Environmental quality assessment method based on three-dimensional grid

Abstract

The embodiment of the invention provides an environment quality evaluation method based on a three-dimensional grid, and is applied to the technical field of environment quality evaluation and monitoring. The method comprises the following steps: acquiring an atmospheric environment measured data sequence of a target grid at the current moment; calculating an air quality index at the current moment according to the measured data sequence of the atmospheric environment at the current moment; determining an environment quality state at the current moment according to the air quality index at the current moment; comparing the environment quality state at the current moment with the environment quality probability sequence at the current moment; the environment quality probability sequence at the current moment is obtained by carrying out environment quality probability prediction according to the environment quality state at the previous moment and diffusion influence data of the target grid; and if the environment quality state at the current moment falls into the maximum probability range in the environment quality probability sequence at the current moment, predicting the environment quality probability sequence at the next moment according to the environment quality probability sequence at the current moment. In this way, the accuracy of environment quality assessment can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of environmental quality assessment and monitoring, and in particular to an environmental quality assessment method based on a three-dimensional grid. Background Art

[0002] Environmental planning is an important part of national land space planning, and its purpose is to rationally utilize and protect environmental resources and ensure the sustainable development of human society. However, traditional environmental planning assessment methods have deficiencies in data acquisition, indicator allocation, and project approval data association, and lack effective technical means to achieve rapid and accurate environmental planning assessments. The spatial expression of risk sources in different environmental media is directly related to the risks that may be caused under the environmental media. For the risks brought by environmental pollution, it is necessary to select appropriate spatial expressions based on the influencing factors of the diffusion process of pollutants in specific environmental media. The commonly used expressions are mainly as follows:

[0003] 1. Spatial contour map: The pollution concentration or risk level of different areas is displayed in the form of contour lines, and the density and distribution of contour lines are used to indicate the diffusion range and degree of pollutants in space. In this way, data is generally organized through shape files. After the data is calculated once, the changed values ​​need to be recalculated. The related data needs to be obtained through complex data exchange and query before calculation, which cannot support the fast computing requirements of current artificial intelligence.

[0004] 2. Spatial statistical analysis: Use spatial statistical methods, such as spatial autocorrelation analysis and spatial kriging analysis, to explore spatial correlation and variability, thereby revealing the laws and trends of pollutant diffusion. This expression data organization is stored through raster storage, which does not support subsequent data changes and modifications, and cannot effectively support the diffusion deduction of pollutants.

[0005] The spatial representation of pollutant distribution and diffusion calculated through spatial contour maps and spatial statistical analysis has nothing to do with the original data structure. The original data is still stored in different relational databases and cannot be effectively mapped to groundwater indicators, solid waste generation indicators, and atmospheric emission indicators. These data are stored in different carriers in structured and unstructured forms. When it is necessary to conduct research and judgment, there are the following deficiencies:

[0006] 1. These data displays need to be superimposed and manually analyzed, which brings difficulties to decision-making in the environmental assessment and approval of the project.

[0007] 2. When superimposing data, various types of business data of the corresponding risk sources are directly operated. When other business systems, such as the atmospheric ecological environment supervision system and the soil ecological environment supervision system, also need to access the pollution source monitoring data, data access deadlock will occur, or the access to the pollution source data will be insecure.

[0008] 3. The deductive calculation of data requires querying and reading data from different storages and systems, which makes the system calculation complex and reduces efficiency. Summary of the invention

[0009] The present invention provides an environmental quality assessment method based on a three-dimensional grid.

[0010] According to a first aspect of the present disclosure, a method for evaluating environmental quality based on a three-dimensional grid is provided. The method comprises:

[0011] Obtain the atmospheric environment measured data sequence of the target grid at the current moment;

[0012] Calculate the air quality index at the current moment according to the atmospheric environment measured data sequence at the current moment;

[0013] Determine the current environmental quality status according to the current air quality index;

[0014] Comparing the environmental quality state at the current moment with the environmental quality probability sequence at the current moment; the environmental quality probability sequence at the current moment is obtained by predicting the environmental quality probability based on the environmental quality state at the previous moment and the diffusion impact data of the target grid;

[0015] If the environmental quality state at the current moment falls within the maximum probability range of the environmental quality probability sequence at the current moment, the environmental quality probability sequence at the next moment is predicted based on the environmental quality probability sequence at the current moment.

[0016] According to the above aspects and any possible implementation, an implementation is further provided, in which an environmental quality probability sequence at a current moment is obtained by performing environmental quality probability prediction based on the environmental quality state and diffusion impact data at a previous moment, including:

[0017] According to the environmental quality state at the previous moment, based on a preset probability distribution, determine the environmental quality probability sequence at the previous moment;

[0018] According to the diffusion impact data, the environmental quality probability sequence at the previous moment is updated;

[0019] The environmental quality probability sequence at the current moment is calculated based on the updated environmental quality probability sequence at the previous moment.

[0020] According to the above aspects and any possible implementation, an implementation is further provided.

[0021] The diffusion impact data include surface roughness grade index, terrain slope grade index, linear corridor and regional wind grade index;

[0022] The updating of the environmental quality probability sequence at the previous moment according to the diffusion impact data includes:

[0023] The product of the surface roughness index, terrain slope index, linear corridor and regional wind index is used as the update coefficient of the environmental quality probability sequence at the previous moment;

[0024] The environmental quality probability sequence at the previous moment is updated according to the update coefficient.

[0025] According to the above aspects and any possible implementation, an implementation is further provided.

[0026] The atmospheric environment measured data sequence includes actual measured concentrations of multiple pollutants, including: formaldehyde, particulate pollutants, sulfur dioxide, nitric oxide, nitrogen dioxide, chlorofluorocarbons, sulfur trioxide, carbon monoxide, ozone, ammonia, fluoride, benzene, PM2.5, PM10, TVOC, methane, ethylene, lead, benzo[a]pyrene, and cyanide.

[0027] According to the above aspects and any possible implementation, an implementation is further provided, wherein the step of calculating the air quality index at the current moment according to the atmospheric environment measured data sequence at the current moment comprises:

[0028] According to the actual measured concentration of the pollutant and the corresponding standard concentration, the air quality index of each pollutant is calculated respectively, and the maximum value of the air quality index is taken as the final air quality index.

[0029] According to the above aspects and any possible implementation manner, an implementation manner is further provided, wherein determining the environmental quality state at the current moment according to the air quality index at the current moment includes:

[0030] According to the preset correspondence between the air quality index and the environmental quality state, the environmental quality state corresponding to the air quality index at the current moment is determined.

[0031] According to the above aspects and any possible implementation manner, an implementation manner is further provided, wherein the method further includes:

[0032] If the environmental quality state at the current moment does not fall within the maximum probability range of the environmental quality probability sequence at the current moment, the atmospheric environment prediction data sequence at the next moment is predicted based on the atmospheric environment measured data sequence at the current moment;

[0033] Calculate the air quality index at the next moment according to the predicted atmospheric environment prediction data sequence at the next moment;

[0034] The environmental quality state at the next moment is determined according to the air quality index at the next moment.

[0035] According to the above aspects and any possible implementation manner, an implementation manner is further provided, wherein the step of predicting the atmospheric environment prediction data sequence at the next moment according to the atmospheric environment measured data sequence at the current moment comprises:

[0036] The atmospheric environment measured data sequence at the current moment is input into the pre-trained LSTM neural network model, and the atmospheric environment prediction data sequence at the next moment is output;

[0037] The LSTM neural network model is trained by the following steps: taking the atmospheric environment prediction data sequence at the previous moment as input and the atmospheric environment measured data sequence at the previous moment as output, the LSTM neural network model is trained, and when the model accuracy of the LSTM neural network model meets a preset threshold, the training is stopped.

[0038] The aspects described above and any possible implementation methods further provide an implementation method, during the training process of the LSTM neural network model, whether the residual sequence trained by the LSTM neural network model has Markov properties is tested, and the parameters of the LSTM neural network model are corrected according to the test results; the residual sequence is obtained by calculating the atmospheric environment prediction data sequence and the atmospheric environment measured data sequence.

[0039] According to the above aspects and any possible implementation manner, an implementation manner is further provided, wherein the method further includes:

[0040] When the environmental quality status exceeds the preset warning line, the corresponding atmospheric monitoring site information is obtained according to the latitude and longitude of the target grid;

[0041] The list of factories in the area covered by the target grid is determined from the business codes in the acquired atmospheric monitoring site information, so that monitoring personnel can monitor the gas emissions of the factories according to the list of factories.

[0042] According to a second aspect of the present disclosure, a three-dimensional grid-based environmental quality assessment device is provided. The device comprises:

[0043] A data acquisition module is used to obtain the atmospheric environment measured data sequence of the target grid at the current moment;

[0044] A calculation module is used to calculate the air quality index at the current moment according to the atmospheric environment measured data sequence at the current moment;

[0045] The calculation module is also used to determine the environmental quality state at the current moment according to the air quality index at the current moment;

[0046] A comparison module is used to compare the environmental quality state at the current moment with the environmental quality probability sequence at the current moment; the environmental quality probability sequence at the current moment is obtained by predicting the environmental quality probability based on the environmental quality state at the previous moment and the diffusion impact data of the target grid;

[0047] The prediction module is used to predict the environmental quality probability sequence at the next moment based on the environmental quality probability sequence at the current moment if the environmental quality state at the current moment falls into the maximum probability range in the environmental quality probability sequence at the current moment.

[0048] According to a third aspect of the present disclosure, an electronic device is provided, which includes a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the program, the method described above is implemented.

[0049] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0050] The embodiment of the present disclosure provides an environmental quality assessment method based on a three-dimensional grid. Through the association of the grid, it is convenient to obtain air quality monitoring stations, water quality monitoring stations, noise monitoring equipment, etc., collect environmental data in real time, and combine remote sensing technology, drone technology and other means to obtain multimedia data such as images and videos in the grid unit. At the same time, in the process of grid coding, business coding and time coding are added, and shared transmission between business data is realized according to business coding categories, supporting the calculation and simulation of pollutant expansion, and realizing rapid and accurate judgment of environmental planning. It has the following specific advantages:

[0051] 1. Efficient data presentation: Three-dimensional spatial grid coding can quickly query the spatial location of monitoring data and generate heat maps, improving data presentation efficiency.

[0052] 2. Fast data sharing: Business type coding enables fast data sharing access proxy and improves data sharing efficiency.

[0053] 3. Accuracy of environmental evaluation: Time coding is combined with space coding and business type coding to realize the fusion and interconnection of multi-source data, and an environmental data deduction and prediction model based on LSTM-Markov chain is constructed to simulate the movement and reduction of environmental data, realize more efficient environmental evaluation, and improve the accuracy of evaluation.

[0054] 4. New evaluation ideas: Using the "source-flow-sink" analysis method, a three-dimensional environmental data deduction and prediction model is established to identify the contribution of pollution sources, providing new ideas and tools for the construction of the urban ecological environmental risk pattern.

[0055] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which:

[0057] Figure 1 A flow chart of an environmental quality assessment method based on a three-dimensional grid according to an embodiment of the present disclosure is shown;

[0058] Figure 2 A schematic diagram of business category coding according to an embodiment of the present disclosure is shown;

[0059] Figure 3 The schematic diagram of GEOSOT earth segmentation encoding according to an embodiment of the present disclosure is shown, wherein (a) is a schematic diagram showing that 180 2°×2° grids segmented from 88° to 90° south (north) latitude are merged into one grid P; (b) is a schematic diagram showing that 360 1°×1° grids segmented from 88° to 89° south (north) latitude are merged into three grids P1, P2, and P3; (c) is a schematic diagram showing that grids segmented from 89°32' to 90° south (north) latitude are merged into one grid, grids segmented from 89° to 89°32' are merged into three grids, and grids from 88° to 89° are segmented according to a quadtree based on the merging of 9 layers of grids; (d) is a schematic diagram showing that grids are encoded after further segmentation based on the aforementioned figure (c);

[0060] Figure 4 An example diagram of merging 8-10 level grids in the range of 88° to 90° in the north (south) latitude according to an embodiment of the present disclosure is shown; wherein (a) represents the 8-level grid in the 88° to 90° area; (b) represents the 9-level grid in the 88° to 90° area; (c) represents the 10-level grid in the 88° to 90° area; in the figure, the symbol E represents the east longitude, W represents the west longitude, and PM represents the 0° longitude;

[0061] Figure 5 A graph showing the relationship between the grid height and the corresponding latitude granularity according to an embodiment of the present disclosure is shown;

[0062] Figure 6 A schematic diagram of a process of organizing and storing environmental assessment data based on a three-dimensional grid according to an embodiment of the present disclosure is shown;

[0063] Figure 7 A schematic diagram of environmental assessment data organization coding based on a three-dimensional grid according to an embodiment of the present disclosure is shown;

[0064] Figure 8 A block diagram of an environment quality assessment device based on a three-dimensional grid according to an embodiment of the present disclosure is shown;

[0065] Fig. 9 A schematic block diagram of an exemplary electronic device capable of implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0067] In addition, the term "and / or" in this article is only a description of the association relationship between the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0068] Figure 1 A flow chart of a method 100 for evaluating environmental quality based on a three-dimensional grid according to an embodiment of the present disclosure is shown. The method 100 includes:

[0069] Step 110, obtaining a sequence of measured atmospheric environment data of the target grid at the current moment.

[0070] In some embodiments, the atmospheric environment measured data sequence includes actual measured concentrations of multiple pollutants, including: formaldehyde, particulate pollutants, sulfur dioxide, nitric oxide, nitrogen dioxide, chlorofluorocarbons, sulfur trioxide, carbon monoxide, ozone, ammonia, fluoride, benzene, PM2.5, PM10, total volatile organic compounds TVOC, methane, ethylene, lead, benzo[a]pyrene, and cyanide.

[0071] Step 120, calculating the air quality index at the current moment according to the atmospheric environment measured data sequence at the current moment.

[0072] In some embodiments, the air quality index of each pollutant is calculated according to the actual measured concentration of the pollutant and the corresponding standard concentration, and the maximum value of the air quality index is taken as the final air quality index. First, the index of each pollutant is calculated. The index of each pollutant is obtained by comparing its concentration with the air quality standard formulated by the country or region. For example, for PM2.5, the following formula can be used to calculate its index: Index = (actual concentration / standard concentration) * 100; standard concentration refers to the limit set by the country or region for the concentration of the pollutant. For example, if the country stipulates that the PM2.5 concentration should not exceed 35 micrograms / cubic meter, the standard concentration is 35. Next, the ambient air quality index (Air Quality Index, referred to as AQI) is calculated according to the index of each pollutant. Then, the air quality index (AQI) is calculated according to the index of each pollutant and its corresponding level. The specific calculation formula varies from country to country, for example: AQI = Max (AQI1, AQI2, ..., AQI20).

[0073] Step 130, determining the current environmental quality state according to the current air quality index.

[0074] In some embodiments, the environmental quality state corresponding to the current air quality index is determined according to the preset correspondence between the air quality index and the environmental quality state. That is, according to the AQI table (Table 1), the AQI is converted into the corresponding air quality level and description. The environmental quality state (one of excellent, good, and poor) can be obtained.

[0075] Table 1: Correspondence between AQI and air quality level

[0076]

[0077] Step 140, comparing the current environmental quality state with the current environmental quality probability sequence.

[0078] In some embodiments, the environmental quality probability sequence at the current moment is obtained by performing environmental quality probability prediction based on the environmental quality state at the previous moment and the diffusion impact data of the target grid. Specifically, performing environmental quality probability prediction based on the environmental quality state at the previous moment and the diffusion impact data to obtain the environmental quality probability sequence at the current moment includes: determining the environmental quality probability sequence at the previous moment based on the environmental quality state at the previous moment and based on a preset probability distribution; updating the environmental quality probability sequence at the previous moment based on the diffusion impact data; and calculating the environmental quality probability sequence at the current moment based on the updated environmental quality probability sequence at the previous moment.

[0079] In some embodiments, the diffusion impact data include a surface roughness level index, a terrain slope level index, a linear corridor and a regional wind level index; updating the environmental quality probability sequence at the previous moment based on the diffusion impact data includes: taking the product of the surface roughness level index, the terrain slope level index, the linear corridor and the regional wind level index as the update coefficient of the environmental quality probability sequence at the previous moment; updating the environmental quality probability sequence at the previous moment according to the update coefficient.

[0080] According to the preset probability distribution (set according to experience), the corresponding environmental quality probability sequence is determined, as shown in Table 2. For example, if the environmental quality state is poor, the corresponding environmental quality probability sequence (arranged in the order of poor, good and excellent) is (0.65, 0.28, 0.07), where poor (lower-class), good (middle-class) and excellent (upper-class) are represented by 1, 2 and 3 respectively.

[0081] Table 2: Correspondence between environmental quality status and probability distribution

[0082] State 1 2 3 1 0.65αβθ 0.28αβθ 0.07αβθ 2 0.15αβθ 0.67αβθ 0.18αβθ 3 0.12αβθ 0.36αβθ 0.52αβθ

[0083] Among them, ɑ represents the surface roughness index, β represents the terrain slope index, and θ represents the linear corridor and regional wind index. The corresponding relationship between each index and the grade is shown in Tables 3 to 5. Influencing factors considered: surface roughness and terrain slope have a negative impact on pollutant diffusion, while linear corridors and regional wind environment characteristics have a positive impact on pollutant diffusion.

[0084] Table 3: ɑ value definition table

[0085]

[0086]

[0087] Table 4: β value definition table

[0088]

[0089] Table 5: Definition of θ values

[0090]

[0091]

[0092] According to these corresponding relationships, the corresponding values ​​of α, β, and θ can be determined. Then the product of α, β, and θ is used as the update coefficient to obtain the updated environmental quality probability sequence, for example (0.65αβθ, 0.28αβθ, 0.07αβθ).

[0093] Step 150: If the current environmental quality state falls within the maximum probability range of the current environmental quality probability sequence, then the next environmental quality probability sequence is predicted based on the current environmental quality probability sequence.

[0094] That is to say, only after verifying whether the results of the environmental quality probability sequence at the current moment are predicted correctly based on the environmental quality status at the current moment, can the environmental quality probability sequence at the next moment be predicted based on the environmental quality probability sequence at the current moment. Otherwise, it is necessary to re-correct, that is, update the environmental quality probability sequence.

[0095] After the hypothesis is verified, it is found that the prediction is correct. The proportion of the three-dimensional environmental data in the bad, good, and excellent state at time π today is the probability distribution vector π t =[π t (1), π t (2), π t (3)]. Then the distribution ratio of tomorrow’s π time will be π t+1 (1) = π t (1)*P, the distribution ratio of the π moment the day after tomorrow will be π t+2 (1) = π t+1 (1)*P=π t (1)*P 2 , the distribution ratio at time π on day n will be π t+n (1) = π t+n-1 (1)*P=π t (1)*P n . Currently assume that the initial probability distribution is π t =[0.21, 0.68, 0.11], then the distribution of the previous n days can be calculated as shown in Table 6 below.

[0096] Table 6: Probability distribution of environmental quality in the previous n days

[0097] Day n inferior good excellent 0 0.210 0.680 0.110 1 0.252 0.554 0.194 2 0.270 0.512 0.218 3 0.278 0.497 0.225 4 0.282 0.490 0.226 5 0.285 0.489 0.225 6 0.286 0.489 0.225 ··· ··· ··· ···

[0098] In some embodiments, the method further includes: if the environmental quality state at the current moment does not fall within the maximum probability range of the environmental quality probability sequence at the current moment, predicting the atmospheric environment prediction data sequence at the next moment based on the atmospheric environment measured data sequence at the current moment; calculating the air quality index at the next moment based on the predicted atmospheric environment prediction data sequence at the next moment; and determining the environmental quality state at the next moment based on the air quality index at the next moment. The method of predicting the atmospheric environment prediction data sequence at the next moment based on the atmospheric environment measured data sequence at the current moment includes: inputting the atmospheric environment measured data sequence at the current moment into a pre-trained LSTM neural network model, and outputting the atmospheric environment prediction data sequence at the next moment; the LSTM neural network model is trained by the following steps: using the atmospheric environment prediction data sequence at the previous moment as input and the atmospheric environment measured data sequence at the previous moment as output, training the LSTM neural network model, and stopping the training when the model accuracy of the LSTM neural network model meets the preset threshold. In the process of training the LSTM neural network model, the residual sequence of the LSTM neural network model training is tested to see whether it has Markov properties, and the parameters of the LSTM neural network model are corrected according to the test results; the residual sequence is obtained by calculating the atmospheric environment prediction data sequence and the atmospheric environment measured data sequence. That is, a three-dimensional environmental data deduction and prediction model is constructed based on the LSTM neural network and the Markov chain, and the change trend and movement cycle of the environmental data under the steady-state probability are deduced, so as to achieve more efficient environmental evaluation and improve the evaluation accuracy.

[0099] In some embodiments, the Markov property is used to define the equation of action and construct a three-dimensional environment simulation data set:

[0100]

[0101] Q π (s t ,r t ) as the grid set at time π, represented by a ternary matrix, s t represents the set of code element grids for the environmental assessment data at time t, r represents the data of other business systems at time t, Indicates the construction of a high-order three-dimensional environmental data Markov model at time t+1, r t-1 It is the data of other business systems needed in the process of environmental deduction as the deduction influencing factor. This formula is a recursive idea, and the influence grid at time t is deduced through the influence grid at time t-1. However, according to the Markov chain convergence theorem, if a non-periodic Markov chain has a transition probability matrix P, and any two of its states are connected, then exists and has nothing to do with i, remember Therefore, the grid deduction process needs to be constantly revised. Taking into account the advantages of selective memory of the LSTM neural network, and the fact that individual environmental data in the current ecological environmental evaluation have no aftereffect, which basically conforms to the characteristics of the Markov chain, a three-dimensional environmental data deduction and prediction model is constructed based on the LSTM neural network and the Markov chain.

[0102] The specific steps for constructing a three-dimensional environmental data deduction prediction model are as follows:

[0103] Step 1: Use LSTM neural network to predict environmental data.

[0104] Step 1.1: Use the standardization method to normalize the original environmental data to eliminate the dimension effect. Shrink the environmental data to the interval so that the loss value of the training model decreases faster. The following max-min normalization formula can be used for normalization:

[0105]

[0106] Among them, x i represents an environmental data of the i-th day, They respectively represent the maximum and minimum values ​​of the environmental data sequence.

[0107] Step 1.2: Use a time sliding window to divide the normalized environmental data into a training set and a test set. Take the environmental data sequence of the previous day as the input sample and the environmental data sequence of the current day as the output sample, train the environmental data based on the LSTM neural network, and obtain the predicted environmental data sequence P by denormalization. t (t=1,2,3,...,N). and the real environment data sequence R t The residual of (t=1,2,3,...,N) is recorded as Y t =R t -P t (t=1,2,3,...,N).

[0108] Step 2: Check whether the residual sequence of LSTM neural network training is Markovian.

[0109] Step 2.1: According to the interval range of the environmental data sequence and its sequence characteristics, it is randomly divided into s small intervals: [y0, y1), [y1, y2), ... [y s-1 ,y s );

[0110] Step 2.2: Let I k =[y k-1 ,y k ), (k=1,2,3,...,s), Yt Considered as I=I k (k=1,2,3,...,s) is a random time series in the state space;

[0111] Step 2.3: Construct χ 2 Statistics, test Y t Whether it has Markov property is as follows:

[0112] If a ij represents Y1, Y2, ..., Y N From State I i The frequency of one-step transfer to state Ij, the frequency matrix can be recorded as:

[0113]

[0114] make

[0115] The one-step transition probability matrix P can be obtained as follows: ij ) 6×6 , then when it is large, the statistic: The degree of freedom is (s-1) 2 The chi-square distribution of .

[0116] Select the confidence level α, if χ 2 > α 2 ((s-1) 2 ), then at this confidence level, we cannot reject Y t The null hypothesis of Markov property is that Y t has the Markov property; otherwise, reject Y t The null hypothesis of Markov property is that Y t It is not Markovian.

[0117] Step 3 If Y t If it has Markov properties, the LSTM neural network is used to predict the test set and obtain K prediction values. Combined with the prediction of the training set, the environmental diffusion sequence P can be obtained. N+K , (K=1,2,3,...). According to the Markov property and the CK equation, the residual of the last environmental observation day in the training set is used as the initial state of the Markov chain, and Y N+K The state probability distribution of is:

[0118] I S K (R N+K ∈(P N+K +y s-1 , P N+K +y s ))=I0PK+1 , (s=1, 2, 3, 4, 5, 6; K=1, 2, 3,...)

[0119] Mean:

[0120]

[0121] The mean series ER N+K , (K = 1, 2, 3, ...) is the environmental data sequence after correction of the test set. If it does not have the Markov property, it means that the environmental data is not suitable for correction using the Markov chain model.

[0122] In some embodiments, the method further includes: when the environmental quality status exceeds the preset warning line, the corresponding atmospheric monitoring site information is obtained according to the latitude and longitude of the target grid; the list of factories in the area covered by the target grid is determined from the business code in the acquired atmospheric monitoring site information, so that the monitoring personnel can monitor the gas emissions of the factories according to the factory list. Among them, the atmospheric monitoring site information is pre-set, and the atmospheric monitoring site information includes the Beidou grid code. According to the latitude and longitude information of the target grid and the Beidou grid code in the atmospheric monitoring site information, the binding relationship between the atmospheric monitoring site information and the grid can be realized, that is, a label of the atmospheric monitoring site information is established in advance for each grid, which is convenient for information query interaction.

[0123] In some embodiments, the process of generating preset atmospheric monitoring site information is as follows:

[0124] 1. The atmospheric monitoring site information = original Internet of Things information + 64-bit Beidou grid code + 27-bit category code + 32-bit time code; wherein the original Internet of Things information is composed of ecological environment monitoring source data such as negative oxygen ions, PM2.5, PM10, temperature and humidity; Beidou grid code is generated by Beidou code calculation based on the latitude and longitude information of the grid; the category code is composed of 5-bit business type code + 18-bit enterprise credit code, 4-bit system internal code; the time code is generated according to the storage time and data collection time. Among them, Beidou grid code: can quickly query the spatial location of the corresponding monitoring data, and form a heat map based on the monitored values ​​to view the impact range of environmental evaluation, thereby improving the data display efficiency. Through the business type code, the influencing factors of the pollutant diffusion process can be quickly requested, the surface roughness, terrain slope, linear corridor and regional wind environment characteristics can be used to determine the ventilation influencing factors, and the above calculation factor data can be applied from other business systems to determine the regional risk process influencing factors. Business type coding: Information resource planning (IRP) refers to the comprehensive and systematic planning and management of information resources within an organization to achieve the rational use and sharing of information resources. Through the coding identification of the business type, when performing shared access proxy, the required environmental impact assessment monitoring information can be quickly obtained through the business type coding and an access request can be initiated. When the original library receives the request, it can quickly generate a backup of the business data according to the coding content of the request, which improves the data sharing efficiency of the business system and can directly realize data sharing at the data structure layer without the need for business programs. In the present invention, the shared association between data is directly realized through the graph database. The business type coding format is shown in Table 7 below:

[0125] Table 7: Encoding format

[0126] Industry Category Code Business area code Business system code <![CDATA[S1]]> <![CDATA[S2S3]]> <![CDATA[S4S5]]>

[0127] The major category code of the S1 industry generally indicates which industry the data belongs to, starting from "1" and arranged in ascending order according to the serial code. Because this article is a technical research combined with the application of the Ecological Window project in a certain place, it only involves the ecological and environmental protection industry. For example, in the construction of a comprehensive smart city, each industry can be sequentially encoded according to the serial code.

[0128] S2 S3 is the middle-class business field code, which classifies the various business fields of a certain industry. This code is represented by two-byte letters, generally using the capitalized combination of the pinyin first letter of the business name, such as: natural ecological supervision, represented by "ZR", atmospheric environment supervision, represented by "DQ"

[0129] S4S5 is the specific business system code of the subcategory, starting from "01" and arranged in ascending order according to the serial code. For example: the second business system in atmospheric environmental supervision, the air quality monitoring subsystem, is represented by "DQ02". After the coding is completed through this coding method, data exchange coding can be established according to different levels, based on the data exchange standards between 6 common ecological environment evaluation related application systems. Because it is mainly a business in the field of ecological and environmental protection, it does not involve information exchange between different industries. For example, the overall smart city can generate a zero-level data flow diagram according to the industry code. According to the theoretical basis of information resource planning, identification and planning and design are carried out through the first-level data flow diagram (1-DFD). The drawn 1-DFD is as follows Figure 2 As shown. GC represents (data flow) inflow, and CG represents (data flow) outflow. For example, the influencing factors provided in the spatiotemporal information system, surface roughness, terrain slope, linear corridors and regional wind environment characteristics determine the ventilation influencing factors. If the atmospheric environment supervision system needs such data, the information exchange code of this construction level 1 is shown in Table 8: Table 8: Information exchange code

[0130] Data Flow System code 1 System code 1 GC 1DQ01 1SK01

[0131] Then, through time coding, combined with three-dimensional space grid coding and business type coding, the movement and reduction of environmental assessment related data can be simulated, and an analysis method based on "source-flow-sink" can be constructed for the urban atmospheric environmental risk pattern to identify the types of landscape land uses that produce characteristic pollutants - industrial areas, living areas, and transportation land as pollution sources, and the contribution of pollution sources can be determined using pollution census data to achieve more efficient environmental evaluation.

[0132] The process of generating Beidou grid code according to longitude and latitude is as follows:

[0133] Beidou grid coding process: When the earth reference ellipsoid is gridded, the latitude is expanded from -90° to 90° to -256° to 256°, and the longitude is expanded from -180° to 180° to -256° to 256°, forming a 512°×512° level 0 network, marked with G. The level 1 network is secondarily differentiated based on the level 0 grid according to equal longitude and latitude; the level 2 to level 9 grids are all recursively secondarily divided step by step based on the upper level grid, except for the level 8 and level 9 grids, the length of the latitude line shrinks sharply when approaching the South Pole and the North Pole of the earth, and the level 8 and level 9 grids in the range of 88° to 90° south and north latitude are merged; the level 10 to level 15 grids are expanded, and the level 10 to level 15 grids in the range of 0° to 88° south and north latitude are in the upper level grid (the level 10 grid is expanded in the level 9 grid) The grids of levels 10 to 15 in the south and north latitudes of 88° to 90° are divided on the merged upper grid; the grids of levels 16 to 21 belong to the second grid expansion, and the south and north latitudes of 0° to 88° are divided into two levels according to the arithmetic progression on the basis of the upper grid, and the south and north latitudes of 88° to 90° are divided on the merged upper grid; the grids of levels 22 to 32 belong to the second and lower grid expansion, except for the south and north latitudes of 88° to 90°, they are divided into two levels according to the principle of step-by-step bisection. The grids of levels 8 and below between the south and north latitudes of 88° to 90° need to be merged. Taking the grids of levels 8, 9, and 10 as an example, the merging rules are as follows:

[0134] In the 8-level grid, 180 2°×2° grids divided from 88° to 90° north (south) latitude are merged into one grid P. The merging example is as follows: Figure 3 As shown in (a); in the 9-level grid, the 360 ​​1°×1° grids divided from 89° south (north) latitude to 90° are merged into one grid P0; the 360 ​​1°×1° grids divided from 88° south (north) latitude to 89° are merged into three grids P1, P2, and P3. The merging example is shown in Figure 3 (b) In the 10-level grid, the grids from 89°32' to 90° south (north) latitude are merged into one grid, and the grids from 89° to 89°32' are merged into three grids. The grids from 88° to 89° are divided into quadtrees based on the 9-level grid merging. The merging example is as follows: Figure 3 (c). For example, Figure 4 (a) represents the 8th level grid in the area of ​​88°~90°, Figure 4 (b) represents the 9-level grid in the area of ​​88° to 90°. Figure 4 (c) represents the 10th level grid in the 88°~90° region, taking the 8th to 10th level grid merged in the range of 88° to 90° south (north) latitude as an example. Figure 3The longitude and latitude of the 10-level grids after merging in (c) are as follows: the grids divided by P0 are P00 (360°×28'), P01 (128°×32'), P02 (104°×32',) P03 (128°×32'); the grids divided by P1 are P10 (64°×32'), P11 (64°×32'), P12 (64°×28'), P13 (64°×28'); the grids divided by P2 are P20 (52°×32'), P21 (52°×32'), P22 (52°×28'), P23 (52°×28'); the grids divided by P3 are P30 (64°×32'), P31 (64°×32'), P32 (64°×28'), P33 (64°×28'). The Beidou grid code 3D code expands the height dimension to form a true three-dimensional network of the sky, land and sea from 500,000 kilometers above the earth's periphery to the center of the earth. The level of height domain subdivision is consistent with the level of earth reference ellipsoid subdivision. Is there any level of subdivision? The height domain is divided into 2 m Floors, 2 underground m-1 Floor, 2 above ground m-1 Layer. The height of each grid at the same level (geodetic granularity) should be equal, and the height should match the granularity of the grid latitude formed by the corresponding level of the equator of the corresponding contour surface of the layer. The relationship between the height of the grid at the same level and the granularity of the grid latitude at the equator of the corresponding contour surface is as follows: Figure 5 shown.

[0135] exist Figure 5 In, L i is the height of the grid on the contour surface, the granularity in the equatorial direction, H i is the grid height, h i is the grid height, the granularity in the geodetic direction, r i is the distance between the grid and the center of the earth on the equatorial plane, and θ is the longitude (latitude) span difference corresponding to the grid.

[0136] The distance between the grid and the center of the earth (calculated on the equatorial plane) r n As shown below:

[0137]

[0138] Grid Earth Height H n The calculation formula is as follows:

[0139]

[0140] Grid height (geostatic granularity) h n As shown below:

[0141]

[0142] The length of the grid on the contour surface (granularity in the equatorial direction) L n As shown below:

[0143]

[0144] Among them, θ0 is the longitude (latitude) span difference corresponding to the basic grid (1° grid) defined in the initial subdivision range, in radians (rad), θ is the longitude (latitude) span difference corresponding to the network, in radians (rad); n is the nth layer of three-dimensional grid from the ground upward (or downward), n is an integer, n ≥ 0 above the ground, and n < 0 below the ground; r0 is the major semi-axis of the earth, in meters (m), r0 = 6378137.

[0145] Beidou grid code has the characteristics of three-dimensional segmentation, multi-scale (globally and minutely), positionability, recursion, globally unique coding, unified spatiotemporal coding, identification, indexability and automatic spatial association. Beidou grid code can be applied to the integration, organization, sequencing, calculation and service of multi-source spatial data.

[0146] In some embodiments, the storage process of environmental monitoring data is as follows: Figure 6 As shown, it is mainly divided into the following related steps:

[0147] 1. The front-end accesses natural ecological monitoring data, water ecological monitoring data, atmospheric ecological monitoring data, climate change response monitoring data, soil ecological environment monitoring data, and nuclear and radiation monitoring data. In order to be able to access the massive IoT data collected by the front-end in large quantities, this method adopts a multi-level caching strategy Kafka cluster, NoSQL data cluster, and data storage to receive massive IoT data.

[0148] 2. The Kafka cluster can regularly check the connection with the front-end monitoring data (such as nuclear radiation monitoring, atmospheric monitoring, and water quality monitoring equipment. If a connection error occurs, it can be reconnected).

[0149] The following is a detailed description of the above encoding process and the process of data exchange based on the encoding with reference to specific examples.

[0150] Assume: There is an atmospheric monitoring site (site address: X area, corresponding enterprise: southeast side of the roof of Building 1 of X company, longitude and latitude: 40.76, 111.61, social credit code: 91150000767877596U), and its business monitoring data includes more than 20 fields such as sulfur dioxide, nitrogen dioxide, sulfur dioxide, carbon monoxide, and ozone. In the Ecological Window platform, it belongs to the business scope of the atmospheric supervision system. It receives the atmospheric pollution index once at 12:15:12:280 milliseconds on June 6, 2024, and sends it to the message bus at 12:15:13:380 milliseconds on June 6, 2024. According to the business requirements of the Ecological Window, there is a need to extract the data of the atmospheric monitoring site in the business environment assessment of the climate change supervision system and the natural ecological supervision system. The climate change supervision system has extraction requirements for fields 2 and 3 of the monitoring data, and the natural ecological supervision system has extraction requirements for fields 3 and 6 of the monitoring data.

[0151] According to the above description, the encoding information of a specific example according to the encoding rule in this article is as follows Figure 7 As shown, the specific steps are as follows:

[0152] Step 1: Add Beidou grid code according to the location information of the atmospheric monitoring station, and add the corresponding 64-bit Beidou grid code information according to the three-dimensional grid representation;

[0153] Atmospheric monitoring site information = original IoT information + 64-bit Beidou grid code + 27-bit category code + 32-bit time code;

[0154] Step 2: Add business coding information based on the data extraction requirements of the climate change supervision system for fields 2 and 3 of the monitoring data, and the data extraction requirements of the natural ecology supervision system for fields 3 and 6 of the monitoring data; this additional code only involves the first type of coding, and 0 is added if it is less than 32 bits. The sequence number indicates the number of the site in the ecological window. The specific field exchange requirements are expressed in detail in the graph database;

[0155] Atmospheric monitoring site information = original Internet of Things information (12065 051 0062+13.1 053.2) G001303111112001 110010 00001012031 0111111111111111111111110001010101DQ0191150000767877596U 0001;

[0156] Step 3: Based on the requirements of climate change supervision system for extracting fields 2 and 3 of monitoring data, and natural ecology supervision system for extracting fields 3 and 6 of monitoring data, the shared graph knowledge information is increased:

[0157] "DQ": {

[0158] {"ZR":

[0159] ["field2":ALL,

[0160] "Field 3": ALL, ]

[0162] },

[0163] {"QH": [

[0165] "Field 3": ALL,

[0166] "Field 6": ALL, ]

[0168] }

[0169] }

[0170] Step 4: Add corresponding time information to facilitate the business system to deduce time, automatically generate samples for neural network calculations, and grasp the diffusion law and trajectory information of pollution sources.

[0171] Atmospheric monitoring site information = original Internet of Things information (12065 051 0062+13.1 053.2) 1DQ0191150000767877596U 0001 2024060612151228 2024060612151338.

[0172] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the order of the actions described, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0173] The above is an introduction to the method embodiment. The following is a further explanation of the scheme disclosed in the present invention through an apparatus embodiment.

[0174] Figure 8 FIG. 8 is a block diagram of an environment quality assessment device 800 based on a three-dimensional grid according to an embodiment of the present disclosure. Figure 8 As shown, the device 800 includes:

[0175] The data acquisition module 810 is used to acquire the atmospheric environment measured data sequence of the target grid at the current moment;

[0176] The calculation module 820 is used to calculate the air quality index at the current moment according to the atmospheric environment measured data sequence at the current moment;

[0177] The calculation module 820 is further used to determine the environmental quality state at the current moment according to the air quality index at the current moment;

[0178] A comparison module 830 is used to compare the current environmental quality state with the current environmental quality probability sequence; the current environmental quality probability sequence is obtained by performing environmental quality probability prediction based on the environmental quality state at the previous moment and the diffusion impact data of the target grid;

[0179] The prediction module 840 is used to predict the environmental quality probability sequence at the next moment according to the environmental quality probability sequence at the current moment if the environmental quality state at the current moment falls into the maximum probability range of the environmental quality probability sequence at the current moment.

[0180] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0181] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device and a readable storage medium.

[0182] Fig. 9 A schematic block diagram of an electronic device 900 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0183] The electronic device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in the ROM 902 or a computer program loaded from the storage unit 908 into the RAM 903. In the RAM 903, various programs and data required for the operation of the electronic device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An I / O interface 905 is also connected to the bus 904.

[0184] Multiple components in the electronic device 900 are connected to the I / O interface 905, including: an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the electronic device 900 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0185] The computing unit 901 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 901 performs the various methods and processes described above, such as method 100. For example, in some embodiments, method 100 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 900 via ROM 902 and / or communication unit 909. When the computer program is loaded into RAM 903 and executed by the computing unit 901, one or more steps of the method 100 described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform method 100 in any other appropriate manner (e.g., by means of firmware).

[0186] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0187] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0188] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0189] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0190] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0191] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0192] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0193] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A three-dimensional grid-based environmental quality assessment method, characterized in that: include: Obtain the atmospheric environment measured data sequence of the target grid at the current moment; Calculate the air quality index at the current moment according to the atmospheric environment measured data sequence at the current moment; Determine the current environmental quality status according to the current air quality index; Comparing the environmental quality state at the current moment with the environmental quality probability sequence at the current moment; the environmental quality probability sequence at the current moment is obtained by predicting the environmental quality probability based on the environmental quality state at the previous moment and the diffusion impact data of the target grid; If the environmental quality state at the current moment falls within the maximum probability range of the environmental quality probability sequence at the current moment, the environmental quality probability sequence at the next moment is predicted based on the environmental quality probability sequence at the current moment.

2. The method according to claim 1, characterized in that According to the environmental quality status and diffusion impact data of the previous moment, the environmental quality probability prediction is carried out to obtain the environmental quality probability sequence at the current moment, including: According to the environmental quality state at the previous moment, based on a preset probability distribution, determine the environmental quality probability sequence at the previous moment; According to the diffusion impact data, the environmental quality probability sequence at the previous moment is updated; The environmental quality probability sequence at the current moment is calculated based on the updated environmental quality probability sequence at the previous moment.

3. The method according to claim 2, characterized in that The diffusion impact data include surface roughness grade index, terrain slope grade index, linear corridor and regional wind grade index; The updating of the environmental quality probability sequence at the previous moment according to the diffusion impact data includes: The product of the surface roughness index, terrain slope index, linear corridor and regional wind index is used as the update coefficient of the environmental quality probability sequence at the previous moment; The environmental quality probability sequence at the previous moment is updated according to the update coefficient.

4. The method according to claim 1, characterized in that The atmospheric environment measured data sequence includes actual measured concentrations of multiple pollutants, including: formaldehyde, particulate pollutants, sulfur dioxide, nitric oxide, nitrogen dioxide, chlorofluorocarbons, sulfur trioxide, carbon monoxide, ozone, ammonia, fluoride, benzene, PM2.5, PM10, TVOC, methane, ethylene, lead, benzo[a]pyrene, and cyanide.

5. The method according to claim 4, characterized in that The step of calculating the air quality index at the current moment according to the atmospheric environment measured data sequence at the current moment includes: According to the actual measured concentration of the pollutant and the corresponding standard concentration, the air quality index of each pollutant is calculated respectively, and the maximum value of the air quality index is taken as the final air quality index.

6. The method according to claim 1, characterized in that Determining the environmental quality state at the current moment according to the air quality index at the current moment includes: According to the preset correspondence between the air quality index and the environmental quality state, the environmental quality state corresponding to the air quality index at the current moment is determined.

7. The method according to claim 1, characterized in that The method further comprises: If the environmental quality state at the current moment does not fall within the maximum probability range of the environmental quality probability sequence at the current moment, the atmospheric environment prediction data sequence at the next moment is predicted based on the atmospheric environment measured data sequence at the current moment; Calculate the air quality index at the next moment according to the predicted atmospheric environment prediction data sequence at the next moment; The environmental quality state at the next moment is determined according to the air quality index at the next moment.

8. The method according to claim 7, characterized in that The method of predicting the atmospheric environment prediction data sequence at the next moment based on the atmospheric environment measured data sequence at the current moment includes: The atmospheric environment measured data sequence at the current moment is input into the pre-trained LSTM neural network model, and the atmospheric environment prediction data sequence at the next moment is output; The LSTM neural network model is trained by the following steps: taking the atmospheric environment prediction data sequence at the previous moment as input and the atmospheric environment measured data sequence at the previous moment as output, the LSTM neural network model is trained, and when the model accuracy of the LSTM neural network model meets a preset threshold, the training is stopped.

9. The method according to claim 8, characterized in that During the training process of the LSTM neural network model, whether the residual sequence of the LSTM neural network model training has Markov property is tested, and the parameters of the LSTM neural network model are corrected according to the test results; the residual sequence is obtained by calculating the atmospheric environment prediction data sequence and the atmospheric environment measured data sequence.

10. The method according to claim 1, characterized in that The method further comprises: When the environmental quality status exceeds the preset warning line, the corresponding atmospheric monitoring site information is obtained according to the latitude and longitude of the target grid; The list of factories in the area covered by the target grid is determined from the business code in the acquired atmospheric monitoring site information, so that monitoring personnel can monitor the gas emissions of the factories according to the factory list.

Images