Atmospheric compound pollutant health influence assessment method and device and early warning system

Abstract

The invention provides an atmospheric composite pollutant health influence assessment method and device and an early warning system. The method comprises the following steps: constructing a composite atmospheric pollutant time sequence database; constructing a comprehensive pollutant index model for representing the composite atmospheric pollutants through an analytic hierarchy process; according to the atmospheric pollutant time sequence database and the comprehensive pollutant index model, constructing a model of influence of composite atmospheric pollutants on crowd health, and extracting an exposure reaction relation coefficient; and according to the exposure reaction relation coefficient, calculating an environment air quality health index and grading the environment air quality health index. According to the invention, health influence assessment is carried out on the composite atmospheric pollutants, and accurate assessment and early warning of air quality health risks are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of health risk assessment, and in particular relates to a health impact assessment method, device and early warning system for atmospheric composite pollutants. Background Art

[0002] In recent years, the problem of air pollution has become increasingly serious, and the number of diseases caused by air pollution has gradually increased. The diseases have brought a high burden to the public, making air quality and its health risks a hot social issue that has attracted widespread public attention.

[0003] Currently, the Air Quality Health Index is used to assess air quality. However, the Air Quality Index only focuses on the mass concentration of major atmospheric pollutants, and does not pay attention to the contribution of multiple complex atmospheric pollutants to air quality and their risks to human health. The existing technology does not include risk assessment of the health effects of complex atmospheric pollutants, and it is impossible to comprehensively and accurately assess the health risks of air quality. Summary of the invention

[0004] In view of the problems in the prior art, the present invention proposes a method, device and early warning system for health impact assessment of composite atmospheric pollutants, which conduct health impact assessment on composite atmospheric pollutants and achieve accurate assessment and early warning of air quality health risks.

[0005] To achieve the above object, the technical solution of the present invention is achieved as follows: A method for assessing the health impact of atmospheric composite pollutants, comprising: S1. Build a time series database of composite air pollutants; S2. Construct a comprehensive pollutant index model to characterize complex air pollutants through the analytic hierarchy process; S3. Constructing a model of the impact of composite air pollutants on human health based on the air pollutant time series database and the comprehensive pollutant index model, and extracting exposure-response relationship coefficients; S4. Calculate and grade the ambient air quality health index based on the exposure-response relationship coefficient.

[0006] Furthermore, step S1 specifically includes: S101. Collect air pollutant data by date; S102, collecting meteorological data by date; S103. Collect population health data by date; S104. Calculate daily life loss years data; S105. Construct a time series database of complex atmospheric pollutants.

[0007] Furthermore, step S2 specifically includes: S201, taking air environment quality as the target layer, taking various air pollutants as various evaluation factors of the criterion layer, taking air environment quality classification as the scheme layer, and constructing an air environment quality hierarchical model; S202, according to the air environment quality hierarchical structure model, taking the air environment quality classification standard as the benchmark, the scale is the ratio of each evaluation factor to the secondary standard, and then constructing a judgment matrix OC; taking the reciprocal of the absolute value of the difference between each evaluation factor and the standards at each level as the scale, constructing a judgment matrix CP; S204, solving the maximum eigenvalue and eigenvector of each judgment matrix, and obtaining the weight according to the eigenvector; S205, performing consistency check on the judgment matrix according to the maximum eigenvalue; and constructing a comprehensive pollutant index model for characterizing complex air pollutants according to the weights of the judgment matrix OC.

[0008] Furthermore, step S3 specifically includes: S301, calculating the comprehensive pollutant index according to the comprehensive pollutant index model characterizing the complex air pollutants and the complex air pollutant time series database, and updating the complex air pollutant time series database; S302. Use time series analysis to construct a comprehensive pollutant impact model on human health based on the updated composite air pollutant time series database; extract exposure-response relationship coefficients; S303. Evaluate the impact of complex air pollutants on human health based on the exposure-response relationship coefficient.

[0009] Furthermore, step S4 specifically includes: S401. Calculate the air quality health index AQHI based on the exposure-response relationship coefficient; S402. Classify the air quality health index according to the health effects of air pollution on the human body.

[0010] Another aspect of the present invention is to provide a device for assessing the health impact of atmospheric composite pollutants, comprising: Database module: build a time series database of composite air pollutants; Hierarchy analysis module: construct a comprehensive pollutant index model to characterize complex air pollutants through the hierarchical analysis method; Health model module: constructs a model of the impact of composite air pollutants on human health based on the air pollutant time series database and the comprehensive pollutant index model, and extracts the exposure-response relationship coefficient; Grading module: Calculates and grades the environmental air quality health index based on the exposure-response relationship coefficient.

[0011] Furthermore, the database module specifically includes: Air pollutant data unit: collect air pollutant data by date; Meteorological data unit: collect meteorological data by date; Health data unit: collect population health data by date; Life loss data unit: calculates daily life loss data; Database unit: Construct a time series database of complex atmospheric pollutants.

[0012] Furthermore, the hierarchical analysis module includes: Hierarchical model unit: Taking air environment quality as the target layer, various air pollutants as evaluation factors of the criterion layer, and air environment quality classification as the scheme layer, a hierarchical model of air environment quality is constructed; Judgment matrix unit: According to the air environment quality hierarchical structure model, the air environment quality classification standard is used as the benchmark, and the scale is the ratio of each evaluation factor to the secondary standard, thereby constructing a judgment matrix OC; the reciprocal of the absolute value of the difference between each evaluation factor and each level of standard is used as the scale to construct a judgment matrix CP; Solving unit: solving the maximum eigenvalue and eigenvector of each judgment matrix, and obtaining the weight according to the eigenvector; Construction unit: Conduct consistency check on the judgment matrix based on the maximum eigenvalue; construct a comprehensive pollutant index model to characterize complex air pollutants based on the weights of the judgment matrix OC.

[0013] Furthermore, the health model module includes: A data updating unit: calculating the comprehensive pollutant index according to the comprehensive pollutant index model characterizing the complex air pollutants and the complex air pollutant time series database, and updating the complex air pollutant time series database; Analysis unit: Using time series analysis, based on the updated composite atmospheric pollutant time series database, a comprehensive pollutant impact model on human health is constructed; exposure-response relationship coefficients are extracted; Assessment unit: Evaluate the impact of complex air pollutants on human health based on the exposure-response relationship coefficient.

[0014] Furthermore, the grading module includes: Index calculation unit: calculate the air quality health index AQHI based on the exposure response relationship coefficient; Classification unit: Classify the air quality health index according to the health effects of air pollution on the human body.

[0015] On the other hand, the present invention also proposes an atmospheric composite pollutant health impact early warning system, which uses the above-mentioned atmospheric composite pollutant health impact assessment method to calculate the air quality health index AQHI, and realizes real-time AQHI value update, health risk advice, AQHI trend analysis, AQHI comparative analysis and AQHI model management functions.

[0016] Furthermore, the atmospheric composite pollutant health impact warning system establishes an air quality and health risk forecasting platform, adopts a B / S system architecture, takes Spring Boot as the basic framework, integrates Spring MVC, Spring JPA, Spring Security, and Spring Cache subframeworks, obtains atmospheric pollutant concentration data monitored by each air quality monitoring station, and implements scheduled scheduling for data update through Spring Quartz Cron timing triggering; when calculating the real-time AQHI value, the concentration within 48 hours, 72 hours, and 96 hours is smoothed according to the impact degree of different pollutants; the visualization of statistical graphics adopts G2 based on the graphic grammar theory as the underlying visualization engine, which is data-driven and provides graphic grammar and interaction grammar; the visualization of geographic spatial data uses L7 as the visualization framework, and intuitively displays the AQHI data and risk level of each area based on WebGL, and performs secondary rendering of the risk level of different areas according to different colors.

[0017] Furthermore, the atmospheric composite pollutant health impact early warning system extracts health risk recommendations into a separate module, divides the risk levels of public health into four levels: low risk, medium risk, high risk and extremely high risk according to AQHI, uniformly manages health recommendations for different groups of people according to different risk levels, and configures health risk recommendations at the data dictionary level, dynamically displays risk levels and health recommendations for different susceptible groups based on real-time AQHI values, and provides personalized health tips for different groups of people.

[0018] Furthermore, the atmospheric composite pollutant health impact early warning system performs AQHI trend analysis at different time granularities, and intuitively reflects the trend of health risks in combination with trend charts; it also provides comparative analysis of AQHI from two dimensions, region and time.

[0019] Compared with the prior art, the present invention has the following beneficial effects: The present invention establishes an exposure-response relationship between composite air pollutants and years of life lost based on the analytic hierarchy process, constructs an air quality health index, and not only focuses on the mass concentration of air pollutants, but also pays attention to the contribution of multiple composite air pollutants to air quality and their risks to human health. At the same time, it can solve the interaction of multiple composite air pollutants on health, and can comprehensively and accurately assess the health risks of air quality. The early warning system established based on this can be integrated into a WeChat public platform or a third-party system to achieve real-time forecasting of AQHI and its health risks in each region, which can realize timely and effective transmission of environmental air quality levels and corresponding health risk level information to the public, provide health advice for daily travel for various groups of people, and promote the forecast of urban air pollution health risks, which has good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic diagram of a flow chart of Embodiment 1 of the present invention; Figure 2 is a schematic diagram of the air environment quality hierarchical structure model of Example 1 of the present invention; Figure 3 It is a schematic diagram of the air quality and health risk forecasting platform architecture of Example 2 of the present invention. DETAILED DESCRIPTION

[0021] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0022] In order to make the purpose and features of the present invention more obvious and understandable, the specific implementation methods of the present invention are further described below in conjunction with the accompanying drawings.

[0023] Embodiment 1:

[0024] This embodiment 1 proposes a method for health impact assessment of composite atmospheric pollutants. The main design concept is to implement health risk assessment of composite atmospheric pollutants based on the analytic hierarchy process and construct an air quality health index.

[0025] Based on the above design concept, the method of this embodiment is as follows Figure 1 As shown, including: 1. Construction of a time series database of composite air pollutants: (1) Air pollutant data collection: The daily average concentrations of various air pollutants at the national air pollutant monitoring stations in the area of ​​interest are collected by date. The various air pollutants in this embodiment include PM 2.5 、PM 10 , SO2, NO2, CO and O3.

[0026] (2) Meteorological data collection: Daily meteorological data of meteorological monitoring points in the area of ​​interest are collected by date, including daily average temperature and daily average relative humidity. Meteorological data will be used as a confounding factor for control when constructing the exposure-response relationship model.

[0027] (3) Population health data collection: non-accidental death data of people of all ages in the area of ​​interest are collected by date.

[0028] (4) Calculation of daily years of life lost: The calculation formula is: YLL=N×L, where: YLL is years of life lost, N is the number of deaths of each age group and gender; L is the number of years of life lost in each age group of death. The number of years of life lost refers to the standard life table of the World Health Organization (WHO) (see Table 1 below), where SEYLL represents the number of years of life lost in the standard life table.

[0029]

[0030] (5) Construction of a time series database of complex air pollutants: Take each of the above data as a variable, use date as the time variable, and summarize them according to the date variable to form a time series database of complex air pollutants.

[0031] 2. AHP method to construct comprehensive pollutants to characterize complex air pollutants: (1) Constructing a hierarchical model: Figure 2 As shown, including: The air environment quality is taken as the target layer, represented by O; Take n kinds of air pollutants as the criterion layer. In this embodiment, there are 6 kinds of air pollutants, that is, n=6. The 6 kinds of air pollutants include PM 2.5 、PM 10 , SO2, NO2, CO and O3 are used as evaluation factors of the criterion layer, and their concentrations are represented by Ci, i=1,2,…,n; The classification of air environment quality is used as the scheme layer, and the classification is represented by Pj, j=1,2,…,m; in this embodiment, according to the classification of air environment quality in the ambient air quality standard (GB3095-2012), it is divided into air quality level 1 and air quality level 2, that is, m=2; Build as Figure 2 A hierarchical model of air quality.

[0032] (2) Constructing the judgment matrix (OC): Based on the hierarchical model constructed above, according to the relative importance of each evaluation factor in the criterion layer, the secondary standard of air quality is used as the benchmark, and the scale is the ratio of each evaluation factor to the secondary standard of air quality, that is, ; In this embodiment, the 6 scales are , , , , , ; Where C represents the concentration of a certain pollutant. 2.5 、PM 10 , SO2 and NO2 are annual average concentrations, CO and O3 have no annual average concentration limits, so CO uses the 24-hour average concentration limit, and O3 uses the daily maximum 8-hour average concentration limit; construct a judgment matrix (OC) for pairwise comparison. The construction method of the judgment matrix (OC) is as follows: Taking 2017 as an example, PM 2.5 、PM 10 The annual average concentrations of SO2 and NO2 were 63.01 μg / m 3 , 104.8μg / m 3 , 16.18 μg / m 3 and 50.1 μg / m 3 The daily average concentration of CO is 1312.6 μg / m 3 The maximum concentration of O3 in 8 hours was 104.18 μg / m 3 . The ratios are calculated with the corresponding standards in the secondary air quality standard, and the scales of the evaluation factors are 1.80, 1.49, 0.27, 1.25, 0.33 and 1.04 respectively. Then the judgment matrix (OC) is constructed as shown in Table 2:

[0033] (3) Constructing the judgment matrix (CP): According to the hierarchical model, the inverse of the absolute value of the difference between each evaluation factor and the standard at each level is used as the scale, that is, , construct a judgment matrix (CP) for pairwise comparison of relative importance between standards at all levels.

[0034] (4) Obtain the maximum eigenvalue and eigenvector of each judgment matrix: First, calculate the product of each row of the judgment matrix Mi, i = 1, 2, ..., n; then calculate the nth root of Mi ; Then with vector As a feature vector, normalize it and get the weight ; Single sorting for each level; Finally, calculate the maximum characteristic root λ of the judgment matrix max .

[0035] (5) Perform consistency check on the constructed judgment matrix: obtain the maximum characteristic root λ of the pairwise judgment matrix max After that, the consistency index CI (Consistency Index) and the average random consistency index RI (Random Index) of each matrix can be calculated and tested. ; Then, according to the order n of the judgment matrix, the average random consistency index table is consulted to obtain the corresponding RI value, and finally the consistency ratio CR (Consistency Rate) is calculated. ,When CR is less than 0.1, it is considered that the consistency of the constructed judgment matrix is ​​good, otherwise the matrix needs to be corrected accordingly.

[0036] (6) Calculate the hierarchical ranking: According to the above steps, the weights of the hierarchical single ranking of each OC matrix and CP matrix are obtained. After that, the total hierarchical ranking of the hierarchical model can be obtained. The weight of the total hierarchical ranking can be the sum of the products of the hierarchical single ranking weight of the OC matrix and the hierarchical single ranking weight corresponding to the CP matrix.

[0037] (7) Construct a new comprehensive pollutant model: According to the weight of the judgment matrix OC As the weights assigned to each air pollutant, after standardizing the concentration of each pollutant, a new comprehensive pollutant index model of composite air pollutants can be constructed based on the obtained weights. The standardization of the concentration of each pollutant can be a Z-Score standardized value of the sliding average of the concentration of each pollutant on the same day and after a period of time.

[0038] 3. Construction and evaluation of the model of impact of complex air pollutants on human health: (1) Calculation of the comprehensive pollutant index model for complex air pollutants: Standardize the concentration of each pollutant based on the data in the time series database of complex air pollutants. Then, calculate the new daily comprehensive pollutant index based on the constructed new comprehensive pollutant index model for complex air pollutants, merge it with the time series database, and update the time series database.

[0039] (2) Use time series analysis to construct a comprehensive pollutant impact model on human health: Based on the constructed practice sequence database, the generalized additive model (GAM) is used to establish the exposure-response relationship between comprehensive pollutants and years of life lost in the population. The Gaussian distribution is selected as the connection function. The specific formula is as follows: ; Where: : The expected value of the resident daily YLL on observation day k; : refers to the daily average concentration of pollutants on observation day k; : The indicator variable coefficient estimated by the regression model, indicating the change in daily YLL of the population for each unit change in pollutant concentration; time: date variable. Choosing a suitable ν value for the date can effectively control the long-term fluctuation and seasonal fluctuation trend of the polluted YLL series data. DOW: dummy variable, eliminating the influence of the natural fluctuation trend of YLL in the short term and the influence of the heating season on YLL fluctuation; ns: natural smoothing spline function, ν is the degree of freedom. It is used to deal with the nonlinear trend and serial correlation of daily YLL on the time axis, that is, to control the confounding factors such as long-term trend and meteorological factors related to long-term variation in time series data.

[0040] Since the health hazards of each pollutant have a lag effect, when establishing the exposure-response relationship, the sliding mean of the average daily concentration of each pollutant is substituted into the above function, and the exposure-response relationship between the sliding mean of the next 1-7 days (mv01-mv07) and YLL is established, and the exposure-response relationship coefficient β is extracted.

[0041] (3) Evaluation of the impact of complex air pollutants on human health: Based on the extracted exposure-response relationship coefficient β, it is converted into the change in the effect value of the impact on human health for every unit increase in the concentration of complex air pollutants and its 95% confidence interval (CI), so as to evaluate and verify the impact of air pollutants on human health.

[0042] 4. Construction and classification of air quality health index: (1) Construction of the Air Quality Health Index: After obtaining the exposure response coefficient β of the new comprehensive pollutant index and daily mortality data, the AQHI is calculated according to the following formula: ; ; in, is the value of the comprehensive pollutant on day t; is the weight of air pollutant a; is the concentration of air pollutant a on day t; β is the exposure-response relationship coefficient; It is the maximum value of (β×Pt) within n days. The purpose of multiplying by 10 and adding 5 in the formula is to normalize AQHI to a value between 0 and 10.

[0043] (2) Classification of Air Quality Health Index: Air pollution is classified into several different levels according to its health effects on the human body. In this embodiment, it is divided into four levels, where AQHI of 0-3 represents low pollution, 4-6 represents moderate pollution, 7-10 represents high pollution, and >10 represents severe pollution.

[0044] According to the above method, the air quality health index AQHI is constructed using the data of a city from 2014 to 2018 as an example: The judgment matrix constructed based on the scale of each evaluation factor in each year from 2014 to 2018 is as follows: The overall OC judgment matrix for the period 2014-2018 is shown in Table 3 below. PM 2.5 、PM 10 , SO2, NO2, CO and O3 obtained weights in the OC judgment matrix are the second to last column in the table, which are 0.319, 0.259, 0.070, 0.193, 0.055 and 0.103 respectively;

[0045] The overall CP judgment matrix for the period 2014-2018 is shown in Table 4 below. The weights calculated after establishing the matrix are the second to last column, including: PM 2.5 The weights for air quality level 1 and level 2 are 0.384 and 0.616 respectively, and PM 10 The weights of SO2 for the two levels are 0.860 and 0.140, NO2 and CO for the two levels are both 0.5; the weights of O3 for the two levels are 0.981 and 0.019;

[0046] Calculate the maximum eigenvalue λ of each judgment matrix max , CI and CR values ​​are further calculated based on the RI values ​​and formulas obtained from the table, and the λ of each OC judgment matrix from 2014 to 2018 max They are 6.0088, 5.9872, 5.9992, 6.0175, 6.0078, and 6.0025 respectively; the lambda values ​​of the CP judgment matrices from 2014 to 2018 max They are 2.0044, 2.0001, 1.9987, 2.0019, 1.9992, and 2.0000 respectively. The final calculation shows that the CR values ​​of the OC judgment matrices from 2014 to 2018 are 0.0014, 0.0021, 0.0001, 0.0028, 0.0013, and 0.0004 respectively; while the CR values ​​of the CP judgment matrices from 2014 to 2018 are all 0. The results show that the CR values ​​of all judgment matrices are less than 0.1, indicating that the above-constructed matrices have good consistency.

[0047] After obtaining the hierarchical single ranking weight Wi of each OC and CP matrix, we can continue to obtain the hierarchical total ranking of the hierarchical model and evaluate the level of atmospheric air quality according to the maximum weight. The hierarchical total ranking of each hierarchical model is shown in Table 5 below. The first-level and second-level weights of the overall air quality in 2014-2018 were 0.503 and 0.496 respectively; the calculation method is the sum of the products of the single ranking weight of the OC matrix and the single ranking weight corresponding to the CP matrix, such as the total ranking first-level weight 0.503=0.319*0.384+0.259*0.366+0.070*0.860+0.193*0.500+0.055*0.500+0.103*0.981; the total ranking second-level weight 0.496=0.319*0.616+0.259*0.634+0.070*0.140+0.193*0.500+0.055*0.500+0.103*0.019.

[0048]

[0049] The above process yielded the PM values ​​of each pollutant. 2.5 、PM 10 The weights of SO2, NO2, CO and O3 are 0.319, 0.259, 0.070, 0.193, 0.055 and 0.103 respectively. According to the exposure-response relationship between each pollutant and the daily death data in the previous step, the effect value with a lag of Lag (0-1) is selected, and the Z-Score standardized value of each pollutant effect value is calculated. Based on this, a new comprehensive pollutant index model is constructed according to the following formula: ; Among them, P represents the new comprehensive pollutant index model, , , , , , They represent the Z-Score standardized values ​​of the moving average of the pollutant concentration on the day and the pollutant concentration lagged by 1 day. The new daily comprehensive pollutant index from 2014 to 2018 was obtained.

[0050] According to the new comprehensive pollutant indicator model, for every unit increase, the daily YLL increases by 43.5370 (95% CI: 12.0891, 74.9849) person-years.

[0051] According to the exposure-response relationship calculated above, the maximum daily YLL for the city during 2014-2018 is expected to be 187.0409 person-years. Therefore, the AQHI is calculated according to the following formula: .

[0052] This embodiment establishes an exposure-response relationship between complex atmospheric pollutants and years of life lost based on the hierarchical analysis method, and constructs an air quality health index. It not only focuses on the mass concentration of air pollutants, but also pays attention to the contribution of multiple complex atmospheric pollutants to air quality and their risks to human health. At the same time, it can solve the interactive effects of multiple complex atmospheric pollutants on health, and can comprehensively and accurately assess the health risks of air quality.

[0053] Embodiment 2: This embodiment calculates and constructs AQHI based on the atmospheric composite pollutant health impact assessment method proposed in Embodiment 1, and develops an atmospheric composite pollutant health impact early warning system.

[0054] 1. Architecture design: The health impact warning system of atmospheric composite pollutants has built an air quality and health risk forecasting platform. The platform is designed with separate development and deployment of the front-end and back-end. The front-end and back-end use Restful API for data interaction, which effectively realizes the decoupling of the front-end and back-end, and further ensures the scalability and multi-terminal service of the platform. The platform includes the basic layer, support layer, application service layer and user layer, among which the support layer and application layer are the core layers of the entire platform. The support layer mainly realizes the raw data collection, processing, structured storage and supports the automatic operation of the system. The application layer mainly realizes the multi-dimensional analysis and display of the data, and has an intuitive human-computer interaction interface. The specific architecture is as follows: Figure 3 The entire platform functional design mainly includes 5 functions: AQHI real-time update, health risk advice, AQHI trend analysis, AQHI comparative analysis and AQHI model management.

[0055] 2. Platform function realization: The platform adopts a browser / server (B / S) system architecture, uses Java as the main development language, and Spring Boot as the basic framework. It integrates sub-frameworks such as Spring MVC, Spring JPA, Spring Security, and SpringCache, and is built in conjunction with big data analysis technologies such as web crawlers, ETL, and data visualization.

[0056] 2.1. Real-time data capture and pre-processing: The calculation of AQHI requires the atmospheric pollutants (PM 2.5 、PM 10, SO2, NO2, CO, O3) concentration data. The platform requests the relevant URL resources through Spring Rest Template, grabs the data in the real-time publishing platform of ambient air quality, and parses the obtained DOM tree to finally obtain the relevant air quality data. The data of the real-time publishing platform of ambient air quality in various places are generally updated on the hour. In order to balance the real-time nature of AQHI analysis and the pressure on the target platform to grab data, the forecast platform adopts the technology of timed scheduling to update the data. The scheduling strategy is generally set to 5 minutes after the hour. If the target platform has not updated the data, it will be scheduled after an interval of 10 minutes. Spring Quartz Cron is used to implement timed triggering.

[0057] 2.2 Calculation of AQHI: The exposure-response relationship between six major air pollutants and the years of life lost due to non-accidental deaths of the population was established using the method described in Example 1, and based on this, an AQHI that conforms to the characteristics of air pollutants and population health risks was established. Through comparative analysis with AQI, it was found that the established AQHI has a good ability to predict population health risks. When calculating the real-time AQHI value, the platform smoothes the concentrations within 48 hours, 72 hours and 96 hours by the degree of influence of different pollutants. The parameters involved in the model are extracted separately into the parameter library for storage in order to quickly respond to adjustments to the model. After the model parameters are adjusted, the platform provides a convenient function of quickly recalculating the AQHI value with one click.

[0058] 2.3 Visual display of air quality and health risk level: The visualization of statistical graphics in the platform uses G2 based on the theory of graphic grammar as the underlying visualization engine. It is data-driven, provides graphic grammar and interaction grammar, and is highly easy to use and scalable. It has the following characteristics: 1. Complete graphic grammar, mapping of data to graphics, and can draw all charts. 2. The new interaction grammar can combine various interactive behaviors through trigger and feedback mechanisms to explore the data and meet the linkage analysis of the AQHI forecast platform. 3. The powerful View module can support the development of personalized data multi-dimensional analysis graphics. 4. Dual-engine rendering, arbitrary switching between Canvas or SVG, can ensure multi-terminal display.

[0059] The visualization of geospatial data combines GIS technology, uses L7 as the visualization framework, is based on WebGL, and supports high performance and 3D high-quality rendering of massive data. In order to intuitively display the AQHI data and risk levels of various areas in the city, the platform performs secondary rendering of the risk levels of different areas in 11 different colors, which can intuitively meet the spatial analysis of the AQHI forecasting platform.

[0060] 2.4. Health advice: In order to better utilize the research results of many experts in the field of public health, the platform has extracted a separate module for health risk recommendations, which is divided into four levels: low risk, medium risk, high risk and extremely high risk according to the risk level of AQHI to public health. The platform uniformly manages health recommendations for different groups of people (including patients with cardiovascular or respiratory diseases, the elderly and children, outdoor workers, and general citizens) according to different risk levels, and configures health risk recommendations at the data dictionary level. The platform dynamically displays risk levels and health recommendations for different susceptible groups based on the real-time calculated AQHI values, thereby providing personalized health tips for different groups of people.

[0061] 2.5 Trend Analysis: Trend analysis is an important means to study the changing trend of AQHI over a period of time. Combined with trend charts, it can intuitively reflect the trend of health risks. In order to study the changes in different time granularities of the city and its regions, the platform conducts AQHI trend analysis at two granularities: hours and days. At the same time, it provides intuitive trend analysis of the city and its regions within different time ranges.

[0062] 2.6. Comparative analysis: The platform provides comparative analysis of AQHI from two dimensions: region and time. Within the selected time range, a comparative analysis is conducted on any selected area, including the number of days with different risk levels in the selected area and their composition ratio. On the other hand, a trend comparative analysis of AQHI in the same period of the whole year in a specific area is conducted to intuitively reflect the air quality of a region and the effectiveness of the fresh air action plan from a more macro perspective.

[0063] The early warning system proposed in this embodiment can be integrated into a WeChat public platform or a third-party system to realize real-time forecasting of AQHI and health risks in various regions, so as to timely and effectively convey the ambient air quality level and corresponding health risk level information to the public, provide health advice for daily travel of various groups of people, and promote the forecast of health risks of urban air pollution, which has good application prospects.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for assessing the health impact of atmospheric composite pollutants, characterized in that: include: S1. Build a time series database of composite air pollutants; S2. Construct a comprehensive pollutant index model to characterize complex air pollutants through the analytic hierarchy process; S3. Constructing a model of the impact of composite air pollutants on human health based on the air pollutant time series database and the comprehensive pollutant index model, and extracting exposure-response relationship coefficients; S4. Calculate and grade the ambient air quality health index based on the exposure-response relationship coefficient.

2. The method for health impact assessment of atmospheric composite pollutants according to claim 1, characterized in that: Step S1 specifically includes: S101. Collect air pollutant data by date; S102, collecting meteorological data by date; S103. Collect population health data by date; S104. Calculate daily life loss years data; S105. Construct a time series database of complex atmospheric pollutants.

3. The method for health impact assessment of atmospheric composite pollutants according to claim 1, characterized in that: Step S2 specifically includes: S201, taking air environment quality as the target layer, taking various air pollutants as various evaluation factors of the criterion layer, taking air environment quality classification as the scheme layer, and constructing an air environment quality hierarchical model; S202, constructing a judgment matrix OC according to the air environment quality hierarchical structure model, taking the air environment quality classification standard as the benchmark, and the scale being the ratio of each evaluation factor to the secondary standard; constructing a judgment matrix CP with the reciprocal of the absolute value of the difference between each evaluation factor and the standards at each level as the scale; S204, solving the maximum eigenvalue and eigenvector of each judgment matrix, and obtaining the weight according to the eigenvector; S205, performing consistency check on the judgment matrix according to the maximum eigenvalue; and constructing a comprehensive pollutant index model for characterizing complex air pollutants according to the weights of the judgment matrix OC.

4. The method for health impact assessment of atmospheric composite pollutants according to claim 1, characterized in that: Step S3 specifically includes: S301, calculating the comprehensive pollutant index according to the comprehensive pollutant index model characterizing the complex air pollutants and the complex air pollutant time series database, and updating the complex air pollutant time series database; S302. Use time series analysis to construct a comprehensive pollutant impact model on human health based on the updated composite air pollutant time series database; extract exposure-response relationship coefficients; S303. Evaluate the impact of complex air pollutants on human health based on the exposure-response relationship coefficient.

5. The method for health impact assessment of atmospheric composite pollutants according to claim 1, characterized in that: Step S4 specifically includes: S401. Calculate the air quality health index AQHI based on the exposure-response relationship coefficient; S402. Classify the air quality health index according to the health effects of air pollution on the human body.

6. A device for assessing the health impact of atmospheric composite pollutants, characterized in that: include: Database module: build a time series database of composite air pollutants; Hierarchy analysis module: construct a comprehensive pollutant index model to characterize complex air pollutants through the hierarchical analysis method; Health model module: constructs a model of the impact of composite air pollutants on human health based on the air pollutant time series database and the comprehensive pollutant index model, and extracts the exposure-response relationship coefficient; Grading module: Calculates and grades the environmental air quality health index based on the exposure-response relationship coefficient.

7. An early warning system for health impacts of atmospheric composite pollutants, characterized in that: The air quality health index AQHI is calculated by applying the atmospheric composite pollutant health impact assessment method as described in any one of claims 1 to 5 to realize real-time AQHI value update, health risk advice, AQHI trend analysis, AQHI comparative analysis and AQHI model management functions.

8. The health impact early warning system of atmospheric composite pollutants according to claim 7 is characterized in that: The atmospheric composite pollutant health impact warning system establishes an air quality and health risk forecasting platform, adopts a B / S system architecture, takes Spring Boot as the basic framework, integrates Spring MVC, Spring JPA, Spring Security, and Spring Cache subframeworks, obtains atmospheric pollutant concentration data monitored by each air quality monitoring station, and implements regular scheduling to update data through Spring Quartz Cron timing triggering; when calculating the real-time AQHI value, the concentration within 48 hours, 72 hours, and 96 hours is smoothed by the impact degree of different pollutants; the visualization of statistical graphics adopts G2 based on the graphic grammar theory as the underlying visualization engine, which is data-driven and provides graphic grammar and interactive grammar; For geospatial data visualization, L7 is used as the visualization framework, and the AQHI data and risk level of each area are intuitively displayed based on WebGL, and the risk level of different areas is re-rendered according to different colors.

9. The health impact early warning system of atmospheric composite pollutants according to claim 7 is characterized in that: The atmospheric composite pollutant health impact early warning system extracts a separate module for health risk recommendations, divides the risk levels of AQHI to public health into four levels: low risk, medium risk, high risk and extremely high risk, uniformly manages health recommendations for different groups of people according to different risk levels, and configures health risk recommendations at the data dictionary level. It dynamically displays risk levels and health recommendations for different susceptible groups based on real-time AQHI values, and provides personalized health tips for different groups of people.

10. The health impact early warning system of atmospheric composite pollutants according to claim 7 is characterized in that: The atmospheric composite pollutant health impact early warning system performs AQHI trend analysis at different time granularities, and intuitively reflects the trend of health risks in combination with trend graphs; it also provides comparative analysis of AQHI from two dimensions, region and time.

Images