban area are acquired, and a comprehensive spatial grid set is established based on the air quality monitoring data and geographic meteorological data at both the urban macro-scale and the street micro-scale; including: S101. Obtain air quality monitoring data from multiple air quality monitoring points within the city, and obtain geographic meteorological data covering the city, including wind speed information, wind direction information, temperature information, humidity information, topographic information, and road network information. S102. Based on the collection coordinates of each air quality monitoring point, perform coordinate transformation on the spatial location corresponding to the air quality monitoring data and geographic meteorological data to obtain multiple collection spatial coordinates, and summarize them into a collection spatial coordinate reference set. S103. Based on the urban area boundary, establish the grid construction range, and in conjunction with the collected spatial coordinate reference set, construct a macro-scale spatial grid for the urban area and a micro-scale spatial grid for the block area, and summarize them into a comprehensive spatial grid set.

[0025] As described in steps S101 to S103 above, a multi-source air quality data system is constructed using fixed standard monitoring stations, mobile sensing equipment, and satellite remote sensing data. Different data sources possess characteristics such as high precision, high coverage, and high timeliness. Simultaneously, geographic and meteorological data such as wind speed, wind direction, temperature, humidity, topography, and road network are acquired to supplement information on pollution diffusion environmental characteristics. Through synchronous acquisition of multi-source data, a comprehensive understanding of the city's air quality status and pollution diffusion background can be obtained. Since multi-source monitoring point data typically use different geographic coordinate systems or acquisition coordinate systems, to ensure the uniformity of subsequent grid construction, coordinate transformation is performed on the data corresponding to each monitoring point to unify them into the same city-level coordinate system, resulting in a collection spatial coordinate reference set. The grid construction scope is delineated based on the city's regional boundaries, and the collection spatial coordinate reference set, city area, building density, and road network distribution are considered. Information such as these are used to adaptively set the grid scale: a macro-scale grid (covering the entire city area, used to depict large-scale air quality change trends) and a micro-scale grid (covering blocks and high-density building areas, used to reflect fine-scale local changes in air pollution). By combining the macro and micro-scale grids to form a comprehensive spatial grid set, it is possible to simultaneously depict large-scale air quality changes at the city level and local pollution details at the block level, improving spatial resolution and analysis accuracy. Each spatial grid serves as the basic unit for carrying and calculating air quality data. It can map air quality monitoring data and geographic meteorological data from different sources to the corresponding grid locations, forming a unified spatial data expression structure. Through coordinate transformation and comprehensive grid construction, it solves the problem of inconsistent spatial systems for multi-source air quality data, providing a unified spatial benchmark for air pollution analysis, monitoring strategy formulation, and intelligent scheduling.

[0026] In a preferred embodiment, the integrated spatial grid set is dynamically adjusted based on pollutant concentration data to obtain a multi-scale spatial grid, including: S201. Obtain pollutant concentration data for each spatial grid in the integrated spatial grid set, and extract pollutant concentration values; S202. Obtain the pollutant concentration value range, determine whether the pollutant concentration value is within the pollutant concentration value range, and make adjustments accordingly; If the pollutant concentration value in each spatial grid is within the pollutant concentration value range, then the pollutant concentration of the spatial grid is determined to meet the requirements. If the pollutant concentration value in a spatial grid is outside the pollutant concentration value range and exceeds the upper limit of the pollutant concentration value range, the pollutant concentration in the spatial grid is determined to be too high. Based on the pollutant concentration exceeding the value of the spatial grid, a corresponding preset spatial grid adjustment strategy is matched to dynamically adjust the comprehensive spatial grid set. If the pollutant concentration value in a spatial grid is not within the pollutant concentration value range and is less than the lower limit of the pollutant concentration value range, the pollutant concentration in the spatial grid is determined to be too low. Based on the lower limit of the pollutant concentration in the spatial grid, a corresponding preset spatial grid adjustment strategy is matched to dynamically adjust the comprehensive spatial grid set. S203. Summarize the adjusted spatial grids to obtain a multi-scale spatial grid.

[0027] As described in steps S201 to S203 above, pollutant concentration values ​​are extracted from each grid cell in the multi-scale spatial grid. A preset pollutant concentration range is used as the evaluation criterion to determine whether the current grid's pollutant concentration is within a reasonable range, thus determining whether the grid needs adjustment. When the pollutant concentration value within the spatial grid exceeds the upper limit of the range, it means that the pollutant distribution within the grid may have a higher spatial gradient or a local pollution source, and the original grid scale is insufficient to reflect the true pollution characteristics. Therefore, by matching the preset grid refinement strategy table with the condition "current concentration value minus upper limit equals the excess value," the appropriate refinement level is automatically selected based on the degree of excess, subdividing the original grid into multiple smaller grids so that the pollutant concentration in the subdivided grids meets the concentration range requirements. To improve spatial resolution, when the pollutant concentration value of the spatial grid is lower than the lower limit of the interval, it indicates that the pollutants in the area are stable and the spatial gradient is low. The pollutant levels of multiple grids are similar, and the original grid scale may be too fragmented. By matching the corresponding grid merging strategy according to the "lower limit minus the current concentration value equals the lower limit value", multiple adjacent low-concentration grids are merged to make the concentration of the merged grid within the normal range, reducing the number of unnecessary spatial computing units and improving the efficiency of spatial modeling. By dynamically adjusting the comprehensive spatial grid set through two strategies of refinement and merging, the adjusted new grid units are re-aggregated to form a multi-scale spatial grid, realizing an adaptive spatial expression of refined high-concentration areas and coarse-grained low-concentration areas, thereby constructing a stable and efficient multi-scale grid system.

[0028] In a preferred embodiment, a pollutant concentration fusion set of a multi-scale spatial grid is obtained based on the spatial location of the multi-scale spatial grid and historical pollution data, including: S301. Obtain the spatial location, current pollutant concentration data, and historical pollution data of each spatial grid in the multi-scale spatial grid; S302. Obtain the spatial distance between each spatial grid and the corresponding air quality monitoring point; S303, Obtain the baseline spatial distance; S304. Obtain the corresponding spatial distance coefficient based on the spatial distance of each spatial grid and the reference spatial distance; S305. Extract the current pollutant concentration value based on the current pollutant concentration data, and extract the corresponding multiple historical pollutant concentration values ​​based on the historical pollution data; S306. Obtain the pollutant concentration fusion value of each spatial grid in the multi-scale spatial grid based on the spatial distance coefficient, the current pollutant concentration value, and multiple historical pollutant concentration values; S307. The pollutant concentration fusion values ​​of multiple spatial grids are summarized to obtain the pollutant concentration fusion set.

[0029] As described in steps S301 to S307 above, each spatial grid corresponds to a specific spatial location. By calculating the spatial distance between this location and each air quality monitoring point, the influence of the monitoring point data on the grid area can be characterized. To ensure the comparability of the fusion contributions of monitoring points at different distances to the grid, a benchmark spatial distance is set. By calculating the ratio of the actual spatial distance to the benchmark spatial distance, a spatial distance coefficient is obtained. This ensures that the contribution of the monitoring points to different spatial grids conforms to the spatial attenuation law, significantly improving the consistency between concentration estimation and actual geographical distribution. For each spatial grid, the current pollutant concentration value (reflecting real-time status) and multiple historical pollutant concentration values ​​(reflecting time trends and typical levels) are extracted. By combining historical data with current data, fluctuation deviations caused by relying solely on instantaneous monitoring data can be avoided, making the fused pollutant concentration more stable and reliable. During the fusion calculation process, the spatial distance coefficient is used as a key weighting factor, participating in the fusion along with the current pollutant concentration value and multiple historical pollutant concentration values. Finally, the fused pollutant concentration value is calculated for each grid, forming a fusion result that combines spatial representativeness and temporal stability. The formula for calculating the fused pollutant concentration value is as follows: In the formula, The value is expressed as the fusion value per pollutant concentration, and w represents the spatial distance coefficient corresponding to each spatial grid. The current pollutant concentration value is represented by , and ... Represented as the i-th historical pollutant concentration value, the pollutant concentration fusion values ​​of all spatial grids are summarized to form a complete pollutant concentration fusion set. In areas with low monitoring point density or uneven distribution, the spatial estimation bias caused by insufficient monitoring data can be effectively compensated by the joint fusion of distance weight and time weight, thereby improving the estimation quality.

[0030] In a preferred embodiment, air quality feature vectors of a multi-scale spatial grid are generated based on pollutant concentration fusion sets and geographic meteorological data. Future air quality change trends are predicted based on changes in these feature vectors, and high-concentration spatial grids are identified, including: S401. Obtain corresponding wind speed, wind direction, temperature, humidity, terrain, and road network information based on geographic meteorological data; S402. Combine the pollutant concentration fusion set of the multi-scale spatial grid with the corresponding wind speed, wind direction, temperature, humidity, terrain and road network information in a preset order to generate the air quality feature vector of each spatial grid in the multi-scale spatial grid. S403. Obtain multiple benchmark monitoring cycles, and obtain the feature vector change information of the air quality feature vector based on the multiple benchmark monitoring cycles. The feature vector change information includes the change amplitude, change rate and change direction. S404. Predict future air quality trends based on feature vector change information, whereby future air quality trends may include rising, falling, or remaining stable. S405. Obtain high-concentration spatial grids based on future air quality change trends and the fusion value of pollutant concentrations for each spatial grid.

[0031] As described in steps S401 to S405 above, the diffusion, accumulation, and migration of different pollutants in the air are closely related to various meteorological and geographical factors, including wind speed and direction (which determine the direction and speed of pollutant migration), temperature and humidity (which affect the rate of chemical reactions and the secondary formation of pollutants), topographic information (which determines areas where airflow is obstructed and pollutants tend to accumulate), and road network information (which helps identify traffic hotspots that contribute to pollution). By extracting these data, key environmental factors affecting air quality changes are constructed for each spatial grid. To comprehensively reflect the air pollution status and environmental background of the spatial grid, pollutant concentration fusion values, wind speed, wind direction, temperature, humidity, topography, and road network information are combined in a preset order. The resulting air quality feature vector has multidimensional attributes; essentially, it is a high-dimensional vector expression that can comprehensively characterize the air quality status of a spatial grid. To capture the dynamic changes in air quality over time, air quality feature vectors are obtained within multiple benchmark monitoring periods, and their changes are calculated, including the magnitude of change (the difference between the components of the feature vectors in two monitoring periods; specifically, different components are calculated based on actual needs. For example, the component in the feature vector is the difference in pollutant concentration fusion values, used to reflect the amount of air quality change), and the change... Rate (the speed of change of the same magnitude within a unit of time) and direction of change (the direction of change of the feature vector, used to characterize the aggregation trend, diffusion trend, etc. of pollutants) upgrade air quality changes from single pollutant changes to multi-factor comprehensive changes, obtaining a more accurate and comprehensive dynamic description of air quality. Utilizing the trend and directionality of feature vector changes, future air quality can be predicted, including upward trends (predicting an increase in pollutant concentration or air quality risk), downward trends (predicting an improvement in air quality), and steady states (predicting that pollutant concentrations will remain at existing levels). Based on the multidimensional change patterns of fused data, it is therefore more suitable for... For complex urban environments and scenarios with multiple sources of pollution, this method combines future trend results with pollutant concentration fusion values ​​to more accurately identify high-risk areas. If the current concentration in a grid is high and the trend is rising, it is considered a high-risk area. If the grid concentration is moderate and the trend is rising significantly, it is considered a potentially high-concentration area. If the concentration is high but the trend is declining, it is considered a low-priority monitoring area. If the concentration is high and environmental conditions are not conducive to diffusion, it is considered a persistently high-concentration area. This results in a high-concentration spatial grid that integrates multi-source data, improving the depth and accuracy of air quality status description. It can accurately capture pollutant accumulation trends and precisely identify the risk of secondary pollution caused by meteorological conditions.

[0032] In a preferred embodiment, based on air quality feature vectors, wind direction information, and wind speed information, the upstream spatial region of the high-concentration spatial grid is identified as a key monitoring area, including: S501. Based on the air quality feature vector of each spatial grid, obtain the dominant wind direction and wind speed parameters of the high-concentration spatial grid in the current monitoring period; S502. Obtain the wind direction relationship of the dominant wind direction in the spatial grid, and obtain multiple upstream spatial grid cells in the upstream direction of the high-concentration spatial grid based on the wind direction relationship; S503. Identify pollutant propagation paths based on the fused pollutant concentration values, air quality feature vectors, and wind speed parameters of multiple upstream spatial grid units; S504. Based on the pollutant propagation path, identify the upstream spatial areas that affect the high-concentration spatial grid and designate the upstream spatial areas as key monitoring areas.

[0033] As described in steps S501 to S504 above, within the current monitoring period, each spatial grid possesses a corresponding air quality feature vector. From this vector, the prevailing wind direction (representing the main direction of movement of pollutants) and the corresponding wind speed parameters (determining the migration capacity and spread range of pollutants) near high-concentration spatial grids are extracted. Based on the directional relationship of the prevailing wind direction in the spatial grid coordinate system (e.g., northerly winds pointing upstream indicate northerly grids, and southwesterly winds pointing upstream indicate southwesterly grids), and based on the geometric relationship between wind direction and spatial grids, grids located in the opposite direction of the prevailing wind, upstream grid units satisfying the wind direction vector projection angle threshold, and areas adjacent to high-concentration grids or areas the wind path must pass through are identified, resulting in multiple potential upstream spatial grid units. To determine whether pollutants truly originate from these upstream areas, further analysis is conducted on the pollutant concentration fusion value of the upstream grids, the air quality feature vectors of the upstream grids (including meteorological, topographical, and road network influence factors), and the wind speed parameters (determining the migration capacity and spread range of pollutants). (Pollutant propagation distance and intensity) By comprehensively analyzing the above information, possible propagation paths of pollutants are constructed. For example, if the pollutant concentration in an upstream grid is high and the prevailing wind direction points to the high-concentration grid, it is highly suspected to be a pollution source area. If the wind speed is high enough and the terrain resistance between the two grids is low, the path is more credible. If the air quality feature vector shows that the upstream grid environment is suitable for pollutant accumulation and downstream migration, then the propagation path is valid. Through multi-factor judgment, possible flow trajectories of pollutants can be accurately identified. After identifying possible propagation paths of pollutants, it is further determined which upstream areas have a significant impact on high-concentration spatial grids, which areas have the potential to continuously supply pollutants, and which areas have a potential impact on future pollution changes. Finally, these upstream areas that have made significant contributions to high-concentration spatial grids are identified as key monitoring areas. This can effectively identify the real pollutant propagation paths, achieve accurate source tracing, maintain stable upstream identification capabilities in complex environments, and improve the overall economy and practicality of the monitoring system.

[0034] In a preferred implementation, alarm information is generated based on key monitored areas, and the sampling frequency and data reporting frequency of all monitoring devices within the key monitored areas are increased, including: S601. Obtain the pollutant concentration fusion value of key regulatory areas, obtain the corresponding pollution risk level based on the pollutant concentration fusion value, and generate the pollutant concentration fusion value. S602. Obtain information on mobile monitoring devices surrounding key monitoring areas and issue patrol instructions to these devices to guide them to concentrate in key monitoring areas. S603. Obtain the corresponding sampling strategy based on the pollution risk level, and adjust the sampling frequency and data reporting frequency of all monitoring equipment in the key regulatory area according to the sampling strategy.

[0035] As described in steps S601 to S603 above, pollutant concentration fusion values ​​are extracted from the multi-scale spatial grid of the key monitoring area to obtain the overall pollution level of the area. According to the preset pollution risk level table, the fusion values ​​are mapped to the corresponding pollution risk levels (such as low, medium, high, and ultra-high), and alarm information for the area is generated, realizing the quantitative characterization of pollution risk in the key monitoring area. Information on mobile monitoring devices around the key monitoring area is obtained, including current coordinates, status, and executable patrol commands. Patrol commands are issued to these mobile monitoring devices to concentrate them along preset paths to the key monitoring area, ensuring that the density and coverage of monitoring devices in the key area are improved in real time, thereby enhancing the ability to capture pollution changes. The corresponding sampling strategy (including sampling frequency, sampling point selection, and data reporting frequency) is obtained by looking up the table according to the pollution risk level. The sampling frequency and data reporting frequency of all fixed and mobile monitoring devices in the key monitoring area are adjusted according to the sampling strategy to achieve refined monitoring of high-risk areas. By dynamically adjusting the monitoring parameters, a rapid response to pollution events is achieved, improving the timeliness and accuracy of data. Air quality changes in key areas can be captured in a timely manner, significantly shortening the response time.

[0036] In a preferred embodiment, the spatial grid is dynamically adjusted based on the actual monitoring results of key monitored areas, the identified pollution source areas and air quality trends are corrected, and alarm information and patrol instructions are updated in real time, including: S701. Obtain the actual monitoring results of key regulatory areas over multiple consecutive monitoring periods, and obtain the corresponding spatial grid adjustment strategy based on the actual monitoring results, and adjust the spatial grid accordingly. S702. Obtain the difference in monitoring values ​​between fixed and mobile monitoring equipment within key regulatory areas; S703. Obtain the corresponding patrol instruction correction strategy based on the difference in monitoring values, and correct the patrol instruction of the mobile monitoring device. The patrol instruction correction strategy includes adjusting the travel direction, stopping point position and patrol frequency of the patrol path. S704. Based on the revised patrol instructions, generate new alarm information in real time and issue updated patrol instructions to relevant mobile monitoring devices.

[0037] As described in steps S701 to S704 above, actual air quality monitoring data for multiple consecutive monitoring cycles are obtained from key monitoring areas, including pollutant concentration data collected by fixed and mobile monitoring devices. Based on a preset spatial grid adjustment strategy, the monitoring results are compared with historical data to determine whether the pollution levels of each grid meet expectations. For grids with extremely high or abnormal pollution levels, the grid division is dynamically adjusted; for example, high-pollution grids are subdivided into smaller grids for more refined monitoring, or low-pollution grids are merged to optimize the distribution of monitoring resources. Data from fixed and mobile monitoring devices within the key monitoring area are compared, and the differences in monitoring values ​​are calculated. By analyzing these differences, blind spots in the coverage of mobile device patrol paths or areas not fully monitored by fixed devices can be identified. Based on the differences in monitoring values ​​and pollution levels... Based on the spatial distribution characteristics of pollutants, a patrol command correction strategy is generated. The correction strategy includes: adjusting the travel direction of the mobile monitoring equipment's patrol path to better cover the upstream of pollution sources and high-risk areas; adjusting the location of stop points to enable higher-frequency monitoring of key grids; and adjusting the patrol frequency to improve the response capability to rapidly changing pollution events. The corrected patrol commands can be directly issued to the mobile monitoring equipment to achieve real-time path optimization. Based on the corrected spatial grid division and patrol strategy, updated alarm information is generated in real time, including the location of pollution sources, risk level, and impact range. The updated patrol commands are issued to the relevant mobile monitoring equipment to ensure that the equipment executes the patrol task according to the optimized strategy, forming a closed-loop control system from data acquisition, grid adjustment, patrol optimization to alarm generation, to achieve dynamic and real-time air quality management.

[0038] Please see the appendix Figure 2 As shown, the present invention also provides a real-time urban air quality monitoring system based on multi-source data fusion, used in the above-mentioned real-time urban air quality monitoring method based on multi-source data fusion, comprising: The spatial grid module is used to acquire air quality monitoring data and geographic meteorological data within the city. Based on the air quality monitoring data and geographic meteorological data, a comprehensive spatial grid set is established at both the macro-scale of the city and the micro-scale of the street blocks. The spatial grid adjustment module is used to dynamically adjust the integrated spatial grid set based on pollutant concentration data to obtain a multi-scale spatial grid. The pollutant concentration fusion module is used to obtain a fused set of pollutant concentrations from a multi-scale spatial grid based on the spatial location of the multi-scale spatial grid and historical pollution data. The high-concentration determination module is used to generate air quality feature vectors of multi-scale spatial grids based on pollutant concentration fusion sets and geographic meteorological data. Based on the changes in air quality feature vectors, it predicts future air quality change trends and determines high-concentration spatial grids. The key monitoring module identifies upstream spatial areas of high-concentration spatial grids as key monitoring areas based on air quality feature vectors, wind direction information, and wind speed information. The alarm module is used to generate alarm information based on key monitoring areas and increase the sampling frequency and data reporting frequency of all monitoring devices in key monitoring areas; The correction module is used to dynamically adjust the spatial grid based on the actual monitoring results of key monitored areas, and to update alarm information and patrol instructions in real time.

[0039] The aforementioned spatial grid module acquires air quality data from fixed monitoring stations, mobile sensors, and satellite remote sensing data. It also acquires geographic and meteorological data covering the urban area, including wind speed, wind direction, temperature, humidity, topography, and road network. The module performs coordinate transformation on the monitoring points and geographic and meteorological data to unify the spatial benchmark. Based on the city area, building density, and road network information, it constructs a macro-grid within the city limits and a micro-grid within city blocks, forming a comprehensive spatial grid set. This set comprehensively represents the spatial distribution of air quality in different urban areas. The spatial grid adjustment module acquires pollutant concentration data for each grid and determines whether it falls within a preset concentration range. Grids with concentrations above the upper limit are subdivided, while those below the lower limit are merged. The adjusted grids are aggregated into a multi-scale spatial grid to adapt to changes in the spatial distribution of pollutant concentrations, achieving dynamic optimization of grid division and improving the accuracy of pollution source monitoring and resource allocation efficiency. The pollutant concentration fusion module acquires current and historical pollutant concentration data, calculates a spatial distance coefficient based on the spatial distance between the grid and monitoring points and the baseline distance, and calculates and aggregates the fusion value using the spatial distance coefficient, current concentration, and historical concentration to form a pollutant concentration fusion set. Historical and spatial information are used to smooth pollution levels, improving the accuracy and reliability of pollutant concentration estimation. The high-concentration determination module combines the pollutant concentration fusion value with data such as wind speed, wind direction, temperature, humidity, topography, and road network to generate air quality data. The feature vector analysis module analyzes the magnitude, rate, and direction of feature vector changes within a continuous monitoring period to predict future air quality trends (rising, falling, or remaining stable). It combines these trends with pollutant fusion values ​​to identify high-concentration grids, enabling air quality trend prediction and early identification of potentially high-pollution areas, providing a basis for prevention and control. The key monitoring module, based on feature vectors, wind direction, and wind speed, obtains dominant wind direction parameters to determine multiple upstream grid units in the direction upstream of high-concentration grids. It analyzes pollutant propagation paths and identifies the upstream areas with the greatest impact on high-concentration grids as key monitoring areas, enabling pollution source tracing and key area location, improving the targeting and scientific nature of air quality management. The alarm module, based on pollutant concentrations in key areas... The system generates pollution risk levels based on the merged values ​​and generates alarm information. It issues patrol instructions to surrounding mobile monitoring devices, guiding them to focus on key areas. Based on the pollution risk level, it adjusts the sampling frequency and data reporting frequency of monitoring devices within the area, enabling real-time monitoring and rapid response in high-risk areas, improving data acquisition accuracy and alarm timeliness. The correction module obtains actual monitoring results within a continuous monitoring cycle, adjusts the spatial grid division, analyzes the differences between fixed and mobile device data, corrects the patrol path, dwell points, and patrol frequency, generates new alarm information in real time, and issues updated patrol instructions, forming a closed-loop dynamic control system. This enables precise tracking of pollution sources and real-time correction of air quality trends, improving the reliability and response speed of the monitoring system.

[0040] And, a real-time urban air quality monitoring terminal based on multi-source data fusion, including: One or more processors; A storage device on which one or more programs are stored; When one or more programs are executed by one or more processors, the one or more processors implement a method for real-time monitoring of urban air quality based on multi-source data fusion.

[0041] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A real-time monitoring method for urban air quality based on multi-source data fusion, characterized in that, The method comprises the following steps: acquiring air quality monitoring data and geographical meteorological data in a city, and establishing a comprehensive spatial grid set on a city macro scale and a street micro scale according to the air quality monitoring data and the geographical meteorological data; adjusting the comprehensive spatial grid set according to pollutant concentration data to obtain a multi-scale spatial grid; obtaining a pollutant concentration fusion set of the multi-scale spatial grid according to the spatial position of the multi-scale spatial grid and historical pollution data; generating an air quality feature vector of the multi-scale spatial grid according to the pollutant concentration fusion set and the geographical meteorological data, predicting a future air quality change trend according to a change of the air quality feature vector, and determining a high-concentration spatial grid; identifying an upstream spatial area of the high-concentration spatial grid as a key supervision area based on the air quality feature vector, wind direction information and wind speed information; generating an alarm information according to the key supervision area, and improving a sampling frequency and a data reporting frequency of all monitoring devices in the key supervision area; dynamically adjusting the spatial grid according to an actual monitoring result of the key supervision area, and updating the alarm information and a patrol instruction in real time. 2.The real-time monitoring method of urban air quality based on multi-source data fusion according to claim 1, characterized in that, The method comprises the following steps: acquiring air quality monitoring data of multiple air quality monitoring points in a city, and acquiring geographical meteorological data covering the city, wherein the geographical meteorological data comprises wind speed information, wind direction information, temperature information, humidity information, terrain information and road network information; performing coordinate conversion on a spatial position corresponding to the air quality monitoring data and the geographical meteorological data according to a collection coordinate of each air quality monitoring point to obtain multiple collection spatial coordinates, and collecting the multiple collection spatial coordinates into a collection spatial coordinate reference set; establishing a grid construction range according to a city area boundary, combining the collection spatial coordinate reference set, constructing a macro scale spatial grid for a city area, constructing a micro scale spatial grid for a street range, and collecting the macro scale spatial grid and the micro scale spatial grid into a comprehensive spatial grid set. 3.The real-time monitoring method of urban air quality based on multi-source data fusion according to claim 1, characterized in that, The method comprises the following steps: acquiring pollutant concentration data of each spatial grid in the comprehensive spatial grid set, and extracting a pollutant concentration value; acquiring a pollutant concentration value interval, and determining whether the pollutant concentration value is within the pollutant concentration value interval, and adjusting the comprehensive spatial grid set; if the pollutant concentration value in each spatial grid is within the pollutant concentration value interval, it is determined that the pollutant concentration of the spatial grid meets the requirements; if the pollutant concentration value in a spatial grid is not within the pollutant concentration value interval and is greater than an upper limit value of the pollutant concentration value interval, it is determined that the pollutant concentration of the spatial grid is too high, a corresponding preset spatial grid adjustment strategy is matched according to an exceeding value of the pollutant concentration of the spatial grid, and the comprehensive spatial grid set is dynamically adjusted. If the pollutant concentration value in the spatial grid is not within the pollutant concentration value interval and is less than the lower limit value of the pollutant concentration value interval, it is determined that the pollutant concentration of the spatial grid is too low, and the comprehensive spatial grid set is dynamically adjusted according to the matching preset spatial grid adjustment strategy corresponding to the lower value of the pollutant concentration of the spatial grid. The adjusted spatial grid is summarized to obtain a multi-scale spatial grid. 4.The method of claim 1, wherein, According to the spatial position and historical pollution data of the multi-scale spatial grid, a pollutant concentration fusion set of the multi-scale spatial grid is obtained, including: Obtain the spatial position, current pollutant concentration data and historical pollution data of each spatial grid in the multi-scale spatial grid; Obtain the spatial distance between each spatial grid and the corresponding air quality monitoring point; Obtain the reference spatial distance; According to the spatial distance and the reference spatial distance of each spatial grid, the corresponding spatial distance coefficient is obtained; According to the current pollutant concentration data, the current pollutant concentration value is extracted, and according to the historical pollution data, a plurality of historical pollutant concentration values are extracted; According to the spatial distance coefficient, the current pollutant concentration value and the plurality of historical pollutant concentration values, a pollutant concentration fusion value of each spatial grid in the multi-scale spatial grid is obtained; The pollutant concentration fusion values of the plurality of spatial grids are summarized to obtain a pollutant concentration fusion set. 5.The real-time monitoring method of urban air quality based on multi-source data fusion according to claim 4, characterized in that, According to the pollutant concentration fusion set and the geographical and meteorological data, an air quality feature vector of the multi-scale spatial grid is generated, and according to the change of the air quality feature vector, a future air quality change trend is predicted, and a high concentration spatial grid is determined, including: Based on the geographical and meteorological data, corresponding wind speed information, wind direction information, temperature information, humidity information, terrain information and road network information are obtained; The pollutant concentration fusion set of the multi-scale spatial grid and the corresponding wind speed information, wind direction information, temperature information, humidity information, terrain information and road network information are combined according to a preset order to generate an air quality feature vector of each spatial grid in the multi-scale spatial grid; A plurality of reference monitoring periods are obtained, and feature vector change information of the air quality feature vector is obtained according to the plurality of reference monitoring periods, wherein the feature vector change information includes change amplitude, change rate and change direction; According to the feature vector change information, a future air quality change trend is predicted, wherein the future air quality change trend includes rising, falling or remaining stable; According to the future air quality change trend and the pollutant concentration fusion value of each spatial grid, a high concentration spatial grid is obtained. 6.The real-time monitoring method of urban air quality based on multi-source data fusion according to claim 1, characterized in that, Based on the air quality feature vector, the wind direction information and the wind speed information, an upstream spatial area of the high concentration spatial grid is identified as a key supervision area, including: Based on the air quality feature vector of each spatial grid, the dominant wind direction and wind speed parameters of the high concentration spatial grid in the current monitoring period are obtained; Obtain the wind direction pointing relationship of the dominant wind direction in the spatial grid, and obtain a plurality of upstream spatial grid units in the upstream direction of the high concentration spatial grid according to the wind direction pointing relationship; According to the pollutant concentration fusion value, the air quality feature vector and the wind speed parameter of the plurality of upstream spatial grid units, the pollutant propagation path is identified; According to the pollution propagation path, an upstream space area affecting the high-concentration space grid is obtained, and the upstream space area is determined as a key supervision area. 7.The real-time monitoring method of urban air quality based on multi-source data fusion according to claim 1, characterized in that, According to the key supervision area, alarm information is generated, and the sampling frequency and data reporting frequency of all monitoring devices in the key supervision area are increased, including: Obtain the pollution concentration fusion value of the key supervision area, obtain the corresponding pollution risk level according to the pollution concentration fusion value, and generate the pollution concentration fusion value; Obtain the surrounding mobile monitoring devices of the key supervision area, and issue a patrol command to the surrounding mobile monitoring devices to guide the concentration to the key supervision area; According to the pollution risk level, the corresponding sampling strategy is obtained, and the sampling frequency and data reporting frequency of all monitoring devices in the key supervision area are adjusted according to the sampling strategy. 8.The method of claim 1, wherein, According to the actual monitoring result of the key supervision area, the space grid is dynamically adjusted, the identified pollution source area and air quality trend are corrected, and the alarm information and the patrol command are updated in real time, including: Obtain the actual monitoring result of the key supervision area in multiple continuous monitoring periods, and obtain the corresponding space grid adjustment strategy according to the actual monitoring result, and adjust the space grid; Obtain the monitoring value difference between the fixed monitoring device and the mobile monitoring device in the key supervision area; According to the monitoring value difference, the corresponding patrol command correction strategy is obtained, and the patrol command of the mobile monitoring device is corrected, wherein the patrol command correction strategy includes adjusting the direction of travel, the position of the stop point and the patrol frequency of the patrol path; According to the corrected patrol command, new alarm information is generated in real time, and the updated patrol command is issued to the related mobile monitoring device.

9. The urban air quality real-time monitoring system based on multi-source data fusion, applied to the urban air quality real-time monitoring method based on multi-source data fusion in any one of claims 1 to 8, characterized in that, It includes: A space grid module for obtaining air quality monitoring data and geographic meteorological data in a city, and establishing a comprehensive space grid set on a city macro scale and a street micro scale according to the air quality monitoring data and the geographic meteorological data; A space grid adjustment module for dynamically adjusting the comprehensive space grid set according to the pollutant concentration data to obtain a multi-scale space grid; A pollutant concentration fusion module for obtaining a pollutant concentration fusion set of the multi-scale space grid according to the spatial position of the multi-scale space grid and historical pollution data; A high-concentration determination module for generating an air quality feature vector of the multi-scale space grid according to the pollutant concentration fusion set and the geographic meteorological data, predicting future air quality change trend according to the change of the air quality feature vector, and determining a high-concentration space grid; A key supervision module for identifying an upstream space area of the high-concentration space grid as a key supervision area based on the air quality feature vector, wind direction information and wind speed information; An alarm module for generating alarm information according to the key supervision area, and increasing the sampling frequency and data reporting frequency of all monitoring devices in the key supervision area; A correction module for dynamically adjusting the space grid according to the actual monitoring result of the key supervision area, and updating the alarm information and the patrol command in real time.

10. The urban air quality real-time monitoring terminal based on multi-source data fusion, characterized in that, It includes: One or more processors; A storage device having one or more programs stored thereon; When one or more programs are executed by one or more processors, so that the one or more processors implement the method for real-time monitoring of urban air quality based on multi-source data fusion according to any one of claims 1 to 8.

Images