Multi-factor Collaborative Processing Method and System for Haze Environment Monitoring and Forecasting Data

Abstract

The present invention relates to the technical field of environmental data processing, and particularly to a multi-factor collaborative processing method and system for haze environment monitoring and forecasting data. The method includes the following steps: S1: Obtain pollutant data under the CUACE model, perform interpolation processing on the pollutant data and ozone data respectively to obtain hourly pollutant concentration data; S2: According to the hourly pollutant concentration data, obtain pollutant IAQI data, and trim the hourly data to obtain adjusted hourly pollutant concentration data; S3: According to the adjusted hourly pollutant concentration data, calculate the primary pollutant and AQI hourly, or calculate the primary pollutant and AQI every 24 hours, and use the genetic algorithm to adjust the primary pollutant and AQI. By using the genetic algorithm to dynamically adjust the ranking of primary pollutants, the present invention adapts to the pollution characteristics under special weather conditions and improves the accuracy and practicability of forecasting.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental data processing, and particularly to a multi-factor collaborative processing method and system for haze environment monitoring and forecasting data. Background Art

[0002] In recent years, with the acceleration of the industrialization process and the increase in transportation, the problem of air pollution has gradually intensified. Especially in some big cities and industrial areas, haze weather occurs frequently, seriously affecting the environmental quality and human health. To solve this problem, various air quality monitoring and forecasting systems have been proposed one after another. These systems usually rely on AQI to evaluate and predict the degree of air pollution.

[0003] Traditional air quality monitoring methods usually rely on fixed models and data processing technologies. For example, pollutant concentration data is obtained through physicochemical sensors, and the data is processed through linear interpolation or simple regression analysis. However, these traditional methods have the following deficiencies: the interpolation processing of pollutant data by traditional methods is usually too simple, without considering the characteristic differences of different pollutants, resulting in inaccurate hourly pollutant concentration data; existing methods are mostly general models and fail to be customized and optimized according to the pollution source distribution and meteorological conditions in the region, resulting in poor adaptability of the system in different regions. Summary of the Invention

[0004] In order to overcome the shortcoming that the primary pollutant is solidified in different seasons, the present invention provides a multi-factor collaborative processing method and system for haze environment monitoring and forecasting data.

[0005] The technical solution of the present invention is: a multi-factor collaborative processing method for haze environment monitoring and forecasting data, including the following steps:

[0006] S1: Obtain pollutant data under the CUACE model, and perform interpolation processing on the pollutant data respectively to obtain hourly pollutant concentration data;

[0007] S2: According to the hourly pollutant concentration data, obtain pollutant IAQI data, and trim the hourly data to obtain adjusted hourly pollutant concentration data;

[0008] S3: According to the adjusted hourly pollutant concentration data, calculate the primary pollutant and AQI hourly, or calculate the primary pollutant and AQI every 24 hours, and use the genetic algorithm to adjust the primary pollutant and AQI;

[0009] S4: After adjusting the primary pollutant index in the grid data, obtain the grid primary pollutant index.

[0010] Preferably, obtaining the pollutant data in the CUACE mode, and respectively performing interpolation processing on the pollutant data to obtain hourly pollutant concentration data, including:

[0011] When the pollutant is ozone data, collect the ozone concentration data at 8 time points every 3 hours, denoted as (y1, y2, y8); calculate the moving average of the ozone concentration data at three consecutive time points. For each time period, select the concentration data at three time points before and after this time period and calculate their average value; take the time period with the largest moving average value during the 8-hour period as the time period with the largest average concentration; according to the largest average concentration, use the first interpolation formula to obtain the first time point concentration assignment, and assign the concentration data at other time points every 3 hours except this time period according to the first time point concentration assignment; for other time points without data, assign them as the first time point concentration assignment or the minimum value of the concentration data at each time point in the time period with the largest moving average value; obtain the hourly pollutant concentration data of ozone, where the first interpolation formula is:

[0012]

[0013] In the formula, K1 is the first time point concentration assignment; is the largest average concentration; c i is the ozone concentration data at the i-th time point; q is the starting time point, and its range is [1, 6].

[0014] Preferably, obtaining the pollutant data in the CUACE mode, and respectively performing interpolation processing on the pollutant data to obtain hourly pollutant concentration data, including:

[0015] When the pollutant is not ozone data, collect the pollutant concentration data at 8 time points every 3 hours, and obtain the 24-hour average concentration of each pollutant according to the deduced pollutant concentration data. According to the average concentration and the pollutant concentration data, use the second interpolation formula to obtain the second time point concentration assignment, and assign the other 16 time points except the time points every 3 hours according to the second time point concentration assignment to obtain the hourly pollutant concentration data of other pollutants, where the second interpolation formula is:

[0016]

[0017] In the formula, K2 is the second time point concentration assignment, x is the 24-hour average concentration of each pollutant, i is the i-th time point in the time points every 3 hours, c i is the pollutant concentration data at the i-th time point, and n is the total number of time points every 3 hours.

[0018] Preferably, obtaining pollutant IAQI data based on the hourly pollutant concentration data and trimming the hourly data includes: performing inverse calculation using the rounding method based on the pollutant IAQI data to obtain the inversely calculated pollutant concentration; and comparing the inversely calculated pollutant concentration with the average value of the hourly pollutant concentration data, and modifying the pollutant concentration at 20:00 when the inversely calculated pollutant concentration is not equal to the average value of the hourly pollutant concentration data.

[0019] Preferably, calculating the primary pollutant and AQI hourly or every 24 hours based on the adjusted hourly pollutant concentration data, and adjusting the primary pollutant and AQI using a genetic algorithm includes: using the O3 / 1h calculation method when calculating the primary pollutant and AQI hourly, and using the O3 / 8h calculation method when calculating the primary pollutant and AQI every 24 hours. Among them, when calculating the primary pollutant and AQI every 24 hours, the O3 / 8h calculation method is not used for the first 7 hourly periods of each daily time period, and the first 7 hourly periods or hourly periods exceeding the forecast time period are supplemented with 9999.

[0020] Preferably, calculating the primary pollutant and AQI hourly based on the adjusted hourly pollutant concentration data includes: taking ozone as the primary pollutant under seasonal background; using a genetic algorithm to optimize the order of PM2.5, PM10, ozone, SO2, CO, and NO2 under non-seasonal background, and dynamically adjusting the priority according to the optimization result of the genetic algorithm under special weather conditions.

[0021] Preferably, using a genetic algorithm to optimize the order of PM2.5, PM10, ozone, SO2, CO, and NO2 under non-seasonal background, and dynamically adjusting the priority according to the optimization result of the genetic algorithm under special weather conditions includes: obtaining real-time pollutant concentration data, historical data of AQI, and the pollution source distribution characteristics of the target area; encoding the priority of the primary pollutant as a chromosome in sequence form, each chromosome contains all pollutants, and the position of the gene represents the priority of the pollutant; defining a fitness function, and optimizing the priority ranking of the primary pollutant based on the actual impact correlation of the pollutant on AQI, the public health risk index, and the pollution source distribution characteristics of the target area; initializing the population, and randomly generating several chromosomes as the initial priority sequence; using the roulette wheel selection or tournament selection method to select chromosomes from the population according to the fitness value; using the partially mapped crossover method to generate a new generation of chromosomes; performing a mutation operation on the chromosomes, randomly swapping the positions of two pollutants; after the fitness function value converges or reaches the preset number of iterations, output the optimal priority ranking.

[0022] Preferably, optimizing the priority ranking of the primary pollutants based on the actual impact correlation of pollutants on AQI, the public health risk index, and the pollution source distribution characteristics of the target area includes: maximizing the correlation between the adjusted priority and the actual AQI change trend; minimizing the public health risk corresponding to the adjusted priority; maximizing the consistency between the priority and the pollution source characteristics of the target area; obtaining the proportion of the main pollution sources in the target area and making a weighted adjustment to the priority of the primary pollutants according to their impacts; simulating the diffusion of pollutants at different time periods based on the meteorological conditions of the target area to optimize the priority ranking; analyzing the historical air quality data of the target area and adjusting the priority according to the dominant pollutants and their impacts on health risks.

[0023] Preferably, after adjusting the primary pollutant index in the grid data, the grid primary pollutant index is obtained, including:

[0024] Adjusting the primary pollutant data of each grid using the grid adjustment formula, where the grid adjustment formula is:

[0025] I adjusted = I * μP;

[0026] In the formula, I adjusted is the adjusted grid primary pollutant index; μ is the adjustment factor; P is the adjustment weight factor; I is the unadjusted grid primary pollutant index;

[0027] Obtaining the adjustment weight factor using the weight adjustment formula, where the weight adjustment formula is:

[0028]

[0029] In the formula, E i is the primary pollutant data of the i-th grid; ∑E is the total sum of the primary pollutant concentration data on all grids; W loc is the geographical attribute weight of the grid location; W hist is the historical proportion weight of the primary pollutant concentration data at the grid location;

[0030] Obtaining the historical proportion weight of the primary pollutant concentration data at the grid location using the historical proportion weight formula, where the historical proportion weight formula is:

[0031]

[0032] In the formula, W hist is the historical proportion weight of the primary pollutant concentration data at the grid location; r1 is the actual primary pollutant concentration data; r0 is the standard primary pollutant concentration data of historical experience; ρ is the adjustment factor.

[0033] Preferably, a multi-factor collaborative processing system for haze environment monitoring and forecasting data includes:

[0034] A data acquisition and preprocessing module, which is used to obtain pollutant data under the CUACE model, perform interpolation processing on ozone and other pollutants respectively, and obtain hourly pollutant concentration data;

[0035] A pollutant index calculation and trimming module, which is used to calculate the pollutant IAQI according to the hourly pollutant concentration data, trim the hourly data, and obtain hourly pollutant concentration data;

[0036] A primary pollutant and AQI calculation module, which is used to calculate the primary pollutant and AQI hourly or every 24 hours, and optimize the primary pollutant and AQI using a genetic algorithm;

[0037] A dynamic adjustment and optimization module, which adjusts the priority of the primary pollutant based on seasonal background and special weather conditions, and dynamically optimizes the sorting of the primary pollutant;

[0038] A grid data adjustment module, which adjusts the primary pollutant index in the grid data to obtain the grid primary pollutant index.

[0039] The beneficial effects are as follows: Through the differential design of the interpolation processing of different pollutants, and by combining the moving average method and the interpolation formula, the present invention significantly improves the accuracy of pollutant concentration data.

[0040] In particular, for ozone concentration data, an interpolation method based on the maximum moving average value is adopted, which ensures the smoothness and accuracy of ozone concentration data during hourly calculation.

[0041] The system optimizes the priority ranking of the primary pollutant by analyzing the pollution source distribution characteristics and meteorological conditions of the target area. By weighted adjustment of the priority, the system can effectively predict and respond to pollutant changes in different regions, and has strong regional adaptability; at the same time, the system combines the historical data and real-time monitoring data of the region to achieve an accurate modeling of the impact of pollutants on the AQI.

[0042] The present invention considers the public health risk index in the priority ranking of pollutants, ensures that the harm of pollutants to health is preferentially evaluated and incorporated into system optimization; through the optimized pollutant priority, the system can better provide decision support for air quality management and public health risk control.

[0043] By using the grid adjustment formula and multi-layer weight factors, such as geographical attributes and historical concentration weights, the spatial resolution and accuracy of grid data are enhanced, which helps improve the monitoring and prediction effects in a large area. Introducing an optimization method based on the characteristics of the target area, such as pollution source distribution and meteorological conditions, can achieve personalized adaptation for different regions and provide more refined data support for air quality management and decision-making. Brief Description of the Drawings

[0044] Figure 1 It is a flowchart of the multi-element collaborative processing method for haze environment monitoring and forecasting data of the present invention;

[0045] Figure 2 It is a schematic structural diagram of the multi-element collaborative processing system for haze environment monitoring and forecasting data of the present invention. Detailed Embodiments

[0046] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0047] Embodiment 1: A multi-element collaborative processing method for haze environment monitoring and forecasting data, as Figure 1 - Figure 2 shown, includes the following steps:

[0048] S1: Obtain pollutant data under the CUACE model, and perform interpolation processing on the pollutant data respectively to obtain hourly pollutant concentration data;

[0049] When the pollutant is ozone data, collect ozone concentration data at 8 time points every 3 hours, denoted as (y1, y2, y8); calculate the moving average value of ozone concentration data for three consecutive time points. Among them, for each time period, select the concentration data of three time points before and after this time period and calculate their average value; take the time period with the largest moving average value during the 8-hour period as the time period with the largest average concentration; according to the maximum average concentration, use the first interpolation formula to obtain the concentration assignment for the first time point, and assign the concentration data of other every-3-hour time points except this time period according to the concentration assignment for the first time point; assign the concentration assignment for the first time point or the minimum value of the concentration data of each time point in the time period with the largest moving average value to other time points without data; obtain the hourly pollutant concentration data of ozone, where the first interpolation formula is:

[0050]

[0051] In the formula, K1 is the concentration assignment for the first time point; is the maximum average concentration; c i is the ozone concentration data at the i-th time; q is the starting time, and its range is [1, 6].

[0052] It should be noted that for data collection: during the monitoring process, the ozone concentration data is recorded once every 3 hours, for a total of 8 time periods, which are respectively recorded as (y1, y2,..., y8). These 8 time period data represent the discontinuous sampling results within 24 hours; for each time period, the ozone concentration data of three consecutive time periods is selected, and its moving average value is calculated. The formula is: The calculated moving average value reflects the change trend of the ozone concentration in each time period. In the formula, y i is the i-th time, AVG is the moving average value; among the 6 moving average values, the maximum value is found, that is, the time period corresponding to the highest ozone concentration; this value will be used as the core parameter for interpolation processing, reflecting the peak time period of ozone pollution; through the maximum moving average value combined with the ozone concentration values of the corresponding three time periods, the first concentration value at the first time is calculated using the first interpolation formula; for the concentration values at other times except the time period corresponding to the maximum moving average value, they are assigned values according to the following rules. If there is no data at a certain time, it is directly assigned as K1; or the minimum value in the time period corresponding to the maximum moving average value is selected for assignment; finally, the hourly ozone concentration data within 24 hours is generated, covering all time periods, providing a complete concentration change trend for subsequent analysis.

[0053] When the pollutant is not ozone data, the pollutant concentration data for 8 time periods is collected every 3 hours, and based on the calculated pollutant concentration data, the 24-hour average concentration of each pollutant is obtained. According to the average concentration and the pollutant concentration data, the second concentration value at the second time is obtained using the second interpolation formula. According to the second concentration value at the second time, the other 16 time periods except the every-3-hour time periods are assigned values to obtain the hourly pollutant concentration data of other pollutants. The second interpolation formula is:

[0054]

[0055] In the formula, K2 is the second concentration value at the second time, x is the 24-hour average concentration of each pollutant, i is the i-th time in the every-3-hour time periods, c i is the pollutant concentration data at the i-th time, and n is the total number of every-3-hour time periods.

[0056] It should be noted that the 24-hour average concentration x is calculated through the 8 time period data. The 24-hour average concentration is combined with the 8 every-3-hour data, and the second interpolation formula is used to calculate the concentration at the other 16 time periods except the every-3-hour time periods. The original concentration values at the every-3-hour sampling time periods are retained, and for the other 16 non-sampled time periods, they are uniformly assigned as K2 to complete the hourly data filling.

[0057] S2: Obtain the pollutant IAQI data based on the hourly pollutant concentration data, and trim the hourly data to obtain the adjusted hourly pollutant concentration data;

[0058] Inverse calculate the pollutant concentration using the rounding method based on the pollutant IAQI data, and compare the inversely calculated pollutant concentration with the average value of the hourly pollutant concentration data. When the inversely calculated pollutant concentration is not equal to the average value of the hourly pollutant concentration data, modify the pollutant concentration at 20:00.

[0059] It should be noted that by using the hourly pollutant concentration data, the IAQI value of the pollutant is obtained through the calculation formula of the air quality sub-index of the pollutant. The calculation of IAQI follows the relevant national standards. The concentration value of the pollutant is re-estimated by inverse calculation using the rounding method based on the IAQI value, which is called the inverse calculation concentration. The inversely calculated pollutant concentration value is compared with the average value of the hourly pollutant concentration data. When the inversely calculated pollutant concentration is not equal to the average value of the hourly concentration, adjust the pollutant concentration value at the 20th hour to make the data more in line with the actual distribution trend of the pollutant concentration.

[0060] S3: According to the adjusted hourly pollutant concentration data, calculate the primary pollutant and AQI hourly, or calculate the primary pollutant and AQI every 24 hours, and use the genetic algorithm to adjust the primary pollutant and AQI;

[0061] When calculating the primary pollutant and AQI hourly, use the O3 / 1h calculation method. When calculating the primary pollutant and AQI every 24 hours, use the O3 / 8h calculation method. Among them, when calculating the primary pollutant and AQI every 24 hours, do not use the O3 / 8h calculation method for the first 7 time periods of each daily time effect, and supplement the first 7 time periods or the time periods beyond the forecast time effect with 9999.

[0062] It should be noted that when calculating hourly, determine the hourly primary pollutant and air quality index according to the concentration of each pollutant and the corresponding air quality sub-index. The methods for calculating the primary pollutant include: the 1-hour average value of O3 (ozone), called O3 / 1h. When calculating every 24 hours, calculate the daily primary pollutant and AQI according to the 8-hour moving average of ozone (O3 / 8h). For the first 7 time periods of each daily time effect, do not use O3 / 8h for calculation, but supplement these time periods or the time periods beyond the forecast time effect with 9999 to reduce the impact on the accuracy of the 24-hour moving average. Use the genetic algorithm to adjust the sorting order of the primary pollutants, where the primary pollutant is the pollutant with the largest IAQI value in that time period.

[0063] Obtain real-time pollutant concentration data, historical data of AQI, and the distribution characteristics of pollution sources in the target area; encode the priority of the primary pollutant as a chromosome in sequence form, where each chromosome contains all pollutants, and the position of the gene represents the priority of the pollutant; define a fitness function, and optimize the priority ranking of the primary pollutant based on the actual impact correlation of the pollutant on AQI, the public health risk index, and the distribution characteristics of pollution sources in the target area; initialize the population by randomly generating several chromosomes as the initial priority sequence; use the roulette wheel selection or tournament selection method to select chromosomes from the population according to the fitness value; adopt the partially mapped crossover method to generate a new generation of chromosomes; perform a mutation operation on the chromosomes by randomly swapping the positions of two pollutants; after the fitness function value converges or reaches the preset number of iteration generations, output the optimal priority ranking.

[0064] It should be noted that the primacy of ozone under seasonal background: In a specific season (May - September), due to high temperature and strong solar radiation, the ozone concentration is usually high and has a significant impact on air quality. Therefore, under such seasonal background, ozone is directly set as the primary pollutant without further adjustment; dynamic adjustment under non-seasonal background: In other seasons (such as autumn or winter), the contribution of ozone to AQI is small, and the original fixed sorting order is PM2.5 > PM10 > ozone > other pollutants, where the other pollutants include SO2, CO, and NO2. The priority of each pollutant needs to be optimized through a genetic algorithm; optimize the pollutant order: including PM2.5, PM10, ozone, and other pollutants, comprehensively consider the actual impact of each pollutant, and adjust their priority order according to the optimization results; dynamic adjustment under special weather conditions: Under special weather conditions such as sandstorms, heavy rainfall, or haze, dynamically adjust the priority of the primary pollutant according to the weather characteristics. For example, give priority to PM10 in sandstorm weather.

[0065] Take ozone as the primary pollutant under seasonal background; use a genetic algorithm to optimize the order of PM2.5, PM10, ozone, SO2, CO, and NO2 under non-seasonal background, and dynamically adjust the priority under special weather conditions according to the optimization results of the genetic algorithm.

[0066] It should be noted that to obtain real-time pollutant concentration data (such as PM2.5, PM10, ozone), historical data of AQI, and the source distribution characteristics of the target area (such as industrial emissions, vehicle exhaust, natural sources), encode the chromosome: Encode the pollutant priorities into chromosomes, where each chromosome represents an order of pollutant priorities, and the position of the gene represents the priority of the pollutant. For example, the chromosome PM2.5, PM10, ozone, other pollutants means that PM2.5 has the highest priority and other pollutants have the lowest priority. Define the fitness function, comprehensively considering the actual impact of pollutants on AQI, the public health risk index (HR index), and the source distribution characteristics of the target area; Population initialization: Randomly generate several chromosomes to form an initial population, and each chromosome is a sequence of pollutant priorities; Selection operation: Use the roulette wheel selection or tournament selection method, and select chromosomes with high fitness values to enter the next generation according to the fitness values of the chromosomes; Use the partially mapped crossover (PMX) method to cross the gene sequences of two chromosomes to generate a new generation of chromosomes. For example: Chromosome 1: PM2.5, ozone, PM10, other; Chromosome 2: ozone, PM10, other, PM2.5; Crossover result: PM2.5, PM10, other, ozone; Mutation operation: Randomly exchange the positions of two genes in the chromosome. For example, mutate PM2.5, PM10, ozone, other to: PM2.5, ozone, PM10, other; Iteration and convergence: Repeat the selection, crossover, and mutation operations until the fitness value converges or reaches the preset number of iterations, and output the optimal chromosome; Dynamic adjustment: Under special weather conditions (such as sandstorms, heavy rainfall), dynamically adjust the priority of the primary pollutant based on the optimization results.

[0067] Maximize the correlation between the adjusted priority and the actual AQI change trend; Minimize the public health risk corresponding to the adjusted priority; Maximize the consistency between the priority and the source characteristics of the target area; Obtain the proportion of the main pollution sources in the target area, and make a weighted adjustment to the priority of the primary pollutant according to its impact; Simulate the diffusion of pollutants at different time periods based on the meteorological conditions of the target area to optimize the priority ranking; Analyze the historical air quality data of the target area, and adjust the priority according to the dominant pollutants and their impact on health risks.

[0068] It should be noted that: maximizing the correlation between the adjusted priority and the actual AQI change trend: by optimizing the sorting of the primary pollutants, the adjusted priority can more accurately reflect the change trend of the AQI; minimizing the public health risk: according to the degree of harm of different pollutants to health, giving priority to the pollutants with a high health risk index; maximizing the matching degree with the characteristics of the pollution sources in the target area: by analyzing the proportion of the main pollution sources in the target area, assigning higher priorities to the high-impact pollutants. According to the impact of each pollutant on the AQI, the health risk index, and the matching degree with the characteristics of the pollution sources, the priority is dynamically adjusted by a weighting method.

[0069] S4: After adjusting the primary pollutant index in the grid data, the grid primary pollutant index is obtained.

[0070] Use the grid adjustment formula to adjust the primary pollutant data of each grid, where the grid adjustment formula is:

[0071] I adjusted = I * μP;

[0072] In the formula, I adjusted is the adjusted grid primary pollutant index; μ is the adjustment factor; P is the adjustment weight factor; I is the unadjusted grid primary pollutant index;

[0073] Use the weight adjustment formula to obtain the adjustment weight factor, where the weight adjustment formula is:

[0074]

[0075] In the formula, E i is the primary pollutant data of the i-th grid; ∑E is the total sum of the primary pollutant concentration data on all grids; W loc is the geographical attribute weight of the grid position; W hist is the historical proportion weight of the primary pollutant concentration data at the grid position;

[0076] Use the historical proportion weight formula to obtain the historical proportion weight of the primary pollutant concentration data at the grid position, where the historical proportion weight formula is:

[0077]

[0078] In the formula, W hist is the historical proportion weight of the primary pollutant concentration data at the grid position; r1 is the actual primary pollutant concentration data; r0 is the standard primary pollutant concentration data of historical experience; ρ is the adjustment factor.

[0079] It should be noted that, first, the acquisition and initialization of the grid's primary pollutant data are carried out. The grid data of the target area are obtained from the monitoring network, including: the primary pollutant data of each grid, the total sum of the primary pollutant concentration data of all grids in the area, the location attributes and their geographical weights of each grid, and the historical primary pollutant concentration weights of each grid. By using the grid adjustment formula to adjust the primary pollutant index of each grid, the actual pollution situation of the grid can be accurately reflected while ensuring the coordination of data among grids. The weight adjustment formula is used to calculate the adjustment weight factor, and the importance of each grid's data is comprehensively weighed through the pollutant concentration ratio, geographical location attributes, and historical pollution characteristics. The historical proportion weight formula is used to quantify the impact of historical pollution data on the adjustment of the current grid index, ensuring that historical characteristics are fully reflected in the adjustment process.

[0080] Embodiment 2: On the basis of Embodiment 1, a multi-element collaborative processing system for haze environment monitoring and forecasting data includes:

[0081] The data acquisition and preprocessing module is used to obtain pollutant data in the CUACE model, perform interpolation processing on ozone and other pollutants respectively, and obtain hourly pollutant concentration data.

[0082] The pollutant index calculation and trimming module is used to calculate the pollutant IAQI according to the hourly pollutant concentration data, and trim the hourly data to obtain hourly pollutant concentration data.

[0083] The primary pollutant and AQI calculation module is used to calculate the primary pollutant and AQI hourly or every 24 hours, and optimize the primary pollutant and AQI using the genetic algorithm.

[0084] The dynamic adjustment and optimization module adjusts the priority of the primary pollutant based on the seasonal background and special weather conditions, and dynamically optimizes the ranking of the primary pollutant.

[0085] The grid data adjustment module adjusts the primary pollutant index in the grid data to obtain the grid primary pollutant index.

[0086] As mentioned above, the above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A multi-factor collaborative processing method for haze environment monitoring and forecasting data, characterized in that It includes the following steps: S1: Obtain pollutant data under the CUACE model, perform interpolation processing on the pollutant data respectively, and obtain hourly pollutant concentration data; S2: According to the hourly pollutant concentration data, obtain pollutant IAQI data, and trim the hourly data to obtain adjusted hourly pollutant concentration data; S3: According to the adjusted hourly pollutant concentration data, calculate the primary pollutant and AQI hourly, or calculate the primary pollutant and AQI every 24 hours, and use the genetic algorithm to adjust the primary pollutant and AQI; S4: After adjusting the primary pollutant index in the grid data, obtain the grid primary pollutant index; The obtaining of the pollutant data under the CUACE model and the performing of interpolation processing on the pollutant data respectively to obtain hourly pollutant concentration data include: When the pollutant is ozone data, collect ozone concentration data at 8 time points every 3 hours, denoted as (y1, y2, ……, y8); calculate the moving average of ozone concentration data at three consecutive time points. Among them, for each time period, select the concentration data at three time points before and after this time period and calculate their average value; take the time period with the largest moving average value during the 8-hour period as the time period with the largest average concentration; according to the largest average concentration, use the first interpolation formula to obtain the concentration assignment for the first time point, and assign the concentration data at other time points every 3 hours except this time period according to the concentration assignment for the first time point; assign the value for other time points without data as the concentration assignment for the first time point or the minimum value of the concentration data at each time point during the time period with the largest moving average value; obtain the hourly pollutant concentration data of ozone, where the first interpolation formula is: Wherein, K1 is the concentration assignment at the first time instance; is the maximum average concentration; c i is the ozone concentration data at the i-th time instance; q is the starting time instance, and its range is [1, 6].

2. The multi-element collaborative processing method for haze environment monitoring and forecasting data according to claim 1, characterized in that, The obtaining of the pollutant data under the CUACE model and the performing of interpolation processing on the pollutant data respectively to obtain hourly pollutant concentration data include: When the pollutant is not ozone data, collect pollutant concentration data at 8 time points every 3 hours, and obtain the 24-hour average concentration of each pollutant according to the deduced pollutant concentration data. According to the average concentration and the pollutant concentration data, use the second interpolation formula to obtain the concentration assignment for the second time point, and assign the values for the other 16 time points except the time points every 3 hours according to the concentration assignment for the second time point, to obtain the hourly pollutant concentration data of other pollutants, where the second interpolation formula is: Wherein, K2 is the concentration assignment at the second time, x is the 24-hour average concentration of each pollutant, i is the i-th time among the three-hour intervals, and c i is the pollutant concentration data at the i-th time, and n is the total number of three-hour intervals.

3. A multi-element collaborative processing method for haze environment monitoring and forecasting data according to claim 1, characterized in that, The obtaining of the pollutant IAQI data according to the hourly pollutant concentration data and the trimming of the hourly data include: perform back-calculation using the rounding method according to the pollutant IAQI data to obtain the back-calculated pollutant concentration; and compare the back-calculated pollutant concentration with the average value of the hourly pollutant concentration data. When the back-calculated pollutant concentration is not equal to the average value of the hourly pollutant concentration data, modify the pollutant concentration at 20:

00.

4. A multi-factor collaborative processing method for haze environment monitoring and forecasting data according to claim 1, characterized in that, Calculating the primary pollutant and AQI hourly or every 24 hours according to the adjusted hourly pollutant concentration data, and using the genetic algorithm to adjust the primary pollutant and AQI, including: when calculating the primary pollutant and AQI hourly, using the O3 / 1h calculation method; when calculating the primary pollutant and AQI every 24 hours, using the O3 / 8h calculation method. Among them, when calculating the primary pollutant and AQI every 24 hours, the O3 / 8h calculation method is not used for the first 7 time periods of each daily time effect, and the first 7 time periods or the time periods exceeding the forecast time effect are supplemented with 9999.

5. The multi-factor collaborative processing method for haze environment monitoring and forecasting data according to claim 4, wherein Calculating the primary pollutant and AQI hourly according to the adjusted hourly pollutant concentration data, including: taking ozone as the primary pollutant under seasonal background; using the genetic algorithm to optimize the order of PM2.5, PM10, ozone, SO2, CO, and NO2 under non-seasonal background, and dynamically adjusting the priority according to the optimization result of the genetic algorithm under special weather conditions.

6. The multi-factor collaborative processing method for haze environment monitoring and forecasting data according to claim 5, characterized in that Using the genetic algorithm to optimize the order of PM2.5, PM10, ozone, SO2, CO, and NO2 under non-seasonal background, and dynamically adjusting the priority according to the optimization result of the genetic algorithm under special weather conditions, including: Obtaining the real-time pollutant concentration data, historical data of AQI, and the pollution source distribution characteristics of the target area; encoding the priority of the primary pollutant as a chromosome in sequence form, each chromosome contains all pollutants, and the position of the gene represents the priority of the pollutant; defining the fitness function, and optimizing the priority ranking of the primary pollutant based on the actual impact correlation of the pollutant on AQI, the public health risk index, and the pollution source distribution characteristics of the target area; initializing the population, and randomly generating several chromosomes as the initial priority sequence; using the roulette wheel selection or tournament selection method to select chromosomes from the population according to the fitness value; using the partially mapped crossover method to generate a new generation of chromosomes; performing mutation operations on the chromosomes, randomly exchanging the positions of two pollutants; after the fitness function value converges or reaches the preset number of iterations, outputting the optimal priority ranking.

7. A multi-factor collaborative processing method for haze environment monitoring and forecasting data according to claim 6, characterized in that Optimizing the priority ranking of the primary pollutant based on the actual impact correlation of the pollutant on AQI, the public health risk index, and the pollution source distribution characteristics of the target area, including: maximizing the correlation between the adjusted priority and the actual AQI change trend; minimizing the public health risk corresponding to the adjusted priority; maximizing the consistency between the priority and the pollution source characteristics of the target area; obtaining the proportion of the main pollution sources in the target area, and making weighted adjustments to the priority of the primary pollutant according to its impact; simulating the diffusion of pollutants in different time periods based on the meteorological conditions of the target area, and optimizing the priority ranking; analyzing the historical air quality data of the target area, and adjusting the priority according to the dominant pollutant and its impact on health risks.

8. A multi-factor collaborative processing method for haze environment monitoring and forecasting data according to claim 1, characterized in that After adjusting the primary pollutant index in the grid data, obtaining the grid primary pollutant index, including: Adjust the primary pollutant data of each grid point using the grid point adjustment formula, where the grid point adjustment formula is: I adjusted = I * μP; Where I adjusted is the adjusted grid primary pollutant index; μ is the adjustment factor; P is the adjusted weight factor; I is the unadjusted grid primary pollutant index; Obtain the adjustment weight factor using the weight adjustment formula, where the weight adjustment formula is: where E i is the primary pollutant data of the i-th grid point; ∑E is the sum of the primary pollutant concentration data at all grid points; W loc is the geographical attribute weight of the grid point location; W hist is the historical proportion weight of the primary pollutant concentration data at the grid point location; Obtain the historical proportion weight of the primary pollutant concentration data at the grid point position using the historical proportion weight formula, where the historical proportion weight formula is: Where W hist is the historical proportion weight of the primary pollutant concentration data at the grid point position; r1 is the actual primary pollutant concentration data; r0 is the standard primary pollutant concentration data of historical experience; and ρ is the adjustment factor.

9. A multi-factor collaborative processing system for haze environment monitoring and forecasting data, which is executed according to the multi-factor collaborative processing method for haze environment monitoring and forecasting data described in any one of claims 1-8, and is characterized in that, Include: A data collection and preprocessing module, which is used to obtain pollutant data under the CUACE model, perform interpolation processing on ozone and other pollutants respectively, and obtain hourly pollutant concentration data; A pollutant index calculation and trimming module, which is used to calculate the pollutant IAQI based on the hourly pollutant concentration data, trim the hourly data, and obtain the hourly pollutant concentration data; A primary pollutant and AQI calculation module, which is used to calculate the primary pollutant and AQI hourly or every 24 hours, and optimize the primary pollutant and AQI using the genetic algorithm; A dynamic adjustment and optimization module, which adjusts the priority of the primary pollutant based on the seasonal background and special weather conditions, and dynamically optimizes the sorting of the primary pollutant; A grid data adjustment module, which obtains the grid primary pollutant index after adjusting the primary pollutant index in the grid data.

Images