Atmospheric environment influence automatic prediction method and equipment based on data driving

Abstract

According to the atmospheric environment influence automatic prediction method and device based on data driving, the problems of accuracy and consistency of data flow transmission between models are solved, and the atmospheric environment influence automatic prediction method and device based on data driving are obtained by accurately analyzing the terrain parameters of AERMAP and the boundary layer meteorological parameters of AERMET and according to the internal atmospheric physical principle requirements of the AERMOD model. The input file is automatically and accurately constructed, the integrity and reliability of the whole simulation process on the physical principle are ensured, and errors possibly introduced by manual splicing are completely eradicated. According to the method, normal form transformation from manual operation to data driving is achieved, high self-consistency of terrain, weather and diffusion models in the physical principle is ensured through automatic data analysis and transmission, and theoretical errors and result distortion caused by manual operation errors are reduced. According to the method, rapid and accurate analysis of complex output files is realized, efficient and error-free integration of massive receptors and meteorological data is realized, and the calculation efficiency and reliability of the whole simulation process are improved.

Description

Technical Field

[0001] This invention relates to the field of atmospheric environmental impact prediction technology, specifically to a data-driven automated prediction method and device for atmospheric environmental impact. Background Technology

[0002] Atmospheric environmental impact prediction and simulation is a crucial technical aspect of environmental impact assessment. This process typically involves extensive data processing, complex physical model calculations, and highly specialized parameter settings, characterized by its technology-intensive nature and frequent repetitive operations. Currently, the AERMOD model system developed by the U.S. Environmental Protection Agency (EPA) is widely used internationally as the standard tool in this field. However, the existing application of this system has many limitations, severely restricting the efficiency and quality of environmental impact assessment work.

[0003] The AERMOD model system is not a single program, but a scientific computing system composed of three core sub-models: the terrain preprocessing model (AERMAP), the meteorological preprocessing model (AERMET), and the atmospheric diffusion concentration prediction model (AERMOD). These sub-models are based on rigorous atmospheric boundary layer physics principles and must be executed sequentially according to strict data dependencies. Its original version was a command-line program, typically running in DOS or similar terminal environments, requiring a very high level of technical background and professional knowledge from the user, and exhibiting a very unfriendly human-computer interaction.

[0004] To address this issue, some commercial graphical user interface (GUI) software has emerged, offering further modifications to simplify the operation process. However, these improvements have not fundamentally resolved the systemic pain points, primarily in the following three aspects: (i) Localized Deployment and Resource Management Issues: Existing solutions are mostly standalone software installed locally. Their operation consumes a large amount of CPU and memory resources on the user's local computer, which can easily cause system lag during large-scale calculations, interfere with the normal operation of other software working in parallel, and affect work efficiency. At the same time, software function updates heavily rely on developers to release offline update packages. Users cannot obtain the latest kernel and function iterations of the model in a timely and automatic manner, which poses a risk of version lag and makes it difficult to meet increasingly stringent regulatory standards and requirements.

[0005] (ii) Complex operation process and low degree of automation: Even with a graphical interface, most existing software is still just a simple encapsulation of command-line parameters, failing to achieve intelligent connection between sub-models. Users must have a deep understanding of the input-output logic relationship between AERMAP, AERMET, and AERMOD, and manually perform each calculation, configure each parameter, and pass intermediate files. This process is not only cumbersome and time-consuming, but also makes it difficult to achieve batch calculations and multi-scenario simulations, resulting in a very low degree of automation.

[0006] (III) High Professional Barrier and Poor Popularity: The accuracy of the model is highly dependent on the rationality of the input parameters, which requires users not only to be familiar with the software operation but also to have a solid theoretical foundation in atmospheric science. The parameter setting interface is complex and has many options, which deters ordinary users without systematic training, making it easy for them to make mistakes in setting parameters, resulting in distorted simulation results. This high professional barrier greatly limits the popularization and application of this technology, becoming a bottleneck for improving the quality and efficiency of environmental impact assessment work.

[0007] In summary, existing technical solutions have significant shortcomings in terms of resource utilization, operational efficiency, user experience, and professional expertise required. Therefore, there is an urgent need in this field for a new method that can achieve fully automated, intelligent, and lightweight computing to address these issues. Summary of the Invention

[0008] The present invention proposes a data-driven automated prediction method, device and storage medium for atmospheric environmental impacts, which can at least solve one of the technical problems in the background art.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: A data-driven automated prediction method for atmospheric environmental impacts includes the following steps: S1. Automated terrain preprocessing: Based on the location of the pollution source and the evaluation range, the input parameters of the AERMAP terrain preprocessing model are automatically generated and driven to run. Then, the output results of AERMAP are automatically parsed to extract receptor data including terrain correction parameters. S2. Automated meteorological preprocessing: Based on the project's geographical information and raw meteorological data, automatically construct the input parameters of the AERMET meteorological preprocessing model and drive its operation. Then, automatically parse the output results of AERMET and extract key physical parameters characterizing the atmospheric boundary layer structure. S3. Automated concentration calculation: The receptor data extracted in step S1 and the key physical parameters extracted in step S2 are automatically assembled to generate an input file that meets the physical requirements of the AERMOD diffusion model, and AERMOD is driven to run to obtain concentration prediction results.

[0010] Furthermore, the step S1, which involves automatically generating input parameters for the AERMAP terrain preprocessing model based on the pollution source location and assessment scope, specifically includes: Based on the location of the pollution source and the evaluation scope of the project, the optimal geographical boundary is automatically determined, and the large-scale digital elevation model (DEM) file is intelligently truncated based on the boundary to generate an efficient DEM file suitable for the project's calculation scope; the input parameters for the AERMAP terrain preprocessing model are generated based on this efficient DEM file.

[0011] Furthermore, the automatic parsing of the AERMAP output in step S1 specifically includes: By identifying specified keywords and data structures in the AERMAP output file, the system can programmatically extract effective parameters such as latitude and longitude coordinates and altitude for each receptor and store these parameters in a structured data object.

[0012] Furthermore, the input parameters for automatically constructing the AERMET meteorological preprocessing model in step S2 are based on the requirements of the Moning-Obukhov similarity theory for meteorological parameters, ensuring the integrity of the input file to meet the physical calculation requirements of its internal solution of friction velocity (u*) and Moning-Obukhov length (L).

[0013] Furthermore, the automatic parsing of the AERMET output in step S2 specifically includes: We parse the surface meteorological files (.SFC) and upper-level profile files (.PFL) output by AERMET, extract time series data of friction velocity (u), Monin-Obukhov length (L), convection velocity scale (w), and mixing layer height (Zi), and perform physical consistency checks on them.

[0014] Furthermore, the receptor data extracted in step S1 and the key physical parameters extracted in step S2 are automatically assembled, specifically including: Write the receptor location information and terrain parameters extracted in step S1 into the receptor definition section of the AERMOD input file. Write the meteorological file path and parameters extracted in step S2 into the meteorological definition section of the AERMOD input file; Ensure that all parameters are consistent in units, time series, and georeference, and meet the physical coupling requirements of the AERMOD model.

[0015] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0016] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0017] As can be seen from the above technical solution, the data-driven automated prediction method for atmospheric environmental impact of the present invention focuses on solving the problem of accuracy and consistency in data flow transmission between models. By accurately analyzing the topographic parameters of AERMAP and the boundary layer meteorological parameters of AERMET, and based on the inherent atmospheric physical principles of the AERMOD model, the input files are automatically and accurately constructed to ensure the integrity and reliability of the entire simulation process in terms of physical principles, and to eliminate errors that may be introduced by manual splicing.

[0018] Compared with existing technologies, this invention realizes a paradigm shift from "manual operation" to "data-driven" approaches, and its beneficial effects are reflected in: (i) Guarantee of consistency of principles: Through automated data parsing and transmission, the terrain, meteorology and diffusion models are highly self-consistent in terms of physical principles, which greatly reduces theoretical errors and result distortions caused by human operation mistakes.

[0019] (ii) High precision and efficiency in data processing: It enables rapid and accurate parsing of complex output files, as well as efficient and error-free integration of massive receptor and meteorological data, significantly improving the computational efficiency and reliability of the entire simulation process.

[0020] (iii) Intelligentization and automation: Integrating professional knowledge and operational experience into automated processes reduces reliance on the professional background of operators, making complex atmospheric simulation work more standardized and easier to use. Attached Figure Description

[0021] Figure 1 This is a flowchart of the present invention; Figure 2 This is a screenshot of the monitoring log from an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0023] like Figure 1 As shown in this embodiment, the core of the data-driven automated prediction method for atmospheric environmental impacts lies in fully utilizing the data dependencies between components of the AERMOD model system based on atmospheric physics, and achieving seamless and accurate transmission of these data streams through automation technology. This includes the following steps: S1. Automated terrain preprocessing: Based on the location of the pollution source and the evaluation range, the input parameters of the AERMAP terrain preprocessing model are automatically generated and the model is driven to run. Then, the output results of the AERMAP terrain preprocessing model are automatically parsed and receptor data including terrain correction parameters are extracted. S2. Automated meteorological preprocessing: Based on the project's geographical information and raw meteorological data, automatically construct the input parameters of the AERMET meteorological preprocessing model and drive its operation. Then, automatically parse the output results of the AERMET meteorological preprocessing model and extract key physical parameters characterizing the atmospheric boundary layer structure. S3. Automated concentration calculation: The receptor data extracted in step S1 and the key physical parameters extracted in step S2 are automatically assembled to generate an input file that meets the physical requirements of the AERMOD diffusion model, and the AERMOD diffusion model is driven to run to obtain concentration prediction results.

[0024] Specifically, step S1, which involves automatically generating the input parameters for the AERMAP terrain preprocessing model based on the pollution source location and assessment scope, includes: Based on the location of the pollution source and the evaluation scope of the project, the optimal geographical boundary is automatically determined, and the large-scale digital elevation model (DEM) file is intelligently truncated based on the boundary to generate an efficient DEM file suitable for the project's calculation scope; the input parameters for the AERMAP terrain preprocessing model are generated based on this efficient DEM file.

[0025] The automatic parsing of the AERMAP output in step S1 specifically includes: By identifying specified keywords and data structures in the AERMAP output file, the system can programmatically extract effective parameters such as latitude and longitude coordinates and altitude for each receptor and store these parameters in a structured data object.

[0026] The input parameters for automatically constructing the AERMET meteorological preprocessing model in step S2 are based on the requirements of the Monin-Obukhov similarity theory for meteorological parameters, ensuring the integrity of the input file to meet the physical calculation requirements of its internal solution of friction velocity (u*) and Monin-Obukhov length (L).

[0027] The automatic parsing of the AERMET output in step S2 specifically includes: We parse the surface meteorological files (.SFC) and upper-level profile files (.PFL) output by AERMET, extract time series data of friction velocity (u), Monin-Obukhov length (L), convection velocity scale (w), and mixing layer height (Zi), and perform physical consistency checks on them.

[0028] The receptor data extracted in step S1 is automatically assembled with the key physical parameters extracted in step S2, specifically including: Write the receptor location information and terrain parameters extracted in step S1 into the receptor definition section of the AERMOD input file. Write the meteorological file path and parameters extracted in step S2 into the meteorological definition section of the AERMOD input file; Ensure that all parameters are consistent in units, time series, and georeference, and meet the physical coupling requirements of the AERMOD model.

[0029] Its basic principles and data processing algorithms are as follows: (I) AERMAP Terrain Data Processing Algorithm AERMAP uses a digital elevation model (DEM) to calculate topographic features of each receiver point relative to the pollution source, such as elevation (Elev), hilltop height (HILLHGT), and effective height (ELEVFT). These parameters directly affect the plume's transport path and diffusion space, and are the physical basis for AERMAP's complex terrain correction.

[0030] The computation time of AERMAP is directly proportional to the size of the DEM file and the number of receptors. This invention first automatically calculates a minimum, fully covering geographic bounding box based on the pollution source location and the assessment area. Based on this boundary, the original large-scale, high-precision DEM topographic file is truncated to generate a new DEM file covering only the target area. This process significantly reduces the amount of data that AERMAP needs to process, fundamentally improving the efficiency of all subsequent computational steps.

[0031] Using the truncated small-scale DEM file, the AERMAP input control file (INP) is compiled, embedding the DEM file, receptor grid, and point information. After driving AERMAP to execute calculations, the output .REC receptor file is automatically parsed. The parsing process accurately extracts the latitude and longitude coordinates, original elevation, and calculated terrain correction parameters for each receptor in a programmatic manner by recognizing specific keywords ELEV, HILLHGT, and format in the file, and stores them in a structured data object, providing accurate terrain input for subsequent meteorological preprocessing and diffusion simulation.

[0032] (II) AERMET meteorological data processing algorithm Based on the Monin-Obukhov similarity theory, AERMET processes raw meteorological observation data into key boundary layer parameters characterizing atmospheric turbulence and thermal structure, such as friction velocity (u), Monin-Obukhov length (L), convection velocity scale (w), and mixing layer height (Zi). These parameters are the physical core of AERMOD in describing diffusion processes.

[0033] This invention utilizes the geographical information obtained in the previous step and the raw meteorological data (surface and radiosonde data) input from external sources. Based on the meteorological requirements of AERMET, it arranges the input files for the second stage, including automatically filling in station parameters, data paths, and output requirements. After AERMET runs, it parses the output surface meteorological files (.SFC) and upper-air profile files (.PFL). The parsing process aims to extract and verify the time series of these key physical parameters, ensuring their continuity and physical rationality. For example, it checks whether the sign of L correctly reflects atmospheric stability, providing AERMOD with meteorological input that strictly conforms to physical laws.

[0034] (III) Algorithm for Calculating AERMOD Concentration AERMOD primarily utilizes topographic dynamic parameters provided by AERMAP and boundary layer turbulence parameters provided by AERMET. Under the steady-state plume assumption, it calculates pollutant concentrations at different receptor sites by combining probability density functions and Gaussian diffusion models. Its control file requires precise coupling of topographic and meteorological data from the first two steps.

[0035]

[0036] Where C s {x r ,y r ,z} represents the pollutant concentration at the receiving point; x r , y r z is the planar coordinate of the receiving point (m); z is the height of the receiving point (m); z ieff For the effective mixing layer height (m); σ zs h is the total vertical diffusion coefficient. es , where is the plume height (m), obtained by superimposing the chimney height on the plume rise height; Q is the pollutant emission rate (g / s), which is the direct input from the pollution source; The average wind speed (m / s) is obtained from meteorological data; F y is the horizontal distribution function; m is the multipath effect index.

[0037] This step is the core of the data algorithm of this invention, which assembles the data output from the first two stages based on physical principles: Receptor data assembly: The receptor location information and its corresponding terrain parameters are written into the RESTARTING path of the AERMOD control file and the receptor table in RE ELEVUNIT METERS.

[0038] Meteorological data link: Write the paths of the .SFC and .PFL files output by AERMET into the corresponding fields such as SURFFILE and PROFFILE in the AERMET input file.

[0039] Physical consistency guarantee: All input parameters are checked to ensure consistency in units, time series, and geographic reference, fully complying with the model requirements. Finally, a complete, error-free, and physically consistent AERMOD input file is generated to drive the calculation and obtain the final atmospheric pollutant concentration distribution results.

[0040] The automated prediction method provided by this invention has demonstrated outstanding performance and value in practical applications. The "Online Prediction System for Atmospheric Environmental Impacts" built based on this invention has been officially put into commercial operation as a Software as a Service (SaaS) platform (accessible at: https: / / airmodel.eecc.cn). Since its launch in December 2024, the actual operational effectiveness of this system has fully verified the beneficial effects of this invention.

[0041] To date, the system has successfully provided stable and efficient atmospheric forecasting services to over 20 environmental consulting agencies, supporting the completion of more than 30 atmospheric environmental impact assessment (EIA) forecasting projects. Practice has shown that using this invention's method shortens the manual operation process, which originally required several days, to an automated process within hours, eliminating calculation interruptions or incorrect results caused by human error. This significantly improves the first-time pass rate of projects and greatly ensures the timeliness and reliability of EIA work.

[0042] like Figure 2 The monitoring log screenshot shown illustrates how, in the actual operation of the automated prediction application built upon this invention, the system automatically invoked and executed the AERMAP terrain preprocessing model, the AERMET meteorological preprocessing model, and the AERMOD atmospheric diffusion model in sequence (see the red box in the figure). Each module started and completed sequentially without human intervention, achieving seamless connection and automated sequential execution between the models.

[0043] The monitoring interface simultaneously and accurately records the actual running time of each model component, providing a quantitative basis for process performance analysis and computational resource evaluation. This running status intuitively verifies the effectiveness of the "data-driven, automatic flow" technology achieved by this invention, completely changing the traditional serial working mode that relies on manual operation and data transfer, and significantly improving the overall efficiency and reliability of simulation prediction.

[0044] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0045] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0046] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0047] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0048] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0049] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0050] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data-driven automated prediction method for atmospheric environmental impacts, characterized in that, Includes the following steps, S1. Automated terrain preprocessing: Based on the location of the pollution source and the evaluation range, the input parameters of the AERMAP terrain preprocessing model are automatically generated and driven to run. Then, the output results of AERMAP are automatically parsed to extract receptor data including terrain correction parameters. S2. Automated meteorological preprocessing: Based on the project's geographical information and raw meteorological data, automatically construct the input parameters of the AERMET meteorological preprocessing model and drive its operation. Then, automatically parse the output results of AERMET and extract key physical parameters characterizing the atmospheric boundary layer structure. S3. Automated concentration calculation: The receptor data extracted in step S1 and the key physical parameters extracted in step S2 are automatically assembled to generate an input file that meets the physical requirements of the AERMOD diffusion model, and AERMOD is driven to run to obtain concentration prediction results.

2. The data-driven automated prediction method for atmospheric environmental impacts according to claim 1, characterized in that: Step S1, which involves automatically generating input parameters for the AERMAP terrain preprocessing model based on the pollution source location and assessment area, specifically includes: Based on the location of the pollution source and the evaluation scope of the project, the optimal geographical boundary is automatically determined, and the large-scale digital elevation model (DEM) file is intelligently truncated based on the boundary to generate an efficient DEM file suitable for the project's calculation scope; the input parameters for the AERMAP terrain preprocessing model are generated based on this efficient DEM file.

3. The data-driven automated prediction method for atmospheric environmental impacts according to claim 2, characterized in that: The automatic parsing of the AERMAP output in step S1 specifically includes: By identifying specified keywords and data structures in the AERMAP output file, the system can programmatically extract effective parameters such as latitude and longitude coordinates and altitude for each receptor and store these parameters in a structured data object.

4. The data-driven automated prediction method for atmospheric environmental impacts according to claim 1, characterized in that: The input parameters for automatically constructing the AERMET meteorological preprocessing model in step S2 are based on the requirements of the Monin-Obukhov similarity theory for meteorological parameters, ensuring the integrity of the input file to meet the physical calculation requirements of its internal solution of friction velocity (u*) and Monin-Obukhov length (L).

5. The data-driven automated prediction method for atmospheric environmental impacts according to claim 4, characterized in that: The automatic parsing of the AERMET output in step S2 specifically includes: We parse the surface meteorological files (.SFC) and upper-level profile files (.PFL) output by AERMET, extract time series data of friction velocity (u), Monin-Obukhov length (L), convection velocity scale (w), and mixing layer height (Zi), and perform physical consistency checks on them.

6. The data-driven automated prediction method for atmospheric environmental impacts according to claim 1, characterized in that: The receptor data extracted in step S1 is automatically assembled with the key physical parameters extracted in step S2, specifically including: Write the receptor location information and terrain parameters extracted in step S1 into the receptor definition section of the AERMOD input file. Write the meteorological file path and parameters extracted in step S2 into the meteorological definition section of the AERMOD input file; Ensure that all parameters are consistent in units, time series, and georeference, and meet the physical coupling requirements of the AERMOD model.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the computer program is executed by the processor, it causes the processor to perform the steps of the method as described in any one of claims 1 to 6.

Images