A model prediction-based pollutant gridding monitoring and evaluation method

Abstract

The application discloses a kind of based on model prediction's pollutant gridding monitoring and evaluation method, belong to environmental protection technical field, including: with the help of industrial internet to the environmental pollutants of monitoring area are monitored and obtain monitoring result;With the pollutant type greater than first threshold value as exceeding standard pollutant;Based on exceeding standard pollutant, obtain the pollution source that other areas except monitoring area produce exceeding standard pollutant, and the emission of pollution source real-time and meteorological data are input to atmospheric quality estimation model, form the diffusion pollutant prediction result inside monitoring area;Based on diffusion pollutant prediction result and the emission of exceeding standard pollutant in monitoring result, obtain the emission of exceeding standard pollutant inside monitoring area;Based on the emission of exceeding standard pollutant inside monitoring area and exceeding standard pollutant, obtain the environmental evaluation value inside monitoring area, so that the treatment effect and treatment efficiency of pollutant are improved to a certain extent.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental protection, and in particular relates to a pollutant grid monitoring and assessment method based on model prediction. Background Art

[0002] In recent years, with the sustained, stable, and rapid development of the domestic economy, people, while meeting their material needs, have become increasingly concerned about their health. The ecological environment is the foundation of human survival, and air pollution directly impacts people's health. With global climate change and severe air pollution, especially during the winter heating season, PM2.5 levels are exceeding standards in many areas, posing a significant threat to people's health. Real-time monitoring and control of air pollution are gaining increasing attention.

[0003] To address pollutant monitoring issues, Chinese invention patent authorization publication number CN111461439B, "Multi-heat Source Heating Load Scheduling Method and System Based on Urban Atmospheric Diffusion Prediction," establishes a heat source pollutant emission model, a heating network model, an air quality prediction model, and an air quality assessment model to determine the optimal air quality objective function and constraints. By analyzing and comparing multiple feasible solutions or using a heuristic algorithm, the optimal multi-heat source heating load scheduling and allocation scheme is determined. This approach can effectively address air quality and heating supply assurance issues during major events. However, the following technical issues exist:

[0004] 1. For pollutants, if the air quality prediction model cannot be simulated specifically based on the pollutant exceeding the standard, it will take up a lot of computing power and memory, and also cause unnecessary energy consumption.

[0005] 2. When pollutants in the pollutant monitoring area exceed the standard, it is impossible to accurately identify whether the source of the pollutants is internal or external transmission, so it is impossible to carry out targeted pollution source control, resulting in unsatisfactory pollutant control effect and efficiency.

[0006] 3. The spatiotemporal distribution results of pollutant concentrations currently obtained using air quality prediction models are often fixed. However, the spatiotemporal distribution results of pollutant concentrations often change over time. Therefore, using the results at a certain moment cannot accurately reflect the actual pollutant concentration situation.

[0007] Based on the above technical problems, it is necessary to design a grid-based pollutant monitoring and assessment method based on model prediction. Summary of the Invention

[0008] The purpose of the present invention is to provide a pollutant grid monitoring and assessment method based on model prediction.

[0009] In order to solve the above technical problems, the first aspect of the present invention provides a grid-based pollutant monitoring and assessment method based on model prediction, comprising:

[0010] S11 sets up pollutant monitoring devices in the pollutant monitoring area to form a pollutant grid monitoring system to monitor pollutants in the environment and obtain monitoring results;

[0011] S12: based on the monitoring results, determining whether there is a pollutant type greater than a first threshold value in the monitoring results; if so, treating the pollutant type greater than the first threshold value as an excessive pollutant;

[0012] S13: Based on the pollutants exceeding the standard, the pollution sources within the city that generate the pollutants exceeding the standard in areas other than the pollutant monitoring area are obtained, and the real-time emissions and meteorological data of the pollution sources are used as an input set. The input set is input into the air quality prediction model to form a prediction result of diffuse pollutants within the pollutant monitoring area; the diffuse pollutant prediction result is updated once every time t, and the spatiotemporal distribution result of the previous excessive pollutant concentration is used as a constraint condition for the spatiotemporal distribution result of the next excessive pollutant concentration;

[0013] S14: obtaining the emission of pollutants exceeding the standard within the pollutant monitoring area based on the prediction result of diffuse pollutants within the pollutant monitoring area and the emission of pollutants exceeding the standard in the monitoring result;

[0014] S15 obtains an environmental assessment value within the pollutant monitoring area based on the emission amount of pollutants exceeding the standard and the pollutants exceeding the standard within the pollutant monitoring area.

[0015] In the above technical solution, further, in step S13, the specific steps of forming the prediction results of diffuse pollutants within the pollutant monitoring area based on the air quality prediction model are:

[0016] Defining constraints for the air quality prediction model, wherein the constraints include at least energy balance, momentum balance, mass balance, and the last diffuse pollutant prediction results;

[0017] Acquire meteorological data and real-time emissions from pollution sources in real time, and input the meteorological data and real-time emissions from pollution sources as initial conditions into an air quality prediction model;

[0018] A numerical iterative solution is performed based on the initial conditions to obtain simulation results, and a flow field analysis is performed using Fluent software based on the simulation results. Based on the results of the flow field analysis, visualization is performed using Tecplot software to obtain the diffuse pollutant prediction results at time t, and the diffuse pollutant prediction results at time t are used as constraint conditions for the diffuse pollutant prediction results at the next moment.

[0019] Furthermore, the meteorological data includes ground meteorological data and high-altitude meteorological data, wherein the ground meteorological data includes wind direction, wind speed, total cloud cover, and temperature; and the high-altitude meteorological data includes air pressure, temperature, wind speed, and wind direction at different altitude layers.

[0020] Furthermore, the initial conditions also include terrain data, and the terrain data of the city and the pollution source are constructed based on the longitude and latitude of the city and the longitude and latitude of the pollution source, and the prediction results of diffused pollutants within the pollutant monitoring area are obtained based on the terrain data, the meteorological data, and the emissions of the pollution source within the first time threshold.

[0021] By first determining whether there are pollutant types exceeding a first threshold, and then establishing an air quality prediction model when such a level is found, this approach reduces unnecessary energy consumption while ensuring accuracy. By establishing an air quality prediction model, the specific sources of pollutant emissions exceeding standards can be determined, enabling targeted pollutant management and control, improving pollutant control and management efficiency.

[0022] By setting the first threshold, the air quality prediction model can be established after the pollutants exceed the standard. This reduces unnecessary memory and computing power consumption while ensuring the quality of the assessment, making the system more stable and further reducing the final energy consumption.

[0023] By measuring the amount of pollutants exceeding the standard and the types of pollutants exceeding the standard within the pollutant monitoring area, the environment within the pollutant monitoring area can be accurately assessed. Since different pollutants exceeding the standard cause different harm to the human body, combining the pollutants exceeding the standard also makes the assessment results more accurate.

[0024] Furthermore, the specific steps for obtaining the pollutants exceeding the standard are:

[0025] Based on the monitoring results, when there is a pollutant type greater than a first threshold, the pollutant type greater than the first threshold is used as a candidate pollutant;

[0026] When the duration of the candidate pollutant is greater than the second time threshold or the number of times the candidate pollutant is monitored is greater than the first number threshold, the candidate pollutant is regarded as an excessive pollutant.

[0027] By setting the first threshold, the second time threshold and the first number threshold, it is possible to prevent misidentification due to fluctuations in monitoring results, which in turn leads to misidentification of pollutants exceeding the standard, resulting in inaccurate final prediction results, and ultimately low efficiency and accuracy in environmental assessment.

[0028] Furthermore, the second time threshold and the first number threshold are determined according to the degree of harm of the candidate pollutants to the human body and the degree of concern of the candidate pollutants.

[0029] Furthermore, the calculation formula of the second time threshold is:

[0030]

[0031] Among them, W is the degree of harm of alternative pollutants to the human body, and the value range is between 0 and 1. B1 is the degree of attention of alternative pollutants, which reflects the degree of public attention paid to alternative pollutants, and the value range is between 0 and 1. T is the benchmark time threshold, and K1 and K2 are weights.

[0032] Furthermore, the step S15 is specifically as follows:

[0033] Obtaining the emission amount and pollutants exceeding the standard within the pollutant monitoring area;

[0034] Obtain the types of other pollutants other than those exceeding the standards and the emissions of other pollutants in the monitoring results;

[0035] Based on other pollutant types and their emissions, the emissions of the pollutants exceeding the standard and the pollutants exceeding the standard, an evaluation model based on the IGWO-GRU algorithm obtains an environmental evaluation value within the pollutant monitoring area.

[0036] By adopting pollutants exceeding the standard and other pollutant types, we can comprehensively construct the environmental assessment value within the pollutant monitoring area, so as to accurately reflect the environmental assessment value within the pollutant monitoring area, and then accurately reflect the environment within the pollutant monitoring area for governance.

[0037] Furthermore, the calculation formula for the environmental assessment value within the pollutant monitoring area is:

[0038]

[0039] Where W i is the degree of harm of the i-th pollutant to the human body, P i is the emission of the i-th pollutant, P ibis the emission standard required by the standard for the i-th pollutant, and N is the total amount of the pollutant.

[0040] Furthermore, when the environmental assessment value within the pollutant monitoring area is greater than the environmental assessment value threshold, targeted treatment is performed on the pollutants within the pollutant monitoring area based on the monitoring result.

[0041] On the other hand, an embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the above-mentioned model-prediction-based gridded pollutant monitoring and assessment method.

[0042] On the other hand, a computer program product is provided in an embodiment of the present application. The computer program product stores instructions, and when the instructions are executed by a computer, the computer implements the above-mentioned model-based prediction-based pollutant grid monitoring and assessment method.

[0043] Other features and advantages will be described in the following description, and in part will become apparent from the description, or understood by practicing the invention. The purpose and other advantages of the invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0044] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 This is a flow chart of a pollutant grid monitoring and assessment method based on model prediction in Example 1;

[0047] Figure 2 A grid layout diagram of the pollutant monitoring device in Example 1;

[0048] Figure 3 This is a flow chart of the specific steps for obtaining pollutants exceeding the standard in Example 1;

[0049] Figure 4 This is a flowchart of the specific steps for forming the prediction results of diffuse pollutants within the pollutant monitoring area in Example 1;

[0050] Figure 5 This is a flow chart of the steps for evaluating the environmental assessment value within the pollutant monitoring area in Example 1. DETAILED DESCRIPTION

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0052] Example 1

[0053] Figure 1 The present invention relates to a grid-based pollutant monitoring and assessment method based on model prediction, comprising:

[0054] S11 sets up pollutant monitoring devices in the pollutant monitoring area to form a pollutant grid monitoring system to monitor pollutants in the environment and obtain monitoring results;

[0055] For example, Figure 2 As shown, in ①, ②, ③, ④, ⑤, ⑥, ⑦, ⑧, ⑨, ⑩, A pollutant monitoring device is installed in the equipment.

[0056] S12: based on the monitoring results, determining whether there is a pollutant type greater than a first threshold value in the monitoring results; if so, treating the pollutant type greater than the first threshold value as an excessive pollutant;

[0057] S13: Based on the pollutants exceeding the standard, the pollution sources in other areas within the city other than the pollutant monitoring area that generate the pollutants exceeding the standard are obtained, and the real-time emissions and meteorological data of the pollution sources are used as an input set. The input set is input into the air quality prediction model to form a diffuse pollutant prediction result within the pollutant monitoring area, wherein the diffuse pollutant prediction result is updated once every time t, and the spatiotemporal distribution result of the previous excessive pollutant concentration is used as a constraint condition for the spatiotemporal distribution result of the next excessive pollutant concentration;

[0058] S14: obtaining the emission of pollutants exceeding the standard within the pollutant monitoring area based on the diffuse pollutant prediction result and the emission of pollutants exceeding the standard in the monitoring result;

[0059] S15 obtains an environmental assessment value within the pollutant monitoring area based on the emission amount of pollutants exceeding the standard and the pollutants exceeding the standard within the pollutant monitoring area.

[0060] By first determining whether there is a pollutant type greater than a first threshold, and then establishing an air quality prediction model when there is a pollutant type greater than the first threshold, unnecessary electricity consumption is reduced while ensuring accuracy. By establishing the air quality prediction model, the specific source of the emission of pollutants exceeding the standard can be obtained, so that pollutant management and control can be carried out in a targeted manner, thereby improving the efficiency of pollutant control and management.

[0061] By setting the first threshold, the air quality prediction model can be established after the pollutants exceed the standard. This reduces unnecessary memory and computing power consumption while ensuring the quality of the assessment, making the system more stable and further reducing the final energy consumption.

[0062] By using the emissions and meteorological data within the first time threshold of the pollution source as an input set, the input set is input into the air quality prediction model. Since the diffusion of excessive pollutants requires a certain amount of time, past emissions and meteorological data are used to more accurately obtain data caused by the diffusion of excessive pollutants generated by the pollution source, thereby ensuring the accuracy of the prediction of the emissions of excessive pollutants within the final pollutant monitoring area.

[0063] By measuring the amount of pollutants exceeding the standard and the types of pollutants exceeding the standard within the pollutant monitoring area, the environment within the pollutant monitoring area can be accurately assessed. Since different pollutants exceeding the standard cause different harm to the human body, combining the pollutants exceeding the standard also makes the assessment results more accurate.

[0064] In another possible embodiment, Figure 2 As shown in Figure 2, the specific steps for forming the prediction results of diffuse pollutants within the pollutant monitoring area are:

[0065] defining constraints for an air quality prediction model, wherein the constraints include at least energy balance, momentum balance, and mass balance;

[0066] Acquire meteorological data and real-time emissions from pollution sources in real time, and input the meteorological data and real-time emissions from pollution sources as initial conditions into an air quality prediction model;

[0067] For example, the WRF model is used for meteorological simulation to obtain multi-layered regional simulated meteorological data, including temperature, relative humidity, wind speed and direction, with a 6-hour temporal resolution and a 1°×1° spatial resolution. This data is then interpolated onto the WRF model grid nodes through four-dimensional assimilation to simulate the local meteorological field and converted into the required gridded meteorological data.

[0068] The natural source emission data in the pollution source inventory are estimated using the natural gas and aerosol emission model MEGAN, and the regional resolution of land cover data can reach 300m; the man-made pollution source inventory of the air quality prediction model uses the heat source emission model as emission information input to quantitatively analyze the impact of various heat source emissions on the air quality of key areas under micrometeorological conditions; the emission source inventory is processed and converted into the format required by the CMAQ model using a written program.

[0069] Pollution sources from the pollution source list were gridded and incorporated into a 0.5°×0.5° resolution database. Multi-layer regional simulated meteorological data and the pollution source list were used as constraints to feed into the air quality prediction model. A Lambert projection coordinate system was used to simulate the spatiotemporal dispersion of pollutants within key protection areas. The air quality prediction model utilizes a three-layer nested gridding technique: the first layer covers all of China, with a coordinate origin of 34°N, 116°E, a grid resolution of 36 km, and 173×136 grid cells; the second layer covers the Yangtze River Delta region, with a coordinate origin of 30°N, 120°E, a grid resolution of 4 km, and 150×174 grid cells; and the third layer covers the urban area, with a coordinate origin of 30°N, 120°E, a grid resolution of 1.6 km, and 140×95 grid cells. The model is divided into 14 vertical layers, with a simulation altitude of 100 hPa. The CMAQ model uses CB05 and AER06 as model parameterization schemes for chemical reactions and aerosol mechanisms.

[0070] The final pollutant concentration distribution results involve two variables, time and space, which can simulate the average concentration of grid points within a specified time period and can be expressed as follows:

[0071]

[0072] In the above formula,

[0073] S m : The area of ​​the city;

[0074] ρ t,i (x,y): mass concentration of air pollutant i at point (x,y);

[0075] The model calculation results were verified using meteorological observation station data. The evaluation indicators were the mean fractional deviation B and the mean fractional error E. When the simulation results B≤±30% and E≤±50%, the atmospheric model was considered to have high accuracy.

[0076]

[0077] In the above formula,

[0078] N: number of simulations;

[0079] ZS : analog value;

[0080] Z R : Observed value.

[0081] A numerical iterative solution is performed based on the initial conditions to obtain simulation results, and a flow field analysis is performed using Fluent software based on the simulation results. Based on the results of the flow field analysis, visualization is performed using Tecplot software to obtain the diffuse pollutant prediction results at time t, and the diffuse pollutant prediction results at time t are used as constraint conditions for the diffuse pollutant prediction results at the next moment.

[0082] In another possible embodiment, the initial conditions also include terrain data, and the terrain data of the city and the pollution source are constructed based on the longitude and latitude of the city and the longitude and latitude of the pollution source, and the prediction results of diffused pollutants within the pollutant monitoring area are obtained based on the terrain data, the meteorological data, and the emissions of the pollution source within the first time threshold.

[0083] For example, the calculation of diffuse pollutant prediction results based on the CALPUFF model primarily consists of pre-processing, CALMET, CALPUFF, and post-processing. The pre-processing module processes geographic and meteorological data. CALMET generates a three-dimensional meteorological field. CALPUFF simulates the atmospheric diffusion and physical and chemical processes of pollutants based on this 3D meteorological field. Post-processing primarily includes post-processors such as CALPOST and CALSUM, as well as common tools such as PRTMET.

[0084] The CALMET module uses mass conservation to diagnose the wind field. It can calculate slope flow, terrain correction, terrain blocking effects, etc.

[0085] The wind field is obtained by correcting the terrain of the simulated area from the initial wind field processed by the WRF model. By calculating the wind field within the simulated grid, the vertical wind speed affected by the terrain is obtained, thereby obtaining the influence of terrain dynamics in the horizontal direction.

[0086] (1) Terrain dynamics effect, the calculation formula is as follows:

[0087] Cartesian vertical velocity:

[0088] W=(V×▽h t )exp(-KZ)

[0089] Where V is the average wind speed in the simulated area, in m / s; h t is the terrain height of the simulated area, in meters; K is the exponential attenuation coefficient; Z is the vertical coordinate.

[0090] Exponential decay coefficient:

[0091]

[0092] Where N is the Brunt-Vecella frequency and V is the regional average wind speed in m / s.

[0093] Brent-Vecella frequency:

[0094]

[0095] Where g is the acceleration due to gravity, in m 2 / s; θ is the potential temperature, unit is ℃.

[0096] (2) Slope flow: Based on Mahrt's hypothesis theory, the thickness of the slope flow layer varies with the height of the slope top. The calculation formula is as follows:

[0097]

[0098]

[0099]

[0100] Where S e is the slope flow equilibrium velocity, x is the distance from the top of the slope, L e is the equilibrium length scale, ▽θ is the ambient potential temperature difference, θ is the ambient potential temperature, C D is the ground resistance coefficient, h is the slope flow width, α is the slope angle relative to the horizontal, K is the entrainment coefficient at the top of the slope flow layer, and g is the acceleration of gravity.

[0101] (3) Terrain blocking effect

[0102]

[0103] Δht=(h max ) ij -Z ijk

[0104] Where Fr is the local Froude number, V is the grid point wind speed, N is the Brundt-Weiser frequency, Δht is the effective blocking height, (h max ) ij is the highest height of the grid point, Z ijk is the height of the grid point (i, j) in the upper layer k.

[0105] The CALMET module can refine the WRF model output meteorological field to generate hourly wind field and temperature field. When modeling, the CALMET module needs to input initial meteorological data (ground station data or numerical weather prediction model simulation data), terrain data, land use data, to provide data support for the diagnosis of wind field.

[0106] It should be noted that the CALPUFF model can simulate multiple heights and multiple pollution sources, and is a non-steady-state Lagrangian diffusion model. When the meteorological field changes with time and space, the CALPUFF model can simulate the diffusion concentration of pollutants in the atmosphere. When calculating the plume lifting, different influences need to be considered, such as plume buoyancy and dynamics, vertical wind shear and stable atmospheric layer structure. The CALPUFF model uses puff mode, which can handle complex meteorological changes and diffusion of pollutants emitted by pollution sources in the atmosphere. The following is the puff mode calculation formula in the CALPUFF model:

[0107] (1) Puff integral method

[0108] Basic concentration equation of a single puff at a receiving point:

[0109]

[0110]

[0111] Where C is the ground concentration, Q is the mass of pollutants in the puff, δ x , δ y is the diffusion coefficient, d a , d c is the distance from the center of the puff, g is the vertical term in the Gaussian direction, H is the effective height of the puff center from the ground, and h is the mixing layer height.

[0112] (2) Diffusion parameter calculation formula

[0113] The diffusion parameter calculation is divided into two parts: horizontal and vertical

[0114]

[0115]

[0116] Where ξ y,n , ξ Zn is the total horizontal and vertical diffusion parameter at a certain position at time step n, ξ y,t , ξ Zt is the turbulent diffusion parameter, ξ y,b , ξ Zb is the buoyancy diffusion parameter, ξ ys is the horizontal diffusion parameter.

[0117] (3) Dry and wet deposition

[0118] The dry and wet deposition of pollutants usually has a certain impact on the diffusion of pollutants. The following is the calculation formula for dry and wet deposition of pollutants:

[0119] Dry deposition calculation formula:

[0120] F=D ht (x m -x s ) / (hZ)=v d x s

[0121] Where F is the deposition flux, x m is the concentration of pollutants in the mixed layer, x s is the concentration of pollutants at the top of the surface layer, h is the height of the mixed layer, Z is the height of the surface layer, and D ht is the boundary layer diffusivity.

[0122] Wet deposition calculation formula:

[0123] x t+dt =x t exp[-∧Δt]

[0124] ∧=λ(R / R1)

[0125] Where x is the concentration (g / m 3 ), is the wet deposition removal coefficient, λ is the removal factor (S -1 ), R is the precipitation rate (mm / hr), and R1 is the reference precipitation rate of 1 mm / hr.

[0126] In the CALPUFF model, there are three main chemical conversion schemes: MESOPUFF II, RIVAD / ARM3, and the secondary organic aerosol (SOA) scheme. The MESOPUFF II scheme primarily involves the conversion of SO2 and nitrogen oxides to sulfate and nitrate. The RIVAD / ARM3 scheme primarily involves the conversion of NO, NO2, and SO2 to NO2, NO3, and SO3. The secondary organic aerosol (SOA) scheme primarily involves the conversion of SOA.

[0127] In another possible embodiment, Figure 3 As shown, the specific steps for obtaining the pollutants exceeding the standard are:

[0128] Based on the monitoring results, when there is a pollutant type greater than a first threshold, the pollutant type greater than the first threshold is used as a candidate pollutant;

[0129] When the duration of the candidate pollutant is greater than a second time threshold or the number of times the candidate pollutant is monitored is greater than a first number threshold;

[0130] The candidate pollutants are regarded as pollutants exceeding the standard.

[0131] By setting the first threshold, the second time threshold and the first number threshold, it is possible to prevent misidentification due to fluctuations in monitoring results, which in turn leads to misidentification of pollutants exceeding the standard, resulting in inaccurate final prediction results, and ultimately low efficiency and accuracy of environmental assessment.

[0132] In another possible embodiment, the second time threshold and the first number threshold are determined according to the degree of harm of the candidate pollutants to the human body and the degree of concern of the candidate pollutants.

[0133] In another possible embodiment, the calculation formula of the second time threshold is:

[0134]

[0135] Among them, W1 is the degree of harm of alternative pollutants to the human body, and the value range is between 0 and 1. B1 is the degree of attention of alternative pollutants, which reflects the degree of public attention to alternative pollutants, and the value range is between 0 and 1. T is the benchmark time threshold, and K1 and K2 are weights.

[0136] In another possible embodiment, the meteorological data includes ground meteorological data and high-altitude meteorological data, wherein the ground meteorological data includes wind direction, wind speed, total cloud cover, and temperature; and the high-altitude meteorological data includes air pressure, temperature, wind speed, and wind direction at different altitude layers.

[0137] In another possible embodiment, Figure 5 As shown, the evaluation steps of the environmental assessment value within the pollutant monitoring area are:

[0138] A further technical solution is that the steps for evaluating the environmental assessment value within the pollutant monitoring area are:

[0139] Obtaining the emission amount and pollutants exceeding the standard within the pollutant monitoring area;

[0140] Obtain the types of other pollutants other than those exceeding the standards and the emissions of other pollutants in the monitoring results;

[0141] Based on other pollutant types and their emissions, the emissions of the pollutants exceeding the standard and the pollutants exceeding the standard, an evaluation model based on the IGWO-GRU algorithm obtains an environmental evaluation value within the pollutant monitoring area.

[0142] Specifically, the specific calculation formula of the GRU algorithm is as follows:

[0143]

[0144] wherein: t is the current time; t-1 is the previous time; r t and z t are reset gate and update gate respectively; x t and h t are the current input capacity value and output capacity value of the battery respectively; h t-1 is the hidden layer state information transmitted from the previous unit node; is the unit to be updated; W r and b r , W z and b z , and are weight matrices and bias parameters required for calculating reset gate output, update gate output and process quantity respectively; is element multiplication; and σ and tanh are sigmoid function and hyperbolic tangent function respectively.

[0145] The prediction model based on the GRU neural network contains a hidden network, and the number of neurons in the hidden network is difficult to determine directly. The iteration number directly affects the prediction effect. If the iteration number cannot meet the requirements, the fitting degree of the prediction result will not be enough, and too many iteration numbers will lead to overfitting of the prediction result. The parameters of the traditional GRU neural network are generally manually set according to experience, which leads to a large randomness of the final estimation result.

[0146] IGWO is obtained by optimizing GWO. GWO is an algorithm inspired by the hierarchical system and hunting behavior of gray wolf groups. Assuming that the total number of wolf groups is N, the search space dimension is d, and the position of the i-th gray wolf is defined as X i =(X i,1 ,X i,2 ,…,X i,d ), the optimization process is as follows:

[0147] 1. Enclosure. The distance D between the gray wolf and the prey is represented as:

[0148] D=|C·X p (t)-X(t)|

[0149] The wolf group updates the gray wolf position according to the distance D:

[0150] X(t+1)=X p (t)-A·D

[0151] Where t is the current iteration number, C is the coefficient, A is the convergence factor, and X p (t) represents the position of the prey at the tth iteration, and X(t) represents the position of the wolf at the tth iteration. A and C are calculated as follows:

[0152] A=2a·r1-a

[0153] C=2·r2

[0154] a=2·(1-t / T)

[0155] Where r1 and r2 are random numbers between [0 and 1], a is a control parameter, and T is the maximum number of iterations. When |A| > 1, the wolf pack expands its search range, emphasizing global search capabilities. When |A| < 1, the wolf pack narrows its search range and conducts detailed optimization in a local area, emphasizing local development capabilities.

[0156] 2. Capture. The capture process is led by three types of alpha wolves (a, β, δ). Ordinary wolves (ω) update their positions based on the alpha wolf's position. The update process is as follows:

[0157]

[0158]

[0159]

[0160] Where D α 、D β 、D δ are the distances between wolf α, wolf β, wolf δ and wolf ω respectively; X(t+1) is the position of the gray wolf after each update.

[0161] The value changes with the linear change of parameter a. Since the linear change of parameter a is not conducive to the actual optimization of the algorithm, a nonlinear change of a is conducive to better optimization. To this end, the present invention adopts a nonlinear parameter control strategy as shown below:

[0162]

[0163] Then the expression of the convergence factor A is:

[0164]

[0165] By adopting pollutants exceeding the standard and other pollutant types, we can comprehensively construct the environmental assessment value within the pollutant monitoring area, so as to accurately reflect the environmental assessment value within the pollutant monitoring area, and then accurately reflect the environment within the pollutant monitoring area for governance.

[0166] In another possible embodiment, the calculation formula of the environmental assessment value within the pollutant monitoring area is:

[0167]

[0168] Where W i is the degree of harm of the i-th pollutant to the human body, P i is the emission of the i-th pollutant, P ib is the emission standard required by the standard for the i-th pollutant, and N is the total amount of the pollutant.

[0169] In another possible embodiment, when the environmental assessment value within the pollutant monitoring area is greater than the environmental assessment value threshold, targeted treatment is performed on the pollutants within the pollutant monitoring area based on the monitoring results.

[0170] Example 2

[0171] In an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the above-mentioned grid-based pollutant monitoring and assessment method based on model prediction.

[0172] Example 3

[0173] In an embodiment of the present application, a computer program product is provided. The computer program product stores instructions that, when executed by a computer, cause the computer to implement the aforementioned model-based prediction-based gridded pollutant monitoring and assessment method.

[0174] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0175] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read Only Memory), mobile access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0176] The above-described embodiments according to the present application are intended to be illustrative only. Changes can be made by those skilled in the art, without departing from the scope of the present application, which is defined by the following claims. The technical scope of the present application is not limited to the above-described embodiments. The technical scope of the present application must be determined based on the scope of the claims.

Claims

1. A grid-based pollutant monitoring and assessment method based on model prediction, characterized in that: Specifically include: S11. Install pollutant monitoring devices in the pollutant monitoring area to form a pollutant grid monitoring system to monitor pollutants in the environment and obtain monitoring results; S12. Based on the monitoring results, when there is a pollutant type greater than a first threshold, the pollutant type greater than the first threshold is used as an alternative pollutant; When the duration of the candidate pollutant is greater than the second time threshold or the number of times the candidate pollutant is monitored is greater than the first number threshold, the candidate pollutant is regarded as an excessive pollutant; The calculation formula of the second time threshold is: ， in, W The degree of harm of the alternative pollutants to the human body is between 0 and 1. B 1 is the degree of attention of the alternative pollutants, reflecting the degree of attention of the public to the alternative pollutants, and the value range is between 0 and 1. T is the benchmark time threshold, K 1. K 2 is the weight; S13. Based on the pollutants exceeding the standard, the pollution sources generating the pollutants exceeding the standard in areas other than the pollutant monitoring area within the city are identified. Real-time emissions and meteorological data from these pollution sources are used as inputs to an air quality prediction model to generate a prediction of diffuse pollutants within the pollutant monitoring area. The prediction of diffuse pollutants is updated every time t, and the spatiotemporal distribution of the previous concentration of pollutants exceeding the standard is used as a constraint for the next spatiotemporal distribution of the concentration of pollutants exceeding the standard. S14. Based on the prediction results of diffuse pollutants within the pollutant monitoring area and the emissions of pollutants exceeding the standard in the monitoring results, obtain the emissions of pollutants exceeding the standard within the pollutant monitoring area; S15. Obtain the emission amount of pollutants exceeding the standard and the pollutants exceeding the standard within the pollutant monitoring area; obtain other pollutant types and the emission amount of other pollutant types other than the pollutants exceeding the standard in the monitoring results; based on the other pollutant types and the emission amount of other pollutant types, the emission amount of the pollutants exceeding the standard and the pollutants exceeding the standard, obtain the environmental assessment value within the pollutant monitoring area based on the assessment model of the IGWO-GRU algorithm.

2. The pollutant grid monitoring and assessment method based on model prediction according to claim 1 is characterized in that: The step S13 predicts the diffuse pollutants in the pollutant monitoring area based on the air quality prediction model, specifically including the following steps: Defining constraints for the air quality prediction model, wherein the constraints include at least energy balance, momentum balance, mass balance, and the last diffuse pollutant prediction results; Acquire meteorological data and real-time emissions from pollution sources in real time, and input the meteorological data and real-time emissions from pollution sources as initial conditions into an air quality prediction model; A numerical iterative solution is performed based on the initial conditions to obtain simulation results, and a flow field analysis is performed using Fluent software based on the simulation results. Based on the results of the flow field analysis, visualization is performed using Tecplot software to obtain the diffuse pollutant prediction results at time t, and the diffuse pollutant prediction results at time t are used as constraint conditions for the diffuse pollutant prediction results at the next moment.

3. The pollutant grid monitoring and assessment method based on model prediction according to claim 2 is characterized in that: The meteorological data includes ground meteorological data and high-altitude meteorological data, wherein the ground meteorological data includes wind direction, wind speed, total cloud cover, and temperature; the high-altitude meteorological data includes air pressure, temperature, wind speed, and wind direction at different altitudes.

4. The pollutant grid monitoring and assessment method based on model prediction according to claim 2 is characterized in that: The initial conditions also include terrain data. The terrain data of the city and the pollution source are constructed based on the longitude and latitude of the city and the longitude and latitude of the pollution source, and the prediction results of diffused pollutants within the pollutant monitoring area are obtained based on the terrain data, the meteorological data, and the emissions of the pollution source within the first time threshold.

5. The pollutant grid monitoring and assessment method based on model prediction according to claim 1 is characterized in that: When the environmental assessment value inside the pollutant monitoring area is greater than the environmental assessment value threshold, the diffused pollutants inside the pollutant monitoring area are targetedly treated based on the monitoring results.

Images